Involvement of α4 integrins in maintenance of cardiac sympathetic axons

Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68 www.elsevier.com/locate/autneu

Involvement of a4 integrins in maintenance of cardiac sympathetic axons Kevin L. Wingerd a,1, William C. Wayne a,1, David Y. Jackson b, Dennis O. Clegg a,* a

Neuroscience Research Institute and Department of Molecular, Cellular and Developmental Biology, University of California, Santa Barbara, CA 93106, United States b Genentech Inc., Department of Bioorganic Chemistry, 1 DNA Way, South San Francisco, CA 94080, United States Received 19 May 2005; received in revised form 1 August 2005; accepted 6 August 2005

Abstract Sympathetic neurons extend and maintain axons that innervate the myocardium, and proper innervation is important for cardiac function. However, the molecular basis for axon outgrowth and maintenance is not well understood. We have shown previously that the integrin a4h1 is expressed on developing axons, and the a4 function is important for the development of innervation in vivo [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]. Here we examine the function of a4h1 integrins in the maintenance of cardiac sympathetic innervation in vitro and in vivo, and investigate integrin expression and function after myocardial infarction and in hypertensive rats. On substrates of vascular cell adhesion molecule-1 (VCAM-1), a4h1 was required for both initial outgrowth and maintenance of neurites in vitro. On fibronectin substrates, initial outgrowth requires only a4 integrins, but maintenance requires both a4 integrins and RGD-dependent integrins. In vivo, in adult Long Evans rats, inhibition of a4 integrins resulted in decreased maintenance of sympathetic fibers innervating the apex of the heart. However, a4 integrins were not detected on most sympathetic axons that sprout after myocardial infarction, and a4 function was not required for sprouting. Spontaneously hypertensive rats (SHR) have increased numbers of cardiac sympathetic fibers compared to the parental Wistar strain, but many of these lack a4 expression, and a4 function is not required for maintenance of these fibers in the heart. These results suggest that developing sympathetic axons and sprouting sympathetic axons use different mechanisms of outgrowth, and that maintenance of cardiac sympathetic innervation involves a4 integrins in some rat strains. D 2005 Published by Elsevier B.V. Keywords: Sympathetic; Integrins; Alpha-4; Beta-1; VLA-4; Myocardial infarction; Sprouting; Neurite outgrowth; Reinnervation; Axon; Synapse; Hypertensive

1. Introduction Sympathetic neurons innervate the heart during the first few postnatal weeks, and axon terminals remain throughout the myocardium in the adult (Lipp and Rudolph, 1972). Maintenance of the distribution of cardiac sympathetic innervation is important, as several pathological states are associated with abnormal sympathetic sprouting. Increased sympathetic function may lead to arrhythmias and contribute to heart attack, hypertension, and, possibly, sudden * Corresponding author. Tel.: +1 805 893 8490; fax: +1 805 893 2005. E-mail address: [email protected] (D.O. Clegg). 1 The contributions of the first two authors were equivalent. 1566-0702/$ - see front matter D 2005 Published by Elsevier B.V. doi:10.1016/j.autneu.2005.08.006

infant death syndrome (Podrid et al., 1990; Schwartz et al., 1998; Palatini and Julius, 1999). After a myocardial infarction (MI) or other myocardial injury, sympathetic innervation is lost in the damaged portion of the heart. A regeneration response ensues, and tissue surrounding the scar is greatly hyper-innervated compared to the surrounding tissue (Chen et al., 2001). Newly formed (improper) sympathetic innervation may lead to ventricular tachycardia, fibrillation and Sudden Cardiac Death (Pugsley et al., 1999; Chen et al., 2001). This hypothesis is supported by studies in dog (Zipes, 1990; Chang et al., 2001), rat (Vracko et al., 1990; Du et al., 1999), and by research on human patients (Vracko et al., 1991). For example, MI leads to sympathetic sprouting in

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

dogs, and if sprouting is augmented by nerve growth factor (NGF), spontaneous ventricular tachycardia, fibrillation, and SCD follows (Cao et al., 2000). Up-regulation of the low affinity NGF receptor and GAP43 appears to drive the sprouting that occurs after MI (Zhou et al., 2004). In rats, Nori et al. (1995) found that necrotic injury of rat myocardium (induced by freeze –thaw) resulted in robust and persistent sympathetic reinnervation. Du et al. (1999) showed that in infarcted rats (coronary artery occlusion), sympathetic activation is a potent trigger for the onset of ventricular tachyarrhythmias. Sympathetic remodeling after MI has been documented in humans (Stanton et al., 1989; Vracko et al., 1991) and h-adrenergic antagonists are known to reduce the incidence of SCD in humans. In fact, recent clinical trials have indicated that h blockers reduce mortality in congestive heart failure (Doggrell, 2001). Treatments to control sympathetic sprouting after MI may be a novel, more effective way of preventing arrhythmias. Sympathetic over-activity may also be involved in hypertension. Studies of spontaneously hypertensive rats have shown that there are increased numbers of sympathetic fibers around blood vessels and in the heart (Kondo et al., 1995; Tabei et al., 1995). Furthermore, sympathetic activity may be augmented by increased firing rates, faulty norepinephrine reuptake, or other factors (Rumantir et al., 2000; Schlaich et al., 2003). Increased sympathetic activity can lead to left ventricular hypertrophy, a risk factor in cardiovascular morbidity (Brum et al., 2002). The adhesive relationship between sympathetic axon terminals and heart tissue and the regulation of sympathetic sprouting are poorly understood. Sympathetic axon terminals often terminate in boutons or varicosities, without recognizable active zones, at some distance from the target cells, and released norepinephrine is thought to diffuse through the tissue and produce a slow, second messengercoupled response (Landis, 1976; Kitajiri et al., 1993). Synaptic structures have also been identified where sympathetic axons terminate on smooth muscle cells of the vasculature and appear to make close cell – cell contacts (Luff, 1996). Integrin receptors mediate cell – cell and cell – ECM interactions and may be involved in maintaining stable connections between neurons and their targets (Clegg et al., 2003). Integrins have also been implicated in neurite sprouting and neuronal development (Ekstrom et al., 2003; Hikita et al., 2003). Furthermore, some anti-adhesive molecules that cause axon retraction, such as ephrins, may function by disrupting integrin signaling pathways (Zou et al., 1999; Kullander and Klien, 2002). Integrins are a family of heterodimeric transmembrane receptors consisting of 18 alpha and 8 beta subunits that form 24 known pairs (Siebers et al., 2005). The integrin a4h1, well known for its role in inflammation and hematopoiesis (Lobb and Hemler, 1994; Arroyo et al., 1996), has also been shown to function in neurons (Vogelezang et al., 2001; Wingerd et al., 2002). The a4h1 integrin binds multiple ligands, including vascular cell

