Effects of Time of Storage, Albumin,
and Osmolality Changes on Outflow
Facility (C) of Bovine Anterior
Segment In Vitro
Arcadi Gual,* Artur Llobet* Rosa Gilabert,^
Miguel Bonds,* Jordi Pales,*
Michael V. W. Bergamini,% and Carlos Belmonte%
Purpose. To analyze the influence of time of storage,
the presence of albumin at physiological concentrations, and the perfusion with anisosmotic media on the
aqueous humor outflow facility (C) of isolated bovine
anterior segments (AS).
Methods. Anterior segments dissected from cow eyes
were perfused at a constant pressure of 10 mm Hg with
Dulbecco's modified Eagle's medium (DMEM; osmolality 300 mOsm/kg), with hyposmotic media (150, 210,
and 270 mOsm/kg), or with hyperosmotic media (360,
420, and 480 mOsm/kg). Outflow facility was calculated
every 5 seconds as the ratio between average inflow
from the reservoir (in microliters per minute) and the
perfusion pressure (in millimeters of mercury). Three
groups were studied: a 0-hour group, with AS perfused
with DMEM 1 to 3 hours after enucleation; a 0-hour
alb-group, with AS perfused with DMEM plus 0.1 mg/
ml albumin 1 to 3 hours after enucleation; and a 24hour group, with AS perfused after storage for 24 hours
in DMEM. In the O-hour groups, perfusion with increasingly hyposmotic or hyperosmotic media was also
made in 30-minute steps, followed by a return to isosmotic medium for 90 minutes.
Results. Perfusion of AS with DMEM for 9 hours caused
a progressive increase in C that was statistically significant at 225 minutes in the 0-hour group perfused with
DMEM and at 195 minutes in the 24-hour group perfused with DMEM. The 0-hour alb-group perfused with
DMEM did not show changes in C throughout the 9hour perfusion period. Perfusion with increasingly hyposmotic media induced a progressive decrease in C
that did not recover on return to isotonic medium.
Hyperosmotic media caused a progressive increase in
From the * Neurophysiology Laboratory, Faculty of Medicina-Fundacid CIJNIC,
University of Barcelona;f Department of Basic Medical Sciences, Faculty of Medicine
and Health Sciences, University Rovira i Virgili, Reus; J Cusi Research and
Development Center, Laboratorios Cusi, El Masnou; and %Neuroscience Institute,
University Miguel Hernandez, Elche, Spain.
Supported in part try Health Research Fund from Social Security (FISSS) grant 94/
1180 and PM9&0058, Ministry of Health, Spain.
Submitted for publication January 31, 1997; revised April 23, 1997; accepted April
24, 1997.
Proprietary interest category: C5, E, C3.
Reprint requests: Arcadi Gual, Laboratori de Neurofisiologia, Facultat de MedicinaUniversital de Barcelona, Casanova, 143, E-08036-Barcelona, Spain.
Cthat returned to control values when isotonic medium
was again perfused.
Conclusions. Preservation of tissue for C measurements
is best achieved with short storage time (1 to 3 hours).
Physiological concentrations of albumin (0.1 mg/ml)
prevent development of washout, suggesting that albumin or an albumin-bound factor in aqueous humor may
play a role in the maintenance of outflow resistance.
Outflow facility also may be influenced by volume
changes in the trabecular meshwork. Invest Ophthalmol Vis Sci. 1997;38;2165-2171.
X erfusion of the anterior chambers has been used to
measure aqueous humor outflow facility (C) and to
assess the influence of drugs on aqueous humor drainage.1 Erickson-Lamy et al2 described a preparation
of the isolated anterior segment (AS) that allows the
measurement of C through conventional pathways,
thus excluding the contribution to outflow resistance
of uveoscleral routes.2 A problem encountered during
perfusion through the conventional aqueous outflow
pathways is the development of washout. Washout, is
defined as a progressive decrease in outflow resistance
as time elapses.3'4 It has been suggested that the phenomenon is mainly caused by the washing out of proteins5 from the trabecular meshwork (TM), thus supporting the hypothesis that some chemical constituents of these structures play a role in the genesis
of aqueous outflow resistance. Recently, it has been
reported that serum or a serum component may contribute to outflow resistance.6 Results of studies in vitro
suggest that the volume changes of TM cells may play
a role in varying tissue permeability.7 However, the
possible functional role of volume variation of the TM
cells in the definition of C has not been established.
