Exp. Eye Res. (1991)
Outflow
52, 723-731
Facility
KRISTINE
Studies
ERICKSON-LAMY”“,
in the Perfused Segment JOHANNES
Human
W. ROHEN*AND
Ocular W. MORTON
Anterior GRANT”
aHowe laboratory of Ophthalmology, Massachusetts Eye and Ear infirmary, Harvard Medical School, Boston, MA, U.S.A. and bDepartment of Anatomy, University of Erlangen, Germany (Received 23 July 1990 and accepted in revised form 17 October 1990) We have recently developed a tissue model of the human aqueous outflow pathway involving placement of the eviscerated anterior corneoscleral shell, [with lens and uveal tissue removed but trabecular meshwork (TM) attached] onto a specialized perfusion apparatus. The TM and associated outflow tissues are perfused with culture medium at a physiologically-relevant perfusion pressure in a 5% CO, environment at 37°C. Under these conditions, the perfused outflow tissues are similar for several days, to the human and/or subhuman primate outflow system in vivo with regard to morphology as well as several functional parameters. Measured facility of outflow (0.271 f0.018 ~1 min’ mmHg-‘, n = 79) is similar to facility values obtained by tonography in living human beings. Moreover, outflow facility decreases in a linear fashion with increased perfusion pressure by 1.4% mmHg-r. Finally the removal of the TM results in a 41 y0 decrease in measured outflow resistance. The ability to study viable human outflow tissue for at least several days and the opportunity to establish a model which serves as an alternative to animal testing, point to the potential importance of this technique in investigating the biology of the aqueous outflow system. Key words: outflow facility: human eye: organ culture ; trabecular meshwork: perfusion.
1. Introduction The perfused human ocular anterior segment offers a unique opportunity to study, for a number of days, the physiology, biochemistry, and morphology of the human outflow pathways in viable outflow tissue. Although the living monkey eye has provided important information about outflow physiology, the model has limitations. First, there appear to be some fundamental differences between monkey and human eyes in the nature of outflow resistance. Monkey eyes, both in vivo and in vitro, show a gradual progressive decrease in resistance during anterior segment perfusion (‘washout’), whereas human eyes do not (Erickson-Lamy et al., 19 90a). Furthermore, uveoscleral flow is said to account for a considerable proportion of total outflow facility in monkey eyes but only a small proportion in human eyes (Bill, 1980). Second, unambiguous assessmentof the effect of drugs and hormones on true outflow facility is not possiblein monkey eyes in vivo because of the contributions of pseudofacility and uveoscleral flow to total measured outflow facility. Finally, the dwindling supply of primates for research along with growing concerns about the useof live animals in medical research point to the advantage of in vitro systemswherever feasible. Therefore, the ability to continuously assessoutflow facility for several days in a simplified system using
human tissue offers advantages over the monkey eyes for certain types of studies. The idea of developing an organ culture model of the outflow pathways is not new. Early studies of Macri and Cerverio (19 73, 19 75) used the arterially perfused cat eye to investigate the effects of various drugs on aqueous humor dynamics. Rohen et al. investigated the morphology and extracellular matrix synthetic capability of explanted outflow tissue (Rohen, Schachtschabel and Wehrmann, 1982; Schachtschabel, Berghoff and Rohen, 1984 ; Rohen, Schachtschabel and Berghoff, 1984). More recently, several laboratories have developed organ culture techniques and have described morphological (Johnson and Tschumper, 1987; Acott et al., 1988; Erickson-Lamy, Rohen and Grant, 1988). biochemical (Acott et al., 1988) and physiological (Erickson-Lamy et al., 1988) characteristics of their respective models. We present preliminary morphological, and for the first time, physiological observations in perfused human outflow pathway tissue which collectively suggest that, as in the calf eye (Erickson-Lamy et al., 1988) the manipulation and dissectionsnecessary to prepare the outflow tissue for perfusion do not greatly disrupt outflow morphology and physiology.
2. Materials and Methods Perfusion Apparatus
* Reprint requests at: Howe Laboratory, Ear Infirmary. 243 Charles Street, Boston,
00144835/91/060723+09
Massachusetts Eye and MA 02114. U.S.A.
$03.00/O
An anterior segment perfusion system, made of 0 1991 Academic
Press Limited 50-2
724
K. ERICKSON-LAMY
3
FIG. 1. A, The perfu lsion apparatus. sake of clarity, one of three clamping
B, Diagrammatic representation of the perfused ocular an1 :rior segm ent posts was omitted from the diagram.
Plexiglas and Teflon, was constructed for the human eye. The essentialcomponents of this system consist of a circular platform (15 mm diameter) cut from a sphere (2 5 mm diameter) upon which the anterior segment of the eye rests, and a spring-loaded ringclamping system (17.5 mm inner diameter) to form a leak-proof seal between the sclera and the shoulder of the platform base. Two central canals in the platform are connected by polyethylene tubing: one to the perfusion reservoir, and the other to a syringe for use during exchange of the anterior chamber contents (Fig. 1). The diameters of the platform and clamping ring and platform base are such that the shoulder of the platform abuts the inner sclera midway between the trabecular meshwork (TM) and the insertion of the extraocular muscles. The inner aspect of the ring clamp pressesagainst the outer aspect of the sclera slightly anterior to the attachments of the extraocular muscles.
ET AL.
:I. For the
Eye Tissue Post-mortem human eyes were obtained from the Seattle Lions Eye Bank (Seattle, WA) or from the National DiseaseResearch Interchange (NDRI, Philadelphia, PA). Eyes were enucleated within 3 hr of death and stored refrigerated in a humid saline environment. Eyes were shipped chilled to Boston via air-freight within 12 hr post-mortem. In caseswhere shipment could not be received within 12 hr, the eyes were dissectedat the eye bank. placed in medium, and shipped chilled via Federal Express. Perfusionof Intact Post-mortem Eyes Prior to perfusion, eyes serving as intact controls were wrapped in moist gauze and placed in a saline bath in a 5 ‘$4 CO, environment at 37°C. After temperature equilibration (approximately 30 min) the
OUTFLOW
FACILITY
IN THE
PERFUSED
ANTERIOR
anterior chamber was cannulated with a 23-gauge needle and the tip of the needle placed through the pupil into the posterior chamber in order to avoid deepening of the chamber (which results in artificially high facilities: Grant, 1958). The 23-gauge needle was connected with polyethylene tubing to a perfusion reservoir of a constant pressure perfusion apparatus similar to that described by Barany (1964).
