Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

Neural regulation of end organ function is dependent not only on impulse activity of central origin but also on interactions with adjacent peripheral nerve terminals. Transmitter release is known to be modulated by substances released from coprojecting axonal populations within the target organ. For example, acetylcholine from parasympathetic nerves can diminish norepinephrine release from sympathetic nerves within the heart (Vanhoutte and Levy, 1980), bladder (de Groat et al., 1999), and tarsal smooth muscle of the orbit (Beauregard and Smith, 1994). Similarly, sympathetic neurotransmitters can prejunctionally inhibit parasympathetically mediated vasodilation in the nasal mucosa (Lacroix et al., 1994). Therefore, transmitter release from heterologous nerves coprojecting to a common target represents an important short-term mechanism for modifying neural control of target activity.

Less is known about the role of heterologous nerves in long-term regulation of neurotransmission. It is known that innervation density is affected by ablation of coprojecting axons. Thus, after sympathectomy numbers of both sensory axons and parasympathetic terminals increase in some targets (Terenghi et al., 1986; Smith and Marzban, 1998), and in at least one instance, this leads to altered neurotransmission. Hence, after long-term sympathectomy of the periorbital tarsal smooth muscle, parasympathetic nerves that normally inhibit sympathetic neurotransmission but do not directly affect muscle tone become excitatory to the smooth muscle cells (Smith and Beauregard, 1993; Krizsan-Agbas et al., 1998). Therefore, long-term disruption of sympathetic nerves can dramatically influence properties of parasympathetic neurotransmission in this nonvascular smooth muscle target.

It is unclear whether sympathetic denervation elicits similar changes in targets other than the tarsal muscle, and whether transmitters other than acetylcholine are affected. To address these questions, we examined the effects of long-term sympathectomy on parasympathetic neuroeffector regulation of vascular smooth muscle function in the eye.

Blood flow to ocular tissues in the rat derives from the choroidal and ciliary vascular plexuses. The choroid consists of a network of arterial and venous vessels that lies just inside the posterior sclera and is responsible for nourishing the photoreceptor and neural cell layers of the retina (Foulds, 1990). Tissues of the anterior eye are perfused by limbal vessels, which extend into the ciliary processes and provide perfusion necessary for aqueous humor formation (Morrison et al., 1995). Regional differences exist in the functional representation of autonomic nerves to these plexuses, with sympathetic noradrenergic vasoconstrictive effects predominating within the posterior choroidal vessels, whereas both sympathetic vasoconstrictor and parasympathetic nitrergic vasodilator nerves regulate anterior blood flow (Koss and Gherezghiher, 1993; Steinle et al., 2000). Blood from throughout the uvea drains into the vortex veins present along the nasolateral and temperolateral aspects of the globe just posterior to the equator. In previous studies, we and others have shown that vortex venous blood flow can be accurately measured using laser Doppler flowmetry, and that recordings of flow through the nasolateral vortex veins reliably reflects the summation of sympathetic and parasympathetic effects on ocular tissues (Best et al., 1972; Bill, 1985; Steinle et al., 2000). Therefore, laser Doppler flowmetry of vortex venous flux represents a convenient means to assess autonomic regulation of blood flow throughout the eye. In this study, we determined the effects of short- and long-term sympathetic denervation on parasympathetically induced nitrergic ocular vasodilation by measuring changes in vortex venous blood flow during parasympathetic stimulation.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Superior Cervical Ganglionectomy. Female Sprague-Dawley rats (Harlan Sprague-Dawley, Inc., Indianapolis, IN) weighing from 180 to 200 g at 60 to 70 days of age were anesthetized by ketamine hydrochloride (27.5 mg/kg, i.p.; Sanofi Winthrop, New York, NY), xylazine hydrochloride (2.5 mg/kg, Rompun; Miles, Shawnee Mission, KS) and atropine sulfate (0.24 mg/kg; Vedco, St. Joseph, MO). A ventral midline incision was made in the neck and the right superior cervical ganglion was removed aseptically (Smith and Beauregard, 1993; Krizsan-Agbas et al., 1998). This produces complete and permanent loss of orbital sympathetic innervation in adult rats (Smith, 1986; Smith and Fan, 1996). All rats recovered without signs of distress. Experiments were conducted in rats with intact innervation (n = 15) and at 2 days (n = 11) or 6 weeks (n = 25) after superior cervical ganglionectomy. All surgical procedures and subsequent experimental manipulations were approved by the Institutional Animal Care and Use Committee of the University of Kansas Medical Center, and conform to local and national statutes.

