Agonist Response of Human Isolated Posterior Ciliary Artery

Doo-Yi Yu, Valerie A. Alder, Er-Ning Su, Edward M. Mele, Stephen J. Cringle, and William H. Morgan

The isometric responses of isolated human posterior ciliary artery to adrenergic agonists, histamine (HIS), and 5-hydroxytryptamine (5-HT) were studied in passively stretched ring segments mounted in a myograph bath. Cumulative dose response curves were measured for nine agonists: HIS, 5-HT, dopamine (DOPA), epinephrine (A), norepinephrine (NA), tyramine (TYR), phenylephrine (PHE), isoproterenol (ISOP), and xylazine (XYL), and the log(molar concentration) at which one half of the maximum active tension was developed (EQo) was estimated. The ring segments were unresponsive to DOPA and XYL; HIS and ISOP produced biphasic responses with a mild relaxation for low concentrations and small contractions for high concentrations of the agonist. The remaining agonists caused contractile responses of magnitude listed in the rank order following compared with the maximum active tension in response to 0.124 M K+-Krebs:

Kmax > A > 5-HT = PHE > NA > TYR It was concluded that functional HIS, a,-adrenergic, and 5-HT receptors were present on human posterior ciliary artery but that there are no a2-adrenergic receptors. Invest Ophthalmol Vis Sci 33:48-54,1992

The regulation of ocular blood flow to ensure that may affect many aspects of ocular function involving all regions receive an adequate supply despite continu- the outer retina and the optic nerve head, the iris, and ously changing local tissue demands is a complex and the ciliary body. There is still disagreement whether hierarchic task, requiring continuous interaction be- vascular disease, particularly of the optic nerve head tween local and global mechanisms. The isolated ves- circulation, plays a role in the etiology of glaucoma. sel preparation in which a small segment of the artery Information about the pharmacologic control of in question is challenged pharmacologically is a pow- smooth muscle in the posterior ciliary artery would be erful approach to understanding the location and type helpful in understanding its normal and pathologic of controlling mechanisms of vascular beds. Because function. There currently have been only two pre- it is probable that there are species differences in the vious reports describing the function of human poste- pharmacologic sensitivity of ocular vessels, donated rior ciliary arteries using an in vitro technique.34 One3 human tissue is an invaluable source for such data if describes the effect of age on spontaneous myogenic we are to extrapolate experimental results to clinical activity, and the other4 used a constant flow technique situations to treat retinal vascular diseases. to ascertain the vascular responses to putative recep- The posterior ciliary arteries have autonomic in- tor agonists. We used ring segments of donated hu- nervation12 and supply blood to the optic nerve head, man posterior ciliary artery to test the pharmacologi- the choroid, the iris, and the ciliary body. Any impair- cal response to adrenergic agonists, histamine (HIS), ment of their ability to control blood flow therefore and 5-hydroxytryptamine (5-HT).

From the Lions Eye Institute and University of Western Austra- Materials and Methods lia, Nedlands, Western Australia, Australia. Supported by the Juvenile Diabetes Foundation International, Thirteen eyes from seven donors were used, and the New York, New York, Medical Research Fund of Western Austra- lia, Perth, Australia, TVW Telethon, Perth, Australia, Clive and experiments were done on 18 ring segments. Human Vera Ramaciotti Foundations, Sydney, Australia, The Ophthalmic donor eyes were obtained either postmortem or post- Research Institute of Australia, The Medical Faculty of the Univer- operatively after consent for use of the tissue for re- sity of Western Australia, Perth, Australia, and the Australian Na- search was obtained according to the guidelines of the tional Health and Medical Research Council, Canberra, Australia. National Health and Medical Research Council and Submitted for publication: May 7,1991; accepted July 25,1991. Reprint requests: Dr. Valerie A. Alder, Lions Eye Institute and under the supervision of the Human Ethics Commit- Department of Surgery, University of Western Australia, Nedlands, tee of the University. Patient identity was not revealed WA 6009, Australia. to the experimenter, but patient history, diagnosis,

Downloaded from iovs.arvojournals.org on 10/02/2021 No. 1 HUMAN POSTERIOR CILIARY ARTERY PHARMACOLOGY / Yu er ol 49