Preparation of Eyes
for
Anterior Segment Perfusion
In preparation for anterior segment perfusion, the eyes were placed in sterile Dulbecco’s Modified Eagle Medium (4.5 mg ml-’ glucose) containing 50 U ml-’ penicillin, 50 pg ml-’ streptomycin, and 5 ,ug ml-’ fungizone (DMEM +PSF) for at least 15 min prior to dissection. Any remaining conjunctiva and extraocular muscles were removed and the eyes were bisected at the equator; the posterior portion was made available to other researchers. The anterior segment was transferred to a new sterile Petri dish (cornea side down) for dissection of the anterior segment. The lens zonule was cut and the lens removed, after which, the iris was grasped at the pupillary margin and gently pulled away. Finally, the choroid and ciliary body were removed by grasping the choroid close to the ora serrata and gently pulling it along with the ciliary body away from the underlying sclera. This procedure generally results in remnants of the ciliary muscle remaining attached to the scleral spur. No further attempt was made to remove the ciliary muscle in the interest of minimizing damage to the outflow tissues. The remaining outflow tissue and sclera was then gently rinsed with medium and placed onto the perfusion apparatus. The three screws were finger tightened (compressing the attached springs) to produce a leak-proof seal between the clamping ring, sclera and platform base. Some studies assessed the effect of removal of the TM. In those studies, with the aid of an operating microscope, a pair of fine jeweler’s forceps was used to peel the TM along with the inner wall of Schlemm’s canal away from the scleral shell. In subsequent portions of the text this procedure will be referred to as a trabeculectomy. All dissections were carried ou.t using sterile technique.
Outflow Facility Measurements Facility of outflow was determined at a constant pressure using a technique similar to that described by Grant (1963). Both intact eyes and anterior segments were perfused with DMEM + PSF at 3 7°C in a 5 y0 CO, environment at a constant manometrically-determined perfusion pressure of 10 mmHg (unless otherwise noted). Under an episcleral venous pressure of 8 mmHg and an intraocular pressure of 15.5 mmHg, this perfusion pressure roughly simulates the pressure
SEGMENT
725
drop across the TM in vivo (Moses, 198 7). Stable baseline facilities can be maintained in the human eye, since ‘washout’ does not occur in human eyes (Erickson-Lamy et al., 1990a). Therefore, a direct comparison can be made between facility measurements obtained in eyes before and after a given manipulation. In both intact eyes and in anterior segment preparations. all determinations of outflow facility consisted of an average of 4-5 consecutive 5min facility determinations, commencing when the rate of fluid flow into the perfusion chamber had reached steady state (usually 20-30 min after onset of perfusion). Outflow facility (C) was calculated as the ratio of the rate of flow of perfusion solution into the eye (~1 min-‘) to the perfusion pressure (mmHg) at steady state. Outflow resistance (R) was calculated as l/C (Moses, 1987). Morphological Examination of Outflow Tissue from Perfused Anterior Segments General methods. Anterior segments were perfused for 6 hr to 4 days at 10 mmHg, after which the perfusion chamber contents were exchanged with fixative and were perfused for 2 hr at 10 mmHg with a 2.5 % glutaraldehyde solution (Ito’s fixative) buffered with cacodylate to pH 7.4 at 4°C. The anterior segments were then fixed by immersion in Ito’s fixative for 3 hr Small pie-shaped sectors (34 mm wide) were post-fixed in osmium tetroxide, dehydrated, embedded in epon, and sectioned according to standard protocols. For light microscopy, semi-thin sections were stained with methylene blue or hematoxylin-eosin. Ultra-thin sections were examined by transmission electron microscopy (Zeiss EM 902). In each experiment, three to four specimens were examined both by light and electron microscopy. In two cases, serial sections were made in order to reconstruct the three dimensional appearance of the inner wall endothelium. Perfusion with cationized ferritin. A preliminary analysis of perfused outflow routes using cationized ferritin as a tracer was performed as described previously (Melamed 1986 ; Epstein and Rohen, 1990). After determination of a baseline facility, the contents of the perfusion chamber were exchanged with isotonic saline (pH 6.18) containing cationized ferritin (10 mg ml-‘). Perfusion with the cationized ferritin solution then continued for another 30 min. The chamber contents were exchanged with fixative and prepared for light and electron microscopy as described above. 3. Results Perfusion Studies Facility in the anterior segment vs. the intact eye. Facility of outflow was measured in the same eyes before and after preparing them for the anterior
726
K. ERICKSON-LAMY
ET AL.
TABLE I
Facility of outflow in the perfusedhuman outflow tissue model Specimen type Anterior segment Intact eye
n 87 8
Cwe 0.271 kO.018 0.480+0.095
Manipulation None Isolate anterior
Cp”ht
cPO’tIC,Iw
0.346 kO.06
0.82 f 0.09*
0.4 59 i: 0.08
1~74+0.01~
Segment Anterior
segment
8
0.265 f0.039
Trabeculectomy
Intact enucleated human eyes and anterior ocular segments with attached outflow tissues were perfused at 10 mmHg. Baseline facility (CD,,) was determined in intact eyes after which the eye was dissected and a second facility determination (C,,,,) was made in the anterior segment preparation. The effect of trabeculectomy was determined in several anterior segment preparations, showing a 41% reduction in outflow different from I.0 using the two-tailed paired t-test: *P < 0.10: j’P < 0.002. resistance. Cpost/CprP is significantly
FIG. 2. Light micrograph of the TM area and outer wall (OW) of Schlemm’s canal after trabeculectomy. wall remains intact. x 560.
segment perfusion model. Similar to our previous finding in the calf eye (Erickson-Lamy et al., 1988) there was an 18 % decreasein facility of outflow after placement onto the perfusion apparatus (Table I). In a larger sample of human eyes, outflow facility in the perfused anterior segment model (0.271 ~1 min-’ mmHg-‘) was similar to the average (0.30 y1 min-’ mmHg-‘) tonographically-determined value in the human eye in vivo (Chandler and Grant, 1979). The facilities in the perfused anterior segmentsranged from 0.06 ~1 min’ mmHg-’ to 0.86 ~1 min-’ mmHg-‘. The average donor age was 71 f 1.3 yr (range 42-94 yr). There was no correlation between donor age and outflow facility (R = 00028). Effect of trabeculectornyon facility of outflow. Facility of outflow was compared in perfused anterior segments before and after removal of the TM. The trabeculectomy technique results in an opening of Schlemm’s
Note that the outer
canal leaving the outer wall of the canal intact (Fig. 2). TM removal resulted in a 41% decrease in outflow resistance (Table I), of the same magnitude (59 % at a perfusion pressure of 10 mmHg) as was previously reported in intact human eyes (Grant, 1963: Rosenquist et al., 1985). Relationship between outflow facility and perlusion pressure.In order to further verify that the measured outflow facility in the perfused anterior segmentmodel was similar in nature to that occurring in intact eyes, we examined the relationship between perfusion pressure and outflow facility. It is known that at perfusion pressures greater than 10 mmHg, an elevation in pressure results in a lowering of outflow facility of approximately 1% mmHg-’ . presumably due to occlusion of Schlemm’s canal (Ellingsen and Grant, 19 7 1; Brubaker, 19 75 ; Hashimoto and Epstein, 19 80 : Erickson-Lamy et al., 198 8). This also held true
OUTFLOW
FACILITY
IN THE
PERFUSED
ANTERIOR
SEGMENT
727
in the perfused human anterior segment where facility decreased approximately 1.4% mmHg (Fig. 3). Morphological Studiesin Perfused Anterior Segments
goG 8
70 -
50
0
’
’ IO
1
’ 20
PerfusIon
1
’ 30
pressure
’
’ 40
’
1 50
1
(mmlig)
FIG. 3. Kelationship between outflow facility and perfusion pressure. Outflow facility (C,) was determined in perfused human ocular anterior segments at an initial perfusion pressure of 10 mmHg. Subsequent determinations of facility were made at stepwise pressure increments to a final pressure of 50 mmHg. Facility data are expressed as a percentage of the facility obtained at 10 mmHg. Error bars represent the standard error of the mean of five eyes.
