Effect of epinephrine in vitro on the morphology, phagocytosis, and mitotic activity of human trabecular endothelium

Exp.

Eye Rm. (1984)

39, 731-744

Effect of Epinephrine In Vitro on the Morphology, Phagocytosis, and Mitotic Activity of Human Trabecular Endothelium BRENDA

J. TRIPATHI

Eyr Research Laboratories,

(Received 23 March

AND

RAMESH

c’. TRIPATHI

The University of Chicago, 939 East 57th Street. Chicago, Illinois 60637. U.S. A. 1984 and accepted 27 July

1984, New

York)

Epinephrine exerts a direct effect on cell morphology, phagocytosis and mitotic activity of human t,rahecular endothelium in primary culture. Its action is probably mediated through both beta and alpha adrenoceptors in a dose-time dependent manner. Younger cells and cells that were loosely attached to the substrate were found to be affected more rapidly and severely than were older cells and those in confluent regions where cell-to-cell attachment and stress fibers were well established. Continuous exposure to epinephrine at a concentration of 10m5 x led to cessation of the normal cytokinetic cell movements, inhibition of mitotic and phagocytic activity, marked cell retraction, separation from the substrate, and, by 4-5 days, degeneration of cells. Similar, but less marked changes were seen with a concentration of 10m6 M, the cell degeneration becoming apparent after one week of exposure. A still weaker concentration of epinephrine, lo-’ M. did not result in cell degeneration even after 10 days of exposure and observation. On complete withdrawal of the drug, the cellular effects were reversible even after 3 days’ exposure to lo+ .M and 5-7 days’ exposure to 1Om6 M epinephrine. The action of epinephrine was partially blocked by pretreatment of cultured trabecular cells with the beta-blocker, timolol. Available evidence suggests that the mechanism of action of epinephrine is mediated through both beta and alpha adrenoceptors. and that it intimately involves the cytoskeletal system of the cells. Extrapolation of our findings in vitro suggests that use of maximal doses of epinephrine over a prolonged time may contribute to tissue damage in certain conditions of glaucoma. Key words: adrenoceptors; cyclic AMP; cell contraction; cell degeneration: calcium ions; glaucoma; intraocular pressure; prostaglandins; beta blockers; tissue culture; trabecular meshwork.

1. Introduction The pressure-lowering effect of epinephrine on the eye is believed to be mediated through adrenergic receptors in the trabecular meshwork. However, experimental adrenergic stimulation ofthe aqueous outflow pathway in vivo has produced equivocal evidence as to whether alpha and/or beta receptors exist in the meshwork, and which of these play a role in lowering intraocular pressure. Overall, data from rabbit, experiments suggest that both alpha and beta agonist activity can increase true facility by egress of aqueous humor through the trabecular meshwork (Sears and Barany, 1960; Sears and Sherk, 1964; Sears, 1975; Potter and Rowland, 1978: Rowland and Potter, 1979; Boas, Messenger, Mittag, and Podos. 1981). In non-human primates, however, the beta response is considered more important than that of alpha (Bill, 1969; Bill and Heilmann, 1975; Thomas, 1980: Potter and Rowland, 1981: Boas et al., 1981; Neufeld and Bartels, 1982). Marked differences among species in ocular structures and in the response t’o It is a privilege to dedicate this paper to the memory of our good friend, the late Professor Cole, whose association and scientific collaboration we so much enjoyed during our decade-long at the Institute of Ophthalmology, University of London. His remarkable original contributions understanding of ocular physiology are most inspiring. 001&4835/84/120731+14$03.00/0

0 1984 Academic

Press Inc. (London)

David tenure to thp

Limited

732

Ii. .I. THI I’il’l’H

I :1X1) IC. (‘. ‘I’IiI l’i\‘[‘Hl

adrenoceptor stimulation of the outflow pathway make it) tliflicult to rstrapolatt* animal data to man (Sears and Neufeld, 1975: Potter. 1981: Potter and Rowlantl. 1981; Mishima, 1982) and the precise mechanism of action of epinephrine as ail anti-glaucoma agent remains poorly understood. It has been considered most likeI> that beta-adrenergic receptors activated by epinephrine are on the endothelial cells lining the outflow pathway: but it is unknown whether these rells undergo alterations in their rate of mueopolysaccharide turnover or phagocytosis. or have increased permeability, or are subject to some other change in response to epinephrine (Neufeld and Bartels, 1982). To evaluate whether epinephrine exerts a direct’ action on the trabecular cells of the aqueous outflow system in man. we studied the morphology, phagocytosis, and mitotic activity of pure human trabecular endothelium cultured in vitro and subjected to various concentrations of epinephrine.

