Biochemical and Electroretinographic Characterization of the Minipig Eye in the Context of Drug Safety Investigations

Statement on Use and Care of Animals

For all investigations, care and use of animal procedures were conducted in accordance with principles outlined in the current Guide to the Care and Use of Experimental Animals published by the Canadian Council on Animal Care and the Guide for the Care and Use of Laboratory Animals published by the Institute of Laboratory Animal Resources. Citoxlab North America’s facility is Association for Assessment and Accredidation for Laboratory Animal Care (AAALAC) accredited, and the procedures were reviewed and approved by the Institutional Animal Care and Use Committee prior to conduct. All procedures were conducted per Standard Operating Procedures in place at the time of conduct.

Ocular Parameters and Intraocular Pressure Measurements

Following euthanasia, the eyes were enucleated, trimmed, and weighed. Aqueous and vitreous humors were collected with a standard sterile needle and syringe and weighed in a pretared tube. Lense was also dissected and weighed. Following the removal of vitreous humor, the remaining tissue was weighed, and vitreous humor weighed was calculated as follows:

Vitreous weight = ocular globe weight − (aqueous humor + lens + remaining tissue weight). (1)

Intraocular pressure (mm Hg) was measured in vivo with a Mentor Tono-Pen XL tonometer. After calibration and bilateral instillation of Alcaine 0.5% to the corneal surface, at least 3 measurements were collected and averaged. Readings >10% of the mean were not used.

Indirect Ophthalmoscopy (Funduscopic)

Prior to fundus imaging, animals were anesthetized with zolazepam (minipigs and dogs) or ketamine and xylazine (primates, rats, and rabbits), positioned, and both eyes received 0.5% tropicamide as a mydriatic. Fundus images were captured using Kowa RC-2 fundus camera with photographic film.

Electrolyte and Thiol Content

Aqueous humor and vitreous humor were collected from 13 Gottingen minipigs (n = 26 eyes), 10 cynomolgus monkeys (n = 20 eyes), 9 beagle dogs (n = 18 eyes), 7 New Zealand White rabbits (n = 14 eyes), and 25 Sprague-Dawley rats (vitreous only). Intra-animal age and body weight ranges were comparable. Prior to collection, animals were deeply anesthetized, and all eyes received 0.5% proparacaine topically, followed by a 10-mL flush with physiological saline. Vitreous humor (targeted 0.2 mL) was then collected using a standard sterile 5-mL syringe with a 25-G needle, and aqueous humor (targeted 0.2 mL) was collected using a standard sterile 1-mL syringe with a 30-G needle. Following collection, the eyes were treated topically with 0.3% Ciloxan antibiotic and a bland ophthalmic lubricant was applied bilaterally to protect the corneas during recovery. Aqueous and vitreous humors were transferred into plain tubes and placed on wet ice pending analysis. Due to their viscosity, vitreous samples were diluted with sterile water (1:2) or as needed. Electrolytes were quantified in aqueous and vitreous humors by a Cobas 6000 module c501 (Roche Diagnostics, Ltd.) set to plasma/serum mode. ⁹ All parameters were expressed in mmol/L. Samples that were contaminated with blood were not considered for the analysis.

Vitreous humor from minipigs was also collected for thiol analysis following the same protocol as described earlier. Immediately after collection, a 10-µL volume in triplicate was mixed with a working solution of Measure-IT (Molecular Probes Inc, Eugene, Oregon). The plate was protected from light (dark) pending analysis. Samples were mixed with the working solution immediately after collection and then analyzed within 20 minutes of collection. Fluorescence quantification was performed using a Spectra Max plate reader (Molecular Devices, LLC) set at excitation/emission 494/517 nm and analyzed by SoftMax Pro GXP (Molecular Devices, LLC).

Electroretinograms

Electroretinograms were recorded using the EPIC-4000 (LKC Technologies, West Midlands, United Kingdom) from all animals while under isoflurane anesthesia following a dark adaptation period of at least 30 minutes prior to evaluation. Three different test items were administered as unilateral intravitreal injections into the posterior chamber at a dose volume of 100 µL. Gentamicin (5 mg/eye, n = 3), indocyanine green (ICG; 0.25 mg/eye, n = 4), and glycine (5 mg/eye, n = 3) were used as positive controls inducing retinal toxicity, light transmission interferences, and neurotransmission inhibition, respectively. The contralateral eye served as an untreated control.