59

adhesion molecule-1 (VCAM-1) (Osborn et al., 1989), fibronectin’s (Fn) connecting sequence-1 LDV motif (Guan and Hynes, 1990), thrombospondin-1 (Yabkowitz et al., 1993), other a4 integrins (Altevogt et al., 1995), the propolypeptide of von Willebrand factor (Isobe et al., 1997), ICAM-4 (Spring et al., 2001), transglutaminase C (Isobe et al., 1999), and osteopontin (Bayless et al., 1998). Both VCAM-1 and FN are expressed in innervated regions of the heart (Sheppard et al., 1994). Neural cells that express a4h1 include neural crest cells (Kil et al., 1998), retinal cells (Sheppard et al., 1994), dorsal root ganglion neurons (Vogelezang et al., 2001), and superior cervical ganglion (SCG) neurons (Vogelezang et al., 2001; Wingerd et al., 2002). We have previously shown that integrins play a crucial role in sympathetic innervation of the heart during development (Wingerd et al., 2002). Blockade of a4 integrins leads to a 50% decrease in the number of fibers that reach the heart in Long Evans rats. The a4h1 integrin is tightly regulated during development. The intracellular distribution and isotype of the integrin change as function is down regulated with age (Wingerd et al., 2004). In the adult, a4 immunoreactivity within the superior cervical ganglion cells and on some of the axons in the myocardium persist, suggesting that a4 could play a role in maintenance of fibers. Here we present evidence that supports the hypothesis that the a4 integrins play a role in the maintenance of sympathetic fibers in the heart. First, an in vitro assay is described that assesses the importance of integrins in maintaining neurites once they have been elaborated on substrates of integrin ligands found in the heart. We show that a4h1 integrins are required for maintaining neuritic projections on VCAM-1 and FN. Next, a4 integrins are shown to be required for maintenance of sympathetic fibers in adult Long Evans rats, but are not required for maintenance in hypertensive rats or for sprouting post MI.

2. Methods 2.1. Quantification of neurite outgrowth Primary cultures of rat SCG cells (P1-3) were isolated and plated on 96 well culture plates (Corning-Costar, Acton, MA) coated with purified proteins, and incubated as described (Choi et al., 1994; Wingerd et al., 2002). Briefly, the cells were manually dissociated in media, trypsinized (0.05% trypsin), and triturated to eliminate clumped cells. The cells were cultured in serum free L15 media supplemented with ITS (insulin, transferin, selenium; Gibco BRL, Carlsbad, CA), PSF (penicillin, streptomycin, fungizone; Gibco BRL) and 100 ng/ml 7S-NGF (Sigma, St. Louis, MO). The wells were coated overnight with either laminin2/4 (10 Ag/ml LN; merosin, Gibco BRL) as a positive control, 1% BSA as a negative control, plasma fibronectin (20 Ag/ml FN, Gibco BRL), or recombinant soluble

60

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

VCAM-1 (5 Ag/ml rsVCAM-1, a kind gift of Roy Lobb, Biogen Inc.). To measure initial outgrowth, cells were incubated at 37 -C in a humidified atmosphere containing 5% CO2 for 24 h, in the presence of either an a4 antagonist (a4a; 40 Ag/ml dissolved in 50% PEG 400 from Sigma), RGD antagonist, RAD control (1 mM, BIOMOL, Plymouth, PA) or anti-h1 (10 Ag/ml, HA2/5, Pharmingen). Integrins that recognize the RGD motif in ECM ligands are blocked by RGD peptide antagonists but not by RAD control peptides. Control cultures received carrier alone (either 50% PEG400 or PBS). The a4 antagonist (similar to the Genentech example compound shown in Table 12 of Jackson (2002)) has been shown to be specific for a4 integrins in binding assays using purified integrins and ligands, and in cell adhesion experiments as described in Jackson (2002). To assay for maintenance of neurites, perturbing agents were added after 24 h, and cells were incubated for an additional 6 h before staining. The neurons and neurites were visualized by staining with the vital dye calcein AM (fluorescein diacetate, Molecular Probes, Eugene, OR) and viewed with an epifluorescence microscope (Culley et al., 2001). Images of > 100 cells from each condition were projected onto a magnetized bit pad, and the average neurite length per cell (including cells that lacked neurites) was scored to quantify the initial retraction response (Dibner et al., 1977; Burns et al., 1991; Culley et al., 2001). A neurite is defined as a visible process emanating from the cell body. Statistical analysis was carried out using a Students’ twotailed T-test.