Table 1. Summary of donor details

Patient Age Delay time number (yr) (hr) History Medications HI 78 9 Not avilable Not available H2 77 6 Hypertension None H3 76 4 Hypertension diabetic (type 2) Daonil, aspirin, aldomet, propranolol H4 63 8 Carcinoma, lung Becotide, ventolin, oxycodone H5 73 6 Hypertension Diltiazem H6 43 2 Enucleation, detached retina, None calcined vitreous H7 78 8 Hypertension Becotide, prednisone, ventolin, spirondactone

All tissue was postmortem except for H6, which was from a surgical enucleation.

and medication information were (Table 1). The pa- the chamber was emptied by suction, and the feed line tients ranged in age from 43-78 yr. Before tissue was for one solution opened until the bath was full to the used, blood samples from donors were screened by level of the suction overflow line. State Health Laboratory Services for human immuno- deficiency virus, hepatitis B, and syphilis, and limbal Isolated Vessel Preparation swabs were cultured for bacterial growth against a The lateral posterior ciliary artery plus connective gentamicin-sensitive disc. The time between death or tissue was carefully dissected free from adherent or- operation before the tissue was placed into oxygen- + bital tissue. The vascular ring segments (1-2 mm in ated Na -Krebs solution at 4°C varied from 2-9 hr. length) were cut and mounted in the myograph sys- All ringsegment s were tested in the myograph bath on tem. One or two segments were obtained from each day 0 and a few on day 1. eye. The mounting of the ring segment was done under microscopic control (Zeiss stereomicroscope Myograph System DR, magnification X40; Zeiss, Oberkochen, Ger- The myograph was modified from another study.5 many). Two 50-/im tungsten wires, coated with gold It was custom-made and consisted of a hollow cylin- paint to render the surface smooth, were threaded dric stainless-steel block in which there was a concen- through opposite ends of the vessel lumen. We used tric cylindric depression (volume, 5 ml) which formed forceps for all vessel and remaining connective tissue the myograph bath. An isometric force transducer handling until the vessel was mounted completely on (Grass FTO3) and an XYZ microdrive system (Nari- the tungsten wires; after this, the connective tissue shige, MM 3, driven in the x direction by a hydraulic was dissected away. The tungsten wires were attached microdrive, MO-8; Narishige, Tokyo, Japan) were rigidly to the transducer and microdrive system, respectively, and the wires were kept parallel to each mounted on a base plate on either side of the organ + bath. These were used to measure the "isometric" other. The bath was filledwit h the Na -Krebs bathing tension developed by the muscle and to apply the pas- solution, and the vessel left to equilibrate for at least sive stretch, respectively. Bath fluid and bathing solu- 30 min under zero tension. The isometric tension out- tions were maintained at 37°C by circulating water put from the transducer was amplified (low level di- from a temperature bath (Techne TE-8D, Cambridge, rect current amplifier, 79122E; Grass, Quincy, MA), UK) through the stainless-steel base block and outer filtered (direct current to 3 Hz), and continuously re- jackets of the glass containers. Bath temperature corded on a six-channel chart recorder (R50; Rika- (LC87 temperature probe; T.P.S., Brisbane, Austra- denki, Tokyo, Japan). The overall compliance of the lia), pH (Activon 209 pH probe and meter; Granville, recording system was 5-10 jtm/mN; this was not corrected for in the data we present. Australia), and Po2 (using a D02M dissolved O2 meter; WPI, New Haven, CT) were monitored continuously and displayed digitally and on a chart re- Solutions and Agonists corder. They were maintained close to 37°C, 7.4, and Solutions: Vessels usually were bathed in normal 95% O2, respectively. The two organ bathing solutions Na+-Krebs solution containing: NaCl 119 mM, KC1 were delivered to the bath through a hydrostatic pres- 4.6 mM, CaCl 1.5 mM, MgCl 1.2 mM, NaHCO 15 sure head by opening a valve in the appropriate line. 2 2 3 mM, NaH2PO4 1.2 mM, and glucose 6 mM. For These solutions plus that in the bath were equilibrated 0.124 M K+-Krebs solution, the composition was KC1 with carbogen (95% O2, 5% CO2), by continuous slow 124 mM, CaCl 1.5 mM, MgCl 1.2 mM, NaHCO 15 bubbling. When a change of bath fluid was required, 2 2 3 mM, NaH2PO4 1.2 mM, and glucose 6 mM.

