Effects of Calcitonin Gene-Related Peptide in the Eye

A Study in Rabbirs and Cars

Olli Oksob*t and Johan Srjernschanrz:}:

Calcitonin gene-related peptide (CGRP) has recently been demonstrated in sensory neurons of the eye. The purpose of the present study was to determine the effects of exogenous CGRP in the rabbit and cat eye. CGRP was injected intracamerally and the intraocular pressure was measured in cannulated eyes. The pupil diameter and the aqueous humor protein concentration were also measured. Indomethacin was used to prevent prostaglandin synthesis and tetrodotoxin (TTX) to block nerve conductance. In the rabbit eye, CGRP caused iridial hyperemia, a breakdown of the blood-aqueous barrier and increased intraocular pressure. These responses were dose-related. The increase in IOP as well as the breakdown of the blood-aqueous barrier could not be blocked with TTX or indomethacin. In cats CGRP caused a decrease in IOP and had only slight effect on the aqueous humor protein concentration. Neither in rabbits nor in cats had CGRP any detectable effect on the pupil size. Intracameral injection of 0.1 Mg (7.4 X 10~n moles) substance P together with 0.1 ng (2.6 X 10"" moles) CGRP in rabbits caused maximal miosis but did not potentiate the intraocular effects of CGRP only. These results indicate that CGRP has marked vascular effects in the rabbit eye, causing a breakdown of the blood-aqueous barrier and increased IOP. The mechanism of this phenomenon does not involve prostaglandins neither nerve conduction, implying most likely a direct effect on the vascular smooth muscle. The mechanism of the decrease of IOP in cats remains unknown. Invest Ophthalmol Vis Sci 29:1006-1011,1988

The eye typically reacts to noxious stimuli with component part of the acute irritative response of the miosis, hyperemia, breakdown of the blood-aqueous rabbit eye5614"17 but the vascular effects of SP in the barrier and increased intraocular pressure (IOP). The anterior uvea of the rabbit have been reported to be mediators behind these responses are not known but much less marked.1518 In the rabbit eye SP can be generally, prostaglandins and sensory neurons are released from sensory nerves by trigeminal nerve thought to be the most important factors.1"6 When stimulation14 and recently it was shown that CGRP the iris is traumatized directly, prostaglandins seem can also be released into the aqueous humor of rab- to be the key factor1-3 probably exerting most of their bits in a similar way.19 Thus, as a strong vasodilatator, effects through an activation of sensory neurons.4"6 In CGRP could very well be involved in the acute irrita- nonperforated eyes the noxious impulse is thought to tive response in the eye causing hyperemia, a break- be propagated into the eye through an axon reflex down of the blood-aqueous barrier and ocular hyper- mechanism.7 In these cases prostaglandins seems to tension. In previous studies CGRP has been demon- be less important. strated immunohistochemically and biochemically in Calcitonin gene-related peptide (CGRP), a novel the eye and the effects of exogenous CGRP have been 37-amino acid polypeptide encoded within the calci- demonstrated in the rabbit eye.1011 The purpose of tonin gene,8 has been shown to occur in the nervous the present study was to analyze further the effects of system.9 CGRP has histochemically been localized exogenous CGRP in the rabbit eye and to investigate to sensory neurons in the anterior segment of the the effects of CGRP in the cat eye with special refer- eye.10"13 Substance P (SP) or a closely related tachy- ence to the irritative response. kinin has been shown to be involved in the miotic Materials and Methods From The Research Department of Pharmaceutical Company Oy Star Ab, Tampere, Finland, and the fDivision of Pharmacol- Albino rabbits weighing 2-3.5 kg and cats weighing ogy, Department of Pharmacy, University of Helsinki, Helsinki, 2-3 kg of either sex were used. The rabbits were an- Finland. aesthetized with 25% urethane (2 g/kg b.w.). The an- X Present address: Pharmacia Ophthalmics Ab, Uppsala, Swe- aesthesia of the cats was initiated with intramuscular den. Submitted for publication: July 6, 1987; accepted March 2, 1988. injection of a combination of ketamine (20 mg/kg Reprint requests: Olli Oksala, Pharmaceutical Company Oy Star b.w.) and xylazine (0.5 mg/kg b.w.) and maintained Ab, P.O. Box 33, 33721 Tampere, Finland. with intravenous infusion of ketamine. In both rab-

