Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

It is generally claimed that the iris dilator, which controls the size of the pupil in concert with the sphincter muscle, is innervated by sympathetic nerves via postsynaptic ₁-adrenoceptors (Hoffman and Taylor, 2001). However, controversy exists concerning the pharmacological properties of these adrenoceptors. For example, in cats, both neural and pharmacological activation of the iris dilator muscle are refractory to antagonism by the prototypical, selective ₁-adrenoceptor antagonist prazosin and the nonselective -adrenoceptor antagonist phentolamine (Hey et al., 1988; Koss and Gherezghiher, 1988; Koss et al., 1988, 1990). In rabbits, pupillary dilation observed after topical application of ₁-adrenoceptor agonists is blocked by phentolamine but not by prazosin (Murray and Leopold, 1985).

Recently, pharmacological and molecular cloning studies have established structural and functional heterogeneity among ₁-adrenoceptors. ₁-Adrenoceptors now are classified into three subtypes denoted by _1A, _1B, and _1D corresponding to the cloned subtypes _1a, _1b, and _1d (for reviews see Hieble et al., 1995; Docherty, 1998; Zhong and Minneman, 1999). All three subtypes exhibit high affinities for prazosin. Based on results of functional studies, the existence of an atypical subtype (_1L), with low affinity for prazosin, has been postulated but not yet cloned (Flavahan and Vanhoutte, 1986; Docherty, 1998). Ensuing with this new classification, efforts have been made to clarify the distribution and function of these subtypes in different tissues, including the iris dilator muscle.

Takayanagi et al. (1992) first suggested that the _1B-adrenoceptor subtype mediates sympathetically evoked mydriasis in rabbits. Their results are based on the susceptibility to chloroethylclonidine, an alkylating agent originally thought to selectively inactivate the _1B-adrenoceptor subtype, but later found also to inactivate _1A- and _1D-adrenoceptor subtypes (Xiao and Jeffries, 1998). Using more specific ligands, Nakamura et al. (1999) reported that, in rabbits, the _1L-adrenoceptor subtype may mediate pupillary dilator responses. In rats, the _1B-adrenoceptor subtype is shown to mediate constriction of irideal blood vessels (Gould and Hill, 1994) and the production of excitatory junction potentials in the iris dilator muscle (Hill et al., 1993). However, a detailed analysis of -adrenoceptors in the mediation of contractile responses of iris dilator has not been conducted in this commonly used species.

The present series of in vivo experiments were performed to determine 1) the pharmacological profiles of -adrenoceptors mediating neurally elicited mydriatic responses and 2) the receptor subtype(s) involved. Our results suggest that, unlike cats and rabbits, adrenoceptors mediating sympathetically evoked mydriasis in rats are more "typical" and can be categorized into the current classification scheme. In addition, _1A-adrenoceptors seem to mediate sympathetic pupillary dilator responses.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Animal Preparation. All studies were approved by the Institutional Animal Care and Use Committee of University of Oklahoma Health Sciences Center and were undertaken in accordance with the NIH Guide for the Care and Use of Laboratory Animals. Adult male Sprague-Dawley rats (312-580 g) were anesthetized with pentobarbital (60 mg/kg i.p. + 5 mg i.v. as needed). A femoral artery and vein were cannulated for monitoring blood pressure (Statham P23 pressure transducer; Statham, Murray Hill, NJ) and for i.v. drug administration, respectively. The trachea was intubated. One cervical preganglionic sympathetic nerve was carefully separated from the vagus nerve and crushed proximally. Rectal temperature was maintained at approximately 37°C with a Deltaphase isothermal pad (Braintree Scientific, Inc., Braintree, MA). Pupillary diameter was measured using a ruler with a 0.1-mm scale under an Olympus surgical microscope (Olympus, Tokyo, Japan) installed with a green light filter. Heart rate was derived from the pressure wave using a Grass tachograph (7P4D; Grass Instruments, Quincy, MA).