FIG. 4. Note the specimen subendotl
Light and electron microscopic analysis of the specimens of perfused outflow tissue of human eyes showed a normal appearance of the trabecular meshwork. Outflow pathway tissue from the eye of a 54-year-old donor is shown in Fig. 4. The eye was enucleated 2 hr post-mortem and perfusion occurred on the second post-mortem day. After 6 hr perfusion (C = 0.25 ,~l min-’ mmHg-l) the inner wall endothelium showed no signs of deterioration. The endothelial cells formed a continuous line of flat cells without intercellular disruptions or degenerative vacuolization. Giant vacuoles were rarely seen. However, at places, the inner wall endothelium appeared to be separated from the underlying connective tissue, thereby forming large pseudovacuoles which bulged into the lumen of Schlemm’s canal.
Electron micrograph of the inner wall of Schlemm’s canal from a non-glaucomatous 54-year-old donor. x 24000. Inormal appearance of the inner wall endothelium (E). Perfusion was performed on the second day post-mortem. The was fixed after 6 hr perfusion at 10 mmHg (C = 0.25 ,uI min-’ mmHg-I). SC, Lumen of Schlemm’s canal; ECM. relial space with extracellular matrix.
728
K. ERICKSON-LAMY
ET AL.
FIG. 5. Light micrograph of the trabecular meshwork from a 51-year-old donor on the second post-mortem day after h hr perfusion (C = 0.32 ,u-’ min mmHg-I). Note the normal appearance of the meshwork and the great number of normal trabecular cells. x 1200. E, Endothelium of Schlemm’s canal: SC. Schlemm’s canal: S. subendothelial space (cribriform layer): T. trabecular beams. Similarly, after 6 hr perfusion on the second postmortem day, the TM of a 51-year-old donor (C = 0.32 ~1 min-’ mmHg-l) showed, under light microscopy, normal appearance of the trabecular beams, meshwork cells and the inner wall of Schlemm’s canal (Fig. 5). No swelling of mitochondria or of other cell organelles was observed by electron microscopy. Figure 6 shows the ultrastructural appearance of the TM of a 54-year-old donor on the fourth day postmortem after 3 days perfusion. The outflow facility on the first day of perfusion was 0.25 ,ul min-’ mmHg-’ and on the third day was 0.46 ,~l min-’ mmHg-I. Morphologically, the trabecular beams appeared relatively normal. The trabecular lamellae were completely covered by trabecular cells which contained a normal set of organelles and a nucleus of normal size. The ultrastructural appearance of the TM of a 59year-old donor was still normal, although the eyes enucleated 4 hr post-mortem were stored in culture medium for 4 days before the perfusion experiment could be started (Fig. 7). There were many cells of normal appearance in the TM and the endothelial lining of Schlemm’s canal consisted of a continuous line of flat cells showing no degenerative changes.
Only a slight vacuolization was seen in cribriform cells (Fig. 7). After repeated perfusions or prolongation of the organ cultures (1 week or more), the trabecular cells began to swell, their cell membranes deteriorated
and their cell organelles appeared as isolated vesicles (e.g. matrix vesicles) within the intertrabecular spaces. In addition, the endothelial cells of the inner wall became detached so that the filtration zone became
more and more incomplete.
Therefore,
the present
study deals only with relatively short-term experiments. In pilot studies made to trace the outflow pathways under various experimental conditions, cationized ferritin was perfused into the outflow tissues. Electron micrographs of these tissues showed that the cell membranes of the trabecular cells were clearly labeled with tracer material indicating that the cells still contained negatively charged molecules and that their membranes had not deteriorated (Fig. 8). Ferritin particles were found in the lumen of Schlemm’s canal. Some of the ferritin particles were phagocytized by trabecular cells. In contrast to observations in the monkey eye (Epstein and Rohen, 1990), the outer surface of the inner wall endothelium was not heavily labeled with the tracer. However, as with the findings in the monkey eye, the giant vacuoles were not filled with ferritin particles. 4. Discussion
Several authors have developed models for the organ culture of the anterior segment and/or segments of the trabecular meshwork describing morphological and biochemical aspects of their respective models (Rohen et al., 1982; Schachtschabel et al., 1984; Rohen et al., 1984 ; Johnson and Tschumper. 1987 : Acott et al., 1988). The present work describes an outflow tissue model using human tissue which, like a similar model developed in the calf eye (Erickson-Lamy et al., 1988), appears to be appropriate for the
OUTFLOW
FACILITY
IN THE
PERFUSED
ANTERIOR
SEGMENT
FIG. 6. Electron micrograph of the TM from the fellow eye of the same donor as in Fig. 4 on the fourth day post-mortem. x 600 10. Note the normal auaearance of the trabecular cells (T) and beams (B) after 3 days of continuous perfusion (C = -0.46 jr1 min. 1mmHg-I). investigations of facility of outflow over a period of at least 4 days. Previous studies have documented the morphological (Johnson and Tschumper, 198 7 ; Acott et al.. 1988) and biochemical (Acott, 1988) integrity of organ cultured human outflow tissues for up to 1 month. Further longer term investigations may indicate the appropriateness of the presently described model for physiological, pharmacological, and biochemical studies of human outflow pathway for a more extended period. Near-normal outflow physiology and morphology of the human outflow pathway tissues was observed during the perfusion periods examined. Similar to our findings in the calf eye (Erickson-Lamy et al., 1988) an 18% reduction in facility of outflow was observed subsequent to preparation of the eye for anterior segment perfusion. It is likely that the observed facility decrease relates to distortion of the sclera introduced by the clamping ring rather than to removal of the iris/ciliary body, since facility of outflow is not significantly affected by either iridectomy or cyclo-
dialysis in the intact enucleated human eye (Grant, 1963). Despite a slight reduction in facility of outflow in the perfused anterior segment, the well-defined characteristics of the aqueous outflow system present in intact eyes including the effect of trabeculectomy on outflow resistance and the relationship of perfusion pressure to outflow facility were also noted in the perfused anterior segment system. Trabeculectomy resulted in a 41% reduction in outflow resistance ; approximately 18 % less than noted at the same perfusion pressure in intact enucleated human eyes after an interno trabeculotomy (Rosenquist et al., 19 8 5). The decreased change in outflow resistance in the perfused outflow tissue model may well relate to a putative increase in scleral resistance induced by the culture technique as noted above. It is probable that perfusate left the aqueous outflow channels through conventional routes of egress as suggested by the decoration of trabecular cells as well as inner wall endothelia by cationized ferritin. Fur-
730
K. ERICKSON-LAMY
ET AL.