1006

Downloaded from iovs.arvojournals.org on 09/27/2021 No. 7 EFFECTS OF CGRP IN THE EYE / Oksalo and Srjernschonrz 1007

bits and cats the level of the anaesthesia was such that ment and one for administration of CGRP/vehicle. the corneal reflex was only weak or absent. All the CGRP at 0.1-50.0 Mg (2.6 X 10~ll-1.3 X 10"8 moles) animals received indomethacin 10 mg/kg b.w. intra- was injected into the experimental eye after a stable venously 20 min before the cannulation of the eyes. baseline IOP was obtained (5-20 min after the second The animals were kept warm by a thermostated heat- cannulation), while the other eye served as a control, ing pad. All the animals were tracheotomized to as- receiving isotonic saline only. sure adequate ventilation and some of them were ar- tificially ventilated. In all experiments the pH, PCo2 Experiments With Intracameral Injection of CGRP and PO2 of the arterial blood were checked routinely a in TTX Pretreated Eyes few times during the course of the experiment (ABL 30; Radiometer, Copenhagen, Denmark). Artificial In these experiments an additional needle was used ventilation was adjusted to give normal or close to for TTX administration. One microgram (3.1 X 10~9 an normal values for Pco2> Po2 d pH. Both in rabbits moles) of TTX was injected intracamerally bilaterally and cats a femoral artery was cannulated and con- 10 min before the CGRP administration. Only rab- nected to a pressure transducer for blood pressure bits were used. The effects of TTX denervation were monitoring and in cats a femoral vein was also can- estimated in six eyes by stimulating the preganglionic nulated for administration of the anaesthetic. part of the sympathetic trunk electrically (10 Hz, 2 The eyes were cannulated with 27 gauge needles msec, 1 mA) before and after the TTX injection. Be- connected to polyethylene tubings. Great care was fore the injection stimulation caused rapidly my- taken not to touch the iris at any time during the driasis, the pupil diameter reaching 10.1 ± 0.4 mm. experiment. The intraocular pressure and the blood The TTX injection itself caused a slight mydriasis, pressure were measured with Statham P-50 pressure the pupil diameter being 8.7 ± 0.4 mm after TTX transducers (Gould, Inc., Oxnard, CA) connected to a adminitration compared with 6.2 ± 0.1 mm prior to Grass Model 79-D polygraph (Quincy, MA). Intra- TTX injection. Ten minutes after the injection, sym- cameral injections were carried out with Hamilton pathetic stimulation had no effect at all on the pupil precision syringes. A volume of 5 /ul was injected size. All animals underwent in addition light stimula- which caused a transient increase in IOP of 2-6 mm tion in order to verify the blockade of cholinergic Hg lasting from 2 to 5 min. miosis. It was assumed that the unmyelinated sensory An intravenous infusion of a 6% Dextran solution neurons are blocked in a similar way to the choliner- (Pharmacia, Uppsala, Sweden) diluted (1:1) with iso- gic and adrenergic neurons. The control eye received tonic saline was used whenever needed to sustain the the vehicle only. blood pressure. The infusion was started as soon as the mean arterial blood pressure fell below a level of Experiments With Intracameral Injection of CGRP about 70 mm Hg and was continued as long as neces- andSP sary. The total amount of dextran solution did not These experiments were performed only in rabbits. exceed 50 ml. A mixture of 0.1 ^g (7.4 X 10"" moles) of SP and 0.1 The pupil diameter was measured with a transpar- Hg (2.6 X 10"" moles) of CGRP was injected intraca- ent millimeter ruler under standard laboratory light. merally into four eyes. The control eye received the Iridial hyperemia was estimated macroscopically in vehicle only. albino animals. The hyperemia of the experimental The results are expressed as the arithmetical mean eye was compared with that of the contralateral con- value ± SEM, and were statistically evaluated with trol eye, and was graded slight, medium or strong. the student t-test for paired data by comparing the At the end of the experiment, paracentesis was per- experimental eye with the control eye. formed with a 27 gauge needle and aqueous humor This research was carried out according to the was stored at -50° until analyzed for protein con- 20 guidelines of the ARVO Resolution on the Use of centration. Animals in Research. Calcitonin gene-related peptide (a-CGRP (rat); Peninsula Laboratories, Belmont, CA), substance P Results (Peninsula), and tetrodotoxin (Sigma Chemical Company, St. Louis, MO) were dissolved in isotonic Intracameral injection of CGRP into the rabbit eye saline and stored at -50°C. caused hyperemia, a breakdown of the blood- aqueous barrier and increased intraocular pressure. Experiments With Intracameral Injection of CGRP CGRP at 0.1 /ig (2.6 X 10"" moles) caused slight in Normal Eyes hyperemia in the iris; a dose of 1-5 fig (2.6 X 10"10- In these experiments the eyes were cannulated with 1.3 X 10"9 moles) caused strong hyperemia. Intra- two needles, one for intraocular pressure measure- cameral injection of 0.1-5.0 ^g (2.6 X 10~"-1.3

