Statins decrease dendritic arborization in rat ... - Wiley Online Library

JOURNAL OF NEUROCHEMISTRY

| 2009 | 108 | 1057–1071

,1,2

doi: 10.1111/j.1471-4159.2008.05854.x

,1 , ,3

,

*Department of Pharmacology and Toxicology, SUNY, Buffalo, New York, USA Center for Research on Occupational and Environmental Toxicology, Oregon Health & Science University, Portland, Oregon, USA àVollum Institute, Oregon Health & Science University, Portland, Oregon, USA §Department of Veterinary and Comparative Anatomy, Pharmacology and Physiology, Washington State University, Pullman, Washington, USA ¶Department of Molecular Biosciences, UC Davis School of Veterinary Medicine, Davis, California, USA

Abstract Clinical and experimental evidence suggest that statins decrease sympathetic activity, but whether peripheral mechanisms involving direct actions on post-ganglionic sympathetic neurons contribute to this effect is not known. Because tonic activity of these neurons is directly correlated with the size of their dendritic arbor, we tested the hypothesis that statins decrease dendritic arborization in sympathetic neurons. Oral administration of atorvastatin (20 mg/kg/day for 7 days) significantly reduced dendritic arborization in vivo in sympathetic ganglia of adult male rats. In cultured sympathetic neurons, statins caused dendrite retraction and reversibly blocked bone morphogenetic protein-induced dendritic growth without altering cell survival or axonal growth. Supplementation with

mevalonate or isoprenoids, but not cholesterol, attenuated the inhibitory effects of statins on dendritic growth, whereas specific inhibition of isoprenoid synthesis mimicked these statin effects. Statins blocked RhoA translocation to the membrane, an event that requires isoprenylation, and constitutively active RhoA reversed statin effects on dendrites. These observations that statins decrease dendritic arborization in sympathetic neurons by blocking RhoA activation suggest a novel mechanism by which statins decrease sympathetic activity and protect against cardiovascular and cerebrovascular disease. Keywords: dendrites, isoprenoids, RhoA, statins, sympathetic neurons. J. Neurochem. (2009) 108, 1057–1071.

Statins, a class of drugs that inhibit 3-hydroxyl-3-methylglutaryl coenzyme A (HMG-CoA) reductase (EC: 1.1.1.34), the rate-limiting enzyme in mevalonate synthesis, are widely used to protect against cerebrovascular (Alvarez-Sabin et al. 2007; Tunon et al. 2007; Zivin 2007) and cardiovascular (Pedersen et al. 1996; Anonymous 2002; Collins et al. 2002, 2004) disease. Increased sympathetic activity is prognostic of poor outcome in stroke (Sander et al. 2001) and congestive heart failure (Francis et al. 1993) and is associated with hypertension and untreated hyperlipidemia (Hachinski et al. 1986; Cechetto et al. 1989; Zucker and Wang 1991; Floras 1993; Melenovsky et al. 2003); therefore, clinical observations that statins decrease sympathetic activity in patients with coronary disease (Pehlivanidis et al. 2001; Hamaad et al. 2005) or untreated hyperlipidemia (Melenovsky et al.

Received November 7, 2008; revised manuscript received December 7, 2008; accepted December 7, 2008. Address correspondence and reprint requests to Pamela J. Lein, Department of Molecular Biosciences, UC Davis School of Veterinary Medicine, One Shields Avenue, Davis, CA 95616, USA. E-mail: [email protected] 1 These authors contributed equally to this work. 2 The present address of Woo-Yang Kim is the Neuroscience Center, University of North Carolina School of Medicine, Chapel Hill, NC 27599, USA. 3 Deceased. Abbreviations used: BMP, bone morphogenetic protein; ca, constitutively active; dn, dominant negative; EGFP, enhanced green fluorescent protein; GST, glutathione-S-transferase; HMG-CoA, 3-hydroxyl-3-methylglutaryl coenzyme A; LVS, lovastatin; SCG, superior cervical ganglia; SDS, sodium dodecyl sulfate; TH, tyrosine hydroxylase; wt, wild type.

2009 The Authors Journal Compilation 2009 International Society for Neurochemistry, J. Neurochem. (2009) 108, 1057–1071

1057

1058 | W.-Y. Kim et al.

2003) have led to the proposal that normalization of sympathetic function contributes to the therapeutic actions of these drugs (Krum and McMurray 2002). Experimental studies demonstrate that statins normalize sympathetic outflow and cardiovascular reflex regulation in a rabbit model of heart failure (Pliquett et al. 2003a,b; Gao et al. 2005), and reduce sympathetic activity coincident with decreased blood pressure in spontaneously hypertensive rats (Kishi et al. 2003). Central mechanisms, including down-regulation of angiotensin II receptors (Gao et al. 2005) and up-regulation of nitric oxide synthase (Kishi et al. 2003; Gao et al. 2008), are proposed to mediate statin effects on sympathetic activity. Whether statins also directly alter the properties of sympathetic nerves that innervate the heart and blood vessels is not known but the therapeutic efficacy of b blockers that function to decrease neurotransmission between post-ganglionic sympathetic neurons and their target tissues support investigation of peripheral mechanisms. The proximal effect of statin inhibition of HMG-CoA reductase is decreased synthesis of not only cholesterol, but also isoprenoids (Lefer et al. 2001; Wierzbicki et al. 2003). Isoprenylation is important for the proper membrane localization and function of the Rho GTPases, including Rho, Rac, and Cdc42 (Bar-Sagi and Hall 2000), and depletion of isoprenoids by statins results in accumulation of nonfunctional Rho GTPases in the cytoplasm (Hori et al. 1991; Koch et al. 1997). Inhibition of Rho GTPase signaling decreases sympathetic activity in vivo (Ito et al. 2003) and has been proposed to mediate the effects of statins on angiotensin signaling (Gao et al. 2005; Higuchi et al. 2007) and nitric oxide synthase expression (Laufs and Liao 1998; Kishi et al. 2003; Gao et al. 2008). In cultured neurons derived from the CNS, altering the activity of RhoA, Rac1, and Cdc42 affects dendritic growth and branching (Redmond and Ghosh 2001; Van Aelst and Cline 2004). While the role of Rho GTPases in modulating dendritic growth in PNS neurons has yet to be established, it is known that in post-ganglionic sympathetic neurons, dendrites are the primary site of synapse formation and the size of the dendritic arbor correlates directly with synaptic density (Purves 1975; Purves and Lichtman 1985; De Castro et al. 1995) and tonic activity (Ivanov and Purves 1989). These observations, coupled with reports that hypertension in rats is associated with dendritic hypertrophy in post-ganglionic sympathetic neurons (Kondo et al. 1990; Peruzzi et al. 1991), suggest the possibility that statins alter sympathetic function by modifying dendritic arborization. This hypothesis is further supported by reports that statins inhibit neurite outgrowth in cultured cortical neurons (Fan et al. 2002; Schulz et al. 2004). In this study, we examine in vivo and in vitro effects of statins on the dendritic morphology of sympathetic neurons from the rat superior cervical ganglia (SCG). Our data demonstrate that statins interact directly with post-ganglionic sympathetic neurons to selectively reduce dendritic arborization. The mechanism

mediating this effect involves disruption of RhoA function as a consequence of isoprenoid depletion.

