The M x Muscarinic Antagonist Pirenzepine Reduces Myopia and Eye Enlargement in the Tree Shrew

Transcription

1 he M x Muscarinic Antagonist Pirenzepine Reduces Myopia and Eye Enlargement in the ree Shrew harles L. ottriall and Neville A. McBrien Purpose. o determine the efficacy of the Mi-selective muscarinic antagonist, pirenzepine, in preventing experimentally induced myopia in a mammalian model, the tree shrew. Methods. ree shrews were monocularly deprived (MD) using translucent goggles or negative lenses for a period of 12 days. In two of the MD groups, tree shrews received daily subconjunctival administration of either pirenzepine (17.7 fimol;re = 9) or vehicle control (re = 6). ontrol groups (re = 6) were used to assess the effects of MD, injection regimen, and drug effects. Results. In sham-injected and saline-injected MD tree shrews, 12 days of MD produced 13.2 D ± 0.8 D and 14.1 D ± 0.5 D of axial myopia, respectively. In pirenzepine-injected MD tree shrews, 12 days of MD induced an axial myopia of only 2.1 D ± 1.4 D. he significant reduction in myopia in pirenzepine-injected MD tree shrews was caused by significantly less vitreous chamber elongation of the deprived eye (0.05 mm ± 0.04 mm) relative to the contralateral control eye when compared to sham-injected and saline-injected MD tree shrews (0.24 mm ± 0.02 mm and 0.29 mm ± 0.01 mm). Mean equatorial enlargement and increased eye weight were prevented in pirenzepine-injected MD tree shrews (P < 0.01). Pirenzepine also was found to reduce myopia and ocular enlargement in lens defocus-induced myopia. ontrol experiments demonstrated that pirenzepine did not cause a significant reduction in amplitude of carbachol-induced accommodation. onclusions. Findings demonstrate that chronic administration of the M,-selective muscarinic antagonist, pirenzepine, prevents experimentally induced myopia in this mammalian model by a nonaccommodative mechanism. Invest Ophthalmol Vis Sci. 1996; 37: Accommodation has long been implicated as a causative mechanism in the development of myopia. he increased prevalence of myopia in college-based samples and occupational groups with a large near-work component is well documented. 1 2 he substantial epidemiologic evidence that some aspect of the near response in humans can lead to myopia does not prove a cause-and-effect relationship but simply indicates an association between near-work and myopia in humans. It could be argued that occupations requiring intensive close work are more suited to individuals with From the artment of Oplometry and Vision Sciences, University of Wales, ardiff, United Kingdom. Supported by grants from the Biotechnology and Biological Sciences Research ouncil (GR/J33265) and the Wellcome rust U.K. (#037531). Submitted for publication fuly 6, 1995; revised February 8, 1996; accepted February 9, hvprietary interest category: N. Reprint requests: Neville A. McBrien, artment of Oploine.lry and Vision Sciences, University of Wales, Redwood Building, King Edward VII Avenue, P.O. Box 905, ardiff F1 3X1% United Kingdom. myopia and thus attract more myopes. However, studies on animal models of refractive development have provided direct evidence that restricting vision to close viewing distances results in adaptive changes to the eye's structure and the development of myopia. 3 Studies have attempted to implicate accommodation specifically, as opposed to near-work, by pharmacologically blocking the accommodative apparatus of the eye using muscarinic antagonists. Daily administration of the cycloplegic agent atropine has been found to prevent or reduce the development or progression of myopia in adolescent humans and mammalian animal models. 4 "*' However, recent studies of refractive development in animal models have provided evidence that question the role of accommodation as a major causative mechanism in the development of myopia. Findings suggest that the visual signals controlling eye growth and refractive development may proceed directly from the retina to the choroid and/or sclera and, not as 1368 Investigative Ophthalmology & Visual Science, June 1996, Vol. 37, No. 7 opyright Association for Research in Vision and Ophthalmology

2 Pirenzepine Reduces Myopia and Eye Enlargement 1369 previously thought, through central visual structures. Investigations have demonstrated that blocking communication between the eye and the central visual processing centers, either by optic nerve section in chicks 7 or monkeys 8 or by blockade of ganglion cell action potentials in tree shrews 9 or chicks 10 does not prevent the development of, or recovery from, experimentally induced myopia. Further evidence against a pivotal role for accommodation in myopia is given by studies demonstrating that axial myopia can be induced in a species that does not possess a functional accommodation system" or in which accommodation has been blocked by bilateral lesioning of the Edinger-Westphal nucleus. 12 he apparent conflict between the success of atropine in preventing the progression or development of myopia in humans and mammalian animal models and the above findings arguing against a role for accommodation as a major causative factor in myopia question the proposed mode of action by which atropine is effective in preventing myopia. Recent studies 13 ' 14 have attempted to address the possibility that atropine may act through a nonaccommodative route. By using the chick model of refractive development, it has been demonstrated that atropine is effective in preventing the ocular enlargement and myopia normally associated with form deprivation without reducing the accommodative amplitude of the striated ciliary muscle (predominantly nicotinic receptors 15 ) of the chick eye. 14 hese findings support a nonaccommodative mechanism for the effect of atropine in the prevention of ocular enlargement and myopia. Atropine is a broad-band muscarinic antagonist that binds to all five identified muscarinic receptors. More selective muscarinic antagonists that have affinity for specific muscarinic receptors 16 make it possible to target a particular class of receptors. It is known that M receptors are usually (although not exclusively) found in neural cells (e.g., retina), whereas M 3 receptors are found in smooth muscle (e.g., mammalian ciliary muscle). Although previous studies 1317 have reported findings using the selective M antagonist pirenzepine to prevent form-deprivation myopia in chick, they will be alluded to here only briefly because we discussed the results of these studies