mouse anti-tyrosine hydroxylase (TH) antibody (Chemicon, CA) and visualized with a Goat anti-mouse Cy3 secondary antibody (Jackson Laboratories). The number of THpositive fibers in the apex of the heart was quantified by directly counting in the microscope using a double blind procedure. From each of the LE and SHR animals, three 16 Am coronal sections (246 Am apart) from the apex of the heart, starting near the most distal portion of the apex, were stained with anti-TH antibodies. TH-positive fibers were counted 2 –3 times for each microscopic field of the section (¨4 fields, 0.28 mm2) for section 1, 6 fields for section 2, and 10 fields for section 3) by 2 observers, using a double blind procedure. Fibers were defined as linear, fine caliber TH immunoreactivity longer than 5 Am (Wingerd et al., 2002). Fibers per section were recorded for each condition. In MI rats, TH+ fibers were counted within areas of infarction and areas of non-infarcted tissue. Infarcted areas were identified by dark field microscopy. These areas were dominated by auto fluorescent particles not seen in the noninfarcted areas. Statistical analysis was carried out using a Students’ two-tailed T-test. 2.3. Analysis of cardiac sympathetic innervation and a4 integrin expression Hearts were collected from LE, SHR, and Wistar rats (P1, P9, P22, and adult), and from rats with induced MI three weeks after the operation. Hearts were fixed and treated as above. Sections were stained with a rabbit anti-TH and mouse anti-rat a4 integrin (TA2). Multiple sections (n = 10) were counted from each animal (n = 3).

2.2. Intrathoracic injection of a4 antagonist The a4 antagonist was dissolved in 50% PEG400 (Sigma) and injected into the thoracic cavity of Long Evans (control n = 6, experimental n = 5) or SHR (control n = 5, experimental n = 4) rats at a concentration of 40 mg/kg animal weight on postnatal days 25 and 28. The vehicle solution (50% PEG400) was injected into littermates as a control. Long Evans rats 5 – 6 weeks of age with MI induced by left anterior ascending coronary artery occlusion were obtained from Charles River Laboratories. Two weeks after the induced MI, the animals were injected with either the antagonist or the PEG control as above (control n = 3, experimental n = 3), followed by dissection 2 days later. The concentration used in the above experiments was approximately ten-fold greater than the amount required for a complete block of neurite outgrowth on rsVCAM-1 in vitro (Wingerd et al., 2002). Similar concentrations have been shown to be effective in other animal models (Jackson, 2002). In both the Long Evans and SHR experiments, P30 hearts were harvested and fixed for sectioning. Hearts were briefly rinsed in PBS then placed in 4% paraformaldehyde for 30 min followed by a PBS rinse, before being placed in 20% sucrose phosphate buffer for cryoprotection (all steps were performed at 4 -C). Sections were stained with a

3. Results 3.1. a4 integrin involvement in both initiation and maintenance of neurites in vitro The role of integrin a4h1 in initiation and maintenance of sympathetic axons was investigated using an in vitro assay where postnatal rat SCG neurons were cultured on relevant purified protein substrates known to induce neurite outgrowth. The importance of a4 integrins in initiation of neurites on LN, VCAM-1, or FN was determined by adding a small molecule a4 integrin antagonist (a4a) to neurons at the start of a 24-h incubation. Fig. 1 shows that the antagonist completely blocked outgrowth on both VCAM-1 and FN, but not on LN. These experiments confirmed that a4 integrin function was required for initial outgrowth on VCAM-1 and FN. a4h1 integrin binds to FN mainly via an LDV motif, but other integrins can interact with the well-characterized RGD adhesion sequence within FN. To determine if RGD binding integrins were needed for initiation of neurites on FN, cultures were challenged with RGD peptides. Fig. 1 shows that neither RGD, nor the control RAD peptide had any

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

61

Fig. 1. Initiation of sympathetic neurite outgrowth is dependent on a4 integrins in vitro. Upper Panel: Sympathetic neurons were incubated 24 h on laminin 2/4 (LN) (A, D, G, J), rsVCAM-1 (B, E, H, K), or fibronectin (FN) (C, F, I, L) in the presence of the a4-antagonist (a4a) (D – F), RGD peptide (G – I), or control RAD peptide (J – L), and then stained with Calcein-AM and photographed. Scale bar = 50 Am. Lower Panel: Average neurite length per cell for each condition was quantified. Error bars represent the SEM. **, p < .01 vs. control.

effect on initiation of neurite outgrowth on any of the substrates tested. This indicated that RGD binding integrins were not required for initial outgrowth. To determine the importance of a4 integrins in maintenance of neurites, neurons were cultured for 24 h, then the antagonist was added and cells were incubated for an additional 6 h. Fig. 2 shows that the antagonist brought about an almost complete retraction of neurites on VCAM-1 and FN, but did not affect maintenance of neurites on LN. These data indicated that a4 integrin function was required for maintenance of neurites on VCAM-1 and FN.

To assay whether RGD binding integrins were involved in maintenance of neurites, RGD peptide inhibitors and RAD control peptides were added to cultures. Surprisingly, RGD brought about retraction of neurites on FN, indicating that at least two integrins were required for maintenance. These results suggested that whereas a single integrin was required for initial outgrowth on FN, at least two integrins were required for maintenance, with inhibition of one or the other bringing about retraction of neurites. The a4 subunit can pair with either the h1 or h7 subunit. To address the beta integrin dependence of neurite maintenance, a blocking anti-h1 antibody was added to neurons

62

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

Fig. 2. Maintenance of established sympathetic neurites is both a4- and RGD-integrin dependent. Upper Panel: Sympathetic neurons were incubated 24 h on LN (A, D, G, J), rsVCAM-1 (B, E, H, K), or FN (C, F, I, L). Cultures were then treated with the a4-antagonist (a4a) (D – F), RGD peptide (G – I), or control RAD peptide (J – L) and incubated for an additional 6 h before cells were stained with Calcein-AM and photographed. Scale bar = 50 Am. Lower Panel: Average neurite length per cell for each condition was quantified. Error bars represent the SEM. **, p < .01 vs. control.