Materials and methods Materials Recombinant human bone morphogenetic proteins (BMPs) were generously provided by Curis (Cambridge, MA, USA). Statins were purchased from LKT laboratories (St Paul, MN, USA); GGTI-298 and FTI-277, from Calbiochem (La Jolla, CA, USA). Mevalonate, cholesterol, geranylgeraniol, mixed isomers of farnesol, manumycin, perillic acid, and zaragozic acid were obtained from Sigma (St Louis, MO, USA). The RhoA inhibitor C3-transferase (CT04) was purchased from Cytoskeleton, Inc. (Denver, CO, USA); the Rho kinase 1 (Rock1) inhibitor Y-27632 and the Rac1 inhibitor NSC23766, from Calbiochem. The Cdc42 inhibitor secramine A was a generous gift from the Kirchhausen lab (Harvard Medical School) and the Hammond lab (University of Louisville) and was synthesized by Bo Yu and G.B. Hammond of the University of Louisville (Xu et al. 2006). Plasmids co-expressing green fluorescent protein and wild type (wt), dominant negative (dn) RhoA (T19N) or constitutively active (ca) RhoA (Q63L) constructs were generously provided by Dr Klaus Hahn (University of North Carolina). Plasmids co-expressing green fluorescent protein and wt, dn (T17N) or ca (Q61L) Cdc42 or Rac1 were generous gifts from Dr William Snider (University of North Carolina). For GTP pulldown assays, Cdc42 (accession number AF205635) was cloned from a cDNA library derived from adult rat hippocampus using primers containing the Myc tag and subcloned into pCAGGS vector using EcoRI/ClaI sites. The pCAGGS-Myc-CDC42 construct was verified by sequencing and expression in heterologous cells. pGEXPAK1-PBD was previously described (Saneyoshi et al. 2008); pCDNA-HA-RhoA and pGEX-Rhotekin-RBD were gifts from Dr Scott Soderling (Duke University). Animals The Institutional Animal Care and Use committees of the State University of New York at Buffalo and the Oregon Health & Science University approved all animal uses reported in these studies. Male (200 g) Holtzman rats (Harlan, Indianapolis, IN, USA) were administered atorvastatin (20 mg/kg/day p.o.) or an equivalent amount of vehicle (20% sucrose) daily for 7 days. All animals were allowed water and food ad libitum and daily monitoring of body weight indicated no significant differences between treatment groups. At the conclusion of the treatment period, rats were killed, SCG excised, immediately fixed and stored in 4% paraformaldehyde at 4C for no more than 30 days until used for morphometric analyses. Cell culture and transfection Post-mitotic sympathetic neurons were dissociated from SCG or stellate ganglia of 20–21 days rat embryos and maintained in the absence of glial cells in serum-free medium supplemented with nerve growth factor as previously described (Higgins et al. 1991). Hippocampal neurons were dissociated from the hippocampi of 18 day old rat embryos and maintained in serum-free medium as previously described (Wayman et al. 2006). Cultured neurons (5–7 days in vitro) were transfected using Lipofectamine 2000


Statins decrease dendritic arborization | 1059

(Invitrogen, Carlsbad, CA, USA) as instructed by the manufacturer. Cultures were allowed to recover for 24 h after transfection before starting experimental treatments. In separate studies, cultured sympathetic neurons were co-transfected with pTLX2-Lux, a plasmid encoding firefly luciferase under control of the Smad1responsive promoter TLX2 (Tang et al. 1998) and pEF-R, a plasmid encoding Renilla luciferase, which were provided by Shao Jun Tang (University of Toronto) and Xin Lin (SUNY at Buffalo), respectively. Following an overnight treatment with BMP7 (25 ng/ mL) ± lovastatin (LVS) (1 lM), luciferase activity in cell lysates was quantified using the dual luciferase reporter assay system (Promega, Madison, WI, USA). Data were collected using an Orion microplate luminometer (Berthold Direct System, Oak Ridge, TN, USA) interfaced to Simplicity 2.0 beta software, and reported as the ratio of firefly luciferase activity to Renilla luciferase activity. Morphological analyses Axonal lengths in short-term sympathetic cultures (15 h in vitro) and dendritic morphology in more mature sympathetic cultures (5–14 days in vitro) were analyzed as previously described (Lein and Higgins 1989; Lein et al. 1995). To quantify axonal morphology in more mature sympathetic cultures exposed to BMP7 for 4 days, cultures were immunostained with anti-tyrosine hydroxylase (TH) antisera (Chemicon, Temecula, CA, USA), which is a marker for noradrenergic sympathetic nerve fibers (Alm and Lundberg 1988), and the density of TH+ fibers determined using a modification of the technique reported by Peter Smith (Blacklock et al. 2004). Briefly, 20X fluorescent images of four fields (0.22 mm2 per field) were randomly selected and captured digitally using a Nikon Eclipse E400 fluorescent microscope and SPOT camera. A stereological grid (NIH ImageJ v.1.33, http:// rsb.info.nih.gov/ij/download.html) with a grid intersection interval of 25 lm was randomly superimposed over the captured images resulting in a total of 352 intersection points per field. The number of TH-positive fibers that crossed grid intersection points were manually counted using the NIH ImageJ point picker plugin. Intersects overlying nerve fibers were counted and divided by total numbers of intersects in the field, providing an estimated percentage of the apparent sectional area occupied by nerve fibers. To measure dendritic morphology in vivo, individual neurons in fixed SCG were labeled with 1,1¢-dioctadecyl-3,3,3¢,3¢-tetramethylindocarbo-cyanine perchlorate (DiI; Molecular Probes, Eugene, OR, USA) using a previously described ballistic delivery system (Grutzendler et al. 2003). Briefly, tungsten particles (1 lm diameter) coated with 1,1¢-dioctadecyl-3,3,3¢,3¢-tetramethylindocarbo-cyanine perchlorate were fired into fixed SCG using a Bio-Rad (Hercules, CA, USA) Helios Gene Gun system with He at 80 psi. Ganglia were mounted in elvanol and immediately imaged using a Bio-Rad 1024 laser scanning confocal microscope. Neurons were selected for analysis if a tungsten particle could be located within the cell body of the neuron and the most distal aspects of the dendritic arbor were labeled as evidenced by tapering of fluorescence to very fine tips of dendritic processes. To capture the entire dendritic arbor, optical sections were collected at 0.5 lm z-steps using a 40x oil immersion objective lens. Three-dimensional images were reconstructed using Voxx 2 software (http://www.nephrology.iupui.edu/imaging/voxx/) and then compressed into two-dimensional images for morphometric analyses.

To visualize the cellular distribution of Smad1, cultures immunostained for Smad1 (Upstate Biotechnology, Lake Placid, NY, USA), confocal images collected at a section thickness of 1 lm using a Bio-Rad MRC 100 laser scanning confocal microscope (Bio-Rad). 1-(4,5-Dimethylthiazol-2-yl)-3,5-diphenylformazan assay of cell viability Primary cultures of sympathetic neurons (2 · 104 cells per well in 24-well plates) were exposed to varying concentrations of LVS ± BMP7 (25 ng/mL) for 4 days beginning on day 8 in vitro. Cultures were incubated at 37C for 4 h in Hanks buffered salt solution supplemented with 1-(4,5-dimethylthiazol-2-yl)-3,5-diphenylformazan (M-5655; Sigma) at a final concentration of 500 lg/mL then solubilized in 5% Triton X-100. Absorbance was measured at 562 nm using a Spectraflour Plus spectrophotometer (Tecan, Research Triangle Park, NC, USA). Western blotting Cell lysates for western blotting were prepared as described (Blanco-Colio et al. 2002) with slight modification. Briefly, cultures were sonicated in lysis buffer (phosphate-buffered saline plus 2 mM EDTA, 2 mM phenylmethylsulfonyl fluoride, and 1 lM pepstatin A), centrifuged at 50 000 g for 30 min and the supernatant collected as the cytosolic fraction. The pellet was resuspended in 100 mM Tris–HCl buffer (pH 7.4) supplemented with 300 mM NaCl, 1% Triton X-100, 0.1% sodium dodecyl sulfate (SDS), 2 mM EDTA, 2 mM phenylmethylsulfonyl fluoride, and 1 lM pepstatin A, centrifuged at 15 000 g for 5 min, the supernatant collected and protein concentration determined using the Bio-Rad protein assay. Samples with equal amounts of protein (50 lg) were separated on 15% SDS–polyacrylamide gel electrophoresis, transferred onto nitrocellulose membranes and probed with RhoA antisera (Cytoskeleton). Immunoreactive bands were detected using enhanced chemiluminescence (Amersham, Piscataway, NJ, USA). Rho GTPase-GTP precipitation assays Cultured sympathetic neurons (7 days in vitro) were transfected with plasmids encoding HA-tagged RhoA or myc-tagged Cdc42 and then maintained for an additional 3 days in serum-free medium ± BMP7 (50 ng/mL). Cell lysates were harvested in ice-cold lysis buffer (25 mM Tris HCl, pH 7.5 containing 125 mM NaCl, 30 mM MgCl2, 1% Triton X-100, 0.5% deoxycholate, 0.1% SDS and EDTA-free protease inhibitor cocktail) and centrifuged at 20 000 g for 10 min to clear insoluble material. Cleared lysates were incubated for 60 min at 4C with pre-loaded glutathione sepharose beads containing 40 lg glutathione-Stransferase (GST)-Rhotekin-RBD for pull-down of GTP-RhoA or 20 lg GST-PAK-PBD for pull-down of GTP-Rac1 or GTP-Cdc42. Resin was washed once with lysis buffer and extracted with 2X SDS sample buffer. Activated RhoA bound by GST-RBD was detected by western blotting using anti-HA Ab (Santa Cruz Biotechnology, Santa Cruz, CA, USA); activated Rac1 and Cdc42 bound to PAK-PBD was detected by western blotting with monoclonal antibody specific for Rac1 (BD Bioscience, San Jose, CA, USA) or myc Ab (purified from 9E10 hybridoma supernatant), respectively. Densitometric analyses of blots were


1060 | W.-Y. Kim et al.

performed using the Odyssey infrared detection system (LiCor Biosciences, Lincoln, NE, USA). Statistical analyses Experiments were performed a minimum of three times, and data are presented as the mean ± SEM. Statistical significance for in vitro experiments was assessed by a one-way ANOVA with p < 0.05 considered significant, followed by Tukey’s post hoc test; for in vivo studies, a two-tailed unpaired Student’s t-test was used.