in a companion article from our laboratory 18 on the dose-dependent effects of pirenzepine in the chick model of myopia. he primary aim of the current study was to determine whether the M selective antagonist pirenzepine 19 effectively reduces or prevents experimentally induced myopia in a mammalian model of refractive development, namely the tree shrew. he study also addresses the possible side effect of pirenzepine on ocular accommodation in a species with a smooth ciliary muscle, as found in humans. Portions of these results have been presented in abstract form. 20 MAERIALS AND MEHODS Animals All tree shrews were supplied from our own breeding colony. Animals were raised in a 15-hour light-9-hour dark cycle. Animals were removed from the maternal breeding cage 15 days after natural eye opening (which was 19 ± 2 days after birth) and were placed in separate primate cages. Food and water were available ad libitum. he luminance in the holding room was 85 cd/m 2, and cage luminance levels varied from 5 to 30 cd/m 2. Animals were assigned to one of six experimental groups on the basis of whether they were monocularly occluded (MD) and/or whether they received subconjunctival injections of either pirenzepine or saline vehicle. Groups were balanced for gender and for which was the experimental eye. All groups consisted of at least six animals. All investigations concerning the use of animals adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Experimental Groups Animals were injected daily with pirenzepine and were monocularly deprived of form vision (pirenzepine MD group). o control for the effects of both the drug and the injection routine, another group of tree shrews was monocularly deprived and treated daily with the saline vehicle (saline MD). A third group of MD tree shrews was included to control for the daily anesthetic regimen; this group underwent MD and the same daily procedures as the two MD treatment groups, but no drug or vehicle was administered (sham-injected MD). o study the effects of daily pirenzepine treatment on an otherwise normal, developing eye, tree shrews were given daily pirenzepine on the same schedule as MD animals but were not form deprived (pirenzepine open). o control for the effect of the injection routine on normal ocular growth, another group of animals was given daily saline vehicle but was not subjected to form deprivation (saline open). A final group of tree shrews underwent neither injection nor monocular deprivation of form vision but was housed in identical conditions and had optical and structural measures taken at the end of the experimental period (normal open). Each group was composed of six animals except the pirenzepine MD group (n = 9), in which additional data were available from animals used for the histologic study. Drug and Injection Protocol Pirenzepine (Sigma hemical, Poole, UK) is a selective muscarinic M! antagonist 19 that has a reported

3 1370 Investigative Ophthalmology 8c Visual Science, June 1996, Vol. 37, No. 7 elimination half-life in human blood of 12 hours. 21 Based on data from a companion study of dose-response curves for intravitreal and subconjunctival injection of pirenzepine in chick ocular tissue, we found an elimination half-life for subconjunctivally injected pirenzepine of approximately 4 hours (unpublished data, 1995). Based on our own data on ocular tissues and the reported data on elimination half-life, it was considered necessary to administer the drug on a daily regimen. Because it has been shown previously that repeated intravitreal injection in tree shrew results in decreased elongation of the vitreous chamber, 9 a subconjunctival route was used, although a companion study in chick reveals that an intravitreal route of application requires a 25-fold lower dose. 18 Based on preliminary studies to establish an effective dose using a subconjunctival injection route, our dose-response results revealed a relatively high threshold (8.85 fimol) before any structural effect was observed, with an ED 50 of //mol and a saturation of /xmol. From these preliminary results, a 10% pirenzepine dose was chosen for the main study. he drug was injected in a total volume of 75 //I, giving a daily administration of 17.7 //mol. o reduce insult to the conjunctiva because of the daily subconjunctival injection, it was decided to alternate between subconjunctival and topical administration over the 12-day treatment period. he subconjunctival injections were given at the beginning of each day (approximately 9:00 AM) starting on the first experimental day. Under anesthesia, topical proxymetacaine H1 (0.5%) was applied to the eye, and a 30-gauge needle was inserted into the bulbar conjunctiva, and the pirenzepine or saline was injected. opical application consisted of five 15 /A drops spaced at 2-minute intervals. It has been reported that pirenzepine, as well as the structurally related drug telenzepine, are found to be as much as 60-fold more potent when administered topically to the eye in acidic solutions than when used in the physiological ph range. 22 It also has been demonstrated that in rat fundus strip, the affinity of pirenzepine for Mi receptors is greater in the protonated form. 23 For these reasons, we decided to administer pirenzepine in a protonated form at the ocular surface (ph 4). All saline vehicle applications were matched to the same ph as pirenzepine. On the first treatment day animals, were anesthetized deeply with a ketamine H1 (90 mg/kg) and xylazine (10 mg/kg) mix. Animals were placed on a heating pad maintained at 37. Axial dimensions of the ocular components were recorded (A-scan ultrasound) to ascertain baseline values before any experimental procedure. A head-mounted goggle, which held the translucent occluder in place over the eye, was affixed to the skull of all MD animals using a technique similar to that described by Siegwart and Norton. 24 his enabled the goggle to be removed to facilitate injections and to be reattached afterward. During the 12-day deprivation period, the goggle and occluder were cleaned regularly. Before daily administration of the drug or vehicle, horizontal and vertical pupil diameters were measured using a small operating microscope with measuring graticule eyepiece at X10 magnification (luminance 100 lux). On each subsequent day, animals were anesthetized with 4% halothane. wo drops of topical anesthetic, proxymetacaine H1 (0.5%), were applied to the eye to be treated to reduce corneal sensation and to facilitate the passage of the predominantly ionic drug through the lipophilic layers of the cornea. 