Assessment of Aqueous Humor Dynamics in the Mouse by a Novel Method of Constant-Flow Infusion

Abstract

Purpose.: To characterize a technique that concurrently assesses all aqueous humor hydrodynamic parameters in mouse eyes.

Methods.: Mouse outflow facility (C) was determined by multiple flow-rate infusion and episcleral venous pressure (Pe) measured by manometry. The animals were then euthanatized, eliminating aqueous formation rate (Fin) and Pe. C was determined again (C_dead) while uveoscleral outflow (Fu_dead) and Fin were deduced. To assess whether Fu_dead would remain the same as Fu_live, the animals were perfused with FITC-dextran and Fu determined. The effects of IOP-lowering drugs on the parameters of aqueous hydrodynamics were also evaluated.

Results.: Under the conditions tested, Fu_live (0.012 ± 0.003 μL/min) was not different from Fu_dead (0.015 ± 0.003 μL/min; P = 0.472). In anesthetized mice, IOP = 11.4 ± 0.2 mm Hg (mean ± SEM, n = 8), C = 0.018 ± 0.0006 μL/min/mm Hg, Pe = 5.4 ± 0.2 mm Hg, Fin = 0.14 ± 0.0007 μL/min, and Fu = 0.029 ± 0.005 μL/min. C_dead was not different from C (P = 0.317). Latanoprost reduced IOP by increasing C by 0.009 ± 0.0003 μL/min/mm Hg (P < 0.001), without affecting Fin or Fu. Betaxolol reduced Fin by 0.075 ± 0.021 μL/min (P = 0.009). Brimonidine increased C by 0.005 ± 0.0005 μL/min/mm Hg (P < 0.001) and Fu by 0.013 ± 0.003 μL/min (P = 0.007).

Conclusions.: In this study, a unique technique was developed to concurrently assess IOP, C, Pe, Fin, and Fu in the mouse eye. This experimental approach should be useful to evaluate effects of pharmacologic agents or genetic manipulations on aqueous humor dynamics in mice and other animal models.

Elevated intraocular pressure (IOP) is a significant risk factor for the development and progression of glaucoma. ^1,2 IOP is determined by aqueous humor production rate (Fin), trabecular outflow facility (C), uveoscleral outflow rate (Fu), and episcleral venous pressure (Pe), as described by the modified Goldmann equation (equation 1 in the Results section). To precisely describe and understand the regulation of IOP in normal, glaucomatous, or drug-treated eyes, accurate assessment of these parameters is very important for glaucoma research. ^3,4  

Experimental techniques for measuring these parameters have been developed and well characterized—for example, the aqueous humor outflow facility, or conductance (C), which is defined as the reciprocal of the resistance to aqueous humor outflow as it exits the anterior chamber of the eye. Techniques for measuring C have been reported for more than 45 years. In the first report, Bárány ⁵ described a two-level, constant-pressure perfusion technique for measuring C in the vervet monkey. In this technique, total C (C_tot) is measured, which is equal to the arithmetic sum of trabecular outflow facility (C_trab), uveoscleral outflow facility (Cu), and inflow facility (Cps is decreased aqueous humor secretion rate with increasing IOP). ⁶ However, in practice, Cu and Cps are generally disregarded because they are reported to represent only approximately 10% of C_tot. ⁶ C_trab is some 10- to 20-fold more pressure dependent than Cu or Cps. Thus, when determined by this method, C_tot is normally assumed to be equal to C_trab, and is usually represented by C. Since publication of Bárány's paper, this approach for measuring C has been adapted for use in several other species—for example, the cynomolgus monkey, ^{7 –14} the rabbit, ^{15 –23} the pig, ²⁴ the rat, ^25,26 and the mouse. ^{27 –29}  

Techniques to determine the other parameters have also been described. Fin may be measured directly, either invasively by the rhodamine-dextran dilution method ²⁸ or noninvasively by a fluorophotometer that measures the rate of decline of fluorescence in the anterior chamber of eyes previously given fluorescein, either by drops or iontophoresis. ^3,4,30,31 However, data obtained by these methods typically exhibit a high degree of variation. For calculation of Fu, ¹³¹I- or ¹²⁵I-labeled albumin or microspheres have been perfused through the anterior chamber, followed by periodic blood draws in the case of the radioisotope method, or by ocular dissection and histology in the case of the microsphere method. ^{32 –34} For Pe, manometric determination has been used successfully. ³⁵  

Although these methods can be used to measure the parameters of aqueous humor hydrodynamics separately, they cannot be used concomitantly in the same eye. Consequently, for any single eye, some of the values in the Goldmann equation have to be derived from eyes of a different group of animals or from the population means, which are assumed to represent or sufficiently approximate the actual values of the studied eye. The approximation may not necessarily be close and can therefore introduce significant errors in the final results and conclusion. 