with established neurites. After 6 h, almost all neurites on LN2, VCAM-1 and FN had retracted, which indicated a requirement for h1 in maintenance (Fig. 3). For all the neurite measurements made in vitro, the percent of cells with neurites was also quantified, and a similar trend was observed in each case (data not shown). 3.2. a4 integrin involvement in neurite maintenance in vivo in adult Long Evans rats The in vitro experiments described above suggested a role for the a4h1 integrin in axon homeostasis. To test

this hypothesis in vivo, the a4 antagonist was injected intrathoracically into adult Long Evans rats. After six days of a4 antagonist treatment and hearts were analyzed for sympathetic fiber density. The apex region was chosen for analysis because it is the last part of the myocardium to be innervated and has a lower density of fibers. The treated animals showed 60% reduction in the number of fibers detected in sections of the apex of the heart, compared to controls that received vehicle alone (Fig. 4). All of the animals receiving the antagonist showed fiber densities lower than controls. These data showed that, like the in vitro model, a4 integrins were important for

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

Fig. 3. h1-integrins are required for maintenance of sympathetic neurites in vitro. Sympathetic neurons were incubated for 24 h on LN, rsVCAM-1, or FN and then treated with the anti-h1 blocking antibody HA-2/5 for an additional 6 h before cultures were stained with Calcein-AM and photographed and neurite lengths were quantified. Black bars, control; white bars, HA-2/5. Error bars represent the SEM. **, p < .01 vs. control.

maintaining at least some sympathetic nerve endings in the myocardium. This result suggested that inhibition of a4 integrins might be useful in controlling excessive cardiac sympathetic innervation associated with disease. Thus, two disease models were investigated to assess expression and function of a4 integrins. First, sprouting sympathetic axons that arose after MI were examined for a4 expression by immunohistochemistry. MIs were induced by permanent

63

ligature of the left anterior ascending coronary artery, and after three weeks of recovery, sections of myocardium were stained for a4 and TH. As in previous studies, large numbers of TH+ fibers were observed at the boundary of the scar tissue, and some sprouted axons displayed a4 immunoreactivity (Fig. 5A). However, quantification showed that very few sympathetic sprouts were immunopositive for a4 integrin (Fig. 5B). Only about 2% of the sprouts showed detectable a4 expression. Consistent with this result, treatment of the MI rats with the a4 antagonist did not decrease the degree of sprouting (data not shown). This result suggested that sprouting sympathetic axons used different integrins (or a non-integrin receptor) to bring about axon extension. a4 expression was also examined in the spontaneously hypertensive rat strain (SHR) and in the related normal Wistar line. To compare cardiac sympathetic innervation in SHR, Wistar, and LE rats, heart sections from the apex were systematically stained for TH and fibers counted (Fig. 6A). At P1, all strains had a similar level of innervation, but as development proceeded, the SHR had significantly more sympathetic innervation than either Wistar or LE rats. a4 expression was assessed in the same three strains (Fig. 6B). Surprisingly, the LE rats had much higher numbers of a4positive fibers than both Wistar and SHR at P1 and P9. At P22, both Wistar and LE had more a4-positive fibers than SHR. These data suggested that a4 integrin expression was altered in these different strains of rat. Consistent with the paucity of a4 positive fibers in the SHR rat, intrathoracic

Fig. 4. a4 integrins are required for maintenance of cardiac sympathetic innervation in adult Long Evans rats. Adult rats were treated with the a4 antagonist (a4a) for 6 days and hearts were sectioned and stained for TH to detect sympathetic axons in the apex region. Fiber numbers for each individual control (C) and treated (a4a) animal are expressed as percent of the average control value. Average control and treated values are shown at the right. Error bars represent the SEM. **, p < .01 vs. control.

64

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

Fig. 5. Most sympathetic fibers that sprout after myocardial infarction lack integrin a4 expression. A) Three weeks after MI, sections from the infarcted region were stained for TH (green) and integrin a4 (red). Arrows indicate a4-positive, TH-positive axons; arrowheads indicate a4 negative, TH-positive axons. Asterisks label acellular autofluorescent granules within the scar. Bar = 20 Am. B) TH-positive sprouting axons at the boundary region of the scar were scored as positive or negative for a4. Numbers of TH-positive fibers (white bars) and TH-positive, a4-positive fibers (black bars) for each infarcted animal are shown, with averages shown at the right. Error bars represent the SEM.

injection of the a4 antagonist did not significantly decrease the maintenance of sympathetic fibers in adult rats (data not shown).

4. Discussion The integrin receptor a4h1 was shown previously to be important for the development of sympathetic innervation of the heart (Wingerd et al., 2002), and evidence suggests that a4h1 on sympathetic growth cones interacts with VCAM-1 in the myocardium to contribute to axon extension.

However, investigation of the developmental time course of a4 expression indicated that a4 integrins are also present on adult sympathetic axons in the heart (Wingerd et al., 2004). What is the function of the a4 integrins on adult axons? In this report, we present evidence that a4 integrins play a role in maintaining axons in adult heart. First, we carried out a series of in vitro experiments to show that a4h1 function was necessary to maintain neurites on FN and VCAM-1. Addition of a small molecule antagonist specific for a4 integrins caused a complete retraction of neurites after six hours. Function blocking antibodies to the integrin

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

Fig. 6. Spontaneously hypertensive rats have increased cardiac sympathetic innervation but low levels of integrin a4 expression. A) The average numbers of TH-positive fibers per field are shown for Wistar (black bars), SHR (gray bars), and Long Evans rats (white bars) at P1, P9, and P22. B) The percentage of TH-positive fibers expressing a4 integrins in Wistar (black bars), SHR (gray bars), and Long Evans rats (white bars) at P1, P9, and P22 are shown. Error bars represent the SEM. *, p < .05 vs. Wistar; **, p < .01 vs. Wistar.