25 Pupil measurements were recorded, and the drug or vehicle was administered. In Vivo Optical and Structural Measures welve days after treatment was initiated, a full set of refractive and biometric measures was recorded on both eyes of an animal. Atropine sulfate 1% was applied topically to the cornea of both eyes 45 minutes before general anesthesia (ketamine H1 and xylazine) was administered. Body temperature and breathing were monitored as before. he animal's head was supported by a dental bite bar to allow measurements along the optic axis of the eye. Measurements of corneal radius, ocular refraction, and axial ocular dimensions were taken as described previously" 14 and will be described only briefly here. orneal radius of curvature was measured using a modified one-position keratometer. Ocular refraction was measured using streak retinoscopy and coincidence optometry. Results are reported as mean spherical equivalent effective at the corneal plane. No correction was made for the small eye artifact, 26 estimated by schematic modeling to be 4.9 D in tree shrews of this age. omparison of ocular refraction values measured by the two techniques revealed good correlation (r 2 = 0.88) for all 78 eyes in the study. Furthermore, there was no significant (P > 0.05) bias when the difference between treated and control eyes was compared. onsequently, only retinoscopy values are reported. Axial ocular dimensions were recorded using A-scan ultrasonography. A 15-MHz focused transducer driven by a Panametrics (Waltham, MA) 5052 pulser/receiver passed waveforms to a Leroy (Geneva, Switzerland) 9400 digital storage oscilloscope. Waveforms were transferred to a computer for later measurement. After all in vivo measurements, the animal was deeply anesthetized with sodium pentobarbital, and the eyes were enucleated and coded. Equatorial and

4 Pirenzepine Reduces Myopia and Eye Enlargement 1371 anterior-posterior measurements were recorded with digital calipers, and wet eye weight was taken to the nearest 10 //g. For these measures, the experimenter was unaware as to which was the treated eye. Accommodation Measures o determine whether pirenzepine, at the dose used in this study, had any effect on ocular accommodation, a control study was performed. ree shrews were given a subconjunctival injection of either pirenzepine (n = 3) or saline vehicle (n = 3) under halothane anesthesia. Ninety minutes after injection, the animals were anesthetized (ketamine-xylazine) and placed on a heating pad (37 ), and baseline measurements of pupil diameter, ocular refraction, and axial ocular dimensions of both the treated and the contralateral control eye were taken. o maintain a sufficient pupil diameter for ultrasound measures after treatment with carbamylcholine chloride (carbachol), animals were pretreated with 5% phenylephrine and 2% epinephrine combined in an 2.5% agar gel button and given iontophoretically. Refractive and ocular dimension measures were repeated after full mydriasis was achieved to control for any drug-induced changes. o stimulate accommodation, carbachol 20% was administered to the treated eye iontophoretically, again incorporated in a 2.5% agar gel button. Previous studies"' 27 have shown this to be a supramaximal dose. All drugs were applied to the eye by positive electrode corneal iontophoresis. 27 Pupil diameter was recorded for the first 10 minutes after carbachol administration before refractive and ocular dimension measures were again taken. he contralateral control eye of each animal was used to monitor for any systemic effects of the topical application of the drugs. Measurements were repeated at 30 and 60 minutes after carbachol administration. Lens Defocus It has been suggested that the mechanisms controlling the enlargement of the eye in form-deprivation myopia may be different from those in lens-induced myopia. 28 o investigate whether pirenzepine prevented the vitreous chamber elongation associated with optical defocus, as opposed to form deprivation, eyes of tree shrews were defocused with a negative lens fitted to the head-mounted goggle instead of a translucent occluder. One group (n = 4) was treated daily with pirenzepine on exactly the same schedule and dose described above and was monocularly deprived of normal pattern vision with a 10 D lens. Another group of animals (n = 4) was monocularly deprived of normal pattern vision with a 10 D lens and anesthetized on a daily basis but did not receive pirenzepine. he stated Sham MD Saline MD Pirenz MD Pirenz Saline Normal FIGURE l. Differences in cycloplegic ocular refraction between open control eyes and deprived eyes in monocularly deprived (MD) tree shrews and control and treated eyes or right and left eyes of nondeprived animals. Daily pirenzepine treatment significantly reduced the experimental myopia produced by monocular deprivation (ukey's HSD test; **P < 0.001). n 6 for all groups except pirenzepine MD, in which n = 9. Error bars = 1 SEM. negative lens powers are the defocus at the corneal plane. Distribution of Pirenzepine In Vivo By using radiolabeled [ 3 H]-pirenzepine (DuPont NEN, Stevenage, UK) in a different group of animals (n = 2), it was possible to investigate the distribution of drug in the ocular tissues. A subconjunctival injection of 17.7 //mol (15 fid activity) was made. Eyes were enucleated after 60 minutes and dissected into anterior and posterior portions (5 mm trephine at posterior pole) before separation into retinal, choroidal, scleral, and vitreal components. issues were weighed and digested before liquid scintillation counting. A standard curve was established for each ocular tissue and used to convert counts per minute to picomoles per milligram of tissue. Molar concentrations were calculated from either our own data on wet and dry weights for both choroid and sclera or on the previously reported volume fraction of extracellular space in the retina, derived from rabbit. 20 Histology Enucleated eyes from selected animals treated with pirenzepine for 12 days were slit around the limbus and fixed in 2.5% glutaraldehyde for 24 hours. Buttons of 1.5 mm diameter were trephined from the posterior pole and the equatorial region and embedded in epoxy (araldite) resin using standard histologic techniques. Semithin sections (1 fim) were cut on a microtome, mounted on glass slides, and stained with toluidine blue. Sections were examined at X1000 magnification on a Leica (Wetzlar, Germany) DM stereomicroscope.