METHODS. Eyes were obtained in a local abattoir from calfs 3 to 6 months old and were enucleated
1 to 3 hours after death. Eyes were immediately submerged in phosphate-buffered saline with antibiotics
(described later). Enucleated eyes were bisected along
the equator to isolate the anterior segment and were
placed in a perfusion chamber.2 Mean volume of the
artificial anterior chamber, formed by the AS and the
polysulfone plastic chamber, was 1854 ± 64 fA (mean
± SEM; n = 52). Perfusion was made with Dulbecco's
modified Eagle's medium (DMEM; Bio-Whitaker, Barcelona, Spain), osmolality 300 mOsm/kg, containing
4.5 mg/ml glucose, 50 IU/ml penicillin, 50 //g/ml
streptomycin, and 5 fJ.g/m\ amphotericin B. All solutions were prepared under sterile conditions in a laminar-flow hood. Three experimental groups were stud-
2165
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
2166
Investigative Ophthalmology & Visual Science, September 1997, Vol. 38, No. 10
Oh-OMEM (n = 15)
24h-DMEM (n=9)
Oh-Alb-DMEM (n=9)
300
TIME (min)
FIGURE l. Effects of 9 hours of perfusion on outflow facility (C) in bovine anterior segments.
•, perfusion with DMEM 1 to 3 hours after enucleation (O-hour DMEM group, n = 15);
A, perfusion with DMEM plus O.I mg/ml albumin 1 to 3 hours after enucleation (0-hour
alb-DMEM group, n = 9); • perfusion with DMEM after 24 hours of storage at 4°C (24hour DMEM group, n — 9). Data are expressed as mean values ± SEM (n = number of
eyes) of the ratio of C at each time (Q) to the baseline C (Co). Statistical significance
(repeated measurements analysis of variance with Bonferroni correction) of paired comparisons between C at each time and baseline Co for each treatment group is indicated for the
period between the arrows. DMEM = Dulbecco's modified Eagle's medium.
ied: AS perfused with DMEM immediately after dissection (0-hour DMEM group), AS perfused with DMEM
plus 0.1 mg/ml bovine albumin (A-3803, Sigma, Madrid, Spain) immediately after dissection (0-hour albDMEM group), and AS stored 24 hours in DMEM
at 4°C and perfused with DMEM afterward (24-hour
DMEM group). Perfusion was carried out in an incubator at 36°C in air with 5% CO2. Perfusion pressure
was monitored throughout the experiment with a
transducer (9162-0, Mallinckrodt, Northhampton,
United Kingdom) and maintained at 10.7 ± 0.2 mm
Hg (n = 65; mean ± SEM) with a suspended reservoir
(1.5 ml). The flow rate was measured using a forcedisplacement transducer (Letica, Barcelona, Spain)
and recorded using specially designed software.
Hyposmotic solutions were prepared by adding
distilled water to the control media to final osmolality values of 270 mOsm/kg (90% of control
value), 210 mOsm/kg (70% of control value), and
150 mOsm/kg (50% of control value). Hyperosmotic solutions were prepared by adding D-sorbitol
to the control media to final osmolality values of
360 mOsm/kg (120% of control value), 420
mOsm/kg (140% of control value), and 480
mOsm/kg (160% of control value). Osmolality values were checked in an osmometer before and after
each experiment. Perfusion with hypo- or hyperosmotic solutions were performed in separate experiments. In either case, after 90 minutes of perfusion
with the control medium (300 mOsm/kg, baseline
facility Co), solutions of decreasing or of increasing
osmolality were introduced successively for 30-minute periods. At the end of a stepwise decrease or
increase in osmolality, perfusion with the control
medium (300 mOsm/kg) was resumed for another
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
90-minute period (post osmolality changes period,
Cp).
Outflow facility C was calculated as the ratio
between inflow of perfusion medium into the eye
(in microliters per minute) and perfusion pressure
(in milliliters of mercury) and was expressed as microliters per minute per milliliters of mercury. C
values were calculated and recorded every 5 seconds, using specially designed computer software.