2. Materials

and Methods

Human eyes from individuals of various ages (2-60 years), obtained 672 hr post mortem, were used as a source of trabecular meshwork explants. Precisely excised tissue blocks of trabecular meshwork were placed either in Falcon tissue culture flasks or on the coverglasses of disposable tissue culture chambers. Cell cultures of human trabecular endothelium were initiated and characterized as described previously (Tripathi and Tripathi, 1982a). The culture medium consisted of MEM Eagle’s medium with 10 o/0 fetal bovine serum (Gibco) and 10% penicillin-streptomycin at pH 7.4. Cultures were incubated at 37°C. As part of the experimental protocol, the medium was replaced every 24 hr. More than 100 primary cell cultures were used for the experiments and for control studies. Pure 1-epinephrine hydrochloride without preservative (Elkins-Sinn, N.J.) was diluted in tissue culture medium to obtain concentrations ranging from 10e5 to lo-’ M. This range was chosen in view of the reported concentrations of epinephrine in the aqueous humor after topical installation of epinephrine hydrochloride or its prodrug in the conjunctival sac (Kaufman and Rentzhog, 1981; Mishima, 1982). Cultures were exposed to epinephrine for various periods, up to a maximum of 10 days. Although the constituents of the tissue culture medium may have prevented oxidative breakdown of epinephrine, the antioxidant superoxide dismutase (1 unit, 3 pg ml-‘) was added to provide some further protection. To evaluate whether the action of epinephrine could be suppressed by beta-blockers. we pretreated selected cultures with pure timolol maleate (Merck, Sharp and Dohme) at a concentration of 10mg M for 2 hr prior to their exposure to epinephrine at a concentration of lop5 M. The response of the cells was monitored under a 10 x objective by phase-contrast, timelapse micrography, and, at selected intervals, the cultures were fixed for detailed morphologic analysis. For time-lapse studies, one frame every minute was recorded either on 16 mm black and white film or, more recently, on NV-T120 video tape with a low light television camera and time-lapse video recorder. The movie films were analyzed at 24 frames per second, and the video tape was studied in real time as well as by frame-hold analysis. A selected area of the culture was recorded continuously for 24 hr prior to drug administration and for up to 10 days during the experimental protocol. For localization of actin filaments in the cells, we used an indirect immunofluorescence staining technique (cf. Tripathi and Tripathi, 1980a) on cultures fixed at selected intervals in either acetone (cells grown on glass) or 10% form01 saline (cells grown on plastic). For conventional bright-field microscopy, fixed cells were stained with methylene blue. For scanning electron microscopy, cells fixed in 3 o/0 buffered glutaraldehyde and postfixed in 1 o/0 buffered osmium tetroxide were dehydrated in ascending grades of ethanol. Following critical point drying, the specimens were coated with gold-palladium. The phagocytic activity of the cells was determined with carmine particles and latex beads which were used as markers and were introduced into the culture chambers as described previously (Tripathi and Tripathi. 1980b, 1982a). To measure cyclic AMP levels in selected trabecular cell cultures as well as in freshly excised human trabecular meshwork explants exposed to epinephrine at a concentration of 10m6 M