Results Statins cause dendrite retraction in vivo To test the hypothesis that statins decrease dendritic arborization in sympathetic neurons, we quantified the dendritic morphology of post-ganglionic sympathetic neurons in adult male rats orally administered the commonly prescribed statin, atorvastatin, at 20 mg/kg/day for 7 days. This dose is within the range shown to result in plasma levels of atorvastatin in rats comparable to those observed in humans following oral administration of therapeutic doses of this drug (Kishi et al. 2003) and to decrease sympathetic nerve activity in humans (Hamaad et al. 2005) and in spontaneously hypertensive rats (Kishi et al. 2003). After a 7-day treatment with either atorvastatin or equal amounts of vehicle (20% sucrose), SCG were harvested, fixed, and individual neurons labeled for morphometric analyses by Diolistics (Grutzendler et al. 2003). We focused these studies on the SCG for technical and scientific reasons: (i) SCG are the largest sympathetic ganglia and the easiest to access in adult rats, which optimizes harvest of structurally intact ganglia and (ii) while stellate ganglia provide the primary sympathetic innervation of the heart, SCG also innervate the heart (Miura and Okada 1981; Korsching and Thoenen 1988; Pardini et al. 1989; Verberne et al. 1999; Hansson 2002; Wingerd et al. 2002) in addition to innervating the cerebrovasculature (Camargos and Machado 1988; Lincoln 1995; Miller et al. 1995; Furuichi et al. 1999). Treatment with atorvastatin did not alter the body weight of adult rats relative to vehicle controls over the 7-day treatment period (data not shown). However, morphometric analyses of individual sympathetic neurons within the SCG (Fig. 1a and b) indicated that atorvastatin significantly reduced the size and complexity of the dendritic arbor of neurons as indicated by a decreased number and length of primary, secondary, and tertiary dendrites (Fig. 1c and d), decreased number of branch points (Fig. 1e), and an approximate 40% decrease in the area of the dendritic arbor (Fig. 1f). Statins selectively inhibit dendritic growth in cultured sympathetic neurons To determine whether statins decrease dendritic arborization via direct interactions with neurons, we evaluated dendritic

morphology in sympathetic neurons cultured from SCG and exposed to statins in vitro in the absence of systemic, target, or glial influences. As previously reported (Lein et al. 1995), sympathetic neurons grown in the absence of serum or glial cells extend only a single axon (Fig. 2a and b); but when grown in the presence of BMP7, these neurons also extend dendrites (Fig. 2c and d). Statins did not induce dendritic growth in control cultures grown in the absence of BMP7 (Fig. 2e and f), but statins significantly inhibited dendritic growth in BMP7-treated cultures (Fig. 2g and h). This effect was concentration dependent with maximal inhibition of BMP7-induced dendritic growth observed at 1–3 lM of either atorvastatin or LVS with an EC50 200 nM (Fig. 2i). Statins also caused dendrite retraction when added to cultured sympathetic neurons that had already extended dendrites (Fig. 2j). Cultured sympathetic neurons were treated with BMP7 for 5 days to induce dendrite formation. During the next 8 days, a subset of cultures continued to receive only BMP7 while another subset was treated with both BMP7 and LVS. In cultures receiving only BMP7 throughout the treatment period, dendrites continue to grow (Fig. 2j). In contrast, in cultures treated with LVS, dendrites begin to retract with elimination of approximately 94% of the pre-existing dendrites within 4 days (Fig. 2j). To determine whether decreased dendritic arborization is a general property of statins we tested three additional statins. Simvastatin and mevastatin, which are lipophilic statins like atorvastatin and LVS, inhibited BMP7-induced dendritic growth in cultured sympathetic neurons with similar efficacy and potency (Fig. 2k). In contrast, pravastatin, which is hydrophilic and poorly absorbed by non-hepatic cells (Corsini et al. 1999), had no effect on BMP7-induced dendritic growth when tested over the same concentration range (Fig. 2k). Dendritic growth in sympathetic neurons is triggered by not only BMP7 but also other BMPs of the dpp and 60A subfamilies (Guo et al. 1998; Beck et al. 2001). LVS blocked the dendrite-promoting activity of representative members of both groups with comparable efficacy (Fig. 2l). Since the stellate ganglia provide the primary sympathetic innervation of the heart, we also quantified the effects of atorvastatin and LVS on dendritic morphology in sympathetic neurons cultured from stellate ganglia. Both statins inhibited BMP7-induced dendritic growth in neurons derived from stellate ganglia (Fig. 2m) with an efficacy and potency comparable to that observed in neuronal cell cultures derived from SCG (Fig. 2i). To address the question of whether statins specifically and selectively target dendrites, we evaluated the effects of statins on axonal growth and cell viability. Exposure to LVS during the first 15 h after plating at concentrations that inhibited dendritic growth had no effect on the extent of axonal growth in cultured sympathetic neurons (Fig. 3a). Similarly, the density of axonal processes was not altered in mature



(a)

Fig. 1 Atorvastatin causes dendrite retraction in sympathetic neurons of the superior cervical ganglia (SCG). (a and b) Photomicrographs of neurons in intact SCG labeled with 1,1¢-dioctadecyl-3,3,3¢,3¢-tetramethylindocarbo-cyanine perchlorate (DiI) using Diolistics (asterisk indicates the axon). SCG were harvested from adult male rats fed vehicle (20% sucrose, a) or atorvastatin (20 mg/kg, b) daily for 7 days. The dendritic arbor of individual neurons was quantified from confocal images with respect to the number (c) and length (d) of primary, secondary and tertiary dendrites, the number of branch points (e) and the area of the dendritic arbor (f). Data presented as the mean ± SEM (n = 50); *significantly different from control at p < 0.05; bar, 50 lm.

(b)

(c)

(d)

(e)

(f)

sympathetic cultures exposed to LVS under the same culture conditions used to examine statin effects on dendritic growth with the exception of decreased axonal density in cultures exposed to only the very highest concentration of LVS (10 lM) in the absence of BMP7 (Fig. 3b). The observation that axonal growth was generally not inhibited by statins suggested that statin effects on dendritic arborization were not because of decreased cell viability. To confirm this, we first examined the effects of LVS on mitochondrial activity in sympathetic cultures using the 1-(4,5-Dimethylthiazol-2-yl)3,5-diphenylformazan assay. With the exception of cultures exposed to LVS at 10 lM in the absence of BMP7, LVS had no deleterious effect on mitochondrial activity under the same culture paradigm used for assays of dendritic growth (Fig. 3c). Next, we examined the reversability of LVS effects on dendritic growth. Sympathetic cultures were treated with BMP7 and LVS for 5 days. During the next 9 days, a subset of cultures continued to be treated with both BMP7 and LVS

while another subset received BMP7 only. In cultures that received LVS throughout the 14 days treatment period, dendritic growth was continuosly repressed (Fig. 3d). In contrast, in cultures from which LVS treatment was discontinued, increased dendritic growth was evident within 3 days after LVS treatments were stopped and recovered to control levels within 9 days (Fig. 3d). Finally, we determined that concentrations of LVS that inhibited dendritic growth had no effect on the number of neurons per culture (Fig. 3e). Statins do not alter Smad1 activation The observation that statins inhibited the dendrite promoting activity of multiple BMPs suggested the possibility that statins inhibit the canonical BMP signaling pathway. Binding of BMPs to BMP receptors activates Smad transcription factors, causing Smads to translocate to the nucleus and alter gene expression (Massague and Gomis 2006). We previously


1062 | W.-Y. Kim et al.