5 1372 Investigative Ophthalmology & Visual Science, June 1996, Vol. 37, No. 7 ABLE l. ycloplegic Ocular Refraction and Axial Ocular omponent Dimensions for Pirenzepine-reated and ontrol Animals Sham-MD (n = 6) Saline-MD (n = 6) Pirenzepine-MD (n = 9) Pirenzepine- (n = 6) Saline- (n = 6) Normal (n = 6) Rednoscopy (D) orneal radius (mm) DO Anterior segment (mm) Lens thickness (mm) Vitreous chamber (mm) Axial length (ultrasound) (mm) Equatorial dimensions (average) (mm) Anterior posterior dimension (mm) Eye weight (mg) -3.9 ± ± ± 0.8* 3.25 ± ± ± nt H i dt d j ± ± ± 0.02* 7.26 ± ± ± 0.02* 8.50 ± ± ± 0.02* 7.42 ± ± ± 0.02* ± ± ± 2.53* -5.5 ± ± ± 0.5* 3.27 ± ± ± : t it it } t jt : t ± ± ± 0.01* 7.36 ± ± ± 0.01* 8.51 ± ± ± 0.02* 7.46: t : t = t 0.04* : : t ± 1.85* +6.3 ± ± ± ± ± ± jt i n i n jt ± ± ± ± ± ± ± ± ± jt it it jt t ± it n n d n jt j i d d nt i ± ± ± 0.01* 7.13 ± ± ± ± ± ± 0.01* 7.33 ± ± ± O.Olf ± ± ± 0.58* +8.2 ± ± ± ± ± ± dt d d d H: d ± ± ± ± ± ± ± ± ± ± nt J d dt ± : t it it it jt : t ± : t : t : t : : t ± ± ± ± ± ± ± ± ± : t : : t : t : t ± 0.67 Mean ± SEM of all deprived and open control eyes of monocularly deprived tree shrews and treated and control or right and left eyes of binocularly open shrews t-test between treated and control eyes: *P < 0.01; fp < Statistical Analysis All data were transferred to a statistical software package (Minitab 9.0) for analysis. For comparison of differences between more than two groups, a single factor analysis of variance (ANOVA) was used. o determine which individual groups were significantly different, the ANOVA was combined with ukey's HSD test. 30 endent or independent -tests were used to examine specific differences within groups. RESULS Ocular Refraction Refractive differences for the six experimental groups in the main study are shown in Figure 1 and able 1. Sham-injected ( 13.2 D) and saline-injected ( 14.1 D) MD tree shrews developed similar degrees of myopia in the deprived eye relative to the contralateral control eye after 12 days. In marked contrast, pirenzepine-treated MD tree shrews developed a relative myopia between deprived and control eyes of only 2.1 D. he observed differences in induced myopia between the pirenzepine MD group and the control MD groups were found to be significant (ANOVA, P < 0.001; ukey's HSD, P < 0.001). No significant difference in refractive state between treated and control eyes of the pirenzepine open and saline open groups or between right and left eyes of the binocularly normal group was ob-

6 Pirenzepine Reduces Myopia and Eye Enlargement 1373 A 0.1 n i - ** i o, ID t- O a. a. * V A* Sham Saline Pirenz Pirenz Saline Normal MD MO MD Sham Saline Pirenz Pirenz Saline Normal MD MD MD Sham Saline Pirenz Pirenz MD MD MD Saline Normal Sham Saline Pirenz Pirenz MD MD MD Saline Normal FIGURE 2. Differences in ocular dimensions between deprived and open control eyes in monocularly deprived (MD) tree shrews and between treated and control eyes or right and left eyes of binocularly open animals. (A) Differences in vitreous chamber depth before treatment (open bars) and after treatment (shaded bars). Daily pirenzepine treatment caused a significant reduction in vitreous chamber elongation in MD tree shrews compared to control animals. In binocularly open animals, treatment to one eye caused a significant reduction to normal vitreous chamber depth. (B) Differences in axial length before treatment (open bars) and after treatment (shaded bars). Daily pirenzepine treatment caused a reduction in both deprived and open injected animals compared to control animals. () Differences in equatorial diameters ((superior-inferior + medial-lateral)/2). Daily pirenzepine treatment produced a marked reduction in mean equatorial eye enlargement in both deprived and open tree shrew eyes compared to control animals. (D) Differences in wet eye weight. Daily pirenzepine treatment caused a reduction in both deprived and open injected animals compared to controls. (ukey's HSD test; *P < 0.05; **P < 0.001). n = 6 for all groups except pirenzepine MD, in which n = 9. Error bars = 1 SEM. served. here was also no significant difference between the relative refractive error between the three binocularly open groups (ANOVA, P = 0.1; Fig. 1). Ocular Dimensions Differences in Monocularly rived Groups. Measures of corneal radius, in vivo axial ocular dimensions, and in vitro equatorial dimensions are shown in able 1. No significant differences in corneal radius, anterior segment depth, or lens thickness were observed for any of the six experimental groups. Elongation of the vitreous chamber in the deprived eye, relative to the contralateral open eye, of the pirenzepinetreated MD animals (0.05 ± 0.04 mm) was significantly less than either sham-injected (0.24 ± 0.02 mm) or saline-injected (0.29 ± 0.01 mm) MD animals (AN- OVA, P < 0.001; ukey's HSD, P < 0.001, Fig. 2A). here was no significant difference in vitreous chamber elongation between the sham-injected and salineinjected MD groups. Analysis of pretreatment ultrasound measures revealed no significant differences in vitreous chamber depth or any other biometric components (Figs. 2A, 2B). aliper measures of the equatorial dimensions of the eye ((superior-inferior + medial-lateral)/2) re-