EPISEPHRINE

ACTION

ONTRABECULAR

ENDOTHELIL-M

533

for up to 2 hr, we used a cyclic AMP radioimmunoassay technique (Becton-Dickinson). Control samples were not exposed to epinephrine. The cyclic AMP was extracted from the pelleted trabecular endothelial cells with 5 y0 trichloroacetic acid containing 3H-labe1ed cyclic the pelleted tissue samples were AMP. Following homogenization and centrifugation, analyzed for protein content according to the Lowry method (Lowry, Rosebrough, Farr, and Randall, 1951). The cyclic nucleotide-containing supernatant was acidified with @1 N hydrochloric acid and extracted with ether. The dried pellet thus obtained was incubated for 15-20 hr at 5°C with Y-labeled antibody. After precipitation with 60% ammonium sulfate, the radioactivity of the pellet was counted in a Beckman scintillation counter. The cyclic AMP levels were determined by comparison to known standard curves. All measurements were carried out in triplicate. Cyclic AMP levels were expressed as picomoles per milligram of protein. 3. Results The effect of the various concentrations of epinephrine on primary cultures of trabecular endothelium was consistent and reproducible, with only minor variations. The proliferating cells at the periphery of the culture demonstrated a more rapid response to epinephrine than did the cells in the central confluent region. Similarly, younger cultures which had not yet reached confluence were more susceptible than were older cultures. We also noted that cells grown on a glass substrate were affected more rapidly than were those on plastic. Documentation by continuous phase-contrast, time-lapse micrography revealed striking changes in the dynamic cellular movement of the cultured trabecular endothelium which were not readily apparent in sequential still photomicrographs. The times given in the description that follows are those for the response of cells propagated in Falcon flasks. Effect of epinephrine at 10w5 M Within 15 min of exposure of the cells to epinephrine, the normal eytokinetic movement of intracellular organelles as well as the morphogenetic movement of the cells ceased, and the cells appeared to be in a state of suspended animation. Mitotic activity also ceased rapidly. Cell contraction gradually caused the cytoplasmic extensions and filopodial processes to retract [Fig. l(a) and (b)]. By 3&48 hr of exposure, the majority of the cells had assumed an oval or round profile. Within 4-5 days, most cells lost their attachment to the substrate and floated into the culture medium [Fig. 2(a) and (b)]. The process of retraction was characterized by loss of the normal flattened profile of the cells and loss of cell-to-cell attachments. Initially, the cells remained elongated and spindly, with slender, spiky projections. The central region of the cell maintained almost normal dimensions but subsequently this region became rounded and appeared more granular as the intracellular organelles accumulated. The retracting cytoplasmic processes along the cell borders became extremely thin and tenuous [Fig. 3(a)]. In the central areas of the cultures, the cells presented a stellate profile, but at the culture periphery they had more bizarre shapes. Finally, as the thin cytoplasmic projections wit.hdrew, the cells assumed a rounded profile and contained small refractile granules and a dense nucleus. As seen by scanning electron microscopy, a remarkable feature of t*he retracted cells was an abundance of microspikes [Fig. 3(b) and (c)]. Rounded cells had a ruffled or crenated surface [Fig. 3(c)]. With the progression of degenerative changes, the cells lost their attachment to the substrate. Indirect immunofluorescence staining for actin revealed fragmentation and dissolution of fiber bundles (Fig. 4).

734

13. .J. ‘l’KIt’.~‘I’Hl

.\Sl)

K. (‘.‘l’K,Il’ATHI

FIG. 1. (a) A vonflurnt cultuw of trithrrular mtlothelial cells prior to rxpowrr to epinrphrinr. The edge of the culture can he seen at the top. Arrow denot,w mitotic figures. Phase-contrast photomicrograph x 126. (h) Appearance of the culturr aftrr 2% hr of rxposuw to 1OF M rpinrphrinn. Cellular retraction is more marked at thr periphery of the cvlturr (top) than in the wnfluent region (hotstom). Phase-cwntrast photomirrograph x 126.

EPIKEPHRINE

ACTION

OR’TRARECULAR

ENDOTHELIUM

735

Fro. 2. (a) Hame culture as in Fig. l(a) after 46 hr of exposure to 10e5 M epinephrine. The cellular retraction process has now involved almost all cells of the culture. Phase-contrast photomicrograph x 126. (h) Same culture aa in (a) after 96 hr of exposure to 1OV M epinephrinr. Most cells show a rounded profile. degenerative changes. and loss of attachment to the substrate. Phase-contrast photomirrograph x 126.

736

13. J. TRIPATHI

AXI)

R. (‘. TKIPATH

I

FIG. 3. (a) The retraction of culturrd trabecular endothelial cells is marked bv thin and twuous cytoplasmic processes along cell borders after 22 hr of exposure to 1O-5 H epineph&. Phase-contrast photomicrograph x 406. (b) and (c) Scanning electron micrographs showing various stages in wllular retraction. The thin cytoplasmic processes denote the retracting cytoplasm. Arrows indicate microspikes on the cell surface. Asterisks indicate rounded profile of cells. Thirty-six hours of exposure to 10-j epinephrine. x 1295.

M

After 8 hr of exposure to epinephrine, the phagocytic capacity of the cells for carmine particles and latex spheres, which were used as markers, was considerably reduced. Quantitative studies on this reduction are in progress. Withdrawal of epinephrine after a 3-day exposure of the culture to the 10P5 M concentration led to a normal morphologic appearance and normal proliferative activity of the cells within 48 hr. Pretreatment of cultures with timolol maleate at a concentration of 10P6 M retarded, but did not completely abolish, the effect of epinephrine. Cell movement and mitotic activity were reduced gradually during the first 24 hr of exposure, and by 48 hr, cell retraction and loss of cell-to-cell attachment became apparent. The long-term blocking effect of timolol, however, remains to be evaluated. Effect of epinephrine

at lop6

M

Exposure to epinephrine at a concentration of 10P6 M caused less marked changes than were seen at 10e5 M, and these changes occurred at a slower pace. Within K-30 min of exposure, the cytokinetic movement of intracellular organelles as well as the morphogenetic movement of the cells slowed down. Time-lapse studies further revealed that those cells undergoing mitosis at the time of administration of epinephrine never completed the cell division, and no new mitotic figures were seen; prior to drug exposure, the mitotic activity of the cells at the periphery of primary confluent cultures was 50 ( f 14) divisions per 500 cells per 24 hr, and that of the central cells was 2.5 (f3) divisions per 500 cells per 24 hr. Following a slow initial cell