(i)

(k)

(a)

(c)

(e)

(g)

(b)

(d)

(f)

(h)

(j)

(l)

(m)

demonstrated that BMP-induced dendritic growth requires Smad1 activation (Guo et al. 2001); here, we investigated the possibility that statins decrease dendritic arborization by disrupting Smad signaling. Consistent with previous reports (Guo et al. 2001), BMP7 caused nuclear translocation of Smad1 in > 90% of neurons, whereas < 10% of neurons exhibited nuclear Smad1 immunoreactivity in control cultures (Fig. 4a). LVS did not alter this pattern of Smad1 immunoreactivity (Fig. 4a): 95% of neurons treated with BMP7 + LVS exhibited nuclear localization of Smad1

Fig. 2 Statins inhibit dendritic growth in cultured sympathetic neurons. Phase-contrast (a, c, e and g) and fluorescence (b, d, f and h) micrographs of sympathetic neurons cultured from superior cervical ganglia (SCG) immunostained for MAP2 after treatment for 4 days with BMP7 (25 ng/ mL) ± lovastatin (LVS, 1 lM); bar, 40 lm. (i) Atorvastatin and lovastatin inhibit BMPinduced dendritic growth in a concentrationdependent manner. (j) Addition of lovastatin (LVS, 1 lM) to cultured sympathetic neurons previously treated with BMP7 (25 ng/ mL) for 5 days to induce dendritic growth caused dendritic retraction. (k) Lipophilic (atorvastatin, mevastatin, and simvastatin), but not hydrophilic (pravastatin) statins inhibit BMP7-induced dendritic growth. (l) Lovastatin inhibits the dendrite promoting activity of BMPs (25 ng/mL) from the dpp (BMP2) and 60A (BMPs 5 and 6) subfamilies. (m) Atorvastatin and lovastatin inhibit BMP7-induced dendritic growth in sympathetic neurons cultured from stellate ganglia. Data presented as the mean ± SEM (n = 60); *significantly different from control at p < 0.05.

compared with 7–8% of neurons in control cultures receiving LVS only. To confirm these morphometric observations, we quantified LVS effects on luciferase activity in sympathetic neurons transfected with pTLX2-lux, a firefly luciferase reporter construct under control of the Smad1-inducible TLX2 promoter (Tang et al. 1998). To account for potential differences in transfection efficiency between cultures, cultures were co-transfected with the Renilla luciferase reporter construct, and firefly luciferase activity was normalized to Renilla luciferase activity. BMP7 treatment significantly



Fig. 3 Statins selectively inhibit dendritic growth in cultured sympathetic neurons independent of effects on axonal growth or cell survival. Axonal growth was assessed in sympathetic neurons 15 h after plating on laminin (2 lg/mL) in medium containing BMP7 ± varying concentrations of lovastatin (LVS) (a) or after a 4-day treatment with BMP7 ± LVS (b). (c) Effects of statin on cell viability as measured using the 1-(4,5-Dimethylthiazol-2-yl)-3,5-diphenylformazan (MTT) assay (n = 4 cultures per treatment). (d) To determine whether statin

effects on dendritic growth were reversible, sympathetic neurons were cultured with BMP7 (25 ng/mL) ± lovastatin (LVS, 1 lM) for 5 days, at which time lovastatin was removed from a subset of cultures and dendritic growth quantified in cultures immunostained for MAP2 at varying times after removal of lovastatin. (e) Effects of lovastatin on neuron number per culture following a 4-day exposure to BMP7 (25 ng/mL) ± lovastatin (n = 3). Data presented as the mean ± SEM (n = 60, unless otherwise noted), *p < 0.05, **p < 0.01.

increased luciferase activity, whereas treatment with LVS alone had no effect (Fig. 4b). Luciferase activity in cultures treated with both BMP7 and LVS was comparable to that observed in cultures treated with BMP7 only (Fig. 4b).

inhibition of these metabolic pathways mediates the effects of statins on dendritic growth, then supplementation with products downstream of HMG-CoA should reverse statin effects. Supplementation with mevalonate had no effect on dendritic growth in control cultures but reversed the inhibitory effects of LVS on dendritic growth in BMP7-treated cultures (Fig. 5b). In contrast, cholesterol supplementation had no effect on LVS inhibition of BMP7-induced dendritic growth (Fig. 5c). Consistent with this latter observation, the cholesterol synthesis inhibitor zaragozic acid (Fig. 5d), which inhibits squalene synthase downstream of HMGCoA (Fig. 5a), did not inhibit BMP7-induced dendritic growth.

Depletion of isoprenoids contributes to statin effects on dendrites Statins inhibit HMG-CoA reductase, the enzyme that catalyzes the synthesis of mevalonate, which is a necessary precursor for synthesis of not only cholesterol but also the isoprenoids farnesyl- and geranylgeranyl-pyrophosphate (Fig. 5a). Thus, statin inhibition of mevalonate synthesis depletes both cholesterol and isoprenoids (Liao 2002). If


1064 | W.-Y. Kim et al.

(a)

(b)

Fig. 4 Lovastatin does not block BMP activation of Smad1. (a) Cultured sympathetic neurons were treated with BMP7 (25 ng/ mL) ± lovastatin (LVS, 1 lM) for 2 h and then immunostained for Smad1. When 1 lm optical sections were examined by confocal microscopy in control cultures not exposed to BMP7, the fluorescence intensity in the cytoplasm and nucleus are comparable. Lovastatin alone had no effect on this pattern of Smad1 immunoreactivity. In contrast, BMP7 causes nuclear Smad1 immunostaining to become more intense than that of the cytoplasm. Simultaneous exposure to

lovastatin did not block the nuclear localization of Smad1. (b) Cultured sympathetic neurons were cotransfected with plasmids encoding either the firefly luciferase gene driven by the Smad1-inducible TLX2 promoter (pTLX2-lux) or Renilla luciferase (pEF-R), then treated for 24 h with BMP7 (25 ng/mL) ± lovastatin (LVS, 1 lM). Smad1 activation was determined by measuring firefly luciferase activity, which was normalized against Renilla luciferase activity to account for differences in transfection efficiency between cultures. Data presented as the mean ± SEM (n = 3); *significantly different from control at p < 0.05.

The observation that mevalonate, but not cholesterol, reversed the dendritic effects of LVS suggested that statins influence dendritic growth via the isoprenoid pathway. To test this, BMP7-induced dendritic growth was quantified in cultures treated with the general isoprenoid synthesis inhibitors, manumycin and perillic acid (Fig. 6a and b), or the more specific geranylgeranyl transferase inhibitor GGTI-298 and farnesyl transferase inhibitor FTI-277 (Fig. 6c and d). All four inhibitors significantly decreased BMP7-induced dendritic growth in a concentration-dependent manner. Conversely, supplementation with the isoprenoid precursors, farnesol, and geranylgeraniol (Fig. 6e and f), or with the isoprenoids, farnesyl pyrophosphate, and geranylgeranyl pyrophosphate (Fig. 6g and h) significantly attenuated the inhibitory effects of LVS on BMP7-induced dendritic growth.

et al. 1991; Koch et al. 1997). Therefore, we considered the possibility that LVS inhibits BMP7-induced dendritic growth by blocking Rho GTPase activation. Studies implicating Rho GTPases in dendritic growth have focused on activitydependent dendritic growth in neurons derived from the CNS (Redmond and Ghosh 2001; Van Aelst and Cline 2004). Thus, we first determined whether Rho GTPases are involved in BMP7-induced dendritic growth in sympathetic neurons. Neurons were treated with BMP7 for 5 days following transfection with plasmids encoding enhanced green fluorescent protein (EGFP) alone (control) or co-expressing EGFP and either wt or dn constructs of RhoA, Rac1, or Cdc42. Quantification of dendritic growth in EGFP-positive neurons revealed that expression of wt constructs had no effect on BMP7-induced dendritic growth (data not shown) but that expression of either dnRhoA or dnCdc42 caused a significant decrease in BMP7-induced dendritic growth (Fig. 7a and b). To address the possibility that effects observed with dnRhoA or dnCdc42 represent off-target effects (Feig 1999), we further investigated the role for specific Rho GTPases in BMP7-induced dendritic growth using pharmacological inhibitors of RhoA (C3-transferase), Rock1 (Y-27632), Cdc42 (secramine A), or Rac1 (NSC23766). Pharmacologically effective concentrations of these inhibitors were confirmed in independent cell spreading assays using rat

RhoA activation is required for statin effects on BMP7-induced dendritic growth Rho GTPases link environmental stimuli to diverse cellular responses (Etienne-Manneville and Hall 2002), including dendritic growth (Redmond and Ghosh 2001; Van Aelst and Cline 2004). Rho GTPase function requires isoprenylation of the C terminus, and depletion of isoprenoids by statins has been linked to functional disruption of Rho GTPases (Hori