Downloaded from iovs.arvojournals.org on 09/27/2021 1008 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / July 1988 Vol. 29

35 35i B

30 30 * * 25 25

-20 -10 10 20 30 40 50 60 70 -20 -10 X) 20 30 40 50 60 70 TUm(mln) Time (min)

mmHfl 35

•!? TP t

-20 -10 10 20 30 40 50 60 70 10 20 30 40 50 60 70 Tim* (min) Tlnwdnln)

Fig. 1. Effect ofintracameral injection of 0.1 /xg(A, n = 6), 0.5 ^g(B, n = 6), l-2/xg(C, n = 5) and 5 ng(D, n = 5) CGRP on the intraocular pressure in rabbits. Mean arterial blood pressure indicated by BP in the lower part of the respective picture. Experimental eyes indicated by solid line and control eyes by dotted line. Bars indicate SEM. * = P < 0.05, ** = P < 0.01 and *** = P < 0.001.

EXPERIMENTAL EYE Illllll CONTROL EYE X 10 9 moles) CGRP caused a dose-dependent increase in IOP (Fig. 1) and in the aqueous humor

RABBIT protein concentration (Fig. 2). Intracameral injection of 10 ng (2.6 X 10~12 moles) had no effect on the IOP or aqueous humor protein concentration. CGRP had no effect on the pupil size. In cats intracameral injection of 5 tig (1.3 X 10~9 moles) CGRP caused a decrease in IOP (Fig. 3). I There were no macroscopic signs of ocular irritation in cats, but a tendency towards an increase (statistically insignificant) in the aqueous humor protein concentration was detected (Fig. 2). CGRP injected intracamerally in as high a dose as 50 Mg (1.3 X 10~8 moles) in three cats caused only a slight increase in CGRP 0.1 uglcGRP 0.5 ug|cGRP 1-2 ug| CGRP 5 ug the aqueous humor protein concentration and did not elevate IOP. Fig. 2. Effect ofintracameral injection of CGRP on the aqueous 9 humor protein concentration in rabbits and cats. Bars indicate Pretreatment of rabbit eyes with 1 /*g (3.1 X 10~" SEM. ** = P< 0.01. moles) intracameral TTX, 10 min before injection of

Downloaded from iovs.arvojournals.org on 09/27/2021 No. 7 EFFECTS OF CGRP IN THE EYE / Oksalo. and Srjernschonrz 1009

mmHg 35 mmHg

BP 120T

"""* T I- I -1 7

-20 -10 0 10 20 30 40 50 60 70 10 20 30 Time (min) Time (min) Fig. 3. Effect of intracameral injection of 5 fig CGRP on the Fig. 4. Effect of intracameral injection of 1 ug CGRP on the intraocular pressure in cats. Experimental eyes indicated by solid intraocular pressure in rabbits which received an intracameral in- line and control eyes by dotted line. Bars indicate SEM. Mean jection of TTX bilaterally 10 min prior to CGRP. Experimental arterial blood pressure indicated as in Figure 1, (n = 6). * = P eyes indicated by solid line and control eyes by dotted line. Bars < 0.05. indicate SEM. Mean arterial blood pressure indicated as in Figure 1, (n = 6). * = P < 0.05, ** = P< 0.01.