Downloaded from iovs.arvojournals.org on 10/02/2021 50 INVESTIGATIVE OPHTHALMOLOGY b VISUAL SCIENCE / January 1992 Vol. 33

Agonists: Agonists used were (±)-norepinephrine HC1, (-)-epinephrine, 5-HT creatinine sulfate, dopamine HC1, (-)-phenylephrine HC1, isoproterenol HC1, tyramine HC1, histamine, xylazine HC1, and acetylcholine chloride (ACH). All chemicals were obtained from Sigma (St. Louis, MO) and were dissolved in 0.9% NaCl. Stock solutions were stored at 0.5 8 1 -70°C, and fresh dilutions of 10" to 10" M were mN/mm made daily. All pharmacologic agents were pipetted directly into the muscle bath in a 50-/ul volume. The concentrations reported (10"'° to 10~4 M) are the cal- culated final concentrations in the bath solution.

Experimental Regimen After equilibration, the amplitudes of three succes- sive activating K+-induced contractions were measured at 10-min intervals to establish preparation via- bility and stability. A passive tension in the middle of the active tension range was set as described previously,5 and the cumulative maximum isometric ten- 10min sion response was determined for one agonist. The + Fig. 2. Raw data traced from the chart recording showing the bath was flushedthre e times with Na -Krebs solution cumulative dose-response curve for histamine. Note that initially to return the tension to baseline values. After 10-20 for low concentrations the response is a relaxation followed by con- min, a control K+-induced contraction was measured traction at higher concentrations. Time passes from right to left. The pre- and post-K+-Krebs contractions are marked by a K, and to ensure preparation stability, and a new agonist was + the flushing of the bath with Na -Krebs with W. Times at which tested. Agonists were tested randomly. In a few exper- histamine was added to the bath in increasing log ([M]) from 10~10 iments, the response to ACH was determined to test to 1O~3 are shown with arrows. Time scale shows 10 min, and the the integrity of the endothelial cell layer.6 In all cases, tension scale is 0.5 mN/mm. vessel relaxation indicated an intact endothelium.

Calculations and Statistical Analysis struction of concentration-response curves. These re- The maximum contractile response induced at sults are presented as average values for each segment each concentration of agonist was used for the con- of the posterior ciliary artery with standard errors. Numbers in brackets represent numbers of segments tested. The Emax% was defined as the contraction produced by the agonist at the maximum concentration 4 w w used (10~ M), and the EC50 was the log(molar concentration) at which the response was half Emax% as determined by linear interpolation. 0.5 mN/mm Results

A typical cumulative dose-response trace for epinephrine is shown in the center portion of Figure 1, -10 with the initial and final control K+-Krebs-induced contractions marked by K. Time is shown from right lOmin to left. The arrows on the dose-response curve repre- Fig. 1. Raw data traced from the chart recording showing in the sent the times at which the epinephrine was added in 10 3 center portion a cumulative dose-response curve for epinephrine. concentrations increasing from 10~ M to 10~ M in Time passes from right to left. The pre- and post-K+-Krebs contrac- + one log unit steps. W marks the times at which the tions are marked by a K, and the flushing of the bath with Na - myograph bath fluid was replaced with Na+-Krebs so- Krebs with W. Times at which epinephrine was added to the bath in 10 3 lution. The initial response is at a concentration of increasing log ([M]) concentrations from 10" to 10" are shown 9 + with arrows. Time scale shows 10 min, and the tension scale is 0.5 10" M in this case. The K -Krebs-induced contrac- mN/mm. tions were characterized by a simple rising phase be-