EPINEPHRINE

ACTION

ON

TRABECULAR

EKDOTHELIUM

737

Fro. 4. Immunofluorescence staining of cultured trabecular end&h&al cells. (a) Fragmentation and dissolution of actin fiber bundles 40 hr after exposure to LO+ M epinephrine. (b) Normal control. Ph~~tomicrographs x 2030.

13. .l.TItIP.-Z’I’HI

ANI)

It. (‘.TRII’ATHI

FIG. 5. (a) Retrttction of cultured trabewlar endothelial cells following 96 hr of exposure to IOF 11 epinephrine. A numhrr of cells have a rounded profile (arrow). Phase-contrast photomirrograph x 252. (h) (‘ultured trahecular cells showing degenrrativv changes and rounded profile as wrll as xrpawtion fiv~rn the substrate after 7 days of rxpowre to IOF M rpinrphrine. I’hasr-wntrast l,hotornic,~[,p~il;lh x 252.

contraction. retraction of the cells became most) apparent by 4X hr of continuous exposure to epinephrine. Cell rounding, however, was not apparent until 06 hr [Fig. 5(a)]. After 5-7 days’ continuous exposure to epinephrine, cell retraction and rounding had progressed to separation of the majorit,y of cells from the substrate [Pig. 5(b)]

EPISEPHRIKE

ACTIOK

OK

TRABECULAR

EN’DOTHELII’M

739

and eventually to cell degeneration. Significant cell loss occurred by 7-10 days. The phagocytic activity ofthe cells for carmine and latex markers was reduced significantly by 16 hr of continuous exposure to epinephrine at this concentration. Even after 5-7 days of exposure, however, recovery of normal morphogenetic movement and of mitotic, phagocytic, and proliferative activity of the cells was observed within 48 hr after withdrawal of epinephrine. Measurement of the level of cyclic AMP on freshly excised explants as well as on confluent cultures exposed to epinephrine for 15 min gave values of 53.1 ( + 1.2) and 65.4 (Ifi2.5) pmol (mg protein)-*, respectively, compared to 122 (+ 1.1) and 14% (+ 1.4) pmol (mg protein)-’ for untreated controls. Effect of epinephrine

at 10e7

M

With this concentration of epinephrine, we observed a reduction in the cytokinetic movements and some minor cellular contraction during the first hour of exposure, as well as some slowing down of mitotic and phagocytic activity, but the marked cellular retraction seen at higher concentrations did not occur. Cellular degeneration was not observed even after 10 days’ exposure of cultures to this concentration of epinephrine. 4. Discussion From the results of our experiments on cultured human trabecular endothelial cells, it is apparent that epinephrine at concentrations of 10e6 M and higher causes inhibition of cytokinetic movement, cellular retraction, cessation of mitosis, and suppression of phagocytic activity. In order to explain the possible mechanisms involved in these processes, we need to review the current concept of the sequence of events in epinephrine action. Epinephrine is believed to stimulate alpha and beta adrenoceptors (cf. Potter and Rowland, 1981; Weiner, 1982; Bylund and U’Prichard. 1983). The beta component, activates the membrane-bound enzyme, adenylate cyclase, which catalyzes the conversion of adenosine triphosphate (ATP) to cyclic adenosine 3’.5’-monophosphate (cyclic AMP). In turn, the cyclic AMP activates certain protein kinases within the cell. and this leads to phosphorylation of various proteins. One of these proteins is t’he myosin light chain that regulates ATPase-actin activity and initiates cell contraction (cf. Dedman, Brinkley, and Means, 1979: Rodger and Bowman, 1983). Phosphorylation of certain membrane proteins also regulates intracellular calcium levels. Under the influence of increased cyclic AMP levels, calcium ions not only are released from intracellular membranes, but’ also enter the cell from the extracellular environment due to membrane hyperpolarization. Calcium ions regulate the prot,ein kinases and therefore reinforce the sequence of events involved in the production of cell contraction (Dedman et al., 1979: Adelstein, Srordilis, and Trotter. 1979: Rodger and Bowman, 1983). Furthermore, and perhaps more importantly, calcium ions influence the function of certain cross-linking and end-binding proteins which affect, the polymerization of actin filaments (Korn, 1982). Cross-linking proteins, which aid in the organization of actin filaments into loose networks or tight bundles. are inhibited by calcium ions. End-binding or capping proteins, however. are dependent on calcium ions, and these proteins cause a redistribution of the a&ion filament lengt,hs which results in more numerous and shorter filaments. In our cultured cells of trabecular endothelium, the levels of cyclic AMP were raised wit,hin 15 minofepinephrineadministration; thisresult parallelspreviousexperimental