Sympathetic Nerve Stimulation. A bipolar electrode was placed under the separated cervical sympathetic nerve (proximal to superior cervical ganglion) and covered with mineral oil. The stimuli, generated by a Grass S88 stimulator (Grass Instruments), consisted of 10-s trains of 5-V pulses (width 2 ms, frequency 1-32 Hz), which elicited reproducible pupillary dilations. Each mydriatic response was allowed to recover to the basal level before the next higher frequency of stimulation was applied. Frequency responses were generated approximately 15 min after i.v. drug or saline administration. No significant alterations of blood pressure or heart rate were observed during nerve stimulation.

Drugs and Data Analysis. Phentolamine hydrochloride, prazosin hydrochloride, phenoxybenzamine hydrochloride, WB-4101, and 5-methylurapidil were purchased from Sigma Chemical Co. (St. Louis, MO). Rauwolscine hydrochloride was obtained from Sigma/RBI (Natick, MA), and BMY-7378 dihydrochloride was obtained from Tocris Cookson Inc. (Ballwin, MO). All solutions of drugs were prepared in sterile physiological saline, with the exception of prazosin [2.5% glucose (w/v), 2.5% glycerol (v/v)]. Doses administered represent the respective salts.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Responses to Sympathetic Nerve Stimulation. A basal pupillary diameter of 0.64 ± 0.05 mm (n = 52) was observed after preganglionic sympathetic nerve section. As shown in Fig. 1, electrical stimulation of sympathetic nerves produced reproducible frequency-response curves of pupillary dilation at 1 to 32 Hz in anesthetized rats. No alterations of the contralateral pupillary diameter were seen, indicating lack of central reflex effects. To avoid rapid blood pressure changes, the doses of antagonists used in this study were administered (i.v.) slowly over 1 to 3 min. All of these drugs produced expected systemic hypotension but were without significant effect on resting pupillary size.

View larger version (16K): [in this window] [in a new window]   Fig. 1.   Pupillary responses to electrical stimulation of the decentralized preganglionic cervical sympathetic nerve in six pentobarbital-anesthetized rats. The frequency of stimulation was varied between 1 and 32 Hz (5-V, 10-s trains, and pulse width 2 ms). Frequency-response curves were generated at 15-min intervals. Values represent means ± S.E.M. of evoked pupillary dilation. Note the stability of evoked pupillary frequency-response curves over time.

Effects of Nonselective ₁-Adrenoceptor Antagonists. Initial experiments were undertaken to determine whether the rat also has "atypical" adrenoceptors in the pupil dilator muscle mediating neurally evoked mydriasis, as is seen in cats. Nonselective, prototypical -adrenergic antagonists, phentolamine (0.3-10 mg/kg) and phenoxybenzamine (0.03-1 mg/kg), as well as the selective ₁-adrenergic antagonist, prazosin (0.01-1 mg/kg), were administered systemically. In other experiments, the ₂-adrenergic antagonist, rauwolscine (0.1-1 mg/kg), was tested. Frequency-response curves were generated before and 15 min after cumulative antagonist administration. As shown in Fig. 2, neurally elicited mydriasis was antagonized by phentolamine, phenoxybenzamine, and prazosin in a dose-dependent fashion. In contrast, rauwolscine produced a modest potentiation of the response. Pupillary dilator responses were not totally blocked by the ₁-antagonists used. Maximal inhibitory effects and ID₅₀ values of these antagonists are summarized in Table 1.

View larger version (39K): [in this window] [in a new window]   Fig. 2.   Effects of increasing cumulative i.v. doses of -adrenoceptor antagonists on sympathetically evoked mydriatic responses in anesthetized rats. Frequency-response curves were generated before and 15 min after drug administration. Each animal received only one drug. Values represent means ± S.E.M. of mydriatic responses for five to seven animals. Asterisks indicate levels of significance at 8 Hz, compared with control values. , P < 0.05; , P < 0.01. Note that the evoked pupillary response was dose-dependently inhibited by the -adrenoceptor antagonists, phentolamine, phenoxybenzamine, and prazosin but not by the 2-adrenoceptor antagonist rauwolscine.