Additionally, C values were averaged during successive 15-minute intervals for statistical analysis and
for graphic representation. As a result, each of
these 15-minute intervals, hereinafter described as
15-minute subperiods, was the mean of 180 C values. To obtain the baseline (Co), six initial 15-minute subperiods were averaged for 90 minutes
(from minute 0 to minute 90). To determine the
validity of the experiment, two criteria were established during the baseline period (initial 90 minutes): the C value of each of these six 15-minute
subperiods should be between 0.4 and 1.3 \x\/
min • mm Hg, and the differences among the C ratios calculated during the 15-minute subperiod
making up (Co) should be less than 10%. Changes
of the perfusion medium were made by rapidly replacing the contents of the anterior chamber with
new medium by opening the exit needle until 200%
of the artificial anterior chamber volume had been
exchanged; this exchange was always made at a
pressure below 10 mm Hg. Recording of C measurements started after stabilization of flow.
The influence of elapsed time on C was evaluated by measuring the ratio between C (obtained
at different times) and Co. Osmolality effects were
expressed as the ratio between facilities at the end
Effect of Time of Perfusion and Albumin on Outflow Facility (C) in Isolated Anterior Segments
EM
EM
EM
EM
n
Co
15 0.73 ± 0.05
(100%)
9 0.50 ± 0.05
(100%)
9 0.51 ± 0.05
(100%)
Cmin iso
15 0.76 ± 0.06
(104%)
9 0.47 ± 0.04
(97%)
9 0.59 ± 0.06
(109%)
Cmin 300
Cmin 15
0.69 ± 0.05
(94%)
0.47 ± 0.03
(97%)
0.57 ± 0.05
(106%)
Cmin 165
0.76 ± 0.06
(104%)
0.48 ± 0.03
(100%)
0.62 ± 0.07
(114%)
Cmin 315
Cmin 30
Cmin 45
Cmin 60
Cmin 75
Cmin 90
0.74 ± 0.05
(102%)
0.50 ± 0.03
(103%)
0.55 ± 0.05
(103%)
0.75 ± 0.06
(102%)
0.47 ± 0.04
(97%)
0.54 ± 0.05
(101%)
0.74 ± 0.06
(102%)
0.49 ± 0.04
(101%)
0.53 ± 0.04
(99%)
0.73 ± 0.05
(99%)
0.49 ± 0.04
(100%)
0.52 ± 0.05
(97%)
0.73 ± 0.05
(101%)
0.50 ± 0.05
(103%)
0.51 ± 0.05
(94%)
Cmin 180
Cmin 195
0.78 ± 0.06
(107%)
0.48 ± 0.03
(99%)
0.63 ± 0.06
(118%)
0.82 ± 0.07
(111%)
0.48 ± 0.04
(98%)
0.65 ± 0.06f
(121%)
Cmin 330
Cmin 345
15 0.89 ± 0.08]: 0.90 ± 0.08]: 0.91 ± 0.08]: 0.92 ± 0.09J
(121%)
9 0.55 ± 0.04
(114%)
9 0.75 ± 0.06|
(142%)
Cmin 450
Cmin 210
Cmin 225
Cmin 240
0.82 ± 0.07 0.85 ± 0.07* 0.86 ± 0.08f
(112%)
(117%)
(115%)
0.49 ± 0.04 0.47 ± 0.04 0.49 ± 0.04
(100%)
(97%)
(99%)
0.68 ± 0.07J 0.72 ± 0.07]: 0.73 ± 0.07]:
(133%)
(137%)
(126%)
Cmin 105
0.72 ± 0.06
(99%)
0.49 ± 0.04
(100%)
0.53 ± 0.05
(99%)
Cmin 120
0.73 ± 0.06
(100%)
0.46 ± 0.04
(94%)
0.54 ± 0.04
(102%)
Cmin 255
Cmin 270
0.88 ± 0.08J
(120%)
0.51 ± 0.05
(103%)
0.74 ± 0.07|
(138%)
0.89 ± 0.08]:
(120%)
0.50 ± 0.05
(101%)
0.75 ± 0.06]:
(142%)
Cmin 360
Cmin 375
Cmin 390
Cmin 405
Cmin 420
0.94 ± 0.09]:
(129%)
0.54 ± 0.04
(112%)
0.76 ± 0.07]:
(143%)
0.92 ± 0.09]:
(126%)
0.53 ± 0.04
(110%)
0.74 ± 0.06]:
(139%)
(124%)
0.55 ± 0.04
(114%)
0.76 ± 0.06]:
(142%)
(124%)
0.55 ± 0.04
(114%)
0.77 ± 0.07]:
(146%)
(126%)
0.55 ± 0.05
(112%)
0.76 ± 0.06]:
(144%)
0.93 ± 0.09]:
(127%)
0.55 ± 0.04
(112%)
0.75 ± 0.06]:
(142%)
0.93 ± 0.09J
(127%)
0.54 ± 0.04
(111%)
0.76 ± 0.06]:
(144%)
0.93 ± 0.09|
(127%)
0.52 ± 0.04
(108%)
0.76 ± 0.06]:
(143%)
Cmin 465
Cmin 480
Cmin 495
Cmin 510
Cmin 525
Cmin 540
Cmin
0.74 ±
(102%
0.46 ±
(95%
0.56 ±
(104%
Cmin
0.89 ±
(121%
0.52 ±
(107%
0.74 ±
(139%
Cmin
0.91 ±
(126%
0.53 ±
(109%
0.74 ±
(139%
15 0.92 ± 0.09]: 0.91 ± 0.09| 0.91 ± 0.09]: 0.90 ± 0.09]: 0.91 ± 0.09]: 0.89 ± 0.09| 0.88 ± 0.09|
(127%)
(126%)
0.54 ± 0.04
(112%)
(112%)
9 0.74 ± 0.06]: 0.74 ± 0.05]:
(139%)
(139%)
9 0.54 ± 0.04
(126%)
0.53 ± 0.03
(111%)
0.76 ± 0.06]:
(143%)
(125%)
0.52 ± 0.03
(108%)
0.76 ± 0.05]:
(143%)
(127%)
0.52 ± 0.04
(108%)
0.77 ± 0.05]:
(146%)
(123%)
0.53 ± 0.04
(110%)
0.75 ± 0.06|
(142%)
(122%)
0.51 ± 0.04
(108%)
0.77 ± 0.05J
(147%)
e facility; Cmin = facility at each time in minutes; Oh-DMEM = anterior segments perfused immediately after enucleation with DMEM; Oh-Alb-DMEM = anterior segmen
mediately after enucleation with DMEM plus 0.1 mg/ml albumin; 24h-DMEM = anterior segments perfused after 24 hours of storage in DMEM at 4°C.
n ± SEM; n = number of eyes. Average of the relation Cmin/Co is shown in parentheses. For the three groups, analysis of variance with Bonferroni correction was perf
nd the C obtained every 15 minutes.
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
2168
Investigative Ophthalmology & Visual Science, September 1997, Vol. 38, No. 10
* T* I * *
in-DMEM (n-15)
of every osmolality step (that is, with hyposmotic
values of C270, C2\o, C150 and with hyperosmotic values of C360, C42o, C48o) anc ^ ^o- Results were expressed as mean values ± SEM. Values of C at each
time versus Co were compared in the control groups
(0-hour DMEM, 24-hour DMEM, and 0-hour albDMEM) using analysis of variance for repeated
measurements with Bonferroni correction. Comparisons between C in the hypo- and the hyperosmolality protocols versus C in the corresponding
control groups measured at the same times were
made using ordinary analysis of variance with Bonferroni correction. Significance was set at P < 0.05 (*);
P< 0.01 (**) a n d P < 0.001 (***).
RESULTS
Effect of Albumin on Changes in Outflow Facility With
Elapsed Time. Figure 1 presents the C values of AS
B
from freshly collected eyes perfused in vitro for 9
hours without (0-hour DMEM group, n = 15) and with
albumin (0-hour alb-DMEM group, n = 9). Figure 1
also shows C values of AS that were stored at 4°C for
24 hours in DMEM, and then perfused with DMEM
(24-hour DMEM group, n = 9). Perfused anterior segments in the 0-hour DMEM group and in the 24-hour
DMEM group displayed a progressive increase in C
with elapsed perfusion time. The difference between
C and CQ was significant at minute 225 (P < 0.05)
in the 0-hour DMEM group, whereas in the 24-hour
DMEM group, the difference was significant at minute
195 (P < 0.05). In contrast, the 0-hour alb-DMEM
group AS did not show any significant difference between C and Co throughout the 9-hour perfusion (Table 1).