In the present study, we devised a novel method in the mouse such that all values contributing to IOP can be assessed in the same eye. Hence, we could construct a complete profile of aqueous humor dynamics for each eye studied, all within a single experimental session. We chose the mouse because, in recent years, this species has been used increasingly in studies of aqueous humor dynamics and glaucoma, mainly due to its considerable structural and functional similarities to the primate eye, ³⁶ and very much better developed gene-based manipulations routinely available in this species. ³⁷ In anesthetized mice, we measured IOP by rebound tonometry. We then determined C by a multiple flow-rate infusion technique and Pe by manometry. The animal was then euthanatized and measurements of IOP and C were repeated. By comparing IOP and C before and after euthanatization, values for Fin, Fu, C, and Pe could all be obtained from the same eye, assuming that Fin and Pe were 0 and Fu was unchanged after euthanatization. This unique experimental approach, although demonstrated only in the mouse, should be applicable in other animal species. It can also be used to assess the effects of various pharmacologic agents or genetic manipulations on aqueous humor dynamics as well as to study and characterize aqueous hydrodynamics in different animal models of glaucoma. In this study, we also evaluated various glaucoma drugs to demonstrate the utility of this approach. 

Materials and Methods

Animal procedures were conducted in compliance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and with all protocols and regulations established by the Animal Care and Use Committee of Alcon Research, Ltd. 

Adult male BALB/cJ mice 30 to 42 weeks of age and weighing between 20 and 35 g were obtained from Jackson Laboratories (Bar Harbor, ME). All animals were maintained on a 12-hour light/12-hour dark cycle (lights on at 0600 hours) with food and water available ad libitum. For drug studies, topical ocular delivery of 1 to 2 drops of the drug formulations, latanoprost 0.005% (Xalatan; Pfizer, New York, NY), betaxolol HCl 0.25%; (Betoptic S; Alcon, Fort Worth, TX), or brimonidine tartrate 0.2%, (Alcon) was administered to the right eye (OD) 25 hours, 19 hours, and 1 hour before measurement of the outflow facility. The contralateral eye (OS) was treated with sterile saline solution (BSS Plus; Alcon) as the control. 

To assess aqueous humor dynamics, animals were anesthetized with an intraperitoneal injection of an anesthetic cocktail comprising ketamine (72.7 mg/kg, Fort Dodge Animal Health, Fort Dodge, IA), xylazine (1.8 mg/kg, Vetus; Butler Animal Health Supply, Westbury, NY) and acepromazine (1.8 mg/kg, Butler Animal Health Supply). Maintenance doses of the anesthetic cocktail (approximately one fourth to one half the initial dose) were given subsequently as required. Each animal was then placed on a heated pad (37°C) for maintenance of body temperature and left for 30 minutes to develop a suitable plane of surgical anesthesia, as judged by the absence of the blink reflex and the hind limb flexor withdrawal response to a toe pinch. After anesthesia, the IOP of each eye was measured with a rebound tonometer (TonoLab; Colonial Medical Supply, Franconia, NH), as described previously. ³⁸ For local anesthesia, 1 to 2 drops proparacaine HCl 0.5% (Alcaine; Alcon) was applied topically to each eye. The eyes were then cannulated with a 30-gauge steel needle and connected via PE60 tubing to a flow-through pressure transducer (BLPR; World Precision Instruments, Sarasota, FL) for the continuous determination of pressure within the system. The opposing terminal of the pressure transducer was connected via further tubing to a three-way valve, the other two ports of which were connected to (1) a 3-mL syringe loaded into a microdialysis infusion pump (SP101I Syringe Pump; World Precision Instruments) and (2) an open-ended, variable-height, raised reservoir manometer (Fig. 1). Altering the height of the reservoir relative to the eye allowed for the artificial manipulation of IOP to values ranging from 0 to 50 mm Hg. The tubing, transducer, syringe, and reservoir were all filled with sterile saline solution. 