Downloaded from iovs.arvojournals.org on 10/02/2021 No. 1 HUMAN POSTERIOR CILIARY ARTERY PHARMACOLOGY / Yu er ol 51

fore a plateau region was reached. Figure 2 shows a 50 similar record for HIS that caused a biphasic re- O—OPHE 40-- • •XYL sponse, relaxing the ring segment at low concentra- A AIS0P 10 -O tions with a threshold of 10~ M and contracting for 30-- concentrations of 10"5 M and greater. E% Cumulative dose-response curves for each of the 20- nine agonists tested are plotted in Figures 3,4, and 5. The ordinate is the ring segment tension plotted as a 10-- + percentage of the control K -Krebs-induced contrac- 0 tion tension, henceforth called the percentage normalized response (E%). The abscissa is the molar agonist -10 concentration in the bath on a logarithmic scale. The -10 -9 -8 -7 -6 -5 -4 -3 three curves in each figure represent the mean re- Log ([M]) sponse (contraction positive, means ± standard error) Fig. 4. Cumulative dose-response curves for phenylephrine for one agonist. Threshold concentrations varied (open circles), xylazine (filled circles), and isoproterenol (open trian- 4 gles), where E% is the tension response normalized to the tension from 10-'°to 10" M. + All dose-response curves showed the conventional generated by a K contraction plotted as a function of log ([M]). sigmoidal shape for low concentrations, although some agonists produced contraction and others, relaxation. The nonspecific adrenergic agonists epineph- caused contraction at higher concentrations of 10 4 rine and norepinephrine had different sensitivities M (Emax, 6.5 ± 4.0%). There was no response to dopa- and potencies. Epinephrine (Fig. 3, filled circles) pro- mine (Fig. 5, open triangles). duced the strongest contraction of all nine agonists The 5-HT (Fig. 5, filled circles) produced a strong 10 9 with a threshold of 10~ M and Emax plateau response and sensitive contraction, with a threshold of 10~ M of 38.8 ± 7.6%. Norepinephrine (Fig. 3, open circles), and Emax of 34.0 ± 9.4%, whereas the HIS response 7 by contrast, had a threshold of 10~ M and an Emax of (Fig. 4, open circles) was biphasic, ie, relaxing at a 20.7 ± 4.2%. The specific a,-adrenergic agonist phen- threshold of 10"8 M and lower concentrations and ylephrine (Fig. 4, open circles) had a threshold of 10~7 starting to contract at a higher concentration of 6 M and Emax of 30.8 ± 5.9%, whereas the specific a2- 10~ M. adrenergic agonist xylazine (Fig. 4, filled circles) did The EC50, the concentration at which the response not stimulate a vascular response. Tyramine (Fig. 3, was one half of its maximal value, was estimated by open triangles) was only effective at eliciting a contrac- linear interpolation from the data points and the Emax. 4 tion at concentrations of 10~ M, and the resulting The threshold concentration, Emax%, and the EC50 for Emaxwas small, 11.8 ± 4.9%. Isoproterenol, a /3-adren- each agonist are given in Table 1. The mean absolute ergic agonist (Fig. 4, open triangles) relaxed the vessels values for the control K+-Krebs-induced contraction at low concentrations with a threshold of 10~8 M but were 1.76 ± 0.05 mN/mm (n = 118).

50 O OHIS 40-- • «5HT A ADOPA 30--

E% E% 20-- 10-- 00 -O -10--

-10 -20 -10 -7 -6 -10 -9 -8 -7 -6 -5 -4 -3 Log ([M]) Log ([M]) Fig. 3. Cumulative dose-response curves for norepinephrine Fig. 5. Cumulative dose-response curves for histamine (open cir- (open circles), epinephrine (filled circles), and tyramine (open trian- cles), 5-hydroxytryptamine (filled circles), and dopamine (open tri- gles), where E% is the tension response normalized to the tension angles), where E% is the tension response normalized to the tension generated by a K+ contraction plotted as a function of log ([M]). generated by a K+ contraction plotted as a function of log ([M]).

Downloaded from iovs.arvojournals.org on 10/02/2021 52 INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / January 1992 Vol. 33