1 tig (2.6 X lO"10 moles) CGRP did not block the group, eg, as compared with the 1 ug group (Fig. 2). effects of CGRP on IOP or the aqueous humor pro- Intracameral administration of as much as 100 /xg tein concentration (Figs. 4, 5). TTX itself induced (7.4 X 10~8 moles) of SP has been shown to have only hyperemia and a slight but insignificant rise in IOP little effect on the vasculature of the anterior uvea and the protein concentration of the aqueous humor. and the blood-aqueous barrier in urethane anaesthe- Intracameral injection of a mixture of 0.1 ug (7.4 tized rabbits.18 Thus, compared with SP, CGRP X 10-" moles) SP and 0.1 /*g (2.6 X lO"11 moles) seems to be of the order of magnitude 1000 times CGRP induced maximal miosis but had no addi- more potent as a vasodilatator in the rabbit eye. This tional effect on IOP (Fig. 6), the hyperemia or the aqueous humor protein concentration (204 ± 58 mg/100 ml m< EXPERIMENTAL EYE mg/100 ml in the experimental eyes and 92 ± 6 IIIIIH CONTROL EYE ** n=6 mg/100 ml in the contralateral control eyes) as com- 1600— pared with the corresponding dose of CGRP only. If anything, the increase in IOP was smaller than with 1400 0.1 ug CGRP only. 1200—

Discussion 1000—

Recent studies demonstrate that CGRP in the ante- 800— rior uvea is confined to sensory neurons,10"13 is re- 19 leased by stimulation of the trigeminal nerve, and is 600— n=3 a strong vasodilatator in the eye.1019 Thus, this potent —i— polypeptide is a good candidate for one of the media- 400— tors of part of the irritative response of the eye, and JDIIil possibly of antidromic vasodilatation in general. 200— ip It seems likely that the threshold dose of CGRP to o > •III induce a change in the parameters studied is in the TTX 1 ug TTX 1 ug range of 10-100 ng (2.6 X 10~12-2.6 X 10"11 moles) CGRP 1 ug NaCl when administered intracamerally in the rabbit. With Fig. 5. Effect of intracameral injection of 1 fig CGRP and iso- a dose of 5 tig the upper limit was reached with a tonic NaCl on the aqueous humor protein concentration in rabbits severe drop in blood pressure and contralateral ef- which received an intracameral injection of TTX bilaterally 10 min fects. Less statistical significance was obtained in this prior to CGRP. Bars indicate SEM. ** = P < 0.01.

Downloaded from iovs.arvojournals.org on 09/27/2021 1010 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / July 1988 Vol. 29

mmHg 8 35i dose as high as 50 ^g (1.3 X 10 moles) intracamerally there was no increase in IOP and no significant barrier breakdown. The mechanism behind the de- 30 crease in IOP remains unknown, and merits further studies. Since the cat eye is considerably larger than 25 the rabbit eye it can be assumed that intracameral injection of 5 ng CGRP in the cat eye corresponds to