740

B. J. TRIPA’I’HI

ASD

R. (‘. TRIPATHI

findings in vivo (Neufeld, Jampol, and Sears, 1972; Neufeld, Chavis, and Sears. 1953: Boas et al., 1981). However, we also found that the effect of epinephrinr was onI> partially blocked by the administration of the non-selective beta-blocker. timolol maleate. This finding has two possible explanations. The first is that the competitive uptake of the antagonist was not sufficient to block all of the agonist sites; the second. and perhaps more likely, explanation is that alpha-adrenoceptors are also present in the cultured trabecular cells. The alpha-adrenergic action of epinephrine is known to have two components, alpha-l and alpha-2. Stimulation of alpha-l receptors increases turnover of the cellular phospholipid, phosphatidylinositol, which allows changes in cellular calcium flux and results in an increase in the number of cytosolic calcium ions (Exton, 1981; Bylund and UPrichard, 1983; Putney, 1983). This increase is probably responsible for the induction of cell contraction, as outlined above for beta-adrenergic stimulation. Activation of alpha-2 receptors leads to a decrease in cyclic-AMP levels and results in an effect opposite to that of beta-adrenergic stimulation (Exton, 1981; Weiner, 1982; Bylund and UPrichard, 1983). The raised levels of cyclic AMP in our investigation suggest that either alpha-2 receptors are not present or the beta action of epinephrine predominates over their inhibitory effect. Further experiments are in progress in which we are using specific alpha and beta agonists and antagonists to determine which alpha and beta receptors are present. The observation by Waitzman, Woods, and Cheek (1979) that indomethacin is capable of suppressing the action of epinephrine in vivo led to the suggestion that alpha-adrenoceptor stimulation is mediated by prostaglandins. However, indomethatin is also a substantial anatagonist of calcium action (Northover, 1977; Potter and Rowland, 1981). Since phosphatidylinositol is enriched with arachidonic acid, it seems likely that this fatty acid serves as the substrate for prostaglandin synthesis during turnover of the phospholipid. In fact, the release of arachidonate appears to be a calcium-mediated event and may, therefore, only participate downstream from the initial calcium-mobilizing mechanism. It is of interest that passaged cultured human trabecular cells have been shown to synthesize large quantities of prostaglandin (Weinreb, Mitchell, and Polansky, 1983). Many other cells in culture, particularly endothelial cells and cells which have been passaged several times, synthesize prostaglandins in greater quantity than they normally do in vivo (Gordon and Pearson, 1982) without any deleterious effect; this may only reflect that the cells have become adapted to their environment in vit,ro. In our experiments, however, we only used primary cell cultures. Moreover, the available evidence, recently reviewed by Putney (1983), suggests that prostaglandin formation is not an early obligatory step in the response of cells to alpha-adrenoceptor stimulation. Since alterations in cell shape are associated with general cellular functions such as locomotion, mitosis, cytokinesis, secretion of macromolecules, and phagocytosis (Adelstein et al., 1979), the involvement of the cytoskeletal system in the action of epinephrine explains the observed response of the cells. Cell-to-cell and cell-to-substrate attachments are mediated through actin filaments (Pegrum and Maroudas, 1975; Goldman, Yerna, and Schloss, 1976; Weihing, 1979). Apparently, epinephrine loosens this attachment as the actin filaments are redeployed to effect cell contraction. The greater effect of epinephrine on the cells at the periphery of the culture, compared with that on the cells in the central, more confluent region, is probably related to differences in the age of the cells, to cell crowding, and to differences in t,he adhesion of the cells to the substrate. The peripheral cells are younger and more mobile, and they adhere more loosely than do the older and confluent cells, which generally have