Fig. 5 Mevalonate, but not cholesterol, reverses the inhibitory effects of lovastatin on dendritic growth. (a) Statins inhibit HMG-CoA reductase, blocking synthesis of mevalonate, a necessary precursor for the synthesis of cholesterol and the isoprenoids, farnesyl pyrophosphate (farnesyl-PP) and geranylgeranylpyrophosphate (geranyl-geranyl-PP). (b) Cultured sympathetic neurons were treated for 4 days with BMP7

(25 ng/mL) ± lovastatin (LVS, 1 lM), and either (b) mevalonate (10 mM); (c) cholesterol (10 lg/mL); or (d) varying concentrations of the specific cholesterol synthesis inhibitor, zaragozic acid (ZGA). The number of dendrites per neuron was quantified in cultures immunostained for MAP2. Data presented as the mean ± SEM (n = 60); *significantly different from BMP7 + lovastatin at p < 0.05.

emybronic fibroblasts (data not shown). As illustrated in Fig. 7(c), only the RhoA inhibitor blocked BMP7-induced dendritic growth. The lack of effect of the Rock1 inhibitor is consistent with a recent report that nerve growth factor, which is present at saturating concentrations in our sympathetic neuronal cell cultures, phosphorylates RhoA on serine188 thereby selectively blocking RhoA interactions with Rho-associated kinase (Nusser et al. 2006). Because molecular and pharmacological approaches yielded discrepant results for Cdc42, and to further substantiate a role for RhoA in BMP7-induced dendritic growth, we next determined whether BMP7 increases the levels of GTP-bound RhoA, Cdc42, and Rac1. Using GTP pull-down assays, we observed that BMP7 significantly increased cellular levels of GTP-RhoA in sympathetic neurons, but caused a concomitant decrease in cellular levels of GTP-Rac1 and GTPCdc42 (Fig. 7d). Our observation that RhoA activation is involved in BMP7-induced dendrite formation is unique and in contradiction to published evidence demonstrating that RhoA activation results in dendrite retraction in neurons cultured from the embryonic rat CNS (Redmond and Ghosh

2001; Van Aelst and Cline 2004). Thus, to further test the effectiveness of our ca and dnRhoA constructs, we evaluated the effects of expressing these constructs on dendritic growth in primary neuronal cell cultures from embryonic rat hippocampi. Consistent with the published literature, we observed that expression of caRhoA decreased whereas dnRhoA increased dendritic complexity in hippocampal neurons as determined by quantification of the total spine number (Fig. 7e) and spine density (Fig. 7f). Consistent with the hypothesis that LVS inhibits BMP7induced dendritic growth in sympathetic neurons by inhibiting RhoA activation, expression of caRhoA, but not control vector, significantly attenuated the inhibitory effects of LVS on dendritic growth (Fig. 8a and b). If statins interfere with RhoA function by depletion of isoprenoids, then LVS should block translocation of RhoA from the cytosol to the membrane. Western blotting revealed that LVS significantly decreased membrane-associated RhoA and significantly increased cytosolic RhoA in the absence or presence of BMP7 (Fig. 8c and d).


1066 | W.-Y. Kim et al.

Fig. 6 Statins inhibit dendritic outgrowth through depletion of isoprenoids. The number of dendrites per neuron was quantified in sympathetic neurons cultured for 4 days with BMP7 (25 ng/mL) ± general inhibitors of isoprenoid biosynthesis, perillic acid (a) or manumycin A (b), the specific geranyl-geranyl transferase inhibitor GGTI-298 (c) or the specific farnesyl transferase inhibitor FTI-277 (d). Dendritic

Discussion Our data support the hypothesis that statins decrease dendritic arborization in post-ganglionic sympathetic neurons. Administration of atorvastatin to adult male rats at therapeutically relevant doses caused significant dendrite retraction in SCG in vivo. These in vivo effects are likely mediated by direct effects of statins on post-ganglionic sympathetic neurons because statins inhibited BMP-induced dendritic growth and caused elimination of existing dendrites when added to primary cultures of sympathetic neurons

number was determined in sympathetic neurons cultured for 4 days with BMP7 (25 ng/mL) ± lovastatin (LVS, 1 lM) and either (e) farnesol, (f) geranylgeraniol (GG-OH), (g) farnesyl pyrophosphate (FPP), or (h) geranylgeranyl pyrophosphate (GG-PP). Data presented as the mean ± SEM (n = 60); *significantly different from control at p < 0.05.

derived from either the SCG or stellate ganglia and grown in the absence of afferent input, ganglionic non-neuronal cells and target tissues. The inhibitory effects of statins on dendritic growth in sympathetic neurons were reversible and occurred in the absence of effects on mitochondrial activity, neuronal cell number or axonal growth, indicating that statins did not reduce dendritic arborization secondary to general deleterious effects on neuronal cell viability, nor did statins act by selectively promoting the survival of a subpopulation of neurons. The observation that axons continued to grow at a rate of several 100 lm/day in the



(b)

(a)

(c)

(d)

(e)

(f)

Fig. 7 RhoA activation contributes to BMP-induced dendritic growth in cultured sympathetic neurons. (a) Fluorescence images of cultured sympathetic neurons transfected with plasmids co-expressing enhanced green fluorescent protein (EGFP) and either wild type (wt) or dominant negative (dn) constructs of RhoA 24 h prior to being exposed to BMP7 (25 ng/mL) for 5 days then immunostained for the dendrite-selective antigen MAP2. (b) Quantification of dendritic growth in green fluorescent protein (GFP)-positive neurons transfected with plasmids co-expressing EGFP and a various dnRho GTPase constructs. (c) Quantification of dendritic growth in sympathetic neurons exposed to BMP7 (25 ng/mL) for 3 days in the absence (vehicle) or presence of inhibitors of RhoA (C3-Transferase, 5 lM), Rock1 (Y27632, 50 lM), Cdc42 (secramine A, 10 lM), or Rac1 (NSC23766,

50 lM). (d) GST pull-down assays of lysates harvested from cultured sympathetic neurons indicate that BMP7 treatment significantly increases levels of GTP-bound RhoA while concomitantly reducing levels of GTP-bound Rac1 and Cdc42. Densitometric values of GTPbound Rho GTPase are normalized to total Rho GTPase and expressed as a % of densitometric values obtained from control cultures not exposed to BMP7 within each experiment (n = 3 independent experiments per Rho GTPase). When expressed in primary cultures of hippocampal neurons, constitutively active (ca) RhoA decreased whereas dnRhoA increased dendritic complexity as measured by total spine number (e) and number of spines per micron (f) (n = 25 per experimental condition). Data presented as the mean ± SEM (n = 60 unless otherwise noted); *p < 0.05.

presence of LVS demonstrated that statins did not globally repress process outgrowth, but rather selectively targeted dendritic growth. Our findings are consistent with previous reports that statins significantly inhibited neurite outgrowth in cultured cortical neurons (Fan et al. 2002; Schulz et al. 2004). While specific effects on axons versus dendrites were not distinguished in one of these reports (Schulz et al. 2004); the other (Fan et al. 2002) identified the affected neurites as

dendrites. In contrast, statins have also been reported to induce neurite outgrowth in cultured hippocampal neurons (Pooler et al. 2006), explants of rat embryonic cortex and post-natal spinal cord (Holmberg et al. 2006) and PC12 cells (Fernandez-Hernando et al. 2005). The observation that neurite outgrowth in hippocampal neurons was induced by pravastatin is puzzling given reports that this statin is not taken up by non-hepatic cells (Liao and Laufs 2005), which is consistent with our observations that this statin had no


1068 | W.-Y. Kim et al.