Discussion high threshold contractile response with no accom- panying low concentration relaxation13 was reported The absolute mean value for maximal active ten- in the bovine isolated long posterior ciliary artery, pos- sion of the human posterior ciliary artery, produced sibly because of insufficient pretension. We showed by a K+-Krebs-mediated contraction of 1.76 ± 0.05 earlier5 that the cat ophthalmociliary artery had a mN/mm (n = 118), was greater than the value quoted graded contractile-only response to HIS, which de- for the distal ophthalmociliary artery of the cat (1 creased in strength at locations close to the eye. Thus, mN/mm)7 and slightly less than that of the ophthal- it would appear that the total ocular vascular response mociliary artery of the dog (2.0 mN/mm),5 obtained to HIS is complex and is location and concentration using the same techniques. The mean value also was dependent. When HIS responses from other vascular less than the 2.46 ± 0.38 mN/mm for human poste- beds were compared between different animals, there rior ciliary artery,3 but this was expected because the was also a large species difference, emphasizing the latter value was measured from the totally relaxed importance of measurements being done on human state (induced by papaverine) to the fully K+-Krebs- tissue. Changes to the HIS response may play a role in activated state. The ring segments were activated par- disease. For example, there is some evidence that an tially in the baseline solution of Na+-Krebs, resulting increase in HIS synthesis in the eye occurs in dia- in intrinsic myogenic tone, as demonstrated by the betes,14 and this may be a direct precursor of the in- relaxation induced by isoproterenol and HIS. The re- crease in vascular permeability,15 which plays an im- sponses to K+-Krebs-induced contractures were portant role in the early manifestations of this disease. monophasic, displaying a simple rising phase, which The ring segment response to 5-HT was contractile 8 reached a plateau. In this respect, the human poste- with an EC50 of 2 X 10" M (Table 2). The plateau rior ciliary artery differed from the cat and dog oph- magnitude was 34.0% of the K+-Krebs contraction; thalmociliary artery57 and bovine retinal arteries.8 In this is similar to that observed in the distal ophthal- these, such K+-Krebs-induced contractions had both mociliary artery for the cat.5 The low threshold for tonic and phasic components.5'7 These two compo- 5-HT was confirmed by measurements made on the nents were attributed to fast and slow stages of entry ophthalmociliary artery of the dog16 and monkey,17 of calcium ion into smooth muscle cells, mediated by the human ciliary artery,2 and the bovine long poste- Reactivated channels.9 We do not believe that the rior ciliary artery.13 Thus, the 5-HT response appears lack of a biphasic response was related to tissue deteri- to be neither location nor species specific for those oration resulting from the inevitable time delay be- sections of the ocular vasculature studied. It is proba- fore the tissue was mounted, because the biphasic na- ble that the release of 5-HT from platelets increases ture of the K+-Krebs response in the cat and dog ocular vessel tone and reduces blood flow after circula- ophthalmociliary artery was not lost even after the tory disturbances. Our findings support the hypothe- tissue had been stored for 2 days (unpublished data). sis that there is a,-adrenergic-mediated constriction in the human posterior ciliary artery, with no «2- Previously, HIS was reported to have two opposite adrenergic response. Epinephrine and norepineph- actions on blood vessels: dilation, mainly involving rine both caused substantial contractions with EC5Os finer vessels, and vasoconstriction, mainly involving of 3.1 X 10"8 M and 3.6 X 10"7 M, respectively; nor- larger vessels. In this study, the human posterior cili- epinephrine was more potent. The magnitude, thresh- ary artery relaxed at low concentrations and con- stricted at higher concentrations. A similar biphasic response was reported10 in perfused canine coronary Table 2. Threshold, E , and EC for all agonists arteries, showing that the low concentration dilator r L 50 tested and high concentration constrictor responses were

mediated by Hi and H2 receptors, respectively. We Agonist Threshold [M] Emax% ECS0 [M] have not used yet specific H, and H antagonists to 2 Histamine lO"9 2.7 ±4.0 (16) _ determine whether the same mechanisms operate in KC1 — 100 — the posterior ciliary artery. We are unaware of any Norepinephrine 10"8 20.7 ±4.2 (15) 3.6 X 10"7 8 data relating the in vivo effect of HIS on choroidal or Epinephrine io-.o 38.8 ±7.6 (12) 3.1 X 10" Dopamine NR 3.6 ± 4.6 (8) — uveal vasculature, although the retinal vasculature is Phenylephrine lO"7 30.8 ±5.9 (11) 5.9 X 10"7 sensitive to this substance. Dilatation of the retinal Tyramine io-4 11.8 ±4.9 (7) 2.0 X 10"5 vasculature occurred after infused histamine,11 and a 5-HT io-9 34.0 ±9.4 (13) 2.0 X 10"8 Isoproterenol 10"8 6.5 ±4.0 (10) — recent report demonstrated H, receptors on the endo- Xylazine NR 2.8 ± 5.4 (9) — thelial cells of the bovine retinal arteries, which, when stimulated, mediate the release of endothelium-de- Threshold in [Molar], Emax (%), and EC50 [Molar] for all agonists tested, 12 means ± SE (number of ring segments in brackets). rived relaxing factor and possibly prostacyclin. A NR means no response, and the dash (—) means not appropriate.