FIG. 7. Electron micrograph of the trabecular meshwork of a 59-year-old donor after 1 hr perfusion on the fourth postmortem day (C = 0.188 ,LL~min-’ mmHg-I). There is only a slight vacuolization of the cribriform cells (C). otherwise the
trabecular meshworkand inner wall endothelium(E) of Schlemm’scanal (SC)show a normal structure. x 4500.
FIG. 8. Electron micrograph of the cribriform region of the trabecular meshwork of a human eye 2 days post-mortem after 6 hr perfusion. The eye was perfused with cationized ferritin prior to fixation as described in Materials and Methods. Ferritin particles are seen within the lumen of Schlemm’s canal (SC) and at the cell surfaces. E, Endothelial lining of SC: T, trabecular cells. Arrowheads indicate ferritin particles phagocytized by a trabecular cell. Asterisks indicate giant vacuoles i free of ferritin particles). x 5000.
thermore, the observed reduction of facility of outflow at elevated perfusion pressure [presumably due to a narrowing of Schlemm’s canal (Johnstone and Grant.
1973)] would be hard to explain if the perfusate was simply leaking through the sclera. In such a case. one would expect the facility to increase or remain
OUTFLOW
FACILITY
IN THE
PERFUSED
ANTERIOR
SEGMENT
with increased perfusion pressure. However, further systematic studies are needed to establish definitely how much of the perfusate leaves the perfused anterior ocular segment through the expected conventional routes. The manipulations necessary for perfusion of the outflow pathways of the human eye do not greatly alter well-established physiological characteristics of outflow resistance nor do they greatly alter normal morphology of the outflow tissues. Further, drug responsiveness has been demonstrated in this preparation. In recent studies, we showed that epinephrine increases facility of outflow in a dose-dependent manner (Erickson-Lamy et al., 1990b). Therefore, this model may have potential usefulnessin determining the site of action of drugs known to influence outflow resistance. Because the ciliary muscle is absent, any drug effect is presumably on the outflow pathway itself rather than via an indirect effect mediated by ciliary muscle contraction. Collectively, these results point to the potential importance of this model in investigating the biology, physiology and pharmacology of the human outflow pathway system. constant
Acknowledgements Supported in part by NIH grant EY0732 1, Massachusetts Lions Eye Research Fund, Inc. and National Glaucoma Research, a program of the American Health Assistance Foundation (K. Erickson-Lamy), and by Deutsche Forschungsgemeinschaft. Bonn-Bad Godesberg, FRG grant RO 81/18/4 (J. W. Rohen).
References Acott, T. S.. Kingsley, P. D., Samples, J. R. and Van Buskirk, E. M. (1988). Human trabecular meshwork organ culture : Morphology and glycosaminoglycan synthesis. Invest. OphthahnoZ. Iris. Sci. 29, 90-100. BBrdny. E. H. (1964). Simultaneous measurement of changing intraocular pressure and outflow facility in the vervet monkey by constant pressure perfusion. Invest. Ophthalmol. 3, 135-143. Bill, A. (1980). The drainage of aqueous humour via Schlemm’s canal and uveoscleral routes. Ophthalmic Res. 12, 130. Brubaker, R. F. ( 1975). The effect of intraocular pressure on conventional outflow resistance in the enucleated human eye. Invest. Ophthalmol. 14, 286-92. Chandler, P. A. and Grant, W. M. (1979). Glaucoma. 2nd edn. P. 23. Lea and Febiger: Philadelphia, PA. Ellingsen. B. A. and Grant, W. M. (1971). The relationship of pressure and aqueous outflow in enucleated human eyes. Invest. Ophthulmol. 10. 430-7.
731
Epstein, D. L. and Rohan, J. W. (1991).
Morphology
of the
trabecular meshworkand inner wall endotheliumafter cationized ferritin perfusion in the monkey eye. Invest. Ophthulmol. Vis. Sci. 32, 160-l 71. Erickson-Lamy K. A., Rohen, J, W. and Grant, W. M. (1988). Outflow facility studies in the perfused bovine aqueous outflow pathways. Curr. Eye Res. 7, 799-807. Erickson-Lamy, K. A., Epstein, D. L.. Schroeder, A. M. and Bassett-Chu, S. (1990a). Absence of time-dependent facility increase (I washout ‘) in the perfused enucleated human eye. Invest. Ophthulmol. Vis. Sci. 31, 2384-8. Erickson-Lamy, K. A., Ostovar, B., Hunnicutt, E. S. and Nathanson, J. A. (1990b). Epinephrine increases facility of outflow and trabecular meshworkCAMP content in the human eye in vitro. Invest. Ophthulmol. Vis. Sci. 31 (ARVO Suppl.), 184. Grant, W. M. (1958). Further studies on facility of flow through the trabecular meshwork. ,4rch. Ophthulmol. 60,
523-33.
Grant, W. M. (1963). Experimental aqueous perfusion in enucleated human eyes. Arch. Ophthulmol. 60, 783801. Hashimoto, J. M. and Epstein, D. L. (1980). Influence of intraocular pressure on aqueous outflow facility in enucleated eyes of different mammals. Invest. Ophthulmol. 19, 1483-9. Johnson, D. H. and Tschumper. R. L. (1987). Human trabecular meshwork organ culture. A new method. Invest. Ophthulmol. Vis. Sci. 28, 945-53. Johnstone, M. A. and Grant, W. M. (1973). Microsurgery of Schlemm’s Canal and the human aqueous outflow system. Am. 1. Ophthulmol. 76, 906-I 7. Macri. F. J., Cevario, S. J. (1973). The induction of aqueous humor formation by the use of Ach + eserine. Invest. Ophthulmol. 12, 910-16. Macri, F. J. and Cevario, S. J. (1975). Ciliary ganglion stimulation. I. Effects on aqueous humor inflow and outflow. Invest. Ophthulmol. 14, 2843. Melamed, S. (1986). Use of cationized ferritin to trace aqueous humor outflow in the monkey eye. Exp. Eye. Res. 43, 273-8. Moses, R. A. (198 7). Intraocular pressure. In Adler’s Physiology of the Eye. Clinical Application. (Ed. Moses, R. A.) 8th edn. P. 224. C. V. Mosby Co: St. Louis, MO. Rohen, J. W., Schachtschabel, D. 0. and Wehrmann, R. (1982). Structural changes of human and monkey trabecular meshwork following in vitro cultivation. Gruefes Arch. Clin. Exp. Ophthulmol. 218. 225-32. Rohan, J. W.. Schachtschabel, D. 0. and Berghoff, K. (1984). Histoautoradiographic and biochemical studies on human and monkey trabecular meshwork and ciliary body in short-term explant culture. Grufes’s Arch. Clin. Exp. Ophthulmol. 221, 199-206. Rosenquist, R. C., Epstein, D. L., Johnson, M., Grant, W. M.
and Melamed, S. (1985). Resistancevariation with perfusion pressure after trabeculotomy. Invest. Ophthulmol. Vis. Sci. 26 (ARVO Suppl.), 26. Schachtschabel, D. O., Berghoff, K. and Rohen, J. W. (1984). Synthesis and composition of glycosaminoglycans by explant cultures of human ciliary body and ciliary processes in serum-containing and serum-free defined media. Grufes’s Arch. Clin. Exp. Ophthulmol. 221, 207-9.