(a)

(b)

(c)

(d)

Fig. 8 Statins inhibit dendritic growth by interfering with RhoA signaling. (a) Fluorescence images of cultured sympathetic neurons transfected with plasmids co-expressing enhanced green fluorescent protein (EGFP) and either wild type (wt) or constitutively active (ca) constructs of RhoA prior to being exposed to BMP7 (25 ng/ mL) ± lovastatin (LVS, 1 lM) for 5 days then immunostained for MAP2 to visualize dendritic arbors. (b) Quantification of dendritic growth in GFP-positive neurons transfected with plasmids expressing

only EGFP (vector) or co-expressing EGFP and caRhoA (n = 60 per experimental condition). (c) Representative western blot of membrane and cytosolic fractions from cultured sympathetic neurons treated for 2 days with BMP7 ± lovastatin probed for RhoA. (d) Densitometric analyses confirm that lovastatin blocks BMP-induced translocation of RhoA from the cytosol to the membrane (n = 3 independent experiments per condition). Data presented as the mean ± SEM; *significantly different from control at p < 0.05.

effect on dendritic growth in sympathetic neurons. The other studies observed neurite induction in response to lipophilic statins and the discrepancy between these findings and our observations that lipophilic statins inhibit dendritic growth may reflect either differences between neuronal cell types or the fact that the neurites extended from explants and from PC12 cells are typically axonal in nature. The latter suggests the possibility that statins exert opposing effects on axons versus dendrites, although we observed that statins did not influence axonal growth either negatively or positively. Inhibition of dendritic growth in sympathetic neurons appears to be a generalized pharmacological property of lipophilic statins. LVS, atorvastatin, mevastatin, and simvastatin all inhibited dendritic growth in a concentrationdependent manner with similar efficacy and potency, suggesting a common mechanism of action. Maximal effects were observed at 1–3 lM with an ED50 200 nM. The concentration of statins needed to inhibit dendritic growth are similar to those required to inhibit HMG CoA reductase or cholesterol synthesis in vivo (Black et al. 1998) and are within the range of steady-state serum levels of LVS in humans being treated for hypercholesterolemia (Pan et al. 1990). In contrast, the hydrophilic statin, pravastatin, which selectively inhibits sterol synthesis in the liver, had no effect on dendritic growth. Non-hepatic cells lack the membrane

carrier protein necessary to transport pravastatin across the cell membrane (Liao and Laufs 2005), suggesting that statins must gain access to the intracellular compartment in order to inhibit dendritic growth. We initially considered Smad1 as a candidate intracellular target in statin-mediated inhibition of dendritic growth, in part because statins inhibited the dendrite promoting activity of BMPs 2, 5, 6, and 7. The canonical signaling pathway triggered by all of these BMPs involves activation and nuclear translocation of cytoplasmic Smad transcription factors (Miyazono et al. 2001; Massague and Gomis 2006), and we have previously demonstrated that BMP-induced dendritic growth is dependent on Smad1 activation (Guo et al. 2001). LVS does not interfere with either BMP-induced translocation of Smad1 to the nucleus or BMP-induced activation of the Smad1 promoter TLX2, suggesting that statins perturb effector molecules either downstream or independent of Smad1 activation. Likely effector molecules include the Rho GTPases. Many of the cholesterol-independent effects of statins involve inhibition of Rho GTPase activity (Liao and Laufs 2005), and Rho GTPases function as central regulators of dendritic morphology, linking extracellular signals to changes in the dendritic actin cytoskeleton (Redmond and Ghosh 2001; Van Aelst and Cline 2004). Studies of cultured hippocampal and cortical neurons suggest that activity-dependent dendritic



growth is positively regulated by Cdc42 and Rac1 activation and negatively regulated by RhoA activation (Ruchhoeft et al. 1999; Li et al. 2000; Nakayama et al. 2000; Wong et al. 2000; Sin et al. 2002; Ahnert-Hilger et al. 2004) and that BMP7-induced dendritic growth is mediated in part by Cdc42 activation (Lee-Hoeflich et al. 2004). In contrast, BMP7-induced dendritic growth in sympathetic neurons appears to require RhoA activation as indicated by our collective data obtained using molecular, pharmacologic and biochemical approaches. That statins block BMP7-induced dendritic growth by interfering with RhoA activation is suggested by several observations. First, the inhibitory effects of LVS on BMP7-induced dendritic growth were significantly attenuated in sympathetic neurons expressing caRhoA. Second, BMP7 triggered translocation of RhoA from the cytoplasm to the membrane and this redistribution of RhoA was blocked by LVS. Translocation of RhoGTPases to the membrane, an event that is required for their functional activation, is regulated by isoprenylation (Bar-Sagi and Hall 2000), and statins interfere with Rho GTPase activity by inhibiting isoprenoid synthesis (Liao and Laufs 2005). We observed that pharmacological inhibition of isoprenoid biosynthesis downstream of HMG-CoA reductase mimicked the effects of statins on BMP7-induced dendritic growth. Conversely, supplementation with mevalonate, isoprenoid precursors, or isoprenoids significantly attenuated the inhibitory effects of statins on dendritic growth. In contrast, pharmacological inhibition of cholesterol synthesis had no effect on BMP7-induced dendritic growth in sympathetic neurons, and supplementation with cholesterol did not alter the inhibitory effects of statins on dendritic growth. These data indicate cholesterol-independent effects of statins on dendritic growth are mediated by depletion of isoprenoid pools secondary to statin inhibition of HMG-CoA reductase, and further support the role of RhoA signaling in mediating statin effects on dendritic growth. Are the effects of statins on dendritic arborization in postganglionic sympathetic neurons functionally linked to the observations that statins and other inhibitors of Rho GTPase signaling decrease sympathetic activity in vivo? While we do not have data that directly address this question, several lines of evidence support a connection. First, dendritic hypertrophy of post-ganglionic sympathetic neurons in both stellate and SCG is observed in the spontaneously hypertensive rat (Kondo et al. 1990; Peruzzi et al. 1991) and is thought to contribute to the pathogenesis of hypertension in this model (Kondo et al. 1990). Second, dendritic morphology is a critical determinant of synaptic connectivity in post-ganglionic sympathetic neurons (Purves 1975, 1988; De Castro et al. 1995), and there is a direct correlation between the size of the dendritic arbor and tonic activity in post-ganglionic sympathetic neurons (Ivanov and Purves 1989). So treatments that decrease dendritic arborization would be expected to reduce synaptic responses in sympathetic neurons. In other studies we have shown that a

30% decrease in the dendritic arborization of neurons in the SCG in vivo (compared with the 40% reduction observed in the current studies) is coincident with a significant decrease in not only synaptic input (Lein et al. 2005) but also in sympathetic ganglionic transmission as evidenced by significant attenuation of the baroreceptor reflex (Lein and Fryer, unpublished observations). Together, these observations suggest a novel mechanism contributing to the protective effects of statins in cardiovascular and cerebrovascular disease that involves modulation of sympathetic function via effects on the dendritic arborization of post-ganglionic sympathetic neurons.

Acknowledgements This work was supported by grants from the National Science Foundation (IBN0121210 to DH) and the National Institutes of Health (NS046649 to PJL).

References Ahnert-Hilger G., Holtje M., Grosse G., Pickert G., Mucke C., NixdorfBergweiler B., Boquet P., Hofmann F. and Just I. (2004) Differential effects of Rho GTPases on axonal and dendritic development in hippocampal neurones. J. Neurochem. 90, 9–18. Alm P. and Lundberg L. M. (1988) Co-existence and origin of peptidergic and adrenergic nerves in the guinea pig uterus. Retrograde tracing and immunocytochemistry, effects of chemical sympathectomy, capsaicin treatment and pregnancy. Cell Tissue Res. 254, 517–530. Alvarez-Sabin J., Huertas R., Quintana M., Rubiera M., Delgado P., Ribo M., Molina C. A. and Montaner J. (2007) Prior statin use may be associated with improved stroke outcome after tissue plasminogen activator. Stroke J. Cerebral Circ. 38, 1076–1078. Anonymous (2002) MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 360, 7–22. Bar-Sagi D. and Hall A. (2000) Ras and Rho GTPases: a family reunion. Cell 103, 227–238. Beck H. N., Drahushuk K., Jacoby D. B., Higgins D. and Lein P. J. (2001) Bone morphogenetic protein-5 (BMP-5) promotes dendritic growth in cultured sympathetic neurons. BMC Neurosci. 2, 12. Black A. E., Sinz M. W., Hayes R. N. and Woolf T. F. (1998) Metabolism and excretion studies in mouse after single and multiple oral doses of the 3-hydroxy-3-methylglutaryl-CoA reductase inhibitor atorvastatin. Drug Metab. Dispos. 26, 755–763. Blacklock A. D., Cauveren J. A. and Smith P. G. (2004) Estrogen selectively increases sensory nociceptor innervation of arterioles in the female rat. Brain Res. 1018, 55–65. Blanco-Colio L. M., Villa A., Ortego M., Hernandez-Presa M. A., Pascual A., Plaza J. J. and Egido J. (2002) 3-Hydroxy-3-methylglutaryl coenzyme A reductase inhibitors, atorvastatin and simvastatin, induce apoptosis of vascular smooth muscle cells by downregulation of Bcl-2 expression and Rho A prenylation. Atherosclerosis 161, 17–26. Camargos E. R. and Machado C. R. (1988) Morphometric and histological analysis of the superior cervical ganglion in experimental Chagas’ disease in rats. Am. J. Trop. Med. Hyg. 39, 456–462. Cechetto D. F., Wilson J. X., Smith K. E., Wolski D., Silver M. D. and Hachinski V. C. (1989) Autonomic and myocardial changes in middle cerebral artery occlusion: stroke models in the rat. Brain Res. 502, 296–305.