Parasympathetic Nerve Stimulation. Intact and sympathectomized rats were anesthetized with urethane (1 g/kg, i.p.; Sigma), and surgical anesthesia confirmed by the absence of deep tendon reflexes. Rectal temperature was maintained at 36°C. A femoral vein and artery were cannulated for drug administration and blood pressure monitoring, respectively. The facial nerve was exposed and transected proximal to the extraorbital lacrimal gland to eliminate orbital motor innervation. The animal was placed in a stereotaxic frame and a scalp incision made over the sagittal suture. A 4-mm diameter craniotomy was performed and a semimicro bipolar concentric electrode (100-µm contact diameter; Rhodes Medical Instruments, Woodland Hills, CA) was stereotaxically positioned within the superior salivatory nucleus (SSN) (Paxinos and Watson, 1986; Spencer et al., 1990) at coordinates (9.5 mm posterior, 9.5 mm ventral, 2.5 mm lateral to bregma) previously shown to elicit selective and complete activation of parasympathetic innervation to the rat orbit (Beauregard and Smith, 1994; Krizsan-Agbas et al., 1998; Steinle et al., 2000). This nucleus is the source of preganglionic innervation to the parasympathetic pterygopalatine ganglion (Spencer et al., 1990). Parasympathetic preganglionic axons originating in the SSN were stimulated electrically at 20 Hz, 0.5-ms pulse duration, 3 V for 40 to 60 s (Grass SD9 stimulator; Grass Instrument Co., Quincy, MA), which has been shown previously to elicit maximal activation of orbital parasympathetic innervation (Beauregard and Smith, 1994; Krizsan-Agbas et al., 1998). Parasympathetic activation of orbital structures during SSN electrical stimulation was confirmed by porphyrin discharge from the Harderian gland (Tashiro et al., 1940), and electrode placement was confirmed by histological examination at the end of the experiment, as described previously (Steinle et al., 2000).

Laser Doppler Flowmetry. Ocular blood flow through the vortex veins was measured using laser Doppler flowmetry (MP3 flow probe, floLAB; Moor Instruments, Devon, England). This method measures a Doppler shift in the laser light (flux), which is determined by erythrocyte number and velocity, and is proportional to the total blood flow within a given volume of tissue (Riva et al., 1994). This method has been shown to provide flux values that correlate linearly with blood flow in vessels of equal or greater size than that of the rat vortex vein (Kajiya et al., 1989). Flux values from the floLab were acquired and displayed as arbitrary units using Polyview software (Astro-Med; Grass Instrument Co.) on a Pentium computer. The pupil was dilated by a 0.01% epinephrine topical solution, a small region of the cornea was excised using micro-Vannas scissors, and the probe was positioned using a micromanipulator in the posterior nasolateral region of the globe corresponding to the vortex veins (Steinle et al., 2000). Mineral oil was applied to prevent drying of the eye. Rats were euthanized by an overdose of urethane (3 g/kg, i.v.) at the end of each experiment. Background flux was then measured and subtracted from all experimental determinations, and the recording site was inspected to confirm correct probe placement and the absence of intraocular bleeding. Because basal flux was increased in the chronically sympathectomized rats, responses to parasympathetic stimulation are presented as percentage change in flux.

Pharmacological Studies. Pharmacological agents were dissolved in distilled water and administered through a femoral venous cannula. Neurotransmitters mediating the effects of parasympathetic stimulation in control and sympathectomized rats were evaluated by administering the nonselective muscarinic antagonist atropine methyl nitrate (0.5 mg/kg i.v.; Sigma) and the selective neuronal nitric-oxide synthetase inhibitor 1-(2-trifluoromethylphenyl) imidazole (TRIM; Research Biochemicals International, Natick, MA) in graded doses (40-60 mg/kg total i.v. dose administered in 8- or 16-mg/kg increments alone or with atropine until maximal blockade was achieved). The dosages of antagonists used in this study have been shown to be effective in blocking ocular parasympathetic neurotransmission (Beauregard and Smith, 1994; Steinle et al., 2000). To further elucidate the role of muscarinic receptors, the nonselective agonist bethanechol (Research Biochemicals International) was administered at a dose (0.1 mg/kg i.v.) that has been shown to be maximally effective in modulating ocular parasympathetic neurotransmission (Beauregard and Smith, 1994; Krizsan-Agbas et al., 1998). Muscarinic receptor subtypes involved in parasympathetic neurotransmission after sympathectomy were assessed by sequential administration of the following selective receptor antagonists: tropicamide (0.2 mg/kg; Research Biochemicals International), a selective M₄ receptor antagonist; pirenzepine dihydrochloride (0.2 mg/kg; Research Biochemicals International), a selective M₁ receptor antagonist; 4-diphenylacetoxy-N-methylpiperdine (4-DAMP) methiodide (0.2 mg/kg; Research Biochemicals International), a selective M₃ receptor antagonist; and methoctramine tetrahydrochloride (0.2 mg/kg; Research Biochemicals International), a selective M₂ receptor antagonist. The dosages for these agents have been shown to be selective for the intended receptors and supramaximal for blocking responses mediated by these receptors (Blanquet and Gonella, 1992; Blanquet et al., 1994).