a1 subunit have been shown to induce neurite retraction in vitro in previous studies (Turner et al., 1989), so it appears that integrins must be constantly bound to ligands to maintain elaborated neurites. The a4 integrin blocking antibody TA-2 also brought about retraction, but only if FN and VCAM-1 were mixed with a low concentration of chondroitin sulfate, an anti-adhesive ligand (data not shown). It is likely that a4a was a more effective inhibitor of a4 integrins, possibly due to better access to the receptor, since steric hindrance of antibody binding may be a factor after receptors have bound ligand in an established neurite. The action of integrin perturbing agents was similar to the effect of anti-adhesive molecules such as ephrins (Himanen et al., 2004). Ephrins function by binding to EPH receptors, which, via a tyrosine kinase activity, are

65

thought to disrupt normal integrin signaling pathways and bring about depolymerization of actin (Zou et al., 1999; Kullander and Klien, 2002). Any agent that disrupts the signaling pathway or binding activity of integrins might be expected to cause retraction of neurites. Interestingly, results showed that different integrins are engaged in initiation of neurites versus maintenance of neurites on FN. Whereas initiation of outgrowth was dependent solely on a4h1, maintenance required both a4 and an RGD dependent integrin. The identity of the RGD binding integrin is not known. SCG neurons express a number of integrins, including RGD dependent FN receptors a3h1, a5h1 (DeFreitas et al., 1995; Wingerd, unpublished). The processes that have been extended may hang on to the FN molecule at two attachment sites — one at the RGD motif, and another, possibly the LDV sequence in the CS-1 region, or another of the known motifs recognized by a4h1 (Mould and Humphries, 1991). There is precedent for such a mechanism: migration of fibroblasts on FN also requires cooperation between multiple integrins that bind to different sites along the molecule (Clark et al., 2003). Results presented here also indicate that a4 integrins mediate interactions that maintain cardiac sympathetic fibers, since the antagonist caused a decrease in sympathetic fibers in the apex region of the heart. The decrease in fibers was unlikely to be caused by cell death, since the antagonist does not give rise to apoptosis of neonatal SCG neurons (Wingerd et al., 2002). Nor is the decrease likely to be due to an effect on myocytes, since they do not express a4 integrins. Based on the in vitro results, the most likely explanation is that inhibition of a4 integrins resulted in a retraction of axon terminals. It will be interesting to determine if longer treatment with higher doses of antagonist have an even greater effect. In the adult, neural interactions with target tissue are dynamic and can increase or decrease, depending on the circumstances. For example, time lapse studies of single neurons in sympathetic ganglia of mice revealed dynamic changes in axon arbors of spinal cord neurons that made synapses on cell bodies of peripheral sympathetic axons (Purves and Lichtman, 1987). Modulation of synaptic strength in the nervous system, which may occur in part by sprouting or retraction of axons, is thought to be the basis of learning and memory. A number of recent reports have implicated integrin receptors in synapse formation and synapse regulation (Grotewiel et al., 1998; Rohrbough et al., 2000; Clegg et al., 2003; Gall and Lynch, 2004). For example, inhibitors of integrin receptors can block LTP in hippocampal slices, and integrins, particularly a5h1, are thought to be involved in the stabilization of LTP by increasing synaptic adhesion events. Thus, inhibition of integrins may interfere with the maintenance of neural contact with the target tissue. While the adhesive relationships between sympathetic axon terminals and heart tissue have not been examined thoroughly, our results suggest that a4 integrins may be involved in maintaining stable

66

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

connections between neuron and target, possibly by binding to FN or VCAM-1, which are known to be expressed in the myocardium (Casscells et al., 1990; Wingerd et al., 2002). Synaptic structures have been identified where sympathetic axons terminate on smooth muscle cells of the vasculature (Luff, 1996). Integrin and integrin ligand distribution in these structures has not been examined thoroughly. Another possible function for a4 integrins expressed on adult sympathetic fibers might be to act as a reserve to allow regeneration of axons after disease or injury in the heart. Fibronectin is known to accumulate around vessels post MI, and VCAM-1 is known to be found on the basolateral surface of blood vessels (Casscells et al., 1990; Wingerd et al., 2002). Both molecules support outgrowth from developing sympathetic neurons and perhaps may play a role in the reinnervation and hyperinnervation of the myocardium (Wingerd et al., 2002). Surprisingly, we found that a4 integrin was not abundantly expressed on sympathetic axons that sprout after MI, and the antagonist did not block the sprouting. This result is unexpected given the fact that at postnatal day 1 (P1), 80% of sympathetic fibers are positive for a4, with 30% maintaining expression at P22 (Wingerd et al., 2004). This result is interesting because it suggests that regenerating axons use a set of integrins (or other receptors) distinct from those used in early development. After MI, extensive remodeling of the cardiac ECM occurs. Increased expression of matrix metalloproteinases brings about degradation of ECM, and collagen is produced to form a scar (Lindsey et al., 2003). Tenascin and fibronectin are expressed transiently in and around the scar tissue (Willems et al., 1996). The changing composition of the ECM may in turn induce changes in integrin expression or localization in both myocytes and sympathetic axons. Nawata et al. (1999) reported induction of a1 and a5 integrins in myocytes post MI. More recently, it has been shown that a7, h1 and h3 integrin subunit expression in myocytes increases after MI or ischemia (Simkhovich et al., 2003; Sun et al., 2003). Integrins are modulated by ECM in neurons as well. Regenerating sensory axons have been shown to up-regulate a6, a7 and h1 (Ekstrom et al., 2003; Wallquist et al., 2004), and other studies have shown that purified ECM components induce changes in the integrin repertoire of cultured neurons (Condic and Letourneau, 1997). The mechanism of outgrowth during sympathetic sprouting in the heart will require further investigation. Hyperinnervation of the heart also occurs in hypertensive rats (Kondo et al., 1995; Tabei et al., 1995), a result we confirmed here. These hyperinnervating axons are similar to those that occur after MI in that they mostly lacked a4 staining. In fact, a lower a4 expression trend in both the SHR and the Wistar parental strain of rats was observed when compared to the LE rats. This may be due to a strain difference in integrin expression, and it is interesting to speculate that the Wistar and SHR rats may have a proclivity for hyperinnervation due to a different complement of integrins on their sympathetic fibers. Further studies are

underway to compare the outgrowth properties of SHR sympathetic neurons to see if they have intrinsically different abilities to extend axons.