20 about 1 Mg in the rabbit. Thus, it seems that there is a species difference in the response to CGRP between cats and rabbits. Since CGRP and SP coexist in sen- 15 sory neurons of the anterior uvea,1113 and SP has very little effect in the cat eye,23 it may simply be that the 10- same is true for CGRP and that other systems are more important in sensory mechanisms in the cat eye. -20 -10 10 20 30 40 50 60 70 Time (min) Simultaneous injection of SP with CGRP did not appreciably increase the response to CGRP in rab- Fig. 6. Effect of intracameral injection of a mixture of 0.1 ng SP and 0.1 ng CGRP on the intraocular pressure in rabbits. Experi- bits; in fact, the increase in IOP and aqueous humor mental eyes indicated by solid line and control eyes by dotted line. protein concentration was less than with CGRP only. Bars indicate SEM. Mean arterial blood pressure indicated as in This may be due to a decreased access of CGRP to the Figure 1, (n = 4). site of action due to instantaneous miosis. It should also be noted that there was a relatively strong decrease in systemic blood pressure of this group of estimation is in relatively good agreement with esti- animals after administration of the SP-CGRP mix- mations of the vasodilatatory effects of SP and CGRP ture (Fig. 6). Thus, it seems that SP in the dose used in the feline tooth pulp.21 does not potentiate the vascular effects of CGRP to The experiments performed in TTX-pretreated any greater extent. Recently it was shown that low rabbits suggest that the effect of exogenous CGRP is doses of CGRP potentiated the miotic effect of a low 19 not dependent on nerve conduction in the uvea. Of dose of SP as well as the weak vascular effects of SP. course, it cannot be excluded that CGRP may exert With the doses used in the present study it was not an effect through a prejunctional receptor directly possible to show any potentiating effect of CGRP on coupled to substances stored in the nerve terminals. the miotic effect of SP since the effect of SP was Strongly irritating agents, such as capsaicin and com- already maximal. The objective was rather to see if SP pound 48/80 seem to act through such a mecha- potentiates the vascular effects of CGRP. Thus, any nism,22 whereas physiologic compounds such as pros- substantial interaction between SP and CGRP in the taglandin E, histamine and bradykinin seem to re- anterior uvea of the rabbit does not seem to exist at quire conduction in sensory nerves for function.617 the dose levels used. Since both cholinergic and adrenergic neurons were The present study confirms that CGRP is a very blocked it is reasonable to assume that the unmyelin- potent vasodilatator in the rabbit anterior uvea, caus- ated sensory neurons were also blocked. Thus, it is ing breakdown of the blood-aqueous barrier. This ef- probable that the effect of CGRP is a direct one on fect does not seem to be mediated by neural elements the vascular smooth muscle. or prostaglandins secondarily. In the cat, on the con- Although TTX in itself caused vasodilatation trary, CGRP seems to have little inflammatory effect, which could have rendered the eye more sensitive to if any, and a decrease was observed in IOP. edema and barrier breakdown, the fact remains that Key words: calcitonin gene-related peptide, intraocular TTX did not induce any marked barrier breakdown pressure, aqueous humor protein concentration, blood- but blocked nerve conduction. Control experiments aqueous barrier, irritative response, nociception, rabbit, cat with intracameral injection of isotonic saline only, in TTX-pretreated eyes, confirmed that the injection Acknowledgment per se had no effect on the blood-aqueous barrier We would like to thank Ms. Tuija Pauhu for technical (Fig. 5). assistance. The effect of intracameral injection of CGRP in cats was surprising, since the effect on the blood- References aqueous barrier was very modest, and there was a 1. Cole DF and Unger WG: Prostaglandins as mediators for the decrease in IOP. Even when CGRP was injected in a response of the eye to trauma. Exp Eye Res 17:357, 1973.

Downloaded from iovs.arvojournals.org on 09/27/2021 No. 7 EFFECTS OF CGRP IN THE EYE / OKsolo and Srjernschanrz 1011