Influence of Osmolality on Outflow Facility. HyFIGURE 2. Effects of different hyposmotic media on outflow
facility (C) in perfused bovine anterior segments. The perfuposmotic Solutions. Anterior segments perfused immedision was divided into three periods of 90 minutes each. The
ately after enucleation without (0-hour HYPO group,
first and last periods were performed with 300 mOsm kg
n = 8) and with albumin (0-hour alb-HYPO group,
DMEM. The second period was performed with three steps
n = 10) showed a progressive decline of Cwhen the
of 30 minutes each of hyposmotic DMEM, 270, 210, and
osmolality of the perfusion fluid was reduced to 90%,
150 mOsm/kg, respectively (open symbols). For comparison,
70%, or 50% of that in controls (270, 210,150 mOsm/
every plot includes in filled symbols each respective control
kg;
Figs. 2A, 2B). Comparison of mean C values in
(Table 1, Fig. 1). The plots also include, expressed as a
these
experiments (Table 2) with the corresponding
solid line, the difference in C values (algebraic subtraction)
C
values
in control experiments (Table 1) shows that
between every hyposmotic group (open symbols) and their
control counterparts (filled symbols). (A) Perfusion with C decreases after reductions in osmolality were statistiDMEM 1 to 3 hours after enucleation (J, n = 8). (B) Perfucally significant from minute 150 until minute 270 (P
sion with DMEM plus 0.1 mg/ml albumin 1 to 3 hours after
< 0.001) for the 0-hour HYPO group and from minute
enucleation (A, n = 10). (C) Perfusion with DMEM after
165 until minute 270 (P < 0.01) for the 0-hour alb24 hours of storage at 4°C (•, n = 9). Data are expressed
HYPO group. The pooled data of C values throughout
as mean values ± SEM (n = number of eyes) of the ratio
the complete perfusion cycle with increasingly hyof Cat each time (Q\) to the baseline C (Co). The statistical
posmotic solutions are presented in Table 2. No sigsignificances of unpaired comparisons between C in each
nificant recovery of C was observed in the 0-hour
hyposmotic group and C in their respective control groups
HYPO group or in the 0-hour alb-HYPO group when
at the same times are indicated by asterisks. Ordinary analythey were reperfused with isosmotic medium from misis of variance was used. DMEM = Dulbecco's modified Eanute
180 to minute 270. The profile of the decline in
gle's medium.
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
Reports
2169
TABLE 2.
Effect of Osmolality Changes of Perfusion Medium on Outflow Facility (C) in
Isolated Anterior Segments
Co
Hypo-osmotic groups
(n)
Oh-Hypo (n = 8)
Oh-Alb-Hypo (n = 10)
24h-Hypo (n = 9)
Cp (,„!„ 270)
Cd: 270
Cd: 210
Cd: 150
0.59 ± 0.11
(100%)
0.66 ± 0.10
(100%)
0.65 ± 0.09
(100%)
0.50 ± 0.11
0.46 ± 0.09
(79%)*
0.52 ± 0.07
(81%)
0.52 ± 0.07
(81%)J
0.44 ± 0.08
(78%)f
0.46 ± 0.06
(73%)|
0.43 ± 0.06
(66%)t
0.50 ± 0 . 1 2
Co
Cd: 360
Cd: 420
Cd: 4S0
Cp (,,,in 270)
0.49 ± 0.06
(100%)
0.55 ± 0.09
(100%)
0.64 ± 0.12
(100%)
0.56 ± 0.08
(113%)
0.60 ± 0.12
(108%)
0.71 ± 0.15
(109%)
0.62 ± 0.07
(128%)*
0.68 ± 0.12
(124%)f
0.76 ± 0.15
(119%)
0.68 ± 0.11
(138%) J
0.73 ± 0.11
(134%)$
0.81 ± 0.16
(126%)
0.52 ± 0.09
(104%)
0.62 ± 0.09
(114%)
0.74 ±0.11
(118%)
(84%)
0.61 ± 0.09
(93%)
0.58 ± 0.08
(87%)
(84%)t
0.51 ± 0.09
(79%)f
0.40 ± 0.07
(60% )X
Hyperosmotic groups
(n)
Oh-Hyper (n = 7)
Oh-Alb-Hyper (n = 7)
24h-Hyper (n = 7)
Mean of outflow facility (C) values expressed for the groups treated with the hypoosmotic or hyperosinotic protocol. Average of the
relation Cti/Co obtained in each experiment is shown in parentheses. Analysis of variance was performed in every group pooling the
values of the experimental groups with their respective controls.