References Altevogt, P., Hubbe, M., Ruppert, M., Lohr, J., von Hoegen, P., Sammar, M., Andrew, D.P., McEvoy, L., Humphries, M.J., Butcher, E.C., 1995. The alpha 4 integrin chain is a ligand for alpha 4 beta 7 and alpha 4 beta 1. J. Exp. Med. 182, 345 – 355. Arroyo, A.G., Yang, J.T., Rayburn, H., Hynes, R.O., 1996. Differential requirements for alpha4 integrins during fetal and adult hematopoiesis. Cell 85, 997 – 1008. Bayless, K.J., Meininger, G.A., Scholtz, J.M., Davis, G.E., 1998. Osteopontin is a ligand for the alpha4beta1 integrin. J. Cell Sci. 111 (Pt. 9), 1165 – 1174. Brum, P.C., Kosek, J., Patterson, A., Bernstein, D., Kobilka, B., 2002. Abnormal cardiac function associated with sympathetic nervous system hyperactivity in mice. Am. J. Physiol, Heart Circ. Physiol. 283, H1838 – H1845. Burns, F.R., von Kannen, S., Guy, L., Raper, J.A., Kamholz, J., Chang, S., 1991. DM-GRASP, a novel immunoglobulin superfamily axonal surface protein that supports neurite extension. Neuron 7, 209 – 220. Cao, J.M., Chen, L.S., KenKnight, B.H., Ohara, T., Lee, M.H., Tsai, J., Lai, W.W., Karagueuzian, H.S., Wolf, P.L., Fishbein, M.C., Chen, P.S., 2000. Nerve sprouting and sudden cardiac death. Circ. Res. 86, 816 – 821. Casscells, W., Kimura, H., Sanchez, J.A., Yu, Z.X., Ferrans, V.J., 1990. Immunohistochemical study of fibronectin in experimental myocardial infarction. Am. J. Pathol. 137, 801 – 810. Chang, C.M., Wu, T.J., Zhou, S., Doshi, R.N., Lee, M.H., Ohara, T., Fishbein, M.C., Karagueuzian, H.S., Chen, P.S., Chen, L.S., 2001. Nerve sprouting and sympathetic hyperinnervation in a canine model of atrial fibrillation produced by prolonged right atrial pacing. Circulation 103, 22 – 25. Chen, P.S., Chen, L.S., Cao, J.M., Sharifi, B., Karagueuzian, H.S., Fishbein, M.C., 2001. Sympathetic nerve sprouting, electrical remodeling and the mechanisms of sudden cardiac death. Cardiovasc. Res. 50, 409 – 416. Choi, E.S., Rettig, W.J., Wayner, E.A., Srour, M.L., Clegg, D.O., 1994. Functional identification of integrin laminin receptors that mediate process outgrowth by human SY5Y neuroblastoma cells. J. Neurosci. Res. 37, 475 – 488. Clark, R.A., An, J.Q., Greiling, D., Khan, A., Schwarzbauer, J.E., 2003. Fibroblast migration on fibronectin requires three distinct functional domains. J. Invest. Dermatol. 121, 695 – 705. Clegg, D.O., Wingerd, K.L., Hikita, S.T., Tolhurst, E.C., 2003. Integrins in the development, function and dysfunction of the nervous system. Front. Biosci. 8, d723 – d750. Condic, M.L., Letourneau, P.C., 1997. Ligand-induced changes in integrin expression regulate neuronal adhesion and neurite outgrowth. Nature 389, 852 – 856. Culley, B., Murphy, J., Babaie, J., Nguyen, D., Pagel, A., Rousselle, P., Clegg, D.O., 2001. Laminin-5 promotes neurite outgrowth from central and peripheral chick embryonic neurons. Neurosci. Lett. 301, 83 – 86. DeFreitas, M.F., Yoshida, C.K., Frazier, W.A., Mendrick, D.L., Kypta, R.M., Reichardt, L.F., 1995. Identification of integrin alpha 3 beta 1 as a neuronal thrombospondin receptor mediating neurite outgrowth. Neuron 15, 333 – 343. Dibner, M.D., Mytilineou, C., Black, I.B., 1977. Target organ regulation of sympathetic neuron development. Brain Res. 123, 301 – 310. Doggrell, S.A., 2001. Moxonidine: some controversy. Expert Opin. Pharmacother. 2, 337 – 350. Du, X.J., Cox, H.S., Dart, A.M., Esler, M.D., 1999. Sympathetic activation triggers ventricular arrhythmias in rat heart with chronic infarction and failure. Cardiovasc. Res. 43, 919 – 929.