Downloaded from iovs.arvojournals.org on 10/02/2021 No. 1 HUMAN POSTERIOR CILIARY ARTERY PHARMACOLOGY / Yu er ol 53

old, and EC50 for phenylephrine, a specific a,-adren- results from fresh tissue on day 0 and tissue kept for 1 ergic agonist, were similar to those of norepinephrine, day showed no discernible difference in magnitude or whereas the failure of xylazine to mediate either con- relative response between agonists. However, day 2 + traction or relaxation excluded the presence of a2- tissue consistently showed larger K -Krebs-induced adrenergic receptors. Experiments on retrobulbar iso- contractions. These were reduced by day 3, and an lated vessels from other species in which the presence alteration in the relative responses from agonists was of adrenergic receptors was investigated also found seen. Because all our results were obtained from day 1 evidence for ax adrenoceptors but minimal a2 adreno- tissue, we believe they represent fresh tissue. The ceptors in the monkey,17 human,2 dog,1618 and cat.5 other unavoidable problem with donor tissue is that In addition, a-adrenergic receptors were found on bo- the patient may have ocular or systemic vascular dis- vine long posterior ciliary arteries.13 This finding eases that alter the pharmacologic responses; these agreed with in vivo data,219 showing a adrenoceptors may be confounded by systemic medications. The in cat short posterior ciliary arteries. The relaxation consistency of our results obtained from our random response to isoproterenol at low concentrations sug- selection of patients suggests that the parameters we gests jS-adrenergic receptors. These data do not agree measured were robust and insensitive to either preex- with previous results that found no evidence for such isting diseases or medication used. receptors in cat and bovine long posterior ciliary ar- In summary, therefore, these results support the ex- 21319 20 5 tery ' or cat ophthalmociliary artery. The con- istence of a,-adrenergic, HIS, and 5-HT receptors in traction observed in response to isoproterenol for mediating vasoconstriction and /3-adrenergic recep- 5 concentrations of 10~ M and more might be analo- tors in mediating vasodilation on the human poste- gous to the results found in cat ophthalmociliary ar- rior ciliary artery. No a2 adrenoceptors were detected. tery where high-concentration contractile responses The species differences in pharmacologic response of 5 to isoproterenol were not receptor specific. the ocular vasculature emphasizes the importance of Tyramine is another sympathomimetic whose using human donor tissue to gain the knowledge mechanism of action is considered to be the displace- needed to improve pathologic ocular circulations us- ment of transmitter from adrenergic axon terminals, ing pharmacologic agents. thus causing indirect adrenergic effects attributable to Key words: human, posterior ciliary artery, isolated vessel, endogenous sympathetic agents. The response to tyra- histamine, 5-HT, adrenergic agonists mine had a high threshold, implying either weak adrenergic innervation or depleted catecholamine Acknowledgments stores in adrenergic nerve endings. This was similar to The authors thank Michael Brown, Julie Crewe, and Wilf the effect of tyramine on dog ophthalmociliary ar- Brewster for their expert technical assistance and Drs. tery21 and simian ophthalmic and ciliary arteries.17 Richard Cooper and Ian McAllister for assistance with tis- These also show a high-threshold contractile response sue supply. for tyramine that could be blocked by imipramine (an References uptake block) when trying to separate endogenous norepinephrine from exogenous norepinephrine re- 1. Weiter J, Schacher RA, and Ernest T: Control of intraocular blood flow: 2. Effects of sympathetic tone. Invest Ophthalmol sponses. However, the cat ophthalmociliary artery 12:332, 1973. had a low-threshold contractile response to tyramine; 2. Chandra SR and Friedman E: Choroidal blood flow: 2. The therefore, the response mechanism is different in this effects of autonomic agents. Arch Ophthalmol 87:67, 1972. species.7 3. Nyborg NC and Nielson PJ: The level of spontaneous myo- There was no vascular response to dopamine. This genic tone in isolated human posterior ciliary arteries decreases with age. Exp Eye Res 512:711, 1990. contrasts with the results in cat ophthalmociliary ar- 7 4. Ohkubo H and Chiba S: Vascular resistivities of isolated and tery where there was a significant but small vasocon- perfused human ciliary arteries. Jpn J Ophthalmol 32:450, striction, although the threshold was high (10~5 M). 1988. The use of postmortem and postoperative tissue 5. Yu D-Y, Alder VA, and Cringle SJ: In vitro characterization of after a relatively long delay between tissue removal the mechanical properties of canine ophthalmic artery. Exp Eye Res 51:729, 1990. and bath mounting always raises the question of the 6. Furchgott RF and Zawadski J V: The obligatory role of endothe- relationship of the acquired data to that which would lial cells in the relaxation of arterial smooth muscle by acetyl- be observed in vivo or in fresh tissue. To address this choline. Nature 288:373, 1980. problem partially, we compared the results for cat 7. Yu D-Y, Su E-N, Alder VA, Cringle SJ, and Mele EM: Pharma- ophthalmociliary artery, taken fresh from the animal cological and mechanical heterogeneity of cat isolated ophthalmociliary artery. Exp Eye Res 1991 (in press). and mounted immediately, with results obtained 8. Hoste AM, Boels PJ, Brutsaert DL, and De Laey JJ: Effect of from the same ophthalmociliary artery, after keeping alpha-1 and beta agonists on contraction of bovine retinal resis- it 1-3 days in oxygenated Na+-Krebs solution. The tance arteries in vitro. Invest Ophthalmol Vis Sci 30:44, 1989.