View this table: [in this window] [in a new window]   TABLE 1 ID50 and maximal inhibition of -adrenoceptor blockers in antagonism of mydriatic responses to preganglionic sympathetic nerve stimulation at 8 Hz Drugs were administered (i.v.) cumulatively. ID50 values were calculated by nonlinear regression analysis with 95% confidence intervals given in parentheses. Maximal inhibition is presented as percentage (means ± S.E.M.), by comparing the maximal reduction of pupillary dilations after antagonists to the control response.

Effects of Selective _1A-Adrenoceptor Antagonists. To determine whether _1A-adrenoceptors were involved, anesthetized animals were given cumulative doses of the partially selective _1A-adrenergic antagonist WB-4101 (0.1-1 mg/kg, i.v.) and the highly selective _1A-adrenergic antagonist 5-methylurapidil (0.1-1 mg/kg, i.v.). As above, frequency-response curves were generated before and 15 min after different doses of these antagonists (Fig. 3). Both agents inhibited evoked pupillary responses, in a dose-related manner. Increase of the dosage of both drugs to 3 mg/kg did not produce further inhibition (data not shown). The degree of maximal blockade achieved by WB-4101 (1 mg/kg) and 5-methylurapidil (1 mg/kg) was similar to that seen with nonselective ₁-adrenergic antagonists (Table 1).

View larger version (23K): [in this window] [in a new window]   Fig. 3.   Effects of increasing cumulative i.v. doses of the selective 1A-adrenoceptor antagonists, WB-4101 and 5-methylurapidil, on sympathetically evoked mydriatic responses in anesthetized rats. Frequency-response curves were generated before and 15 min after drug administration. Each animal received only one drug. Values represent means ± S.E.M. of pupillary dilation for five to six animals. Asterisks indicate levels of significance at 8 Hz, compared with control values. , P < 0.05; , P < 0.01. Note that the evoked pupillary response was dose-dependently inhibited by both subtype-selective antagonists.

View larger version (38K): [in this window] [in a new window]   Fig. 4.   Effect of combination of the selective 1A-adrenoceptor antagonists, 5-methylurapidil (5-MU), and the nonselective -adrenoceptor antagonist, prazosin and phenoxybenzamine (PBZ), on sympathetically evoked mydriatic responses in anesthetized rats. Each preparation was given sequentially (i.v.): 5-methylurapidil (1 mg/kg), prazosin (1 mg/kg), and phenoxybenzamine (3 mg/kg). Evoked pupillary responses were recorded before and 15 min after each drug administration. Only values of 8 to 16 Hz are shown in the figure as means ± S.E.M. Note that the residual pupillary dilation after 1 mg/kg 5-methylurapidil was not further blocked by addition of prazosin or, subsequently, phenoxybenzamine.

Effect of the Selective _1D-Adrenoceptor Antagonist. The potent and selective _1D-antagonist BMY-7378 (1-3 mg/kg) was administered to determine whether _1D-adrenoceptors were involved in the mediation of sympathetically evoked pupillary dilations. As shown in Fig. 5, BMY-7378 had no effect on the evoked pupillary frequency-response curve.

View larger version (16K): [in this window] [in a new window]   Fig. 5.   Effects of increasing cumulative i.v. doses of the selective 1D-adrenoceptor antagonist, BMY-7378, on sympathetically evoked mydriatic responses in anesthetized rats. Frequency-response curves were generated before and 15 min after drug administration. Values represent means ± S.E.M. of pupillary dilation for six animals. Note that BMY-7378 did not antagonize neurally evoked pupillary responses.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

Molecular and pharmacological investigations indicate a mixed population of ₁-adrenoceptor subtypes in the iris dilator muscle. Analysis of mRNA levels in the iris of rats (Vidovic and Hill, 1995) and rabbits (Nakamura et al., 1999) suggests that the most abundant transcript is that for the _1A-adrenoceptor gene, which is followed by the gene for the _1B-adrenoceptor subtype. Only minimal _1D-adrenoceptor gene transcription is seen. In accordance with these findings, binding studies also demonstrate a predominance of the _1A-adrenoceptor over the _1B-adrenoceptor subtype in the iris of rabbits, with no _1D-adrenoceptors found (Nakamura et al., 1999; Wikberg-Matsson et al., 2000). However, functional studies concerning the role of these adrenoceptor subtypes are incomplete.