2. Neufeld AH, Jampol LM, and Sears ML: Aspirin prevents the tonin gene-related peptide (CGRP) immunoreactivity in the disruption of the blood aqueous barrier in rabbit. Nature mammalian eye. J Comp Neurol 233:506, 1985. 238:158, 1972. 14. Bill A, Stjernschantz J, and Mandahl A: Substance P: Release 3. Jampol LM, Neufeld AH, and Sears ML: Pathways for the on trigeminal nerve stimulation, effects in the eye. Acta Phys- response of the eye to injury. Invest Ophthalmol 14:184, 1975. iol Scand 106:371, 1979. 4. Butler JM and Hammond BR: Effect of sensory denervation 15. Stjernschantz J, Sears ML, and Stjernschantz L: Intraocular on the response of the rabbit eye to bradykinin and PGEi. effects of substance P in the rabbit. Invest Ophthalmol Vis Sci Trans Ophthalmol Soc UK 97:668, 1977. 20:53, 1981. 5. Butler JM and Hammond BR: The effects of sensory denerva- 16. Leander S, Hakanson R, Rosell S, Folkers K, Sundler F, and tion on the responses of the rabbit eye to prostaglandin Ei, Tornquist K: A Specific substance P antagonist blocks smooth bradykinin and substance P. Br J Pharmacol 69:495, 1980. muscle contractions induced by non-cholinergic, non-adren- 6. Mandahl A and Bill A: Ocular responses to antidromic trigem- ergic nerve stimulation. Nature 294:467, 1981. inal stimulation, intracameral prostaglandin E, and E2, cap- 17. Mandahl A: Release of substance P and its role in the acute saicin and substance P. Acta Physiol Scand 112:331, 1981. inflammatory response in the rabbit eye. Acta Universitatis 7. Butler JM, Unger WG, and Cole DF: Axon reflex in ocular Upsaliensis 470, (Academic dissertation), 1983. injury: Sensory mediation of the response of the rabbit eye to 18. Stjernschantz J, Sears ML, and Mishima H: Role of substance laser irradiation of the iris. Q J Exp Physiol 65:1981, 1980. P in the antidromic vasodilatation, neurogenic plasma extra- 8. Rosenfeld MG, Mermod J-J, Amara SG, Swanson LW, Saw- vasation and disruption of the blood-aqueous barrier in the chenco PE, Rivier J, Vale WW, and Evans RM: Production of rabbit eye. Naunyn Schmiedebergs Arch Pharmacol 321:329, a novel neuropeptide by the calcitonin gene via tissue specific 1982. RNA processing. Nature 304:129, 1983. 19. Wahlestedt C, Beding B, Ekman R, Oksala O, Stjernschantz J, 9. Gibson SJ, Polak JM, Bloom SR, Sabate IM, Mulderry PM, and Hakanson R: Calcitonin gene-related peptide in the eye: Ghatei MA, McGregor GP, Morrison JFB, Evans RM, and Release by sensory nerve stimulation and effects associated Rosenfeld MG: Calcitonin gene-related peptide immunoreac- with neurogenic inflammation. Regul Pept 16:107, 1986. tivity in the spinal cord of man and of eight species. J Neurosci 4:3101, 1984. 20. Lowry OM, Rosebrough NJ, Farr AJ, and Randall RS: Protein measurement with folin phenol reagent. J Biol Chem 193:265, 10. Unger WG, Terenghi G, Ghatei MA, Ennis KW, Butler JM, 1951. Zhang SQ, Too HP, Polak JM, and Bloom SR: Calcitonin gene-related polypeptide as a mediator of the neurogenic ocu- 21. Gazelius B, Edwall B, Olgart L, Lundberg JM, Hokfelt T, and lar injury response. J Ocular Pharmacol 1:189, 1985. Fischer JA: Vasodilatatory effects and coexistence of calcitonin 11. Wahlestedt C, Hakanson R, Beding B, Brodin E, Ekman R, gene-related peptide (CGRP) and substance P in sensory and Sundler F: Intraocular effects of CGRP, a neuropeptide in nerves of cat dental pulp. Acta Physiol Scand 130:33, 1987. sensory nerves. In Tachykinin Antagonists, Hakanson R and 22. Mandahl A and Bill A: Effects of substance P antagonist (D- 1 2 79 Sundler F, editors. Amsterdam, New York, Oxford, Elsevier, Arg , D-Pro , D-Trp , Leu")-SP on the miotic response to 1985, pp. 137-146. substance P, antidromic trigeminal nerve stimulation, capsai- 12. Stone RA, Kuwayama Y, Terenghi G, and Polak JM: Calci- cin, prostaglandin E, compound 48/80, and histamine. Acta tonin gene-related peptide: Occurrence in corneal sensory Physiol Scand 117:139, 1983. nerves. Exp Eye Res 43:279, 1986. 23. Unger WG and Tighe J: The response of the isolated iris 13. Terenghi G, Polak JM, Ghatei MA, Mulderry PK, Butler JM, sphincter muscle of various mammalian species to substance Unger WG, and Bloom SR: Distribution and origin of calci- P. Exp Eye Res 39:465, 1984.

Downloaded from iovs.arvojournals.org on 09/27/2021