Downloaded from iovs.arvojournals.org on 10/02/2021 54 INVESTIGATIVE OPHTHALMOLOGY G VISUAL SCIENCE / January 1992 Vol. 33

9. Hogestatt ED and Andersson KE: Mechanisms behind the bi- 15. Cunha-Vaz JG: Blood-retinal barriers in health and disease. phasic contractile response to potassium depolarization in iso- Trans Ophthalmol Soc UK 100:337, 1980. lated rat cerebral arteries. J Pharmacol Exp Ther 228:187, 16. Ohkubo H and Chiba S: Pharmacological analysis of vasocon- 1984. striction of isolated canine ophthalmic and ciliary arteries to 10. Nakane T and Chiba S: Characteristics of histamine receptors alpha-adrenoceptor agonists. Exp Eye Res 45:263, 1987. in the isolated and perfused canine coronary arteries. Arch Int 17. Ohkubo H and Chiba S: Vascular reactivity of simian ophthal- Pharmacodyn Ther 290:92, 1987. mic and ciliary arteries. Curr Eye Res 6:1197, 1987. 11. Regnualt R and Riva C: A study of vasoactive drugs in the 18. Ohkubo H and Chiba S: Vascular responsiveness to periarterial retinal circulation of a extracorporeal perfused eye. Exp Eye electrical nerve stimulation on canine ophthalmic arteries. Res 11:144, 1971. Curr Eye Res 7:607, 1988. 12. Benedito S, Prieto D, Nielson PJ, and Nyborg NC: Histamine 19. Morgan WE and Macri FJ: Vascular response of the posterior induces endothelial-dependent relaxation of bovine retinal ar- segments of the cat eye. Arch Ophthalmol 79:779, 1968. teries. Invest Ophthalmol Vis Sci 32:32, 1991. 20. Dawis SM and Niemeyer G: Dopamine influences the light 13. Dalski HF: Pharmacological reactivity of isolated ciliary arter- peak in perfused mammalian eye. Invest Ophthalmol Vis Sci ies. Invest Ophthalmol 13:389, 1974. 27:330, 1986. 14. Carroll WJ, Hollis TM, and Gardner TW: Retinal histamine 21. Ohkubo H and Chiba S: Vascular responses of ophthalmic ar- synthesis is increased in experimental diabetes. Invest Ophthal- teries to exogenous and endogenous norepinephrine. Exp Eye mol Vis Sci 29:1201, 1988. Res 48:539, 1989.

Downloaded from iovs.arvojournals.org on 10/02/2021