Figure 1.

 

Schematic representation of the apparatus for assessments of aqueous humor hydrodynamics and episcleral venous pressure. The anterior chamber of the mouse eye is cannulated by using a 30-gauge steel needle, which was connected to a flow-through pressure transducer for the continuous determination of pressure within the system. The opposing end of the pressure transducer was connected to a three-way valve, which in turn was connected to (1) a 3-mL syringe loaded into a microdialysis infusion pump and (2) an open-ended, variable-height, raised reservoir manometer. The pressure transducer was connected electrically to a bridge amplifier. An analog-to-digital converter was included in the circuit, and pressure data were displayed and stored by a computer.

The following steps were then implemented in succession: (1) measurement of Pe, (2) measurement of C, (3) euthanatization of the animal, and (4) measurement of C in the euthanatized animal. 

Measurement of Pe

An estimate of Pe was obtained via an adaptation of the method of Aihara et al. ²⁸ Briefly, the three-way valve was switched such that the manometer reservoir was open to the eye. The reservoir was then lowered slowly until there was a visible reflux of blood from the episcleral veins into Schlemm's canal as visualized via direct ophthalmoscopy with a handheld ophthalmoscope (Professional; Keeler Instruments USA, Inc., Broomhall, PA). At normal IOP in the BALB/cJ mouse (∼12 mm Hg), Schlemm's canal is difficult to visualize. However, when IOP is equal to Pe, Schlemm's canal swells with blood due to reflux from the valveless episcleral venous system and rapidly develops a rich scarlet coloration. At this point, the indicated manometric pressure was assumed to be equal to Pe. 

Measurement of C

The three-way valve was switched to close the reservoir and open the microdialysis infusion pump. The pump was set to a flow rate of 0.1 μL/min, and the system was allowed time (typically, 10–15 minutes) to equilibrate to a stable pressure. Flow rate was then increased sequentially to 0.2, 0.3, 0.4, and 0.5 μL/min, and the stabilized pressures at each flow rate were recorded. An absolute flat plateau at each flow rate was in practice very difficult to obtain, perhaps due to small but continuous variations in C secondary to changing autonomic effects (which could continue for a brief time after death) on the trabecular meshwork. With experience, we determined as a working compromise that if the initial sharp rate of increase in pressure at each new flow rate had tailed off to <15% of the original increase, it would represent an acceptable degree of error in the final calculation of facility of <5%. The computed mean pressure over this part of the curve after the initial sharp increase was used to compute facility. For each eye, C (μL/min/mm Hg) was calculated as the reciprocal of the slope of the respective pressure-flow rate curves. 

Euthanatization and Measurement of C in Dead Animals

After measurement of C, the animal was euthanatized via anesthetic overdose, and after approximately 30 minutes, C was measured again (C_dead). It was assumed that after death (absence of heartbeat), aqueous production and Pe would both be reduced to 0. 

Estimation of Fu and Fin

As mentioned earlier, to calculate Fu and Fin, we used a modified Goldmann equation. The equation is normally expressed in the following manner^3,4:  

 

As just described, the IOP and Pe in each eye were measured directly by rebound tonometry and manometric methods, respectively. When the mouse eye was cannulated and the infusion pump was switched on, the equation became:  

 

where IOPp is IOP when the infusion pump is on, and Fp is the flow rate incurred by the infusion pump. By adjusting Fp to different values (0.1–0.5 μL/min), C was determined by the reciprocal of slope of the IOPp/Fp regression line. 

After euthanatization, assuming that Fu was unchanged and, because of the cessation of vascular flow, that Fin and Pe were both 0, the modified Goldmann equation may be rewritten:  

 

Algebraic manipulation of equation 3 yields:  

 

By adjusting Fp, C_dead was again determined by the reciprocal of slope of the IOPp/Fp regression line. Fu could then be calculated. 

Algebraic manipulation of equation 1 produces:  

 

The mean Fu, IOP (TonoLab; Colonial Medical Supply), C, and Pe were then substituted into equation 5, and Fin was calculated for each living eye. 