7 1374 Investigative Ophthalmology & Visual Science, June 1996, Vol. 37, No. 7 ABLE 2. Differences in ycloplegic Ocular Refraction and Ocular Dimensions Between reated and ontralateral ontrol Eyes for ree Shrews Undergoing Lens Defocus -10 D Lens (n = 4) 10D Lens + Pirenzepine (n = 4) Retinoscopy (D) orneal radius (mm) Anterior segment (mm) Lens thickness (mm) Vitreous chamber (mm) Axial length (mm) Equatorial dimensions (average) (mm) Anterior posterior dimension (mm) Eye weight (mg) -9.4 ± 0.9* 0.01 ± ± ± ± 0.01* 0.18 ± 0.02* 0.14 ± 0.01* 0.23 ± 0.03* ± 1.7* f * 0.01* O.Olf O.Olf 0.6f reatment for 12 days. Mean ± SEM of all treated minus control eyes. /= test: *P< 0.01; fp < vealed that pirenzepine also was effective in preventing equatorial enlargement and reducing the axial elongation normally associated with form-deprivation myopia (see able 1, Fig. 2). he reduction in equatorial diameter in deprived eyes of pirenzepine MD animals ( 0.07 ± 0.05 mm) compared to the equatorial enlargement in sham-injected (0.17 ± 0.02 mm) and saline-injected (0.15 ± 0.02 mm) MD animals was found to be significant (ANOVA, P = 0.001; Fig. 2). Wet eye weight of deprived eyes was increased for sham-injected and saline MD groups by ± 2.53 mg and ± 1.85 mg respectively when compared to contralateral control eyes, whereas for pirenzepinetreated MD animals, there was a reduction of 6.02 ± 2.82 mg in wet eye weight. his difference between MD groups was significant (ANOVA, P < 0.001; Fig. 2D). Differences in Binocularly Groups. In animals not form deprived, significant differences were found between pirenzepine-treated animals and control animals (able 1). here was a small but consistent reduction in the vitreous chamber depth in all the pirenzepine-treated open eyes compared to their contralateral control eye ( 0.06 ± 0.01 mm), whereas no significant differences were observed between right and left eyes for the other two open groups (able 1, Fig. 2A). Vitreous chamber depth differences between the three open groups were found to be significant (AN- OVA, P < 0.001). Analysis of ultrasound measures of ocular dimensions between the three groups before the start of the study found no significant differences in vitreous chamber depth (ANOVA, P = 0.53) or any of the other biometric components. Pirenzepinetreated open eyes were found to have reduced equatorial dimensions when compared to contralateral open eyes ( 0.08 ± 0.01 mm). Saline-treated and normal open groups had no significant interocular differences in equatorial dimensions (able 1). Differences in equatorial dimensions between the three open groups were significant (ANOVA, P < 0.001). Reductions in eye size were reflected in the wet eye weights, with pirenzepine-treated animals having a significant reduction in the treated eye compared to control groups (ANOVA, P < 0.02; able 1, Fig. 2D). Lens-Induced Myopia Lens-induced defocus in tree shrews for 12 days using a 10 D lens positioned over one eye produced significant myopia compared to the control eye ( 1.4 ± 1.0 versus 8.0 ± 0.2 D; diff -9.4 ± 0.9 D; P < 0.01). When lens defocus was combined with daily pirenzepine treatment, significantly less myopia was induced (3.9 ± 1.0 versus 8.1 ± 0.2 D; diff -4.2 ± 0.8 D; P < 0.01, able 2). his reduction in induced myopia was caused primarily by a reduction in the degree of vitreous chamber elongation, which was significantly different between the two groups (P < 0.001; able 2) < ESS ^M SALINE INJEED PIRENZEPINE INJEED LENS HIKNESS N = 3 (each group) INDUED AOMMODAION FIGURE 3. Measurement of the amount of accommodation induced by corneal iontophoresis of 20% carbachol in tree shrew eyes after treatment with either 10% pirenzepine or saline vehicle. No significant difference in induced accommodation was observed between pirenzepine-treated and saline-treated tree shrew eyes, n = 3 in each group. Error bars = 1 SEM

8 Pirenzepine Reduces Myopia and Eye Enlargement A B D FIGURE 4. Histologic appearance of the retina at the posterior pole from the treated (A) and control (B) eyes of a tree shrew that underwent daily injections of 10% pirenzepine. No signs of toxic damage (X1000) to the retina of the pirenzepine-treated eye could be observed when compared to the control eye. Stained with toluidine blue. Scale bar = 50 fim. Histologic appearance of the sclera at the posterior pole from the treated () and control (D) eyes of a tree shrew that underwent daily injections of 10% pirenzepine. No signs of toxic damage (XlOOO) to the sclera of the pirenzepine-treated eye could be observed when compared to the control eye. Stained with toluidine blue. Scale bar = 20 fim. Pupil and Accommodation Pirenzepine, at the dose used in the current study, was found to cause a significant increase in pupil diameter when compared to saline-treated control animals (P < 0.01; n = 6). Pirenzepine-treated eyes showed a relative increase in pupil diameter of 0.37 ± 0.06 mm, whereas saline-treated eyes underwent an increase of only 0.03 ± 0.04 mm. he level of carbachol-induced accommodation in pirenzepine-treated (7.2 ± 0.3 D; 1375 n = 3) and saline-treated (7.8 ± 0.2 D; n = 3) animals was not significantly different (P = 0.13, Fig. 3). arbachol-induced lens thickness changes also were not significantly different between pirenzepine- (0.03 ± 0,01 mm) and saline-treated eyes (0.03 ± 0.01 mm; P = 0.47; Fig. 3). Distribution of Pirenzepine In Vivo After subconjunctival injection of sh-pirenzepine, the amount of the drug in ocular tissues of the injected eye was determined, and from this molar concentrations were estimated, as detailed in Materials and Methods. oncentrations were calculated to be sclera 10 ' M, choroid lo"4 M, retina 10~5 M, and vitreous 10~f1 M, 1 hour after administration (in chick ocular tissue, it was found that peak concentration of the drug occurred 1 hour after subconjunctival injection). In terms of the amount of drug (moles) in the whole tissue, the levels found represented only a small fraction of the drug injected; sclera (6 nmol), retinal pigment epithelium + choroid (3 nmol), retina (800 pmol), and vitreous (350 pmol). Also of interest was the distribution of the drug in anterior.and posterior tissues. he concentration of the drug per milligram of tissue was found to be approximately 35% higher in the anterior compared to the posterior portion of both retina and sclera. Although this was not surprising because the site of injection was the bulbar conjunctiva, it is noteworthy that the greatest structural effects were seen in the anterior region, where equatorial enlargement was prevented completely. Histology Histologic evaluation of the retina and sclera at the light microscopic level (XlOOO) revealed no apparent toxic effect on either the neural retina or the sclera of a tree shrew eye that received 12 days of daily pirenzepine treatment when compared to contralateral control eyes or saline-treated eyes (Figs. 4A, 4B). DISUSSION he results demonstrate that the Mrselective muscarinic antagonist, pirenzepine, was effective in markedly reducing both form-deprivation and lens-defocus myopia in a mammalian model, namely the tree shrew, and it did so through a nonaccommodative mechanism. hese findings add further support to a role for cholinergic antagonists in the control of ocular growth and myopia that has been indicated from studies using atropine in mammals5b and atropine and pirenzepine in chicks.18""'17-18 he reduction in induced axial myopia was accounted for by a reduction in the major structural consequence of both form-deprivation and lens-defo-