Outflow
52, 723-731
Facility
KRISTINE
Studies
ERICKSON-LAMY”“,
in the Perfused Segment JOHANNES
Human
W. ROHEN*AND
Ocular W. MORTON
Anterior GRANT”
aHowe laboratory of Ophthalmology, Massachusetts Eye and Ear infirmary, Harvard Medical School, Boston, MA, U.S.A. and bDepartment of Anatomy, University of Erlangen, Germany (Received 23 July 1990 and accepted in revised form 17 October 1990) We have recently developed a tissue model of the human aqueous outflow pathway involving placement of the eviscerated anterior corneoscleral shell, [with lens and uveal tissue removed but trabecular meshwork (TM) attached] onto a specialized perfusion apparatus. The TM and associated outflow tissues are perfused with culture medium at a physiologically-relevant perfusion pressure in a 5% CO, environment at 37°C. Under these conditions, the perfused outflow tissues are similar for several days, to the human and/or subhuman primate outflow system in vivo with regard to morphology as well as several functional parameters. Measured facility of outflow (0.271 f0.018 ~1 min’ mmHg-‘, n = 79) is similar to facility values obtained by tonography in living human beings. Moreover, outflow facility decreases in a linear fashion with increased perfusion pressure by 1.4% mmHg-r. Finally the removal of the TM results in a 41 y0 decrease in measured outflow resistance. The ability to study viable human outflow tissue for at least several days and the opportunity to establish a model which serves as an alternative to animal testing, point to the potential importance of this technique in investigating the biology of the aqueous outflow system. Key words: outflow facility: human eye: organ culture ; trabecular meshwork: perfusion.
1. Introduction The perfused human ocular anterior segment offers a unique opportunity to study, for a number of days, the physiology, biochemistry, and morphology of the human outflow pathways in viable outflow tissue. Although the living monkey eye has provided important information about outflow physiology, the model has limitations. First, there appear to be some fundamental differences between monkey and human eyes in the nature of outflow resistance. Monkey eyes, both in vivo and in vitro, show a gradual progressive decrease in resistance during anterior segment perfusion (‘washout’), whereas human eyes do not (Erickson-Lamy et al., 19 90a). Furthermore, uveoscleral flow is said to account for a considerable proportion of total outflow facility in monkey eyes but only a small proportion in human eyes (Bill, 1980). Second, unambiguous assessmentof the effect of drugs and hormones on true outflow facility is not possiblein monkey eyes in vivo because of the contributions of pseudofacility and uveoscleral flow to total measured outflow facility. Finally, the dwindling supply of primates for research along with growing concerns about the useof live animals in medical research point to the advantage of in vitro systemswherever feasible. Therefore, the ability to continuously assessoutflow facility for several days in a simplified system using
human tissue offers advantages over the monkey eyes for certain types of studies. The idea of developing an organ culture model of the outflow pathways is not new. Early studies of Macri and Cerverio (19 73, 19 75) used the arterially perfused cat eye to investigate the effects of various drugs on aqueous humor dynamics. Rohen et al. investigated the morphology and extracellular matrix synthetic capability of explanted outflow tissue (Rohen, Schachtschabel and Wehrmann, 1982; Schachtschabel, Berghoff and Rohen, 1984 ; Rohen, Schachtschabel and Berghoff, 1984). More recently, several laboratories have developed organ culture techniques and have described morphological (Johnson and Tschumper, 1987; Acott et al., 1988; Erickson-Lamy, Rohen and Grant, 1988). biochemical (Acott et al., 1988) and physiological (Erickson-Lamy et al., 1988) characteristics of their respective models. We present preliminary morphological, and for the first time, physiological observations in perfused human outflow pathway tissue which collectively suggest that, as in the calf eye (Erickson-Lamy et al., 1988) the manipulation and dissectionsnecessary to prepare the outflow tissue for perfusion do not greatly disrupt outflow morphology and physiology.
2. Materials and Methods Perfusion Apparatus
* Reprint requests at: Howe Laboratory, Ear Infirmary. 243 Charles Street, Boston,
00144835/91/060723+09
Massachusetts Eye and MA 02114. U.S.A.
$03.00/O
An anterior segment perfusion system, made of 0 1991 Academic
Press Limited 50-2
724
K. ERICKSON-LAMY
3
FIG. 1. A, The perfu lsion apparatus. sake of clarity, one of three clamping
B, Diagrammatic representation of the perfused ocular an1 :rior segm ent posts was omitted from the diagram.
Plexiglas and Teflon, was constructed for the human eye. The essentialcomponents of this system consist of a circular platform (15 mm diameter) cut from a sphere (2 5 mm diameter) upon which the anterior segment of the eye rests, and a spring-loaded ringclamping system (17.5 mm inner diameter) to form a leak-proof seal between the sclera and the shoulder of the platform base. Two central canals in the platform are connected by polyethylene tubing: one to the perfusion reservoir, and the other to a syringe for use during exchange of the anterior chamber contents (Fig. 1). The diameters of the platform and clamping ring and platform base are such that the shoulder of the platform abuts the inner sclera midway between the trabecular meshwork (TM) and the insertion of the extraocular muscles. The inner aspect of the ring clamp pressesagainst the outer aspect of the sclera slightly anterior to the attachments of the extraocular muscles.
ET AL.
:I. For the
Eye Tissue Post-mortem human eyes were obtained from the Seattle Lions Eye Bank (Seattle, WA) or from the National DiseaseResearch Interchange (NDRI, Philadelphia, PA). Eyes were enucleated within 3 hr of death and stored refrigerated in a humid saline environment. Eyes were shipped chilled to Boston via air-freight within 12 hr post-mortem. In caseswhere shipment could not be received within 12 hr, the eyes were dissectedat the eye bank. placed in medium, and shipped chilled via Federal Express. Perfusionof Intact Post-mortem Eyes Prior to perfusion, eyes serving as intact controls were wrapped in moist gauze and placed in a saline bath in a 5 ‘$4 CO, environment at 37°C. After temperature equilibration (approximately 30 min) the
OUTFLOW
FACILITY
IN THE
PERFUSED
ANTERIOR
anterior chamber was cannulated with a 23-gauge needle and the tip of the needle placed through the pupil into the posterior chamber in order to avoid deepening of the chamber (which results in artificially high facilities: Grant, 1958). The 23-gauge needle was connected with polyethylene tubing to a perfusion reservoir of a constant pressure perfusion apparatus similar to that described by Barany (1964).