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68 Ekstrom, P.A., Mayer, U., Panjwani, A., Pountney, D., Pizzey, J., Tonge, D.A., 2003. Involvement of alpha7beta1 integrin in the conditioninglesion effect on sensory axon regeneration. Mol. Cell. Neurosci. 22, 383 – 395. Gall, C.M., Lynch, G., 2004. Integrins, synaptic plasticity and epileptogenesis. Adv. Exp. Med. Biol. 548, 12 – 33. Grotewiel, M.S., Beck, C.D., Wu, K.H., Zhu, X.R., Davis, R.L., 1998. Integrin-mediated short-term memory in Drosophila. Nature 391, 455 – 460. Guan, J.L., Hynes, R.O., 1990. Lymphoid cells recognize an alternatively spliced segment of fibronectin via the integrin receptor alpha 4 beta 1. Cell 60, 53 – 61. Hikita, S.T., Cann, G.M., Wingerd, K.L., Mullick, L.H., Wayne, W.C., Webb, S.W., Clegg, D.O., 2003. Integrin alpha4beta1 (VLA-4) expression and activity in retinal and peripheral neurons. Mol. Cell. Neurosci. 23, 427 – 439. Himanen, J.P., Chumley, M.J., Lackmann, M., Li, C., Barton, W.A., Jeffrey, P.D., Vearing, C., Geleick, D., Feldheim, D.A., Boyd, A.W., Henkemeyer, M., Nikolov, D.B., 2004. Repelling class discrimination: ephrin-A5 binds to and activates EphB2 receptor signaling. Nat. Neurosci. 7, 501 – 509. Isobe, T., Hisaoka, T., Shimizu, A., Okuno, M., Aimoto, S., Takada, Y., Saito, Y., Takagi, J., 1997. Propolypeptide of von Willebrand factor is a novel ligand for very late antigen-4 integrin. J. Biol. Chem. 272, 8447 – 8453. Isobe, T., Takahashi, H., Ueki, S., Takagi, J., Saito, Y., 1999. Activityindependent cell adhesion to tissue-type transglutaminase is mediated by alpha4beta1 integrin. Eur. J. Cell Biol. 78, 876 – 883. Jackson, D.Y., 2002. Alpha 4 integrin antagonists. Curr. Pharm. Des. 8, 1229 – 1253. Kil, S.H., Krull, C.E., Cann, G., Clegg, D., Bronner-Fraser, M., 1998. The alpha4 subunit of integrin is important for neural crest cell migration. Dev. Biol. 202, 29 – 42. Kitajiri, M., Kubo, N., Tomoda, K., Maeda, N., Yamashita, T., Kumazawa, T., 1993. Sympathetic nerve disturbance and fine structure of adrenergic nerve terminals in nasal mucosa after experimentally-induced nasal hypersensitivity. Acta Oto-laryngol., Suppl. 506, 47 – 50. Kondo, M., Terada, M., Fujiwara, T., Arita, N., Yano, A., Tabei, R., 1995. Noradrenergic hyperinnervation in the heart of stroke-prone spontaneously hypertensive rats. Clin. Exp. Pharmacol. Physiol. Suppl. 22, S75 – S76. Kullander, K., Klein, R., 2002. Mechanisms and functions of Eph and ephrin signalling. Nat. Rev., Mol. Cell Biol. 3, 475 – 486. Landis, S.C., 1976. Rat sympathetic neurons and cardiac myocytes developing in microcultures: correlation of the fine structure of endings with neurotransmitter function in single neurons. Proc. Natl. Acad. Sci. U. S. A. 73, 4220 – 4224. Lindsey, M.L., Mann, D.L., Entman, M.L., Spinale, F.G., 2003. Extracellular matrix remodeling following myocardial injury. Ann. Med. 35, 316 – 326. Lipp, J.A., Rudolph, A.M., 1972. Sympathetic nerve development in the rat and guinea-pig heart. Biol. Neonate 21, 76 – 82. Lobb, R.R., Hemler, M.E., 1994. The pathophysiologic role of alpha 4 integrins in vivo. J. Clin. Invest. 94, 1722 – 1728. Luff, S.E., 1996. Ultrastructure of sympathetic axons and their structural relationship with vascular smooth muscle. Anat. Embryol. (Berl.) 193, 515 – 531. Mould, A.P., Humphries, M.J., 1991. Identification of a novel recognition sequence for the integrin alpha 4 beta 1 in the COOH-terminal heparinbinding domain of fibronectin. EMBO J. 10, 4089 – 4095. Nawata, J., Ohno, I., Isoyama, S., Suzuki, J., Miura, S., Ikeda, J., Shirato, K., 1999. Differential expression of alpha 1, alpha 3 and alpha 5 integrin subunits in acute and chronic stages of myocardial infarction in rats. Cardiovasc. Res. 43, 371 – 381. Nori, S.L., Gaudino, M., Alessandrini, F., Bronzetti, E., Santarelli, P., 1995. Immunohistochemical evidence for sympathetic denervation and rein-