The goal of the present study, conducted in anesthetized rats, was to functionally characterize the -adrenoceptors in the iris dilator muscle mediating neurogenic mydriasis. We observed evoked mydriatic responses to be dose-dependently blocked by prazosin (₁), phentolamine (₁, ₂), and phenoxybenzamine (₁ > ₂), but not by rauwolscine (₂). These results support the conclusion that a "typical" adrenoceptor subtype(s) is involved. This is consistent with the in vitro finding that contraction of the rat iris dilator is sensitive to blockade by phentolamine (Narita and Watanabe, 1982) and that constriction of blood vessels, as well as production of excitatory junction potentials in the rat iris, are sensitive to prazosin (Hill et al., 1993; Gould and Hill, 1994). Similarly, in mice, topical application of ₁-adrenoceptor agonists, methoxamine and phenylephrine, causes pupillary dilations that are readily inhibited by both prazosin and phentolamine (Fawcett et al., 1993).

Using a similar sympathetic nerve stimulation technique, it is apparent that the cat iris dilator is endowed with "atypical" adrenoceptors that are resistant to blockade by both prazosin and phentolamine (Koss and Gherezghiher, 1988; Koss et al., 1990). Schaeppi et al. (1966) also were unable to inhibit electrical activation of the isolated iris dilator muscle of cats with phentolamine. However, these adrenoceptors are readily antagonized by phenoxybenzamine and WB-4101 but not by either rauwolscine or yohimbine (Koss and Gherezghiher, 1988; Koss et al., 1990). Under the current classification, it seems that the proposed _1L-adrenoceptor may mediate sympathetically evoked mydriasis in the cat.

Insensitivity to prazosin is also seen in rabbits in that topical and systemic administration of prazosin does not antagonize the mydriasis induced by ₁-adrenoceptor agonists (Murray and Leopold, 1985). Unlike other -adrenergic antagonists such as phentolamine and phenoxybenzamine, which cause a pronounced miotic response, prazosin only has a minimal effect on pupillary size of rabbits (Mittag, 1983). Nakamura et al. (1999) recently reported that _1L-adrenoceptors, rather than previously suggested _1B-adrenoceptors (Takayanagi et al., 1992), may mediate contraction of the rabbit iris dilator muscle in response to both exogenous and endogenous norepinephrine. It is of interest, however, that _1L-adrenoceptor sites are not detected in binding studies (Nakamura et al., 1999).

To assess which subtype(s) was involved in our model, three currently available subtype-selective antagonists were used. WB-4101 is moderately (about 20 times) selective for the _1A- over the _1B-adrenoceptor subtype (Zhong and Minneman, 1999). 5-Methylurapidil is reported to have 75- to 140-fold higher affinity for _1A- than for _1B-adrenoceptors (Hanft and Gross, 1989; Eltze, 1997) and shows in vivo selectivity for the _1A- over the _1D-adrenoceptor subtype at 1 to 3 mg/kg, in rats (Castillo et al., 1998). By inference, 5-methylurapidil may also be selective for _1A- over _1B-adrenoceptors at this dose range, in view of the fact that it exhibits lower affinity for _1B- than for _1D-adrenoceptors (Eltze, 1997). BMY-7378 displays 126- and 100-fold affinity for _1D-adrenoceptors over that for the _1A- and _1B-adrenoceptor subtypes, respectively (Goetz et al., 1995). The latter two drugs have been repeatedly used in vivo to characterize the functional role of the respective subtypes (Zhou and Vargas, 1996; Ibarra et al., 1997; Castillo et al., 1998). Currently, no selective antagonists for the _1B-adrenoceptor are available.