Comparison of Fu_live and Fu_dead

A pivotal assumption of this analysis is that Fu does not change significantly immediately after death. To investigate this assumption, we perfused four anesthetized living mice (eight live eyes) and four mice freshly euthanatized by anesthetic overdose (eight dead eyes) through a 30-guage needle inserted into the anterior chamber with fluorescein isothiocyanate (FITC)-dextran (10⁻⁴ M; 7000 ng/μL; Sigma Chemical Co., St. Louis, MO) dissolved in sterile saline, at a flow rate of 0.5 μL/min (live or dead), according to the methodologies previously reported. ^{2,39 –41} To obtain a value for Fu during the period 80 to 90 minutes after death, mice were euthanatized by anesthetic overdose, and then 30 minutes later the carcasses were perfused with sterile saline for 50 minutes, followed by sterile saline incorporating FITC-dextran for an additional 10 minutes. Live mice were perfused in a similar manner. After 10 minutes of perfusion with FITC-dextran, the live animals were euthanatized by anesthetic overdose, and then all eyes were enucleated and promptly dissected to isolate (1) the retina/choroid/iris–ciliary body/scleral shell, (2) the cornea/trabecular meshwork, (3) the lens, and (4) the vitreous. Each fraction was placed in a 1-mL tube with 150 μL PBS and homogenized. Homogenates were then centrifuged at 3000g for 5 minutes, 100 μL of each supernatant was decanted from each tube and placed in one well of a 96-well plate, and the fluorescence was read (excitation 492 nm, emission 518 nm). Fu for each eye was calculated as:  

 

where a is the volume of each sample (mL), b is the concentration of FITC-dextran in each sample (ng/mL), T is the time of perfusion (10 minutes). 

Fluorescence in the cornea/trabecular meshwork was excluded from the calculation, as this tissue does not form part of the uveoscleral outflow pathway. 

Statistical Analysis

The means of all drug-treated and respective paired vehicle control groups were compared by using the paired Student's t-test. Comparison between uveoscleral flow in live eyes and dead eyes was conducted with the unpaired Student's t-test. P <0.05 was considered significant. All data are expressed as the mean ± SEM. 

Results

Comparison of Fu in Live and Dead Eyes

An important assumption of the current method is that Fu is unchanged after euthanatization. The FITC-dextran perfusion methodology showed Fu_live to be 0.012 ± 0.003 μL/min (mean ± SEM, n = 8 eyes), which was not significantly different from Fu_dead (0.015 ± 0.003 μL/min, n = 8 eyes; P = 0.472; Fig. 2). These results provide a critical justification of assessing parameters of aqueous hydrodynamics using the same eye before and after euthanatization. 

Figure 2.

 

Comparison of Fu in eyes of anesthetized (live) versus euthanatized (dead) mice as determined by perfusion of FITC-dextran. Bars represent mean ± SEM (n = 8 eyes per group). There was no statistically significant difference between the two groups (P = 0.472).

Aqueous Humor Hydrodynamics in Untreated Control Eyes

To establish and characterize baseline values, untreated mouse eyes were evaluated. An example of a pressure–flow rate relationship in a mouse eye before and after euthanatization is shown in Figure 3. As seen in the pressure trace (Fig. 3A) and the corresponding pressure-flow rate curves in the live (Fig. 3B) and euthanatized (Fig. 3C) animal, there was a linear relationship (r ² > 0.99) between pressure and infusion rate (within the range of 0.1–0.5 μL/min). Before euthanatization, the slope of the regression line was 58.7, the reciprocal (outflow facility; C) of which is 0.017 μL/min/mm Hg. After euthanatization, the slope of the regression line was 55.2; the corresponding C was 0.018 μL/min/mm Hg, suggesting that outflow facility did not change in the study period immediately after euthanatization. Furthermore, the y-intercept of the regression line before euthanatization was 17.6 mm Hg, representing the IOP of this eye under anesthesia. After euthanatization, the y-intercept became almost 0 mm Hg, indicating that aqueous production stops after cessation of heart beat. Similar results from eight control eyes (y-intercept = −0.44 ± 2.44 mmHg, mean ± SEM) corroborated these conclusions. 