Co = average of C during baseline period; C270, C«.i0, and Ci50 = C values obtained after treatment with hyposmotic solutions of 270,
210, and' 150 mOsm/kg, respectively; CSCo, C420, and C480 = C values obtained after treatment with hyperosinotic solutions of 360, 420,
and 480 mOsm/kg respectively; Cp = C at the end of the experiments (min 270).
* P < 0.05.
f P< 0.01.
t P < 0.001.
C throughout the 270 minutes was similar in both
groups (compare Figs. 2A and 2B). Anterior segments
stored for 24 hours in DMEM and perfused with hyposmotic solutions (24-hour HYPO group, n = 9)
showed a progressive decrease in C that continued
when reperfusion with isosmotic medium was resumed (Table 2, Fig. 2C). The Cdecreases were statistically significant from minute 150 through minute 270
(P < 0.001) in comparison with the equivalent C values in the 24-hour DMEM group at the same times.
Hyperosmotic Solutions. Anterior segments perfused
immediately after enucleation with (0-hour albHYPER group, n = 7) and without albumin (0-hour
HYPER group, n = 7) and subjected to stepwise increases in osmolality showed a progressive increase
in C when the osmolality of the perfusion fluid was
increased to 120%, 140%, and 160% of control osmolality (360, 420, 480 mOsm/kg, respectively; Figs. 3A,
3B). Comparison with control values at similar times
showed that the C increases were statistically significant from minute 135 (P < 0.001) in the 0-hour
HYPER and 0-hour alb-HYPER groups. The C values
recovered gradually after the onset of reperfusion with
isosmotic medium, returning approximately to the initial values at minute 270 (Figs. 3A, 3B). No differences
in the response to the various hyperosmotic solutions
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
were observed between 0-hour HYPER and 0-hour
alb-HYPER groups. Table 2 presents the pooled data
of C throughout the complete perfusion cycle with
increasingly hyperosmotic solutions. Anterior segments stored for 24 hours in DMEM and perfused
with hyperosmotic media (24-hour HYPER group, n
= 7) showed a progressive increase in Cthat appeared
to be reversed by reperfusion with isosmotic medium
(Table 2, Fig. 3C). The C increases seen in the 24hour HYPER group did not differ significantly from
those obtained in the control group (24-hour DMEM
group), in which a gradual rise of C was observed
during long-lasting perfusion with control medium.
The apparent reversal after reperfusion with isosmotic
media did not reach statistical significance during the
90-minute period; however, by minute 270, Cwas substantially (24%) lower than that in the the corresponding control C27oDISCUSSION. In isolated AS, it has been reported that prolonged anterior chamber perfusion
causes a time-dependent increase in C that is known
as the washout effect. Similarly, in our experiments, AS
in the 24-hour DMEM group showed significant increases in C that appeared after 3 hours of perfusion
and were similar to values of washout reported by
2170
Investigative Ophthalmology 8c Visual Science, September 1997, Vol. 38, No. 10
A
for C measurements in vitro. A short interval between
enucleation and the onset of perfusion (1 to 3 hours)
partially prevented the development of the washout
effect. Nevertheless, a 9-hour perfusion period in the
presence of 0.1 mg/ml albumin (0-hour alb-DMEM
group) completely prevented the development of the
washout effect. Johnson et al5 and Kee et al6 prevented
the development of the washout adding different concentrations of serum to the perfusion media. In many
species, the total protein concentration in aqueous
humor is low, 1/500 of the plasma proteins levels.8
B
The albumin represents approximately one half of the
total protein amount in aqueous humor8 and reaches
concentrations of 0.05 to 0.07 mg/ml. In our experiments, we used 0.1 mg/ml albumin, a concentration
that can be considered to be within the physiological
range. In nonhuman species, the washout effect has
been associated with a progressive depletion of
plasma-derived proteins entering the TM rather than
with a decrease of glycosaminoglycans in the aqueous
humor pathway.5 In our experiments, the addition of
albumin to the perfusion medium successfully counteracted the increase in C caused by prolonged perfusion; thus albumin at a physiological concentration of
0.1 mg/ml prevents washout during at least 9 hours
of perfusion. This observation lends support to the
hypothesis that plasma proteins may contribute to the
resistance of the TM to the passage of aqueous humor.5'6 Nevertheless, the findings in our experiments
do not clarify whether this effect is related to albumin
FIGURE 3. Effects of difiFerent hyperosmotic media on outflow
or to an albumin-bound factor, or whether the effect
facility (C) in perfused bovine anterior segments. The perfusion
of plasma proteins is mediated by an action on the
was divided into diree periods of 90 minutes each; die first and
extracellular matrix or by an action on cell membrane
the last periods were performed with 300 mOsm/kg DMEM.