Preparation of Eyes
for
Anterior Segment Perfusion
In preparation for anterior segment perfusion, the eyes were placed in sterile Dulbecco’s Modified Eagle Medium (4.5 mg ml-’ glucose) containing 50 U ml-’ penicillin, 50 pg ml-’ streptomycin, and 5 ,ug ml-’ fungizone (DMEM +PSF) for at least 15 min prior to dissection. Any remaining conjunctiva and extraocular muscles were removed and the eyes were bisected at the equator; the posterior portion was made available to other researchers. The anterior segment was transferred to a new sterile Petri dish (cornea side down) for dissection of the anterior segment. The lens zonule was cut and the lens removed, after which, the iris was grasped at the pupillary margin and gently pulled away. Finally, the choroid and ciliary body were removed by grasping the choroid close to the ora serrata and gently pulling it along with the ciliary body away from the underlying sclera. This procedure generally results in remnants of the ciliary muscle remaining attached to the scleral spur. No further attempt was made to remove the ciliary muscle in the interest of minimizing damage to the outflow tissues. The remaining outflow tissue and sclera was then gently rinsed with medium and placed onto the perfusion apparatus. The three screws were finger tightened (compressing the attached springs) to produce a leak-proof seal between the clamping ring, sclera and platform base. Some studies assessed the effect of removal of the TM. In those studies, with the aid of an operating microscope, a pair of fine jeweler’s forceps was used to peel the TM along with the inner wall of Schlemm’s canal away from the scleral shell. In subsequent portions of the text this procedure will be referred to as a trabeculectomy. All dissections were carried ou.t using sterile technique.
Outflow Facility Measurements Facility of outflow was determined at a constant pressure using a technique similar to that described by Grant (1963). Both intact eyes and anterior segments were perfused with DMEM + PSF at 3 7°C in a 5 y0 CO, environment at a constant manometrically-determined perfusion pressure of 10 mmHg (unless otherwise noted). Under an episcleral venous pressure of 8 mmHg and an intraocular pressure of 15.5 mmHg, this perfusion pressure roughly simulates the pressure
SEGMENT
725
drop across the TM in vivo (Moses, 198 7). Stable baseline facilities can be maintained in the human eye, since ‘washout’ does not occur in human eyes (Erickson-Lamy et al., 1990a). Therefore, a direct comparison can be made between facility measurements obtained in eyes before and after a given manipulation. In both intact eyes and in anterior segment preparations. all determinations of outflow facility consisted of an average of 4-5 consecutive 5min facility determinations, commencing when the rate of fluid flow into the perfusion chamber had reached steady state (usually 20-30 min after onset of perfusion). Outflow facility (C) was calculated as the ratio of the rate of flow of perfusion solution into the eye (~1 min-‘) to the perfusion pressure (mmHg) at steady state. Outflow resistance (R) was calculated as l/C (Moses, 1987). Morphological Examination of Outflow Tissue from Perfused Anterior Segments General methods. Anterior segments were perfused for 6 hr to 4 days at 10 mmHg, after which the perfusion chamber contents were exchanged with fixative and were perfused for 2 hr at 10 mmHg with a 2.5 % glutaraldehyde solution (Ito’s fixative) buffered with cacodylate to pH 7.4 at 4°C. The anterior segments were then fixed by immersion in Ito’s fixative for 3 hr Small pie-shaped sectors (34 mm wide) were post-fixed in osmium tetroxide, dehydrated, embedded in epon, and sectioned according to standard protocols. For light microscopy, semi-thin sections were stained with methylene blue or hematoxylin-eosin. Ultra-thin sections were examined by transmission electron microscopy (Zeiss EM 902). In each experiment, three to four specimens were examined both by light and electron microscopy. In two cases, serial sections were made in order to reconstruct the three dimensional appearance of the inner wall endothelium. Perfusion with cationized ferritin. A preliminary analysis of perfused outflow routes using cationized ferritin as a tracer was performed as described previously (Melamed 1986 ; Epstein and Rohen, 1990). After determination of a baseline facility, the contents of the perfusion chamber were exchanged with isotonic saline (pH 6.18) containing cationized ferritin (10 mg ml-‘). Perfusion with the cationized ferritin solution then continued for another 30 min. The chamber contents were exchanged with fixative and prepared for light and electron microscopy as described above. 3. Results Perfusion Studies Facility in the anterior segment vs. the intact eye. Facility of outflow was measured in the same eyes before and after preparing them for the anterior
726
K. ERICKSON-LAMY
ET AL.
TABLE I
Facility of outflow in the perfusedhuman outflow tissue model Specimen type Anterior segment Intact eye
n 87 8
Cwe 0.271 kO.018 0.480+0.095
Manipulation None Isolate anterior
Cp”ht
cPO’tIC,Iw
0.346 kO.06
0.82 f 0.09*
0.4 59 i: 0.08
1~74+0.01~
Segment Anterior
segment
8
0.265 f0.039
Trabeculectomy
Intact enucleated human eyes and anterior ocular segments with attached outflow tissues were perfused at 10 mmHg. Baseline facility (CD,,) was determined in intact eyes after which the eye was dissected and a second facility determination (C,,,,) was made in the anterior segment preparation. The effect of trabeculectomy was determined in several anterior segment preparations, showing a 41% reduction in outflow different from I.0 using the two-tailed paired t-test: *P < 0.10: j’P < 0.002. resistance. Cpost/CprP is significantly
FIG. 2. Light micrograph of the TM area and outer wall (OW) of Schlemm’s canal after trabeculectomy. wall remains intact. x 560.
segment perfusion model. Similar to our previous finding in the calf eye (Erickson-Lamy et al., 1988) there was an 18 % decreasein facility of outflow after placement onto the perfusion apparatus (Table I). In a larger sample of human eyes, outflow facility in the perfused anterior segment model (0.271 ~1 min-’ mmHg-‘) was similar to the average (0.30 y1 min-’ mmHg-‘) tonographically-determined value in the human eye in vivo (Chandler and Grant, 1979). The facilities in the perfused anterior segmentsranged from 0.06 ~1 min’ mmHg-’ to 0.86 ~1 min-’ mmHg-‘. The average donor age was 71 f 1.3 yr (range 42-94 yr). There was no correlation between donor age and outflow facility (R = 00028). Effect of trabeculectornyon facility of outflow. Facility of outflow was compared in perfused anterior segments before and after removal of the TM. The trabeculectomy technique results in an opening of Schlemm’s
Note that the outer
canal leaving the outer wall of the canal intact (Fig. 2). TM removal resulted in a 41% decrease in outflow resistance (Table I), of the same magnitude (59 % at a perfusion pressure of 10 mmHg) as was previously reported in intact human eyes (Grant, 1963: Rosenquist et al., 1985). Relationship between outflow facility and perlusion pressure.In order to further verify that the measured outflow facility in the perfused anterior segmentmodel was similar in nature to that occurring in intact eyes, we examined the relationship between perfusion pressure and outflow facility. It is known that at perfusion pressures greater than 10 mmHg, an elevation in pressure results in a lowering of outflow facility of approximately 1% mmHg-’ . presumably due to occlusion of Schlemm’s canal (Ellingsen and Grant, 19 7 1; Brubaker, 19 75 ; Hashimoto and Epstein, 19 80 : Erickson-Lamy et al., 198 8). This also held true
OUTFLOW
FACILITY
IN THE
PERFUSED
ANTERIOR
SEGMENT
727
in the perfused human anterior segment where facility decreased approximately 1.4% mmHg (Fig. 3). Morphological Studiesin Perfused Anterior Segments
goG 8
70 -
50
0
’
’ IO
1
’ 20
PerfusIon
1
’ 30
pressure
’
’ 40
’
1 50
1
(mmlig)
FIG. 3. Kelationship between outflow facility and perfusion pressure. Outflow facility (C,) was determined in perfused human ocular anterior segments at an initial perfusion pressure of 10 mmHg. Subsequent determinations of facility were made at stepwise pressure increments to a final pressure of 50 mmHg. Facility data are expressed as a percentage of the facility obtained at 10 mmHg. Error bars represent the standard error of the mean of five eyes.