1070 | W.-Y. Kim et al.

Collins R., Peto R. and Armitage J. (2002) The MRC/BHF Heart Protection Study: preliminary results. Int. J. Clin. Pract. 56, 53–56. Collins R., Armitage J., Parish S., Sleight P. and Peto R. (2004) Effects of cholesterol-lowering with simvastatin on stroke and other major vascular events in 20536 people with cerebrovascular disease or other high-risk conditions. Lancet 363, 757–767. Corsini A., Bellosta S., Baetta R., Fumagalli R., Paoletti R. and Bernini F. (1999) New insights into the pharmacodynamic and pharmacokinetic properties of statins. Pharmacol. Ther. 84, 413–428. De Castro F., Sanchez-Vives M. V., Munoz-Martinez E. J. and Gallego R. (1995) Effects of postganglionic nerve section on synaptic transmission in the superior cervical ganglion of the guinea-pig. Neuroscience 67, 689–695. Etienne-Manneville S. and Hall A. (2002) Rho GTPases in cell biology. Nature 420, 629–635. Fan Q. W., Yu W., Gong J. S., Zou K., Sawamura N., Senda T., Yanagisawa K. and Michikawa M. (2002) Cholesterol-dependent modulation of dendrite outgrowth and microtubule stability in cultured neurons. J. Neurochem. 80, 178–190. Feig L. A. (1999) Tools of the trade: use of dominant-inhibitory mutants of Ras-family GTPases. Nat. Cell Biol. 1, E25–E27. Fernandez-Hernando C., Suarez Y. and Lasuncion M. A. (2005) Lovastatin-induced PC-12 cell differentiation is associated with RhoA/ RhoA kinase pathway inactivation. Mol. Cell. Neurosci. 29, 591– 602. Floras J. S. (1993) Clinical aspects of sympathetic activation and parasympathetic withdrawal in heart failure. J. Am. Coll. Cardiol. 22, 72A–84A. Francis G. S., Cohn J. N., Johnson G., Rector T. S., Goldman S. and Simon A. (1993) Plasma norepinephrine, plasma renin activity, and congestive heart failure. Relations to survival and the effects of therapy in V-HeFT II. The V-HeFT VA Cooperative Studies Group. Circulation 87, VI40–VI48. Furuichi S., Endo S., Haji A., Takeda R., Nisijima M. and Takaku A. (1999) Related changes in sympathetic activity, cerebral blood flow and intracranial pressure, and effect of an alpha-blocker in experimental subarachnoid haemorrhage. Acta Neurochir. (Wien) 141, 415–423; discussion 423–414. Gao L., Wang W., Li Y. L., Schultz H. D., Liu D., Cornish K. G. and Zucker I. H. (2005) Simvastatin therapy normalizes sympathetic neural control in experimental heart failure: roles of angiotensin II type 1 receptors and NAD(P)H oxidase. Circulation 112, 1763– 1770. Gao L., Wang W. and Zucker I. H. (2008) Simvastatin inhibits central sympathetic outflow in heart failure by a nitric-oxide synthase mechanism. J. Pharmacol. Exp. Ther. 326, 278–285. Grutzendler J., Tsai J. and Gan W. B. (2003) Rapid labeling of neuronal populations by ballistic delivery of fluorescent dyes. Methods 30, 79–85. Guo X., Rueger D. and Higgins D. (1998) Osteogenic protein-1 and related bone morphogenetic proteins regulate dendritic growth and the expression of microtubule-associated protein-2 in rat sympathetic neurons. Neurosci. Lett. 245, 131–134. Guo X., Lin Y., Horbinski C., Drahushuk K. M., Kim I. J., Kaplan P. L., Lein P., Wang T. and Higgins D. (2001) Dendritic growth induced by BMP-7 requires Smad1 and proteasome activity. J. Neurobiol. 48, 120–130. Hachinski V. C., Smith K. E., Silver M. D., Gibson C. J. and Ciriello J. (1986) Acute myocardial and plasma catecholamine changes in experimental stroke. Stroke J. Cerebral Circ. 17, 387–390. Hamaad A., Sosin M., Lip G. Y. and MacFadyen R. J. (2005) Short-term adjuvant atorvastatin improves frequency domain indices of heart rate variability in stable systolic heart failure. Cardiovasc. Drugs Ther. 19, 183–187.

Hansson M. (2002) Natriuretic peptides in relation to the cardiac innervation and conduction system. Microsc. Res. Tech. 58, 378– 386. Higgins D., Lein P. J., Osterhout D. J. and Johnson M. I. (1991) Tissue culture of mammalian autonomic neurons, in Culturing Nerve Cells, 1st edn (Banker G. and Goslin K., eds), pp. 177–205. MIT Press, Cambridge, MA. Higuchi S., Ohtsu H., Suzuki H., Shirai H., Frank G. D. and Eguchi S. (2007) Angiotensin II signal transduction through the AT1 receptor: novel insights into mechanisms and pathophysiology. Clin. Sci. (Lond) 112, 417–428. Holmberg E., Nordstrom T., Gross M., Kluge B., Zhang S. X. and Doolen S. (2006) Simvastatin promotes neurite outgrowth in the presence of inhibitory molecules found in central nervous system injury. J. Neurotrauma 23, 1366–1378. Hori Y., Kikuchi A., Isomura M., Katayama M., Miura Y., Fujioka H., Kaibuchi K. and Takai Y. (1991) Post-translational modifications of the C-terminal region of the rho protein are important for its interaction with membranes and the stimulatory and inhibitory GDP/GTP exchange proteins. Oncogene 6, 515– 522. Ito K., Hirooka Y., Sakai K., Kishi T., Kaibuchi K., Shimokawa H. and Takeshita A. (2003) Rho/Rho-kinase pathway in brain stem contributes to blood pressure regulation via sympathetic nervous system: possible involvement in neural mechanisms of hypertension. Circ. Res. 92, 1337–1343. Ivanov A. and Purves D. (1989) Ongoing electrical activity of superior cervical ganglion cells in mammals of different size. J. Comp. Neurol. 284, 398–404. Kishi T., Hirooka Y., Mukai Y., Shimokawa H. and Takeshita A. (2003) Atorvastatin causes depressor and sympatho-inhibitory effects with upregulation of nitric oxide synthases in stroke-prone spontaneously hypertensive rats. J. Hypertens. 21, 379–386. Koch G., Benz C., Schmidt G., Olenik C. and Aktories K. (1997) Role of Rho protein in lovastatin-induced breakdown of actin cytoskeleton. J. Pharmacol. Exp. Ther. 283, 901–909. Kondo M., Terada M., Shimizu D., Fujiwara T. and Tabei R. (1990) Morphometric study of the superior cervical and stellate ganglia of spontaneously hypertensive rats during the prehypertensive stage. Virchows Arch. B Cell Pathol. Incl. Mol. Pathol. 58, 371–376. Korsching S. and Thoenen H. (1988) Developmental changes of nerve growth factor levels in sympathetic ganglia and their target organs. Dev. Biol. 126, 40–46. Krum H. and McMurray J. J. (2002) Statins and chronic heart failure: do we need a large-scale outcome trial? J. Am. Coll. Cardiol. 39, 1567–1573. Laufs U. and Liao J. K. (1998) Post-transcriptional regulation of endothelial nitric oxide synthase mRNA stability by Rho GTPase. J. Biol. Chem. 273, 24266–24271. Lee-Hoeflich S. T., Causing C. G., Podkowa M., Zhao X., Wrana J. L. and Attisano L. (2004) Activation of LIMK1 by binding to the BMP receptor, BMPRII, regulates BMP-dependent dendritogenesis. EMBO J. 23, 4792–4801. Lefer A. M., Scalia R. and Lefer D. J. (2001) Vascular effects of HMG CoA-reductase inhibitors (statins) unrelated to cholesterol lowering: new concepts for cardiovascular disease. Cardiovasc. Res. 49, 281–287. Lein P. J. and Higgins D. (1989) Laminin and a basement membrane extract have different effects on axonal and dendritic outgrowth from embryonic rat sympathetic neurons in vitro. Dev. Biol. 136, 330–345. Lein P., Johnson M., Guo X., Rueger D. and Higgins D. (1995) Osteogenic protein-1 induces dendritic growth in rat sympathetic neurons. Neuron 15, 597–605.