Figure 3.

 

An example of aqueous outflow facility determinations in an eye of an anesthetized and subsequently euthanatized mouse. (A) Pressure–time trace obtained from a single experiment. In the anesthetized animal, IOPp was monitored continuously as the Fp was set at different rates. The pump was stopped and the circuit opened to the manometer, which was then rapidly lowered to re-establish baseline pressure. After this, the circuit was closed again, the animal was euthanatized, and perfusion was resumed and IOPp monitored. (B) Linear relationship of IOPp versus Fp in the anesthetized (live) animal eye. C, which equals the inverse of the slope of the regression line, C_live = 0.017 μL/min/mm Hg. (C) Linear relationship of IOPp versus Fp the same animal eye post mortem. C_dead = 0.018 μL/min/mm Hg.

The IOP, Pe, and C (before and after euthanatization) of the eight control eyes were measured, and Fin and Fu were calculated. A summary of the results is presented in Table 1. Thus, this cohort of anesthetized BALB/cJ mice had a mean IOP of 11.4 mm Hg, a Pe of 5.4 mm Hg, an Fin of 0.14 μL/min, a pressure-dependent C of 0.018 μL/min/mm Hg, and an Fu of 0.029 μL/min, which accounted for 20.5% of Fin. Moreover, confirming the results in Figure 3, facility after euthanatization (C_dead) was not significantly different from facility before death (C_live; P = 0.317). 

Table 1.

 

IOP and Aqueous Humor Hydrodynamics Parameters for Untreated Control Mouse Eyes

IOP	11.4 ± 0.2 mm Hg
C	0.018 ± 0.0006 μL/min/mm Hg
Cdead	0.018 ± 0.0003 μL/min/mm Hg
Pe	5.4 ± 0.2 mm Hg
Fin	0.140 ± 0.0007 μL/min
Fu	0.029 ± 0.005 μL/min
	20.5 ± 3.3 % of Fin

 All values are expressed as the mean ± SEM (n = 8 eyes). C and C_dead were not significantly different (P = 0.317, paired t-test).

Effects of Latanoprost

Under the present study conditions, topical ocular administration of latanoprost reduced IOP by 33.6% ± 1.2% compared with that in the contralateral vehicle-treated control eyes (from 11.8 ± 0.2 to 7.8 ± 0.2 mm Hg; n = 6; P < 0.001), increased C by 55.3% ± 1.8% (from 0.016 ± 0.0003 to 0.025 ± 0.0004 μL/min/mm Hg; P < 0.001), and reduced Pe by 50.0% ± 5.0% (from 5.7 ± 0.2 to 2.8 ± 0.3 mm Hg; P < 0.001; Fig. 4). It did not significantly affect Fin (vehicle, 0.168 ± 0.008 μL/min; treated, 0.170 ± 0.011 μL/min; P > 0.05) or Fu (vehicle, 0.023 ± 0.007 μL/min; treated, 0.031 ± 0.007 μL/min; P > 0.05). 

Figure 4.

 

Effects of topically administered latanoprost on mouse IOP, aqueous humor hydrodynamics, and Pe. All values were obtained as described in Methods. Bars represent mean ± SEM (n = 6). ***P < 0.001 versus sterile saline vehicle control.

Effects of Betaxolol

Topical ocular administration of betaxolol reduced mouse IOP by 38.1% ± 2.0% (from 13.4 ± 0.6 to 8.2 ± 0.4 mm Hg; n = 8; P < 0.001), mainly due to a decrease in Fin by 40.5% ± 10.5% compared to contralateral control eyes (from 0.166 ± 0.091 to 0.091 ± 0.016 μL/min; P = 0.009; Fig. 5). Betaxolol had no significant effect on C (vehicle, 0.018 ± 0.002 μL/min/mm Hg; treated, 0.017 ± 0.001 μL/min/mm Hg; P > 0.05), Fu (vehicle, 0.015 ± 0.008 μL/min; treated, 0.020 ± 0.010 μL/min; P > 0.05), or Pe (vehicle, 4.6 ± 0.3 mm Hg; treated, 4.2 ± 0.2 mm Hg; P > 0.05). 

Figure 5.