EPISEPHRINE

ACTION

ON

TRABECULAR

ENDOTHELIUM

741

greater adhesion not only to the substrate, but also to each other. The role of substrate adhesion in the response of the cells to epinephrine is further apparent from our observations that the cultures harvested on glass showed relatively rapid retraction compared with those harvested on plastic; evidently, the adhesion to glass is more tenuous than that to plastic. Our studies of live cultures and of indirect immunofluorescence staining for actin filaments revealed that high doses of epinephrine and prolonged exposure lead to fragmentation and destruction of the cytoskeletal system. Our observation of epinephrine-induced cessation of mitotic activity can be related to the increased levels of calcium ions, since raised calcium levels are known to cause depolymerization of microtubules, which are essential to the formation of the mitotir spindle (Weisenberg, 1972; Lee and Timasheff, 1977; Adelstein et al., 1979; Sanger and Sanger, 1979). Epinephrine has also been shown to inhibit mitosis of other cells in the eye, such as those of the cornea1 epithelium (Krejci and Harrison, 1970). Normally, the endothelium lining the trabecular meshwork does not undergo cell division in vivo, but in a tissue culture environment the cells are stimulated t’o multiply by mitosis until a confluent monolayer culture is established. At the periphery of primary cultures, however, where contact inhibition is less, mitotic activity continues at a slow rate for several months. This may also be a factor in determining the more marked response to epinephrine by the peripheral cells compared to the established confluent cells. The fact that administration of epinephrine inhibits mitotic activity is further indicative of its effect on the cytoskeletal system of the cultured trabecular endothelial cells. The cytotoxic effect of epinephrine at high concentration observed in our experiments in vitro may be unrelated to its mechanism of action mediated through cyclir nucleotides, calcium ions, and the contractile protein system of the cells. Even though we took precautions to prevent oxidative breakdown of epinephrine, it is likely that with prolonged exposure and repeated administration, oxidative free radicals were produced. The lack of sufficient protective enzymes, especially glutathione, to combat the oxidative insult is one possible explanation for the death of the trabecular cells exposed to 10e6 M and higher concentrations of epinephrine. Nevertheless, the striking cessation of cytokinetic movement within the cells after only 15-30 min of exposure to epinephrine, as observed by time-lapse photomicrography, would suggest an early and dramatic effect on the cellular kinetics and cytoskeletal system. This experiment in vitro does not explain how epinephrine increases the facility of aqueous outflow in vivo, but raises the possibility of an increase in trabecular porosit’y as a consequence of direct modulation of cell shape. The normal function of the endothelial cells lining the meshwork and trabecular wall of Schlemm’s canal is intimately related to their cytoskeletal system (Tripathi, 1977 ; Tripathi and Tripathi, 1982b). Our present study shows that epinephrine indirectly exerts an effect on the cellular contractile proteins. It may thus influence many contractile protein mediated properties of the trabecular cells such as synthetic activity, phagocytosis, maintenance of cell shape, and reactivity to neuronal and hormonal agents (Tripathi and Tripat.hi. 1980a). The implications of our present findings for the clinical situation can only be arrived at by extrapolation and conjecture. If a concentration of epinephrine greater than 10P6 M were to be maintained in the trabecular meshwork in vivo (despite a washout effect by the aqueous humor and detoxification by adjacent tissues), this could increase the risk of separation of those trabecular cells which are not firmly attached to the trabecular beams, or it may even cause cell degeneration and eventual cell loss. 26-2

742

I<..I.TKIPBTHI

ASI)

K.('.'T'l~IPATHI

The latter may partly explain the loss of’ cellularity in glauc*omatous patients trewt~rtl with a maximal therapeutic dose of epinephrine for ext,ended periods. Prolonged IISC of epinephrine at high concentration therefore may be detrimental to the t,raberulal, system especially in those conditions where trabecular cell adhesion t,o the hrarns has already been diminished, for example, in inflammatory and traumatic insults to the anterior segment of the eye. The epinephrine-mediated suppression of phagocytic activity of the t.rabecular cells that we observed could further imply that this drug at a high concentration may not be suitable for prolonged use in those patients with glaucoma in whom increased phagocytic activity of the trabecular cells is considered beneficial (Tripat#hi and Tripathi, 1980a). Our findings, however, do not imply that epinephrine should not be used at all in the therapy of glaucoma; but they merely add a note of caution on the possible cytotoxic effect of maximal concentrations of this drug. ACKNOWLEDGMENTS This work was supported by a grant (EY-03747) from the National Eye Institute, Bethesda, Maryland. Preliminary investigations on cyclic AMP were carried out with the help of Dr Meena Rao. Professors E. W. Taylor and J. D. Kohli provided helpful suggestions for the discussion. The continued help of the National Diabetes Research Interchange and of the Illinois Society for the Prevention of Blindness in providing human tissue for our culture studies is gratefully acknowledged. The electron microscope facilities were kindly made available by Professor Hewson Swift. REFERENCES Adelstein, R. S., Scordilis, S. P. and Trotter, J. A. (1979). The cytoskeleton and cell movement: general considerations. Meth. Achiev. Exp. Pathol. 8, l-41. Bill, A. (1969). Early effects of epinephrine on aqueous humor dynamics in vervet, monkeys (Cercopithecus

Exp.

ethiops).