receptors.
The second period was performed widi diree steps of 30 minThe possibility that outflow resistance was modiutes each of hyperosmotic DMEM, 360, 420, and 480 mOsm/
kg, respectively {open symbob). For comparison, ever)' plot in- fied by volume changes in TM tissues was also excludes in filled symbols each respective control (Table 1, Fig.
plored by changing the osmolality of the perfusion
1). The plots also include, expressed as a solid line, the mean
medium. Hyposmotic solutions induced a prodifference in C values (algebraic subtraction) between every
nounced decrease in C, with no recovery of C values
hyperosmotic group (open symbols) and dieir control {filled sym- after reperfusion with isosmotic medium; conversely,
bols). (A) Perfusion with DMEM 1 to 3 hours after enucleation
hyperosmotic solutions elicited a marked and progres(J, n = 7). (B) Perfusion widi DMEM plus 0.1 mg/ml albumin
sive increase in C that was reversed when isosmotic
1 to 3 hours after enucleation (A, n = 7). (C) Perfusion with
medium was again perfused. It is conceivable that
DMEM after 24 hours of storage at 4°C ( • , n = 7). Data are
these changes in outflow resistance were caused by
expressed as mean values ± SEM (n = number of eyes) of the
volume modifications of TM. Under physiological cirratio of Cat each time (Q) to the baseline C (Co). The statistical
cumstances, the participation of volume regulation
significances of unpaired comparisons between Cin each hyperosmotic group and C in dieir respective control groups at die
mechanisms in the control of C is presumably insigsame times are indicated by asterisks. Ordinary analysis of varinificant. Nevertheless, the possibility that cell volume
ance was used. DMEM = Dulbecco's modified Eagle's medium.
changes contribute to the final outflow resistance in
some pathologic conditions7'9 should be considered.
4
other authors. In comparison, AS perfused immediThe available information about the effects in vitro of
ately after dissection showed a smaller and more deanisosmotic media on TM cells in culture do not allow
layed C elevation during equivalent perfusion periods.
conclusions to be made about the correlation between
The differences in response between AS in the 0-hour
C and perfusate osmolality.7 However, because paraDMEM group and those in the 24-hour DMEM group
cellular pathways participate in the passage of aqueous
stresses the importance of using fresh ocular tissues
humor through the trabecular tissue,10 an increase of
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
2171
Reports
flow is expected to occur when TM and the endothelial cells of the aqueous plexus shrink. Conversely,
cells that swell with hyposmotic solutions would increase the resistance to aqueous humor flow through
paracellular pathways, thus decreasing C. Although
the osmolality of aqueous humor does not change
under physiological conditions, there is indirect experimental evidence that volume regulation mechanisms may contribute to the adjustment of aqueous
outflow resistance. Ethacrynic acid, a nonspecific
blocker of the N a - K - C l cotransporter, increases Cin
different species.11 Thus, regulatory mechanisms of
cell volume in trabecular tissues, by maintaining the
size and hydraulic resistance of paracellular pathways,
may play a role in the determination of aqueous humor drainage.
3.
4.
5.
6.
7.
Key Words
albumin, aqueous humor, osmolality, outflow facility,
trabecular meshwork
Acknowledgments
8.
9.
The authors thank Prof. M. Wiederholt for his valuable comments.
10.
References
1. Hashimoto JM, Epstein DL, Inoue I, Shimahara T.
Influence of intraocular pressure on aqueous outflow
facility in enucleated eyes of different mammals. Invest
Ophthalmol Vis Sci. 1980; 19:1483-1489.