A mydriatic agent (Mydriacyl) was administered to both eyes at least 30 minutes before ERG monitoring. A topical ophthalmic anesthetic (Alcaine) was applied to both corneas. The animals were instrumented with corneal electrodes (ERG-Jet), which were prewetted with 0.5% methylcellulose to ensure corneal hydration and electrical conduction. The ground electrode was placed approximately at Cz (mid-point on top of the cranium), and reference electrodes were placed under each lower eyelid. The animal’s head was then centered into the Ganzfeld dome stimulator. The ERG recording commenced following test item administration at 0.5 to 1 hour postdose for gentamycin; 2.5 to 3.5 hours postdose for ICG, and 1.5 to 2 hours postdose for glycine. The ERG protocol included the steps shown in Table 2.

A prestimulus baseline of 20 milliseconds was recorded with a total sweep length of 256 milliseconds. No prestimulus baseline was included for the photopic flicker stimuli (step 12).

Steps 1 to 7, 9, and 10 were used to evaluate scotopic luminance response and oscillatory potentials. A minimum of 10 seconds was allowed between each flash for steps 1 to 7, 9, and 10 to allow the retina to recover its sensitivity. Following step 10, animals were subjected to light adaptation, using the Ganzfeld background light, for at least 5 minutes.

Electroretinogram Analysis

The ERG waveforms were transformed with the Fast Fourier method to determine the frequency components of waveform data, which were then used to qualitatively assess the presence of electromagnetic interferences (60 Hz and harmonics). The ERG waveforms were averaged by the system for each light intensity from −4.09 log cds/m² to 0.90 log cds/m² (except for step 8 which is a single flash). Analysis of ERGs included measurement of amplitudes and latency of a-waves and b-waves for the ERG waveform average of step 4 for low intensity and step 9 for high intensity, respectively. The maximal b-wave amplitude at the saturation point of steps 1 to 10 was determined (V_max). Retinal sensitivity (Log K) was calculated using the Naka-Rushton equation (semisaturation constant) ¹⁰ and applied to the results from steps 1 to 8. The amplitude of oscillatory potentials on the ascending limb of the b-wave was extracted from the ERG traces recorded in steps 8 or 9. The waveforms from the photopic flicker stimulation were used to calculate mean latency and amplitude (Table 2).

OPEN IN VIEWER

Table 2.

Electroretinogram Protocol.

Step	Stimulus
1	Scotopic single flash repeated 3 times and averaged −4.09 log cds/m2
2	Scotopic single flash repeated 3 times and averaged −3.09 log cds/m2
3	Scotopic single flash repeated 3 times and averaged −2.49 log cds/m2
4	Scotopic single flash repeated 3 times and averaged −2.09 log cds/m2
5	Scotopic single flash repeated 3 times and averaged −1.59 log cds/m2
6	Scotopic single flash repeated 3 times and averaged −0.59 log cds/m2
7	Scotopic single flash repeated 3 times and averaged −0.09 log cds/m2
8	Scotopic single flash once for oscillatory potentials 0.41 log cds/m2
9	Scotopic single flash repeated 3 times and averaged 0.41 log cds/m2
10	Scotopic single flash repeated 3 times and averaged 0.90 log cds/m2
11	Photopic single flash of high intensity 0.41 log cds/m2 for evaluation of cone response.
12	Photopic flicker stimuli at high intensity 0.41 log cds/m2 at a frequency of 30.3 Hz used to evaluate latency and amplitude of the cone response.

Statistical Analysis

Numerical values were expressed as mean ± standard error of the mean. Statistical significance was determined using an analysis of variance test followed by a Bonferroni correction for multiple comparisons. For each test item, data were obtained from a minimum of at least 3 animals. Charts and figures were prepared with Microsoft Excel and Power Point.