67

nervation after necrotic injury in rat myocardium. Cell. Mol. Biol. (Noisy-le-grand) 41, 799 – 807. Osborn, L., Hession, C., Tizard, R., Vassallo, C., Luhowskyj, S., ChiRosso, G., Lobb, R., 1989. Direct expression cloning of vascular cell adhesion molecule 1, a cytokine-induced endothelial protein that binds to lymphocytes. Cell 59, 1203 – 1211. Palatini, P., Julius, S., 1999. Relevance of heart rate as a risk factor in hypertension. Curr. Hypertens. Rep. 1, 219 – 224. Podrid, P.J., Fuchs, T., Candinas, R., 1990. Role of the sympathetic nervous system in the genesis of ventricular arrhythmia. Circulation 82, I103 – I113. Pugsley, M.K., Walker, M.J., Yong, S.L., 1999. Are the arrhythmias due to myocardial ischaemia and infarction dependent upon the sympathetic system? Cardiovasc. Res. 43, 830 – 831. Purves, D., Lichtman, J.W., 1987. Synaptic sites on reinnervated nerve cells visualized at two different times in living mice. J. Neurosci. 7, 1492 – 1497. Rohrbough, J., Grotewiel, M.S., Davis, R.L., Broadie, K., 2000. Integrinmediated regulation of synaptic morphology, transmission, and plasticity. J. Neurosci. 20, 6868 – 6878. Rumantir, M.S., Jennings, G.L., Lambert, G.W., Kaye, D.M., Seals, D.R., Esler, M.D., 2000. The Fadrenaline hypothesis_ of hypertension revisited: evidence for adrenaline release from the heart of patients with essential hypertension. J. Hypertens. 18, 717 – 723. Schlaich, M.P., Kaye, D.M., Lambert, E., Sommerville, M., Socratous, F., Esler, M.D., 2003. Relation between cardiac sympathetic activity and hypertensive left ventricular hypertrophy. Circulation 108, 560 – 565. Schwartz, P.J., Stramba-Badiale, M., Segantini, A., Austoni, P., Bosi, G., Giorgetti, R., Grancini, F., Marni, E.D., Perticone, F., Rosti, D., Salice, P., 1998. Prolongation of the QT interval and the sudden infant death syndrome. N. Engl. J. Med. 338, 1709 – 1714. Sheppard, A.M., Onken, M.D., Rosen, G.D., Noakes, P.G., Dean, D.C., 1994. Expanding roles for alpha 4 integrin and its ligands in development. Cell Adhes. Commun. 2, 27 – 43. Siebers, M.C., Ter Brugge, P.J., Walboomers, X.F., Jansen, J.A., 2005. Integrins as linker proteins between osteoblasts and bone replacing materials. A critical review. Biomaterials 26, 137 – 146. Simkhovich, B.Z., Marjoram, P., Poizat, C., Kedes, L., Kloner, R.A., 2003. Age-related changes of cardiac gene expression following myocardial ischemia/reperfusion. Arch. Biochem. Biophys. 420, 268 – 278. Spring, F.A., Parsons, S.F., Ortlepp, S., Olsson, M.L., Sessions, R., Brady, R.L., Anstee, D.J., 2001. Intercellular adhesion molecule-4 binds alpha(4)beta(1) and alpha(V)-family integrins through novel integrinbinding mechanisms. Blood 98, 458 – 466. Stanton, M.S., Tuli, M.M., Radtke, N.L., Heger, J.J., Miles, W.M., Mock, B.H., Burt, R.W., Wellman, H.N., Zipes, D.P., 1989. Regional sympathetic denervation after myocardial infarction in humans detected noninvasively using I-123-metaiodobenzylguanidine. J. Am. Coll. Cardiol. 14, 1519 – 1526. Sun, M., Opavsky, M.A., Stewart, D.J., Rabinovitch, M., Dawood, F., Wen, W.H., Liu, P.P., 2003. Temporal response and localization of integrins beta1 and beta3 in the heart after myocardial infarction: regulation by cytokines. Circulation 107, 1046 – 1052. Tabei, R., Kondo, M., Terada, M., Miyazaki, T., Watanabe, Y., Shimizu, D., Yamamoto, I., 1995. Noradrenergic hyperinnervation in the caval vein of stroke-prone spontaneously hypertensive rats. Clin. Exp. Pharmacol. Physiol. Suppl. 22, S73 – S74. Turner, D.C., Flier, L.A., Carbonetto, S., 1989. Identification of a cellsurface protein involved in PC12 cell-substratum adhesion and neurite outgrowth on laminin and collagen. J. Neurosci. 9, 3287 – 3296. Vogelezang, M.G., Liu, Z., Relvas, J.B., Raivich, G., Scherer, S.S., ffrenchConstant, C., 2001. Alpha4 integrin is expressed during peripheral nerve regeneration and enhances neurite outgrowth. J. Neurosci. 21, 6732 – 6744. Vracko, R., Thorning, D., Frederickson, R.G., 1990. Fate of nerve fibers in necrotic, healing, and healed rat myocardium. Lab. Invest. 63, 490 – 501.

68

K.L. Wingerd et al. / Autonomic Neuroscience: Basic and Clinical 122 (2005) 58 – 68

Vracko, R., Thorning, D., Frederickson, R.G., 1991. Nerve fibers in human myocardial scars. Hum. Pathol. 22, 138 – 146. Wallquist, W., Zelano, J., Plantman, S., Kaufman, S.J., Cullheim, S., Hammarberg, H., 2004. Dorsal root ganglion neurons up-regulate the expression of laminin-associated integrins after peripheral but not central axotomy. J. Comp. Neurol. 480, 162 – 169. Willems, I.E., Arends, J.W., Daemen, M.J., 1996. Tenascin and fibronectin expression in healing human myocardial scars. J. Pathol. 179, 321 – 325. Wingerd, K.L., Goodman, N.L., Tresser, J.W., Smail, M.M., Leu, S.T., Rohan, S.J., Pring, J.L., Jackson, D.Y., 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. Wingerd, K.L., Goodman, N.L., Leu, S.T., Clegg, D.O., 2004. Expression and function of integrin alpha4beta1 and vascular cell adhesion

molecule-1 (VCAM-1) during sympathetic innervation of the heart. Dev. Dyn. 231, 359 – 369. Yabkowitz, R., Dixit, V.M., Guo, N., Roberts, D.D., Shimizu, Y., 1993. Activated T-cell adhesion to thrombospondin is mediated by the alpha 4 beta 1 (VLA-4) and alpha 5 beta 1 (VLA-5) integrins. J. Immunol. 151, 149 – 158. Zhou, S., Chen, L.S., Miyauchi, Y., Miyauchi, M., Kar, S., Kangavari, S., Fishbein, M.C., Sharifi, B., Chen, P.S., 2004. Mechanisms of cardiac nerve sprouting after myocardial infarction in dogs. Circ. Res. 95, 76 – 83. Zipes, D.P., 1990. Influence of myocardial ischemia and infarction on autonomic innervation of heart. Circulation 82, 1095 – 1105. Zou, J.X., Wang, B., Kalo, M.S., Zisch, A.H., Pasquale, E.B., Ruoslahti, E., 1999. An Eph receptor regulates integrin activity through R-Ras. Proc. Natl. Acad. Sci. U. S. A. 96, 13813 – 13818.