Statistics. Responses to stimulation and drugs were compared statistically by one-way ANOVA, two-way ANOVA, or one-way repeated measures ANOVA, with post hoc analysis using Student Newman-Keuls. All values are presented as mean ± S.E., with P < .05 taken as statistically significant.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Effect of Acute and Chronic Sympathectomy on Ocular Blood Flow. SSN stimulation in control rats elicited a 1.0- to 1.5-fold increase in vortex venous blood flux. Atropine methyl nitrate had no significant effect on choroidal blood flow in either unstimulated or stimulated conditions. TRIM did not alter vortex venous flux in unstimulated conditions but blocked the increase during stimulation (P < .001, n = 6, Fig. 1). Neither drug produced a significant change in mean arterial pressure (Steinle et al., 2000).

View larger version (18K): [in this window] [in a new window]   Fig. 1.   Changes in vortex venous flux during SSN stimulation (Control) alone and after the administration of atropine only (Atropine alone, 0.5 mg/kg), atropine followed by TRIM (Atropine + TRIM), and TRIM only (TRIM alone, 40-60 mg/kg) in control rats, and 2 days and 6 weeks after sympathectomy (SNX). Data are displayed as mean ± S.E. for six rats. *P < .05 versus prestimulation flux, **P < .05 versus Control, ***P < .005 versus control and Atropine alone; , not significant versus baseline flux.

Characterization of Muscarinic Transmission after Sympathectomy. To determine the site at which muscarinic receptors act to increase ocular flux at 6 weeks postsympathectomy, the nonselective muscarinic agonist, bethanechol, was administered i.v. Bethanechol (0.1 mg/kg) before SSN stimulation produced a transient decrease in mean arterial blood pressure that quickly returned to baseline, but had no effect on vortex venous flux in either the control (n = 5) or the 6-week (n = 5) ganglionectomized animals. In the rats with intact sympathetic innervation, SSN stimulation 100 s after bethanechol administration increased ocular flux to a degree that was comparable to that observed in the absence of bethanechol. However, at 6 weeks postsympathectomy, bethanechol administration increased vortex venous flux during SSN stimulation by approximately 30% (P = .041, Fig. 2).

View larger version (11K): [in this window] [in a new window]   Fig. 2.   Effect of bethanechol (Bethanechol, 0.1 mg/kg1) on vortex venous flux during SSN stimulation (Control) in animals with intact (Intact) innervation and 6 weeks after sympathectomy (6-week SNX). Data are expressed as mean ± S.E. for five rats. *P < .05 versus basal flux after 6-week sympathectomy.

View larger version (25K): [in this window] [in a new window]   Fig. 3.   Effect of SSN stimulation on vortex venous flux under control conditions (Control) and after the administration of tropicamide (Tropicamide, M4 antagonist, 0.2 mg/kg), pirenzepine (Pirenzepine, M1 antagonist, 0.2 mg/kg), 4-DAMP (M3 antagonist, 0.2 mg/kg), methoctramine (Methoctramine, M2 antagonist, 0.2 mg/kg), and atropine (Atropine, nonselective antagonist, 0.5 mg/kg). Data are expressed as mean ± S.E. for four rats. *P = .001 versus Control and tropicamide, **P < .05 versus pirenzepine, ***P < .001 versus Control and all other drugs.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Parasympathetic Nerves and Control of Choroidal Blood Flow. Parasympathetic innervation from the SSN and pterygopalatine ganglion plays a major role in regulating ocular blood flow. Stimulation of this pathway increases uveal blood flow substantially in rats, cats, and rabbits as demonstrated by both labeled microspheres and laser Doppler flowmetry (Bill, 1985; Nilsson et al., 1985; Steinle et al., 2000), providing evidence that parasympathetic nerves subserve a major vasodilatory role. There is general agreement that parasympathetic vasodilation is not cholinergic in nature, as atropine normally does not affect uveal blood flow (Bill, 1985; Nilsson et al., 1985), and there is strong evidence from studies using both nonselective and neuronally selective blockers of nitric-oxide synthetase that the vasodilatory transmitter is in fact nitric-oxide (Deussen et al., 1993; Jacot et al., 1998; Steinle et al., 2000). Observations from control preparations in this study confirm a strong nitrergically mediated vasodilation whereas cholinergic effects on ocular blood flow were absent.