 

Effects of topically administered betaxolol on mouse IOP, aqueous humor hydrodynamics, and Pe. All values were obtained as described in Methods. Bars represent mean ± SEM (n = 8). **P < 0.01, ***P < 0.001 versus sterile saline vehicle control.

Effects of Brimonidine

Topical ocular administration of brimonidine reduced IOP by 15.8% ± 1.2% (from 11.8 ± 0.1 to 9.9 ± 0.1 mm Hg; n = 6; P < 0.001). Compared to contralateral control eyes, it increased C by 29.3% ± 4.1%, from 0.017 ± 0.0006 to 0.022 ± 0.0003 μL/min/mm Hg (P < 0.001), and Fu by 58.4% ± 19.9%, from 0.022 ± 0.006 to 0.035 ± 0.008 μL/min (P = 0.007; Fig. 6), but did not significantly affect Fin (vehicle, 0.144 ± 0.011 μL/min; treated, 0.154 ± 0.015 μL/min; P > 0.05) or Pe (vehicle, 4.7 ± 0.3 mm Hg; treated, 4.5 ± 0.3 mm Hg; P > 0.05). 

Figure 6.

 

Effects of topically administered brimonidine on mouse IOP, aqueous humor hydrodynamics, and Pe. All values were obtained as described in Methods. Bars represent mean ± SEM (n = 6). **P < 0.01, ***P < 0.001 versus sterile saline vehicle control.

Discussion

In the present study, a novel technique of constant-flow infusion is presented as a means of measuring C in the mouse. The technique was combined with tonometry and manometry so that preperfusion measurements of IOP and Pe could be made. By repeating perfusions in the same animals post mortem, where both Pe and Fin would be reduced to 0, values for Fu and Fin could be deduced. 

The advantages of a constant-flow infusion method over the more traditional Bárány's constant-pressure perfusion technique ⁵ are minimal cost and ease of data acquisition. To obtain required precise measurements, an adjustable height reservoir suspended from a very sensitive strain gauge is required in Bárány's technique. The apparatus must be set up on a vibration-dampened bench top, with restricted movement of laboratory staff. Because of the very small changes in voltage output (typically in the μV range), it may be necessary to enclose the apparatus within a Faraday cage to shield from external electrical interference. Despite these precautions, a slight disturbance in electrical fields or mechanical disturbance sensed by the apparatus may compromise an entire experiment. In addition, the small size of the mouse eye with minute changes in fluid volume movement requires greater sensitivity and further magnifies the problem. 

For the constant-flow infusion used in the present study, a strain gauge is not required, because flow is induced with an inexpensive and reliable microdialysis infusion pump at the desired rates. The resultant pressure generated is determined with a pressure transducer with an output signal in volts (rather than microvolts) with a baseline sufficiently stable for the study period. The apparatus can be set up in a limited bench space, and elaborate vibration-dampening and electrical shielding are unnecessary. The procedure is technically straightforward, with a low failure rate (∼5%) once the necessary skills have been acquired. This method also allows determination of the pressure-flow rate relationship using several infusion flow rates (we used five different flow rates) to ascertain linearity of the pressure-flow rate response. In numerous other studies of aqueous outflow facility reported in the literature, only two flow rates have been used, which raises the question of whether the response is linear. Our present study has confirmed a linear relationship. 