Eye Res.

8, 35-43.

Bill, A. and Heilmann,

K. (1975). Ocular effect of clonidine in cats and monkeys (&zcaca iv-us). Exp. Eye Res. 21, 481-8. Boas, R. S., Messenger, M. J., Mittag, T. W. and Podos, 6. M. (1981). The effect of topically applied epinephrine and timolol on intraocular pressure and aqueous humor cyclic-AMP in the rabbit. Exp. Eye Res. 32, 681-90. Bylund, D. B. and U’Prichard, D. C. (1983). Characterization of alpha-l and alpha-2 adrenergic receptors. Int. Rev. Neurobiol. 24, 343-431. Dedman, J. R., Brinkley, B. R. and Means, A. R. (1979). Regulation of microfilaments and microtubules by calcium and cyclic AMP. In Advances in Cyclic Nucleotide Research, Vol. 11 (Eds Greengard, P. and Robison, G. A.). Pp. 131-74. Raven Press, New York. Exton, ,J. (1981). Molecular mechanisms involved in alpha-adrenergie receptors. Mol. Cell. Endocrin.

23, 23.3-64.

Goldman, R. D., Yerna, M.-J. and Schloss, ,J. A. (1976). Localization microfilaments and related proteins in normal and virus transformed Struct.

Gordon.

and organization cells. J. Supramokc.

of

5, 155-83.

J. L. and Pearson,

J. D. (1982). Responses of endothelial cells to injury. In Cell (Eds Nossel, H. L. and Vogel. H. J.). Pp. 433-54. Academic Press, New York. Kaufman, P. L. and Rentzhog. L. (1981). Effect of total iridectomy on outflow facility responses to adrenergic drugs in cynomologus monkeys. Exp. Eye Res. 33, 65-74. Korn, E. D. (1982). Actin polymerization and its regulation by proteins from non-muscle cells. Pathobiology

Physiol.

Krejci.

of the

Endothelial

Revs 62, 672-737.

L. and Harrison, R. (1970). Epinephrine effects on cornea1 cells in tissue culture. Arch. Ophthalmol. 83, 451-6. Lee, J. C. and Timasheff, S. N. (1977). In vitro reconstitution of calf brain microtubules: effects of solution variables. Biochemistry 16. 1754-64.