Ocular Parameters and Electrolyte Characterization

Table 3 summarizes the ocular structure weights, tonometry, and thiol concentration in the vitreous humor of 7 minipigs (n = 14 eyes) of approximately equal age and body weight. Gender combined values are presented.

OPEN IN VIEWER

Table 3.

Ocular Parameters in Minipigs.^a

Item	Values
Body weight	16.8 ± 0.8 kg
Ocular globe (all structures)	4.54 ± 0.23 g
Aqueous humor	0.25 ± 0.05 g
Lens	0.31 ± 0.07 g
Vitreous humor	2.97 ± 0.24 g
Intraocular pressure (IOP)	10.6 ± 2.2 mm Hg
Thiol (vitreous humor)	8.68 ± 3.08 µM/L

^a n = 7 (5 males and 2 females).

Figure 1 shows representative pictures of fundus from different species. In addition to the differences described in Table 1 (presence of macula lutea, fovea centralis, and tapetum lucidum), minipig eyes have a slightly horizontal oval optic disc and blood vessels arising from its center, while primates have 4 pairs of venules and arterioles extending from an oval optic disc. Dogs normally have 3 to 4 retinal veins and several small arteries that emerge from the central area of the optic disc that is located at the tapetal–nontapetal junction. In rats, all veins and arteries radiate from the optic disc (center to periphery). Finally, rabbits differ the most with respect to the optic disc structure (merangiotic), which is horizontal with a deeply cupped appearance and with myelin rays extending nasally and temporally from the optic disc.

OPEN IN VIEWER

Figure 1.

Representative fundus images from different species. A, Minipig. B, Primate (Rhesus). C, Dog. D, Rabbit. E, Rat.

As shown in Table 4, when comparing across species, the rank order of electrolyte concentrations in both aqueous and vitreous humors was sodium, chloride, potassium, calcium, and phosphorus, except in rats where potassium was greater than calcium and phosphorus. Sodium and chloride concentrations were generally higher in aqueous and/or vitreous samples across species when compared to calcium, potassium, and phosphorus. For each species except rats, electrolyte concentrations were generally comparable between their respective aqueous and vitreous humors, with the exception of phosphorus, where the aqueous–electrolyte concentration was slightly higher than vitreous humor levels. Aqueous and/or vitreous humor concentrations of sodium, potassium, and chloride were generally similar to species-specific serum levels; however, phosphorus and calcium concentrations were slightly below serum levels, and vitreous potassium concentrations in rats were 2.7-fold higher than serum potassium concentrations. This difference is likely due to terminal sample collection, as potassium levels have been reported as 5.07 ± 0.09 mmol/L following in vivo collection, ¹⁴ which is comparable to levels in other species reported here.

OPEN IN VIEWER

Table 4.

Electrolyte Concentrations of the Common Preclinical Species and Humans in Aqueous and Vitreous Humors.