The most unique feature of the current technique is measurement of C in eyes of euthanatized animals. Together with IOP and Pe measurements, we were able to calculate Fin and Fu. The distinctive advantage of this approach is that all aqueous humor hydrodynamics parameters described in the Goldmann equation can be characterized in the same eye. This eliminates the need to derive some of the parameters from eyes of a different group of animals or from the population means, thus eliminating the often erroneous supposition that different eyes have similar numerical values. However, for the current method to be useful, several assumptions are essential. Specifically, within the time period studied, euthanatization does not affect C and Fu, but reduces Fin and Pe to 0. In this report, we demonstrated empirically the apparent correctness of most of these assumptions. We showed that the slope of the pressure-flow rate regression line, and thus C, did not change due to euthanatization (Table 1, Fig. 3). We showed that Fu was not affected significantly by euthanatization with the FITC-dextran perfusion method (Fig. 2), but the mean Fu_dead was 25% greater than the corresponding Fu_live. However, variability in Fu actually leads to a much smaller percentage change in the calculated value for Fin. Using the parameters shown in Table 1 as an example, where IOP = 11.4 mm Hg, C = 0.018 μL/min/mm Hg, Pe = 5.4 mm Hg, Fu = 0.029 μL/min, then the calculated value for Fin is: Fin = {[0.018 × (11.4 – 5.4)] + 0.029} μL/min = 0.137 μL/min. If Fu was overestimated and the correct value was 25% smaller, or 0.0217 μL/min, then, with all other factors remaining constant, Fin = {[0.018 × (11.4 – 5.4)] + 0.0217} μL/min = 0.130 μL/min, a decrease of only 5.1%. A survey of all the calculated Fin values using the parameters generated in Figures 4, 5, and 6, with both the vehicle and drug-treated groups, shows that a 25% overestimation of Fu will reduce Fin values by only 2.3% to 5.1%. We also showed that the y-intercept of the pressure-flow rate regression line, and thus Fin, approached 0 after euthanatization (Fig. 3). For obvious reasons, we were not able to measure Pe directly after euthanatization. However, the assumption that Pe became 0 after the cessation of heart beat and loss of blood pressure should be reasonable and appropriate. Therefore, the approach to aqueous hydrodynamics study reported herein appears valid and meaningful. 

A further assumption underlying the present study would be that Pe remains constant in the face of different perfusion pressures. However, we predict that if Pe increases with increasing flow rate, then, assuming C to be constant, the pressure-flow rate line will become a curve, with different slopes at different flow rates. Our observed linearity of the pressure-flow rate line argues that Pe does not change significantly in the face of increasing flow rates/perfusion pressures. Furthermore, the actual perfusion pressures at each flow rate match closely with the predicted pressure using the modified Goldmann equation, assuming a constant value for Pe at each flow rate. Thus, we believe that in the present studies in mice, perfusion at various flow rates and perfusion pressures is likely to have only a minimal effect on Pe. In addition, Mäepea and Bill ⁴² reported that, in anesthetized monkeys, in which the episcleral veins, Schlemm's canal, and the trabecular meshwork were cannulated with a micropuncture technique and in which IOP was measured, a stepwise increase in IOP to 50 cm H₂O (36.8 mm Hg) corresponded with an increase in Pe of only 1.5 cm H₂O (1.1 mm Hg). This result suggests that Pe is modulated (probably by vasoactive neurotransmitters released by nerves innervating the episcleral veins) and thus has the capacity to remain reasonably constant, even in the face of artificially increased IOP. 

By using this methodology, some values of our assessments in mouse aqueous hydrodynamics are similar to those reported by others. For example, the mean IOP in this study ranged 11.4 to 13.8 mm Hg in the anesthetized mice, which correlates well with those previously published (see review in Ref. 43). Similarly, the mean Fin values in this study were 0.14 to 0.17 μL/min, which are similar to the 0.18 ± 0.05 μL/min reported by Aihara et al. ²⁸ in NIH Swiss White mice. Nonetheless, it is important to note that Fin is likely to be influenced by the measurement technique and anesthesia. Toris et al., ³⁰ using fluorophotometry, found Fin of the CD1 mouse to be 0.09 ± 0.01 μL/min with ketamine/xylazine anesthesia, and 0.20 μL/min with Avertin anesthesia. They also described a Fin value of 0.38 ± 0.07 μL/min when obtained by the fluorescein-dextran method. 

However, despite the similarities, some values observed in the present study differ from those reported by Aihara et al. ²⁸ in NIH Swiss White mice. The mean C values in this report were 0.016 to 0.018 μL/min/mm Hg, compared with 0.0051 ± 0.0006 μL/min/mm Hg shown by Aihara et al. The mean Fu values in this report were 0.012 to 0.029 μL/min, compared with 0.148 μL/min shown by Aihara et al. The mean Pe in this study ranged from 4.6 to 5.7 mm Hg; Aihara et al., reported a Pe of 9.5 ± 1.2 mm Hg. ²⁸ Several reasons may explain, at least partially, these differences. Different strains of mice (BALB/cJ mice in the present study versus NIH Swiss White mice) and different anesthesia cocktails (ketamine/xylazine/acepromazine in the present study versus ketamine/xylazine) were used. These may affect the measured parameters. Furthermore, in our study, we waited for approximately 30 minutes after administration of anesthesia before cannulation of the eyes and measurement of Pe, by which time, the systemic drop in blood pressure after xylazine administration ⁴⁴ should have occurred and stabilized, which could account for the lower Pe values in our study. 