Acute Sympathectomy. Observations from other systems indicate that sympathetic nerve degeneration can elicit rapid effects on parasympathetic nerves. In the tarsal smooth muscle of the eyelid with intact innervation, parasympathetic nerves act to inhibit sympathetic excitatory nerves prejunctionally, but do not exert detectable effects on the smooth muscle itself (Beauregard and Smith, 1994). However, within 2 days after superior cervical ganglionectomy, parasympathetic stimulation evokes smooth muscle contraction, apparently through enhanced acetylcholine release via excitatory prejunctional  receptors, which are activated by circulating adrenal medullary catecholamines (Krizsan-Agbas et al., 1998). In contrast, short-term sympathectomy in this study had no obvious effects on parasympathetic modulation of uveal blood flow despite a common origin from pterygopalatine ganglion neurons. In this regard, it is interesting to note that there appears to be a close functional relationship between sympathetic and parasympathetic nerves in the tarsal muscle (Beauregard and Smith, 1994), although we have been unable to demonstrate prejunctional interactions between sympathetic and parasympathetic nerves modulating uveal blood flow (Steinle et al., 2000). In any event, findings from these studies provide evidence that different populations of parasympathetic nerves do not respond uniformly to acute sympathectomy.

Long-Term Sympathectomy. In contrast to acute sympathectomy, chronic sympathetic denervation elicited significant changes in vortex venous flux and in parasympathetic neurotransmission properties. Flux was consistently elevated in the eyes of rats at 6 weeks after sympathectomy and, although it is not possible to directly infer altered blood flow simply on the basis of flux (Riva et al., 1994), microsphere measurements of blood flow and morphometric analysis of vessel numbers support the idea that uveal blood flow and vessel numbers are increased in chronically sympathectomized eyes (J. J. Steinle, J. D. Pierce, R. L. Clancy, and P. G. Smith, manuscript in preparation).

Prejunctional Facilitation of Parasympathetic Vasodilation Is Mediated by the M₃ and M₅ Receptor. There are five currently recognized muscarinic receptor subtypes, and four of these are amenable at this time to selective pharmacological blockade. Antagonism of the M₄ receptor by tropicamide or the M₂ receptor by methoctramine did not affect the vasodilatory response, thus excluding these receptors as candidates for cholinergic modulation of parasympathetic vasodilation after sympathectomy. The M₁ receptor antagonist pirenzepine, however, produced a significant increase in dilation. Although the M₁ receptor is excitatory at many sites, it does act as a prejunctional inhibitory autoreceptor in the guinea pig ileum (Lambrecht et al., 1999), and this study provides evidence that the M₁ receptor is a prejunctional inhibitory autoreceptor on parasympathetic choroidal vascular nerves in the chronically sympathectomized preparation.

View larger version (48K): [in this window] [in a new window]   Fig. 4.   Schematic diagram of changes in parasympathetic transmission after long-term sympathectomy. Under conditions of intact sympathetic innervation (left), nitric oxide generated by the enzyme nitric-oxide synthetase (NOS) mediates parasympathetic vasodilation; this is not modulated ostensibly by cholinergic muscarinic mechanisms. After long-term superior cervical ganglionectomy (SNX, right), sympathetic nerves degenerate. In the remaining parasympathetic nerves, NOS immunoreactivity is decreased. Prejunctional muscarinic receptors are now activated by acetylcholine (Ach) released from the parasympathetic nerve and are pharmacologically demonstrable. Although M1 receptors act to inhibit nitrergic transmission, the excitatory effects of M3 and presumptive M5 receptors predominate, thereby facilitating nitrergic transmission.