Species	Matrix	n	Sodium, mmol/L	Potassium, mmol/L	Chloride, mmol/L	Calcium, mmol/L	Phosphorus, mmol/L
NHP	Aqueous	8	152.00 ± 3.21	3.78 ± 0.11	116.08 ± 3.24	1.37 ± 0.06	0.36 ± 0.04
	Vitreous	10	159.00 ± 12.16	4.40 ± 0.27	123.90 ± 9.57	1.59 ± 0.16	0.24 ± 0.03
	Serum	169	151.00 ± 3.10	4.73 ± 0.60	106.00 ± 3.20	2.58 ± 0.10	2.12 ± 0.30
Dog	Aqueous	9	153.67 ± 4.58	4.93 ± 0.28	116.40 ± 2.97	1.50 ± 0.28	0.43 ± 0.13
	Vitreous	7	146.86 ± 7.82	5.30 ± 0.36	112.74 ± 6.83	1.70 ± 0.07	0.27 ± 0.02
	Serum	178	150.00 ± 2.00	4.77 ± 0.30	112.00 ± 2.00	2.73 ± 0.10	1.96 ± 0.20
Rabbit	Aqueous	7	145.29 ± 3.20	4.42 ± 0.27	107.46 ± 3.91	2.18 ± 0.04	0.72 ± 0.05
	Vitreous	7	157.00 ± 17.58	5.15 ± 0.57	113.97 ± 12.64	2.34 ± 0.25	0.33 ± 0.19
	Serum	208	143.88 ± 2.91	3.86 ± 0.38	103.11 ± 2.59	3.38 ± 0.11	1.89 ± 0.33
Pig	Aqueous	13	142.54 ± 4.36	4.69 ± 0.90	102.76 ± 6.46	1.29 ± 0.09	0.85 ± 0.08
	Vitreous	11	137.82 ± 12.25	4.84 ± 0.73	100.95 ± 11.48	1.50 ± 0.15	0.26 ± 0.09
	Serum	200	143.46 ± 3.71	4.50 ± 0.81	100.88 ± 2.76	2.61 ± 0.31	2.40 ± 0.28
Rat	Aqueous	-	-	-	-	-	-
	Vitreous	25	132.70 ± 3.25	13.00 ± 1.55	105.98 ± 3.89	1.62 ± 0.06	0.82 ± 0.13
	Serum	317	143.70 ± 1.70	4.80 ± 0.41	102.40 ± 2.70	2.62 ± 0.12	2.32 ± 0.51
Humana	Aqueous	-	147.00 ± 1.40 11	4.40 ± 0.20 11	127.00 ± 1.40 11	1.44 ± 0.15 12	0.71 ± 0.15 12
	Vitreous	-	146.70 ± 3.30 13	5.73 ± 0.86 13	121.60 ± 2.60 13	1.13 ± 0.52 13	-

Abbreviation: NHP, nonhuman primates.

^aHuman parameters are from the indicated references. Samples were collected during eye surgery.

Electroretinogram

Electroretinography was evaluated in isoflurane anesthetized minipigs treated with test items known to alter retinal function. Standard ERG waveforms from untreated eyes are presented in Figure 2. An increase in scotopic a-wave amplitudes in eyes receiving 5 mg Gentamicin (−0.59 to 0.90 log cds/m²) and eyes receiving 5 mg Glycine (0.41 to 0.90 log cds/m²) was observed when compared to control eyes (Figure 3A). Similarly, scotopic b-wave amplitudes were increased for glycine-treated eyes (−2.09 to 0.90 log cds/m²), whereas decreased amplitudes were observed in eyes given gentamicin (−1.59 to 0.90 log cds/m²; Figure 3B). At lower light intensities (2.09 to −0.90 log cds/m²), b-wave amplitude was mildly decreased after administration of 0.25 mg ICG, but at greater light intensities (≥0.41 log cds/m²), the amplitudes were generally within the control range (Figure 3B). A decrease in photopic b-wave amplitudes was noted for the glycine- and gentamicin-treated eyes. There were no treatment-related effects on scotopic or photopic a- or b-wave implicit time and latency (Figures 4A and B). As expected, marked reductions in oscillatory potentials (∼4-fold) and flicker response (∼3-fold) were noted in glycine- and gentamicin-treated eyes as well (Figure 5A).

OPEN IN VIEWER

Figure 2.

Standard scotopic waveforms in Göttingen minipigs.

OPEN IN VIEWER

Figure 3.

A, Scotopic a-wave amplitude of Göttingen minipigs treated with positive control agents. B, Scotopic b-wave amplitude of Göttingen minipigs treated with positive control agents.

OPEN IN VIEWER

Figure 4.

A, Photopic a-wave amplitude of Göttingen minipigs at 0.41 log cds/m² treated with positive control agents. B, Photopic b-wave amplitude of Göttingen minipigs at 0.41 log cds/m² treated with positive control agents.

OPEN IN VIEWER

Figure 5.

A, Oscillatory potentials of Göttingen minipigs at 0.41 log cds/m² treated with positive control agents. B, Retinal sensitivity of Göttingen minipigs at 0.41 log cds/m² treated with positive control agents.

Log K, which is representative of retinal sensitivity, was reduced in gentamicin-treated eyes. In contrast, eyes treated with glycine had a slight increase in the Log K (Figure 5B).