2. Erickson-Lamy K, Rohen JW, Grant WM. Outflow
11.
facility studies in the perfused bovine aqueous outflow
pathways. Curr Eye Res. 1988; 7:799-807.
Erickson KA, Kaufman PL. Comparative effects of
three ocular perfusates on outflow facility in the cynomolgus monkey. Cwr Eye Res. 1981; 1:211-216.
Erickson-Lamy K, Schroeder AM, Bassett-Chu S, Epstein DL. Absence of time-dependent facility increase
('washout') in the perfused enucleated human eye.
Invest Ophthalmol Vis Sci. 1990;31:2384-2388.
Johnson M, Gong H, Freddo TF, Ritter N, Kamm R.
Serum proteins and aqueous outflow resistance in bovine eyes. Invest Ophthalmol Vis Sci. 1993; 34:35493557.
Kee C, Gabelt BT, Gange SJ, Kaufman PL. Serum effects on aqueous outflow during anterior chamber
perfusion in monkeys. Invest Ophthalmol Vis Sci.
1996; 37:1840-1848.
O'Donnell ME, Brandt JD, Cuny FRE. Na-K-CI cotransport regulates intracellular volume and monolayer permeability of trabecular meshwork cells. AmJ
Physiol 1995;268:C1067-C1074.
Bermann ER. Biochemishy of the Eye. New York: Plenum
Press; 1991:151-200.
Gonsalves L, Brandt JD, O'Donnell ME. Effects of dexamethasone on Na-K-Cl cotransport activity and protein expression in trabecular meshwork cells. ARVO
Abstracts. Invest Ophthalmol Vis Sci. 1996;37:S894.
Epstein DL, Rohen JW. Morphology of the trabecular
meshwork and die inner-wall endothelium after cationized ferritin perfusion in the monkey eye. Invest
Ophthalmol Vis Sci. 1991; 32:160-171.
Epstein DL, Freddo TF, Bassett-Chu S, Chung M,
Karageuzian L. Influence of ethacrynic acid on outflow facility in the monkey and calf eye. Invest. Ophthalmol Vis Sci. 1987;28:2067-2075.
Induction of Experimental
Autoimmune Anterior Uveitis by a
Self-Antigen
Melanin Complex Without Adjuvant
(EAAU) is an organ-specific autoimmune disease induced by immunization with bovine melanin-associated
antigen (MAA) and two adjuvants (complete Freund's
adjuvant and purified pertussis toxin). This study was
undertaken to explore whether an adjuvant is required
in the induction of EEAU.
Nalini S. Bora, Ming-Dar Woon,
Michael T. Tandhasetti, Thomas P. Cirrito,
and Henry J. Kaplan
Methods. Insoluble MAA was extracted from the bovine
iris and ciliary body. Soluble bovine MAA was derived
by treatment of insoluble MAA with the proteolytic enzyme, V8 protease. Lewis rats were immunized with the
insoluble or soluble antigen, with or without adjuvant
(complete Freund's adjuvant and purified pertussis
toxin). Adoptive transfer of CD4+ and CD8+ T cells
was performed to investigate the pathogenesis of EAAU.
Purpose. Experimental autoimmune anterior uveitis
From the Department of Ophthalmology and Visual Sciences, Washington University
School of Medicine, St. Louis, Missouri.
Supported in part by gmnts from Research to Prevent Blindness and by National
Institutes of Health grants EY10543, EY09730, and EY026S7.
Submitted for publication August 2, 1996; revised March 24, 1997; accepted April
17, 1997.
Proprietary interest category: N.
Reprint requests: Nalini S. Bora, Department of Ophthalmology and Visual Sciences,
Box 8096, Washington University School of Medicine, 660 South Euclid, St. Louis,
MO 63110.
Downloaded From: http://iovs.arvojournals.org/ on 06/16/2017
Results. Experimental autoimmune anterior uveitis can be
induced in Lewisratsby immunization with 100 /u,g insoluble bovine MAA alone without die use of adjuvants. The
disease can be adoptively transferred to naive syngenic rats
by primed CD4+ T cells. In contrast, soluble bovine MAA
was not uveitogenic unless adjuvants were employed.