The present combination experiments demonstrated that the residual mydriatic response seen after 1 mg/kg 5-methylurapidil was not further antagonized by high doses of either prazosin or phenoxybenzamine. This suggests that 1 mg/kg 5-methylurapidil blocked all the response mediated by ₁-adrenoceptors and that _1A-adrenoceptors play a major role. The doses of BMY-7378 at 0.1 to 1 mg/kg have been clearly shown to be selective for the _1D-adrenoceptor subtype (Zhou and Vargas, 1996). The lack of effect of BMY-7378 (1-3 mg/kg) suggests that _1D-adrenoceptors are not involved. Obviously, one limitation of our conclusion is the lack of selective ligands for the _1B-adrenergic subtype to directly evaluate the role of this subtype in the iris dilator muscle.

Our observation that _1A-adrenoceptors almost totally mediated the mydriatic response in rats seems to contradict results from a previous study by Hill et al. (1993), showing that excitatory junction potentials, measured in the rat iris dilator muscle using an intracellular recording technique, are mediated by the _1B-adrenoceptor subtype. This conclusion is based primarily on the effect of chloroethylclonidine, the effectiveness of which to discriminate different -adrenergic subtypes is controversial (Xiao and Jeffries, 1998). It is possible that two different subtypes (_1A and _1B) may mediate these two processes (muscular contraction versus excitatory junction potential, respectively), because it has been shown that the amplitude of excitatory junction potentials does not necessarily reflect the degree of the muscular contraction (Bolton and Large, 1986; Hill et al., 1993).

Neurally evoked pupillary responses are not completely blocked by - and -adrenoceptor antagonists (Hill et al., 1991). In the present study, we confirmed this observation in that neither nonselective nor subtype-selective -adrenoceptor antagonists, given either singly or in combination, could totally antagonize the evoked pupillary response. A similar situation was also seen in the vasculature in that -adrenergic antagonists do not completely block the pressor response to spinal stimulation in pithed rats (Castillo et al., 1998). The residual component may be due to corelease of other transmitters (e.g., neuropeptide Y, ATP, and others) from sympathetic nerve terminals. Indeed, neuropeptide Y released from sympathetic nerves during long-duration, high-frequency stimulation produces irideal vasoconstriction in rats (Newhouse and Hill, 1997). In addition, purinergic transmission mediates contraction of the iris dilator muscle in rabbits (Muramatsu et al., 1994).

In humans, the adrenoceptor subtype(s) mediating neurally evoked mydriasis is still controversial. Phentolamine blocks contraction of isolated human iris dilator muscle evoked by electrical stimulation (Yoshitomi et al., 1985). Prazosin inhibits neurogenic, as well as phenylephrine-induced, pupillary dilations (Mortlock et al., 1996). Based on the common clinical use of the miotic drug dapiprazole, which shows higher affinities for _1A- and _1D-adrenoceptors than for the _1B-adrenoceptor subtype, it has been proposed that the _1A-adrenoceptor subtype mediates neurally induced mydriasis (Eltze, 1997). However, others have suggested that -adrenoceptors of human iris dilator may be characterized as the _1L-adrenoceptor subtype, because prazosin only exhibits low potency in antagonizing phenylephrine-evoked contraction of isolated iris dilator muscle (Ishikawa et al., 1996). Taken together, these observations suggest that _1A- or _1L-adrenoceptors mediate neurally evoked mydriasis in humans.

Species variations appear to exist with regard to involvement of _1A- or _1L-adrenoceptor subtypes between rabbits, cats, rats, and humans. Interestingly, the _1L-adrenoceptor gene still eludes identification (for review see Docherty, 1998). In the iris dilator muscle of rabbits, binding and molecular studies show a predominance of the _1A-adrenoceptor subtype, with no _1L-adrenoceptors present. In the same study, however, the functional characterization is that of the _1L-adrenoceptor subtype (Nakamura et al., 1999). These observations are consistent with findings by Ford et al. (1997) that when analyzed functionally, cloned _1a-adrenoceptors display pharmacological profiles of the putative _1L-adrenceptor subtype. If true, the dominant adrenoceptors mediating neurogenic mydriatic responses in the iris dilator muscle could all be of the _1a-adrenoceptor subtype, which may exhibit higher or lower affinity for prazosin in different species.