9 1376 Investigative Ophthalmology & Visual Science, June 1996, Vol. 37, No. 7 cus myopia, namely vitreous chamber elongation in the treated eye. Daily pirenzepine also completely prevented the equatorial enlargement associated with form-deprivation and lens-defocus myopia in the tree shrew. his finding is in keeping with two previous studies investigating the effect of atropine on formdeprivation myopia in chick 14 ' 31 and, although not observed by all previous studies, 13 seems to be consistent across avian and mammalian species. he current findings on tree shrew using an M r selective antagonist are in agreement with studies on chick using the nonselective muscarinic antagonist, atropine, given either intravitreally or subconjunctivally, in demonstrating its effectiveness in preventing both form-deprivation myopia and lens-defocus myopia It was found also that pirenzepine caused a small but significant reduction in vitreous chamber depth and mean equatorial diameter in tree shrew eyes treated with the M r receptor antagonist but not form deprived. his finding, which has not been observed with atropine treatment in chick, implies that endogenous acetylcholine is involved in the control of normal ocular growth as well as axial myopia development. Schematic modeling predicted that the reduced vitreous chamber depth alone would produce 2.5 D of hyperopia, yet no significant refractive error was observed. his was caused by the remaining ocular structures in the treated open eye undergoing compensatory changes that accounted for ~2 D, indicating that the eye had actively emmetropized to overcome the shallower vitreous chamber depth resulting in no significant relative refractive error. Reports have suggested the possibility of retinal and scleral toxicity to muscarinic antagonists Histologic evidence from the current study found no observable differences between pirenzepine-treated and control eyes and no signs of toxic damage to the retina or sclera at the dose used. In chick, we have found that daily 5 mg subconjunctival and 500 fig intravitreal injections of pirenzepine also caused no toxic damage to the retina. 18 Rickers et al 17 reported that intravitreal pirenzepine was only effective in preventing axial myopia in chick at doses of 2000 ng, which was found to cause retinal damage. However, examination of their data indicates that a dose-dependent reduction in induced myopia was found with an ED 50 of approximately 500 /itg. he difference in results may, in part, be attributed to the fact that in the study by Rickers et al, injections were made every 3 days, and, because we found the elimination half-life in chick ocular tissue to be approximately 4 to 5 hours, it would seem likely that for most of the period pirenzepine levels would be very low. In addition, all measures of retinal toxicity (electroretinography, histology) were reported for a dose of 2000 //g, and no evidence of retinal toxicity was found at doses lower than 1500 fig (Schaeffel, personal communication, 1995). Further evidence of normal retinal and scleral function in pirenzepine-treated tree shrews was provided by the finding that it was still possible to induce form-deprivation myopia in tree shrews that had received daily application of pirenzepine for 12 days and were form deprived for another 18 days. his period of form deprivation resulted in an axial myopia of 10.7 D, which is comparable to that of age-matched tree shrews over a similar developmental period (30 to 45 days MD). 33 A similar functional response has been found subsequent to pirenzepine treatment in chicks. 18 Using a subconjunctival injection, the relatively high dose of pirenzepine required to initiate some prevention of vitreous chamber elongation may be a consequence of the reported poor penetration of pirenzepine at neural sites in the brain when compared to other muscarinic antagonists. 34 he large dose administered also may reflect the relatively poor distribution of the drug to the various ocular tissues with the administration technique used. By using radiolabeled pirenzepine, it was found that subconjunctival injections of pirenzepine resulted in only 0.032%, 0.018%, and 0.005% of the injected dose of pirenzepine being detected in the sclera, retinal pigment epithelium plus choroid, and retina, respectively, 1 hour after injection. Despite this large dilution of the drug, our subconjunctival injection still resulted in micromolar concentrations of freely available pirenzepine in retina, retinal pigment epithelium (RPE) plus choroid, and sclera. It has been reported in other tissues (e.g., cerebral cortex) that the I 50 for pirenzepine at M, receptor sites is in the nanomolar range, 35 although in ocular tissue, micromolar values have been found. 36 However, these values are based on the levels of receptor-bound drug from studies using in vitro models. It is likely that differences in tissue-receptor pharmacokinetics will exist between in vivo and in vitro models, and there is evidence that the tissue dose required for functional effects in vivo may be higher than they are in vitro. 37 Although drug toxicity has been addressed, the question of drug selectivity must be considered at such concentrations. A companion study that investigated the dose-response curves of intravitreally and subconjunctivally injected pirenzepine in chick revealed that equivalent doses of M 2 and M 3 selective antagonists were not effective in reducing form-deprivation myopia, a finding previously reported by Stone et al. 14 hese findings suggest that the selectivity of the drug, with respect to ocular growth mechanisms, was maintained despite the high dose, although, at such doses, the possibility of pirenzepine acting at nonmuscarinic receptors cannot ruled out. o understand the mechanism by which pirenzep-