EPINEPHRINE Lowry,

ACTION

OK

TRABEC’ULAR

EiKDOTHELIUM

713

0. H., Rosebrough, N. J., Farr, A. L. and Randall, R. J. (1951). Protein measurement with the Folin phenol reagent. J. Biol. Chem. 193, 265-75. Mishima, S. (1982). Ocular effects of beta-adrenergic agents. Surv. Ophthalmol. 27. 187-208. Neufeld, A. H. and Bartels, S. P. (1982). Receptor mechanisms for epinephrine and timolol. In Basic Aspects of Glaucoma Research (Ed. Lutjen-Drecoll, E.). Pp. 113-22. Schattauer. Stuttgart. Neufeld, A. H., Chavis, R. M. and Sears, M. L. (1973). Cyclic-AMP in the aqueous humor: the effects of repeated topical epinephrine administration and sympathetic denervation. Exp. Eye Res. 16, 265-72. Neufeld, A. H., ,Jampol, L. M. and Sears, M. 1~. (1972). Cyclic-AMP in the aqueous humor. The effects of adrenergic agents. Exp. Eye Res. 14. 242-W. Northover. B. J. (1977). Indomethacin - a calcium antagonist. Gen. Pharmac. 8, 293-6. Pegrum. S. M. and Maroudas, N. G. (1975). Early events in fibroblast adhesion to glass. An electron microscopic study. Exp, Cell Res. 96, 416-22. Pott’er, D. E. (1981). Adrenergic pharmacology of aqueous humor dynamics. f’h’h.armaco/. R~K 33. 133-53. Pott’er, D. E. and Rowland, J. M. (1978). Adrenergic drugs and intraocular pressure: effec+s of select,ive beta-adrenergic agonists. Exp. Eye Res. 27. 615-25. Potter. D. E. and Rowland, J. M. (1981). Adrenergic drugs and intraocular pressure. Gr//. Ph,armcol. 12, l-13. Putney. J. W. (1983). Phosphatidylinositol metabolism and alpha-adrenoceptor mechanisms. In Adrenoceptors and Cat&olamine Artion. Vol. 1. Part B (Ed. Kunas, G.). Pp. 51-64. John Wiley. New York. Rodger. I. W. and Bowman, \il’. C. (1983). Adrenoceptors in skeletal muscle. In Adrenoceptors anal fktwholamine Action, Vol. 1. Part B (Ed. Tunas. G.). Pp. 123-55. ,John Wiley. Se% York. Rowland. J. M. and Potter, D. E. (1979). Effects of adrenergic drugs on aqueous c*\MP and cGMP and intraocular pressure. Albrerht c:on Graefes Arch. KlirL. Exp. Ophthalmol. 212. 65-75. Sanger. ,J. W. and Sanger, J. M. (1979). The cytoskeleton and cell division. Meth. Achirv. Exp. Pathol. 8, 110-42. Sears. M. L. (1975). Catecholamines in relation to the eye. In Handbook of Physiolog,q. section on endocrinology (Eds Astwood, E. and Creep. R.). Pp. 552-90. American Physiologic+al Society, Washington. Sears. M. L. and Barany, E. H. (1960). Outflow resistance and adrenergic mechanisms: erects of spmpathectomy. N-(2-chlorethyl) dibenzylamine hydrochloride (dibenamine) and dichloroisoproterenol on the outflow resistance of t,he rabbit rye. birch. Ophthalmol. 64. 839-48. Sears, M. L. and Neufeld, A. H. (1975). Adrenergic modulat,ion of t)hr outflow of aqurc~us humor. Inurst. Ophthalmol. 14, 83-6. Sears. 11. I,. and Sherk, T. E. (1964). The trabecular effect of noradrenaline in the rabbit ryr. Inwst. Ophthalrnol. 3. 157-63. Thomas. *J. V. (1980). Ocular adrenergic receptor sites pertinent to aqueous humor dynamicx Ann. Ophthalmol. 12, 96-8. Tripathi. B. ,J. and Tripathi, R. c’. (1980b). Effect of arachidonic acid on normal and dystrophic retinal pigment epithelium in tissue culture. fnwvt. Ophthal. C’is. Sri. 20. 553- rI Tripathi. IX. (‘. (1977). The functional morphology of the outflow systems of oc.ular anti c*errbrospinal fluids. Exp. Eye Res. 25 (Suppl.). 65-l 17. Tripathi. R. (‘. and Tripathi, 13. J. (1980a). Contractile protein alteration in trabrcular rndothelium in primary open-angle glaucoma. Exp. Eye Rrs. 31. 721-1. Tripathi. R. (1. and Tripathi. B. J. (1982a). Human trabecular end&helium, cornea1 entlot,hrlium. keratocytes. and scleral fibroblasts in primary (.elI culture. A comparative stud) of growt,h characteristics, morphology. and phagocytic activit>by light and scanning electron microscopy. Exp. Eye Res. 35, 61 l-24. Tripathi. R. C’. and Tripathi. B. ,J. (1982h). Functional anatomy of the anterior chamber angle. In Biomedical Foundations of Ophthalmology. Vol. 1 (Eds Duane. T. I). and ,J;trg,rrr. E. A.). Pp. l-88. Harper and Row, Philadelphia.

744

H. J. TRIPATHI

AXI)

R. (‘. TRII’ATHI

Waitzman. M. B.. Woods, W. D. and Cheek, W. V. (1979). Effects of prostaglandins and norepinephrine on ocular pressure and pupil size in rabbits following bilateral (ervi(.aI ganglionectomy. Invest. Ophthalmol. Vis. Sci. 18, 52-60. Weihing, R. R. (1979). The cytoskeleton and plasma membrane. Meth. Achier?. h’xp. t’uthol. 8, 42-109. Weiner, N. (1982). Norepinephrine, epinephrine, and the sympathomimetic amines. In 7% Pharmacological Basis of Therapeutics (6th Edn) (Eds Goodman. L. and Gilman. A.). Chap 8. Macmillan, New York. Weinreb, R. N., Mitchell, M. D. and Polansky, J. R. (1983). Prostaglandin production by human trabecular cells: in vitro inhibition by dexamethasone. Invest. Ophthalmol. Vis. Sci. 24, 1541-5. Weisenberg, R. C. (1972). Microtubule formation in vitro in solutions containing low calcium concentrations. Science 177. 1104-5.