Several IOP-lowering glaucoma medications were also evaluated for their effects on aqueous hydrodynamics in the mouse. We confirmed that the prostaglandin FP agonist latanoprost, the β-blocker betaxolol, and the α₂-adrenergic agonist brimonidine lowered IOP, similarly to previous studies. ^{27,45 –51} We also found that latanoprost increased C without affecting Fin or Fu in the present study (Fig. 4). These results are in agreement with the findings of Avila et al., ³¹ who showed that latanoprost increases the rate of fluorescein decay in the aqueous of three strains of mice (Black Swiss, DBA/2J, C57BL/6), suggesting improvement of aqueous outflow facility. Crowston et al. ²⁷ also reported that latanoprost, when applied topically to the mouse eye, lowered IOP by ∼11% and increased C by ∼40% without affecting Fu. Fin measured by the rhodamine-dextran dilution method was unaffected by latanoprost treatment. ^27,28 The lack of effect of latanoprost on mouse Fu is different from that on primate Fu. Prostaglandin FP agonists significantly increase monkey and human Fu, presumably via their actions on the ciliary muscle. ^32,33,52,53 However, the ciliary muscle is not as well developed in the mouse eye and is unlikely to be involved in the regulation of aqueous humor outflow compared with the primates. ³⁶ Interestingly, we also observed that latanoprost induced a significant reduction in mouse Pe (Fig. 4). This finding is of interest and warrants further investigation. Brubaker ⁵⁴ cited a human study in which the prostanoid bimatoprost reduced tonographic resistance to aqueous outflow by 26%. However, this reduction in outflow resistance was inadequate to account for the observed reduction in IOP, and thus it was concluded that bimatoprost may also reduce Pe (although in this study no attempt to measure Pe was made). However, no reports exist as yet in the literature indicating any effects of other prostanoids, such as latanoprost or travoprost, on this parameter. ^{55 –57} The biological significance of these findings has yet to be elucidated. 

We found that betaxolol lowered mouse IOP by decreasing Fin without affecting other parameters (Fig. 5). These results corroborate our previous understanding of the mechanism of action of β-blockers. ⁵⁸ The effect is similar to that of another β-antagonist levobunolol, which decreased the rate of fluorescein decay in the aqueous, indicating a reduction in the rate of aqueous secretion in the mouse. ³¹  

In the mouse, the α₂-agonist brimonidine lowered IOP by increasing C and Fu without producing significant changes in Fin or Pe (Fig. 6). In humans, brimonidine reduces IOP via a reduction in Fin and an increase in Fu. ⁵⁹ However, the related compound apraclonidine (also an α₂-agonist and also reduces IOP in humans) decreases both Fin and Fu. ⁶⁰ These apparent differences between two drugs within a specific pharmacologic class and between species are interesting and merit further investigation. 

In conclusion, we have devised a novel method to evaluate IOP, Fin, C, Fu, and Pe in the same eye. C and Pe can be easily and reproducibly measured with a constant-flow infusion method in the mouse, and in combination with accurate and reproducible IOP measurements, ^{38,61 –64} Fu and Fin can be estimated, all in a single session and in the same eye. An ability to assess all major elements of aqueous humor dynamics in a single eye in a single experimental session represents an improvement over other techniques reported previously, where not all parameters can be determined simultaneously. ^{5,7 –24} In addition, the apparatus required is minimal and can be set up in any conventional laboratory. The resultant data are reproducible, are sensitive in the expected manner to standard drug treatments, and compare favorably with data reported previously by other laboratories attempting to determine these parameters in the mouse eye using somewhat more complex methodologies. 

Footnotes

 Supported by Alcon Research, Ltd.

Footnotes

 Disclosure: J.C. Millar, Alcon Research, Ltd. (E); A.F. Clark, None; I.-H. Pang, Alcon Research, Ltd. (E)

The authors thank Allan R. Shepard for assistance with the FITC-dextran assay.