FIG. 4. Note the specimen subendotl
Light and electron microscopic analysis of the specimens of perfused outflow tissue of human eyes showed a normal appearance of the trabecular meshwork. Outflow pathway tissue from the eye of a 54-year-old donor is shown in Fig. 4. The eye was enucleated 2 hr post-mortem and perfusion occurred on the second post-mortem day. After 6 hr perfusion (C = 0.25 ,~l min-’ mmHg-l) the inner wall endothelium showed no signs of deterioration. The endothelial cells formed a continuous line of flat cells without intercellular disruptions or degenerative vacuolization. Giant vacuoles were rarely seen. However, at places, the inner wall endothelium appeared to be separated from the underlying connective tissue, thereby forming large pseudovacuoles which bulged into the lumen of Schlemm’s canal.
Electron micrograph of the inner wall of Schlemm’s canal from a non-glaucomatous 54-year-old donor. x 24000. Inormal appearance of the inner wall endothelium (E). Perfusion was performed on the second day post-mortem. The was fixed after 6 hr perfusion at 10 mmHg (C = 0.25 ,uI min-’ mmHg-I). SC, Lumen of Schlemm’s canal; ECM. relial space with extracellular matrix.
728
K. ERICKSON-LAMY
ET AL.
FIG. 5. Light micrograph of the trabecular meshwork from a 51-year-old donor on the second post-mortem day after h hr perfusion (C = 0.32 ,u-’ min mmHg-I). Note the normal appearance of the meshwork and the great number of normal trabecular cells. x 1200. E, Endothelium of Schlemm’s canal: SC. Schlemm’s canal: S. subendothelial space (cribriform layer): T. trabecular beams. Similarly, after 6 hr perfusion on the second postmortem day, the TM of a 51-year-old donor (C = 0.32 ~1 min-’ mmHg-l) showed, under light microscopy, normal appearance of the trabecular beams, meshwork cells and the inner wall of Schlemm’s canal (Fig. 5). No swelling of mitochondria or of other cell organelles was observed by electron microscopy. Figure 6 shows the ultrastructural appearance of the TM of a 54-year-old donor on the fourth day postmortem after 3 days perfusion. The outflow facility on the first day of perfusion was 0.25 ,ul min-’ mmHg-’ and on the third day was 0.46 ,~l min-’ mmHg-I. Morphologically, the trabecular beams appeared relatively normal. The trabecular lamellae were completely covered by trabecular cells which contained a normal set of organelles and a nucleus of normal size. The ultrastructural appearance of the TM of a 59year-old donor was still normal, although the eyes enucleated 4 hr post-mortem were stored in culture medium for 4 days before the perfusion experiment could be started (Fig. 7). There were many cells of normal appearance in the TM and the endothelial lining of Schlemm’s canal consisted of a continuous line of flat cells showing no degenerative changes.
Only a slight vacuolization was seen in cribriform cells (Fig. 7). After repeated perfusions or prolongation of the organ cultures (1 week or more), the trabecular cells began to swell, their cell membranes deteriorated
and their cell organelles appeared as isolated vesicles (e.g. matrix vesicles) within the intertrabecular spaces. In addition, the endothelial cells of the inner wall became detached so that the filtration zone became
more and more incomplete.
Therefore,
the present
study deals only with relatively short-term experiments. In pilot studies made to trace the outflow pathways under various experimental conditions, cationized ferritin was perfused into the outflow tissues. Electron micrographs of these tissues showed that the cell membranes of the trabecular cells were clearly labeled with tracer material indicating that the cells still contained negatively charged molecules and that their membranes had not deteriorated (Fig. 8). Ferritin particles were found in the lumen of Schlemm’s canal. Some of the ferritin particles were phagocytized by trabecular cells. In contrast to observations in the monkey eye (Epstein and Rohen, 1990), the outer surface of the inner wall endothelium was not heavily labeled with the tracer. However, as with the findings in the monkey eye, the giant vacuoles were not filled with ferritin particles. 4. Discussion
Several authors have developed models for the organ culture of the anterior segment and/or segments of the trabecular meshwork describing morphological and biochemical aspects of their respective models (Rohen et al., 1982; Schachtschabel et al., 1984; Rohen et al., 1984 ; Johnson and Tschumper. 1987 : Acott et al., 1988). The present work describes an outflow tissue model using human tissue which, like a similar model developed in the calf eye (Erickson-Lamy et al., 1988), appears to be appropriate for the
OUTFLOW
FACILITY
IN THE
PERFUSED
ANTERIOR
SEGMENT
FIG. 6. Electron micrograph of the TM from the fellow eye of the same donor as in Fig. 4 on the fourth day post-mortem. x 600 10. Note the normal auaearance of the trabecular cells (T) and beams (B) after 3 days of continuous perfusion (C = -0.46 jr1 min. 1mmHg-I). investigations of facility of outflow over a period of at least 4 days. Previous studies have documented the morphological (Johnson and Tschumper, 198 7 ; Acott et al.. 1988) and biochemical (Acott, 1988) integrity of organ cultured human outflow tissues for up to 1 month. Further longer term investigations may indicate the appropriateness of the presently described model for physiological, pharmacological, and biochemical studies of human outflow pathway for a more extended period. Near-normal outflow physiology and morphology of the human outflow pathway tissues was observed during the perfusion periods examined. Similar to our findings in the calf eye (Erickson-Lamy et al., 1988) an 18% reduction in facility of outflow was observed subsequent to preparation of the eye for anterior segment perfusion. It is likely that the observed facility decrease relates to distortion of the sclera introduced by the clamping ring rather than to removal of the iris/ciliary body, since facility of outflow is not significantly affected by either iridectomy or cyclo-
dialysis in the intact enucleated human eye (Grant, 1963). Despite a slight reduction in facility of outflow in the perfused anterior segment, the well-defined characteristics of the aqueous outflow system present in intact eyes including the effect of trabeculectomy on outflow resistance and the relationship of perfusion pressure to outflow facility were also noted in the perfused anterior segment system. Trabeculectomy resulted in a 41% reduction in outflow resistance ; approximately 18 % less than noted at the same perfusion pressure in intact enucleated human eyes after an interno trabeculotomy (Rosenquist et al., 19 8 5). The decreased change in outflow resistance in the perfused outflow tissue model may well relate to a putative increase in scleral resistance induced by the culture technique as noted above. It is probable that perfusate left the aqueous outflow channels through conventional routes of egress as suggested by the decoration of trabecular cells as well as inner wall endothelia by cationized ferritin. Fur-
730
K. ERICKSON-LAMY
ET AL.
FIG. 7. Electron micrograph of the trabecular meshwork of a 59-year-old donor after 1 hr perfusion on the fourth postmortem day (C = 0.188 ,LL~min-’ mmHg-I). There is only a slight vacuolization of the cribriform cells (C). otherwise the
trabecular meshworkand inner wall endothelium(E) of Schlemm’scanal (SC)show a normal structure. x 4500.