10 Pirenzepine Reduces Myopia and Eye Enlargement 1377 ine is effective in preventing experimentally induced myopia, it is necessary to know its site of action. Previous studies provide evidence implicating retinal amacrine cells, such as dopaminergic and enkephalinergic amacrine cells, in modulating ocular growth. 38 ' 39 Acetylcholine is a known retinal neurotransmitter, 40 and cholinergic amacrine cells that contain muscarinic M, receptors have been identified. 41 It seems feasible that pirenzepine may affect ocular growth regulation by altering retinal acetylcholine release 42 and subsequent cholinergic input to ganglion cells 43 and other neurotransmitter pathways, such as dopaminergic, 44 GA- BAergic, 45 glycinergic, 45 and enkephalinergic, 40 ' assuming basic similarities between species. Although the neural retina is a possible site of action, alternative sites of action could be the RPE and choroid, which are known to possess muscarinic receptors. 47 ' 48 Our radiolabeled data revealed that the RPE plus choroid tissue contained a substantial amount of the drug found in ocular tissue, and a recent study 3 ' 1 has reported that the RPE has a higher density of M receptors than neural retina. It is feasible that muscarinic receptors in the RPE and/or choroid are involved in regulating scleral growth and that subconjunctivally injected pirenzepine alters this regulation to prevent or reduce axial elongation. Recently, it was proposed that muscarinic antagonists may produce their effect by acting directly on scleral M receptors. 451 It is suggested that the increased scleral growth in chick models of myopia 50 may be explained by acetylcholine promoting cellular proliferation and growth in chondrocytes through a phospholipase-/ipn/-fos pathway. As a consequence, the muscarinic antagonist, pirenzepine, could initiate a downregulation of the pathway and possibly epidermal growth factor receptors, thereby reducing scleral growth. However, in tree shrew sclera (and possibly chick fibrous sclera 51 ), a decrease in proteoglycan synthesis during myopia development 52 and an increased synthesis in recovery from myopia 53 have been reported. Based on the above findings, it would be difficult to argue a role for pirenzepine in downregulating scleral growth to prevent myopia in mammalian/primate species. However, this does not rule out an alternative mechanism of direct cholinergic control on scleral metabolism, and data on muscarinic antagonist effects on mammalian scleral fibroblasts are needed. However, because apparently functional doses (based on reported I r,,> values for M, receptors) for pirenzepine were found in retina, choroid, and sclera, we could not, in the current study, isolate one particular tissue as the site of action for pirenzepine in the control of ocular growth. It will be important to investigate further the mechanism by which pirenzepine produces its effect on eye enlargement and myopia. It will be important to determine whether muscarinic antagonists interact with other mechanisms and pathways implicated in the regulation of ocular growth. Key Words accommodation, muscarinic antagonist, myopia, pirenzepine, tree shrew Acknowledgments he authors thank Lynn Marie ornell for technical assistance regarding retinal and scleral histology. References 1. Myopia: Prevalence and Progression. Working Group on Myopia Prevalence and Progression, ommittee on Vision, ommission on Behavioral and Social Sciences and Education, National Research ouncil. Washington D: National Academy Press; Adams DA, McBrien NA. Prevalence of myopia and myopic progression in a population of clinical microscopists. Invest Ophthalmol Vis Sci. 1992; 69: Young FA. he effect of restricted visual space on the refractive error of the young monkey eye. Int Ophthalmol. 1963; 2: Bedrossian RH. he effect of atropine on myopia. Am J Ophthalmol. 1979;86: Raviola E, Wiesel N. An animal model of myopia. N EnglJMed. 1985; 312: McKanna JA, asagrande VA. Atropine affects lid-suture myopia development: experimental studies of chronic atropinization in tree shrews. Doc Ophthalmol. 1981;28: roilo D, Gottlieb MD, Wallman J. Visual deprivation causes myopia in chicks with optic nerve section. uir Eye Res. 1987; 6: Wiesel N, Raviola E. Myopia and eye enlargement after neonatal lid fusion in monkeys. Nature. 1977; 266: Norton, Essinger JA, McBrien NA. Lid-suture myopia in tree shrews with retinal ganglion cell blockade. VisNeurosd. 1994; 11: McBrien NA, Moghaddam HO, ottriall L, Leech EM, ornell LM. he effects of blockade of retinal cell action potentials on ocular growth, emmetropization and form deprivation myopia in young chicks. Vision Res. 1995;35: McBrien NA, Moghaddam HO, New R, Williams LR. Experimental myopia in a diurnal mammal (Sciurus carolinensis) with no accommodative ability. / Physiol. 1993; 469: Schaeffel F, roilo D, Wallman J, Howland H. Developing eyes that lack accommodation grow to compensate for imposed defocus. Vis Neurosci. 199O;4: Stone RA, Lin, Laties AM. Muscarinic antagonist effects on experimental chick myopia. Exp Eye Res. 1991;52:

11 1378 Investigative Ophthalmology & Visual Science, June 1996, Vol. 37, No McBrien NA, Moghaddam HO, Reeder AP. Atropine reduces experimental myopia and eye enlargement via a nonaccommodative mechanism. Invest Ophthalmol Vis Sci. 1993;34: Pilar G, Nunez R, McLennan IS, Meriney SD. Muscarinic and nicotinic synaptic activation of the developing chicken iris. JNeurosci. 1987;7: Goyal RK. Muscarinic receptor subtypes: Physiology and clinical implications. N Engl J Med. 1989; 321: Rickers M, Schaeffel F. Dose-dependent effects of intravitreal pirenzepine on deprivation myopia and lensinduced refractive errors in chickens. Exp Eye Res. 1995;61: Leech EM, ottriall L, McBrien NA. Pirenzepine prevents form deprivation myopia in a dose dependent manner. Ophthalmic Physiol Opt. 1995; 15: Hammer R, Berrie P, Birdsall NJM, Burgen ASV, Hulme E. Pirenzepine distinguishes between different sub-classes of muscarinic receptor. Nature. 1980; 283: McBrien NA, ottriall L. Pirenzepine reduces axial elongation and myopia in monocularly deprived tree shrews. ARVO Abstracts. Invest Ophthalmol Vis Sci. 1993;34: armine AA, Brogden RN. Pirenzepine, a review of its pharmacodynamic and pharmacokinetic properties and therapeutic efficacy in peptic ulcer disease and other allied diseases. Drugs. 1985;30: Hagan JJ, van der Heijden B, Broekkamp LE. he relative potencies of cholinomimetics and muscarinic antagonists on the rat iris in vivo: Effects of ph on potency of pirenzepine and telenzepine. Naunyn- Schmiedeberg's Arch Pharmacol. 1988; 338: Barlow RB, han M. he effects of ph on the affinity of pirenzepine for muscarinic receptors in the guinea pig ileum and rat fundus strip. Br J Pharmacol. 1982; 77: Siegwart J, Norton. Goggles for controlling the visual environment of small animals. Lab Anim Sci. 1994; 44: Brewitt M, Bonatz E, Honegger J. Morphological changes to the corneal epithelium after application of topical anaesthetic ointment. Ophthalmologica. 1980; 180: Glickstein M, Millodot M. Retinoscopy and eye size. Science. 1970; 168: Koretz JF, Bertasso AM, Neider MW, rue-gabelt B, Kaufman PL. Slit-lamp studies of the rhesus monkey eye: II: hanges in the crystalline lens shape, thickness and position during accommodation and aging. Exp Eye Res. 1987;45:3l Schaeffel F, Hagel G, Bartmann M, Kohler K, Zrenner E. 6-Hydroxy dopamine does not affect lens-induced refractive errors but suppresses deprivation myopia. Vision Res. 1994;34: Ames A III, Hastings AB. Studies on water and electrolytes in nervous tissue: I. Rabbit retina: Methods and interpretation of data. / Neurophysiol. 1956; 19: Daniels WW. Biostatistics: A Foundation for Analysis in the Health Sciences. New York: John Wiley; Wildsoet F, McBrien NA, lark IQ. Atropine inhibition of lens-induced effects in chick-evidence for similar mechanisms underlying form deprivation and lens induced myopia. ARVO Abstracts. Invest Ophthalmol Vis Sci. 1994; 35: hew SJ, Beuerman RW, Kaufman HE. In vivo assessment of corneal stromal toxicity by tandem scanning confocal microscopy. Lens Eye oxicity Res. 1992;9: McBrien NA, Norton. he development of experimental myopia and ocular component dimensions in monocularly lid sutured tree shrews (upaia belangeri). Vision Res. 1992;32: Bymaster FP, Heath I, Hendricks J, Shannon HE. omparative behavioral and neurochemical activities of cholinergic antagonists in rats. J Pharmacol Exp her. 1993;267: Mohr K, rankle. Allosteric effects of the alkanebis-ammonium compound w84 and of tacrine on ( 3 H) pirenzepine binding at Mpreceptors in rat cerebral cortex. Pharmacol oxicol. 1994;75: Salceda R. Muscarinic receptors binding in retinal pigment epithelium during rat development. Neurochem Res. 1994; 19: Prast H, Fischer HP, Prast M, Philippu A. In vivo modulation of histamine release by autoreceptors and muscarinic acetylcholine receptors in the rat anterior hypothalamus. Naunyn-Schmiedeberg's Arch Pharmacol. 1994; 350: Stone RA, Lin, Laties AM, Iuvone PM. Retinal dopamine and form-deprivation myopia. Proc Natl Acad Sci USA. 1989;86: Seltner RLP, Grant V, Stell WK [ME 5 ] Enkephalin and form deprivation myopia. Invest Ophthalmol Vis Sci. ARVO Abstracts. 1994; 35: Hutchins JB. Review: Acetylcholine as a neurotransmitter in the vertebrate retina. Exp Eye Res. 1987; 45: Vanderheyden P, Ebinger G, Vauquelin G. haracterization of M r and M.2-muscarinic receptors in calf retina membranes. Vision Res. 1988; 28: unningham JR, Dawson, Neal MJ. Evidence for a cholinergic inhibitory feed-back mechanism in the rabbit retina. /Physiol. 1983; Ariel M, Daw NW. Pharmacological analysis of directionally sensitive rabbit ganglion cells. / Physiol. 1982;324: Brown JH, Rietow M. Muscarinic-dopaminergic synergism on retinal cyclic AMP formation. Brain Res. 1981;215: unningham JR, Neal MJ. Effect of gama-aminobutyric acid agonist, glycine, taurine and neuropeptides on acetylcholine release from the rabbit retina. J Physiol. 1983; 336: Neal MJ, Paterson SJ, unningham JR. Enhancement of retinal acetylcholine release by DAMGO: Possibly

12 Pirenzepine Reduces Myopia and Eye Enlargement 1379 a direct opiod receptor-mediated excitatory effect. Br J Pharmacol. 1994; 113: Friedman Z, Hachett SF, ampochiario PA. Human retinal pigment epithelium cells possess muscarinic receptors coupled to calcium mobilization. Brain Res. 1988;446: Meriney SD, Pilar G. holinergic innervation of the smooth muscle cells in the choroid coat of the chick eye and its development. / Neurvsri. 1987; 7: Marzani D, Lind GJ, hew SJ, Wallman J. he reduction of myopia by muscarinic antagonists may involve a direct effect on scleral cells. ARVO Abstracts. Invest Ophthalmol Vis Sti. 1994;35: hristensen AM, Wallman J. Evidence that increased scleral growth underlines visual deprivation myopia in chicks. Invest Ophthalmol Vis Sti. 1991; 32: Wallman J. Nature and nurture of myopia. Nature. 1994;37l: Reeder AP, McBrien NA. Investigation of scleral metabolism during the development of experimentally induced myopia in tree shrew. ARVO Abstracts. Invest Ophthalmol Vis Sti. 1994;35: McBrien NA, Lawlor P. Increased proteoglycan synthesis in the sclera of tree shrew eyes recovering from form deprivation myopia. ARVO Abstracts. Invest Ophthalmol Vis Sti. 1995;36:S760.