Lein P. J., Gonsiorek E. A., Garred M. and Higgins D. (2005) Gammainterferon causes dendritic retraction in sympathetic neurons in vivo, Program No. 600.11, Abstract Viewer/Itinerary Planner, p Online. Society for Neuroscience, Washington, DC. Li Z., Van Aelst L. and Cline H. T. (2000) Rho GTPases regulate distinct aspects of dendritic arbor growth in Xenopus central neurons in vivo. Nat. Neurosci. 3, 217–225. Liao J. K. (2002) Isoprenoids as mediators of the biological effects of statins. J. Clin. Invest. 110, 285–288. Liao J. K. and Laufs U. (2005) Pleiotropic effects of statins. Annu. Rev. Pharmacol. Toxicol. 45, 89–118. Lincoln J. (1995) Innervation of cerebral arteries by nerves containing 5-hydroxytryptamine and noradrenaline. Pharmacol. Ther. 68, 473–501. Massague J. and Gomis R. R. (2006) The logic of TGFbeta signaling. FEBS Lett. 580, 2811–2820. Melenovsky V., Wichterle D., Simek J., Malik J., Haas T., Ceska R. and Malik M. (2003) Effect of atorvastatin and fenofibrate on autonomic tone in subjects with combined hyperlipidemia. Am. J. Cardiol. 92, 337–341. Miller J. M., Ren T. Y. and Nuttall A. L. (1995) Studies of inner ear blood flow in animals and human beings. Otolaryngol. Head Neck Surg. 112, 101–113. Miura M. and Okada J. (1981) Cardiac and non-cardiac preganglionic neurones of the thoracic vagus nerve: an HRP study in the cat. Jpn. J. Physiol. 31, 53–66. Miyazono K., Kusanagi K. and Inoue H. (2001) Divergence and convergence of TGF-beta/BMP signaling. J. Cell. Physiol. 187, 265–276. Nakayama A. Y., Harms M. B. and Luo L. (2000) Small GTPases Rac and Rho in the maintenance of dendritic spines and branches in hippocampal pyramidal neurons. J. Neurosci. 20, 5329–5338. Nusser N., Gosmanova E., Makarova N., Fujiwara Y., Yang L., Guo F., Luo Y., Zheng Y. and Tigyi G. (2006) Serine phosphorylation differentially affects RhoA binding to effectors: implications to NGF-induced neurite outgrowth. Cell. Signal. 18, 704–714. Pan H. Y., DeVault A. R., Wang-Iverson D., Ivashkiv E., Swanson B. N. and Sugerman A. A. (1990) Comparative pharmacokinetics and pharmacodynamics of pravastatin and lovastatin. J. Clin. Pharmacol. 30, 1128–1135. Pardini B. J., Lund D. D. and Schmid P. G. (1989) Organization of the sympathetic postganglionic innervation of the rat heart. J. Auton. Nerv. Syst. 28, 193–201. Pedersen T. R., Berg K., Cook T. J. et al. (1996) Safety and tolerability of cholesterol lowering with simvastatin during 5 years in the Scandinavian Simvastatin Survival Study. Arch. Intern. Med. 156, 2085–2092. Pehlivanidis A. N., Athyros V. G., Demitriadis D. S., Papageorgiou A. A., Bouloukos V. J. and Kontopoulos A. G. (2001) Heart rate variability after long-term treatment with atorvastatin in hypercholesterolaemic patients with or without coronary artery disease. Atherosclerosis 157, 463–469. Peruzzi D., Hendley E. D. and Forehand C. J. (1991) Hypertrophy of stellate ganglion cells in hypertensive, but not hyperactive, rats. Am. J. Physiol. 261, R979–R984. Pliquett R. U., Cornish K. G. and Zucker I. H. (2003a) Statin therapy restores sympathovagal balance in experimental heart failure. J. Appl. Physiol. 95, 700–704. Pliquett R. U., Cornish K. G., Peuler J. D. and Zucker I. H. (2003b) Simvastatin normalizes autonomic neural control in experimental heart failure. Circulation 107, 2493–2498. Pooler A. M., Xi S. C. and Wurtman R. J. (2006) The 3-hydroxy-3methylglutaryl co-enzyme A reductase inhibitor pravastatin enhances neurite outgrowth in hippocampal neurons. J. Neurochem. 97, 716–723.

Purves D. (1975) Functional and structural changes in mammalian sympathetic neurones following interruption of their axons. J. Physiol. 252, 429–463. Purves D. (1988) Body and Brain: a Trophic Theory of Neural Connections. Harvard University Press, Cambridge, MA. Purves D. and Lichtman J. W. (1985) Geometrical differences among homologous neurons in mammals. Science 228, 298–302. Redmond L. and Ghosh A. (2001) The role of Notch and Rho GTPase signaling in the control of dendritic development. Curr. Opin. Neurobiol. 11, 111–117. Ruchhoeft M. L., Ohnuma S., McNeill L., Holt C. E. and Harris W. A. (1999) The neuronal architecture of Xenopus retinal ganglion cells is sculpted by rho-family GTPases in vivo. J. Neurosci. 19, 8454– 8463. Sander D., Winbeck K., Klingelhofer J., Etgen T. and Conrad B. (2001) Prognostic relevance of pathological sympathetic activation after acute thromboembolic stroke. Neurology 57, 833–838. Saneyoshi T., Wayman G., Fortin D., Davare M., Hoshi N., Nozaki N., Natsume T. and Soderling T. R. (2008) Activity-dependent synaptogenesis: regulation by a CaM-kinase kinase/CaM-kinase I/ betaPIX signaling complex. Neuron 57, 94–107. Schulz J. G., Bosel J., Stoeckel M., Megow D., Dirnagl U. and Endres M. (2004) HMG-CoA reductase inhibition causes neurite loss by interfering with geranylgeranylpyrophosphate synthesis. J. Neurochem. 89, 24–32. Sin W. C., Haas K., Ruthazer E. S. and Cline H. T. (2002) Dendrite growth increased by visual activity requires NMDA receptor and Rho GTPases. Nature 419, 475–480. Tang S. J., Hoodless P. A., Lu Z., Breitman M. L., McInnes R. R., Wrana J. L. and Buchwald M. (1998) The Tlx-2 homeobox gene is a downstream target of BMP signalling and is required for mouse mesoderm development. Development 125, 1877–1887. Tunon J., Martin-Ventura J. L., Blanco-Colio L. M. and Egido J. (2007) Mechanisms of action of statins in stroke. Expert Opin. Ther. Targets 11, 273–278. Van Aelst L. and Cline H. T. (2004) Rho GTPases and activitydependent dendrite development. Curr. Opin. Neurobiol. 14, 297–304. Verberne M. E., Gittenberger-De Groot A. C., Van Iperen L. and Poelmann R. E. (1999) Contribution of the cervical sympathetic ganglia to the innervation of the pharyngeal arch arteries and the heart in the chick embryo. Anat. Rec. 255, 407–419. Wayman G. A., Impey S., Marks D., Saneyoshi T., Grant W. F., Derkach V. and Soderling T. R. (2006) Activity-dependent dendritic arborization mediated by CaM-kinase I activation and enhanced CREB-dependent transcription of Wnt-2. Neuron 50, 897–909. Wierzbicki A. S., Poston R. and Ferro A. (2003) The lipid and non-lipid effects of statins. Pharmacol. Ther. 99, 95–112. Wingerd K. L., Goodman N. L., Tresser J. W., Smail M. M., Leu S. T., Rohan S. J., Pring J. L., Jackson D. Y. and Clegg D. O. (2002) Alpha 4 integrins and vascular cell adhesion molecule-1 play a role in sympathetic innervation of the heart. J. Neurosci. 22, 10772– 10780. Wong W. T., Faulkner-Jones B. E., Sanes J. R. and Wong R. O. (2000) Rapid dendritic remodeling in the developing retina: dependence on neurotransmission and reciprocal regulation by Rac and Rho. J. Neurosci. 20, 5024–5036. Xu B., Pelish H., Kirchhausen T. and Hammond G. B. (2006) Large scale synthesis of the Cdc42 inhibitor secramine A and its inhibition of cell spreading. Org. Biomol. Chem. 4, 4149–4157. Zivin J. A. (2007) Clinical trials of neuroprotective therapies. Stroke J. Cerebral Circ. 38, 791–793. Zucker I. H. and Wang W. (1991) Reflex control of renal sympathetic nervous activity in heart failure. Herz 16, 82–91.