FIG. 8. Electron micrograph of the cribriform region of the trabecular meshwork of a human eye 2 days post-mortem after 6 hr perfusion. The eye was perfused with cationized ferritin prior to fixation as described in Materials and Methods. Ferritin particles are seen within the lumen of Schlemm’s canal (SC) and at the cell surfaces. E, Endothelial lining of SC: T, trabecular cells. Arrowheads indicate ferritin particles phagocytized by a trabecular cell. Asterisks indicate giant vacuoles i free of ferritin particles). x 5000.
thermore, the observed reduction of facility of outflow at elevated perfusion pressure [presumably due to a narrowing of Schlemm’s canal (Johnstone and Grant.
1973)] would be hard to explain if the perfusate was simply leaking through the sclera. In such a case. one would expect the facility to increase or remain
OUTFLOW
FACILITY
IN THE
PERFUSED
ANTERIOR
SEGMENT
with increased perfusion pressure. However, further systematic studies are needed to establish definitely how much of the perfusate leaves the perfused anterior ocular segment through the expected conventional routes. The manipulations necessary for perfusion of the outflow pathways of the human eye do not greatly alter well-established physiological characteristics of outflow resistance nor do they greatly alter normal morphology of the outflow tissues. Further, drug responsiveness has been demonstrated in this preparation. In recent studies, we showed that epinephrine increases facility of outflow in a dose-dependent manner (Erickson-Lamy et al., 1990b). Therefore, this model may have potential usefulnessin determining the site of action of drugs known to influence outflow resistance. Because the ciliary muscle is absent, any drug effect is presumably on the outflow pathway itself rather than via an indirect effect mediated by ciliary muscle contraction. Collectively, these results point to the potential importance of this model in investigating the biology, physiology and pharmacology of the human outflow pathway system. constant
Acknowledgements Supported in part by NIH grant EY0732 1, Massachusetts Lions Eye Research Fund, Inc. and National Glaucoma Research, a program of the American Health Assistance Foundation (K. Erickson-Lamy), and by Deutsche Forschungsgemeinschaft. Bonn-Bad Godesberg, FRG grant RO 81/18/4 (J. W. Rohen).
References Acott, T. S.. Kingsley, P. D., Samples, J. R. and Van Buskirk, E. M. (1988). Human trabecular meshwork organ culture : Morphology and glycosaminoglycan synthesis. Invest. OphthahnoZ. Iris. Sci. 29, 90-100. BBrdny. E. H. (1964). Simultaneous measurement of changing intraocular pressure and outflow facility in the vervet monkey by constant pressure perfusion. Invest. Ophthalmol. 3, 135-143. Bill, A. (1980). The drainage of aqueous humour via Schlemm’s canal and uveoscleral routes. Ophthalmic Res. 12, 130. Brubaker, R. F. ( 1975). The effect of intraocular pressure on conventional outflow resistance in the enucleated human eye. Invest. Ophthalmol. 14, 286-92. Chandler, P. A. and Grant, W. M. (1979). Glaucoma. 2nd edn. P. 23. Lea and Febiger: Philadelphia, PA. Ellingsen. B. A. and Grant, W. M. (1971). The relationship of pressure and aqueous outflow in enucleated human eyes. Invest. Ophthulmol. 10. 430-7.
731
Epstein, D. L. and Rohan, J. W. (1991).
Morphology
of the
trabecular meshworkand inner wall endotheliumafter cationized ferritin perfusion in the monkey eye. Invest. Ophthulmol. Vis. Sci. 32, 160-l 71. Erickson-Lamy K. A., Rohen, J, W. and Grant, W. M. (1988). Outflow facility studies in the perfused bovine aqueous outflow pathways. Curr. Eye Res. 7, 799-807. Erickson-Lamy, K. A., Epstein, D. L.. Schroeder, A. M. and Bassett-Chu, S. (1990a). Absence of time-dependent facility increase (I washout ‘) in the perfused enucleated human eye. Invest. Ophthulmol. Vis. Sci. 31, 2384-8. Erickson-Lamy, K. A., Ostovar, B., Hunnicutt, E. S. and Nathanson, J. A. (1990b). Epinephrine increases facility of outflow and trabecular meshworkCAMP content in the human eye in vitro. Invest. Ophthulmol. Vis. Sci. 31 (ARVO Suppl.), 184. Grant, W. M. (1958). Further studies on facility of flow through the trabecular meshwork. ,4rch. Ophthulmol. 60,
523-33.
Grant, W. M. (1963). Experimental aqueous perfusion in enucleated human eyes. Arch. Ophthulmol. 60, 783801. Hashimoto, J. M. and Epstein, D. L. (1980). Influence of intraocular pressure on aqueous outflow facility in enucleated eyes of different mammals. Invest. Ophthulmol. 19, 1483-9. Johnson, D. H. and Tschumper. R. L. (1987). Human trabecular meshwork organ culture. A new method. Invest. Ophthulmol. Vis. Sci. 28, 945-53. Johnstone, M. A. and Grant, W. M. (1973). Microsurgery of Schlemm’s Canal and the human aqueous outflow system. Am. 1. Ophthulmol. 76, 906-I 7. Macri. F. J., Cevario, S. J. (1973). The induction of aqueous humor formation by the use of Ach + eserine. Invest. Ophthulmol. 12, 910-16. Macri, F. J. and Cevario, S. J. (1975). Ciliary ganglion stimulation. I. Effects on aqueous humor inflow and outflow. Invest. Ophthulmol. 14, 2843. Melamed, S. (1986). Use of cationized ferritin to trace aqueous humor outflow in the monkey eye. Exp. Eye. Res. 43, 273-8. Moses, R. A. (198 7). Intraocular pressure. In Adler’s Physiology of the Eye. Clinical Application. (Ed. Moses, R. A.) 8th edn. P. 224. C. V. Mosby Co: St. Louis, MO. Rohen, J. W., Schachtschabel, D. 0. and Wehrmann, R. (1982). Structural changes of human and monkey trabecular meshwork following in vitro cultivation. Gruefes Arch. Clin. Exp. Ophthulmol. 218. 225-32. Rohan, J. W.. Schachtschabel, D. 0. and Berghoff, K. (1984). Histoautoradiographic and biochemical studies on human and monkey trabecular meshwork and ciliary body in short-term explant culture. Grufes’s Arch. Clin. Exp. Ophthulmol. 221, 199-206. Rosenquist, R. C., Epstein, D. L., Johnson, M., Grant, W. M.
and Melamed, S. (1985). Resistancevariation with perfusion pressure after trabeculotomy. Invest. Ophthulmol. Vis. Sci. 26 (ARVO Suppl.), 26. Schachtschabel, D. O., Berghoff, K. and Rohen, J. W. (1984). Synthesis and composition of glycosaminoglycans by explant cultures of human ciliary body and ciliary processes in serum-containing and serum-free defined media. Grufes’s Arch. Clin. Exp. Ophthulmol. 221, 207-9.
