Disruption of Emmetropization and High Susceptibility to Deprivation Myopia in Albino Guinea Pigs

Abstract

Purpose.: To compare emmetropization in albino and pigmented guinea pigs.

Methods.: Distributions of refractive state were examined in 214 albino and 234 pigmented guinea pigs. Albino (A) and pigmented (P) guinea pigs were divided into two groups, hyperopic (H) and myopic (M). Eye development was separately followed in 10 randomly selected animals from each group (AH, AM, PH, PM) from 2 to 10 weeks of age. In addition, deprivation myopia was induced in 36 age-matched albino (18 AH and 18 AM) and 36 pigmented (18 PH and 18 PM) guinea pigs by diffusers that were worn from 2 to 6 weeks of age. Finally, sclera fibril diameters were measured using transmission electron microscopy.

Results.: Strikingly, the distributions of refractive errors were bimodal at 2 weeks of age, both in albino and pigmented animals, with clearly different averages (−2.86 ± 5.60 diopters [D] vs. 2.13 ± 5.27 D respectively; t = 9.712; P < 0.001). Spontaneous myopia was more common in albino animals: 70.1% were myopic (AM) and 29.9% hyperopic (AH), whereas only 28.6% were myopic (PM) and 71.4% hyperopic (PH) in pigmented guinea pigs. Different from PM and AM did not show any recovery from myopia. With diffusers, AH became more myopic (−7.61 ± 2.71 D and −11.17 ± 2.55 D) than PH (−4.48 ± 1.46 D and −8.28 ± 2.13 D) after 2 and 4 weeks, respectively. Deprivation myopia could still be induced in PM (−1.64 ± 1.44 D and −5.17 ± 1.88 D after 2 and 4 weeks, respectively; P < 0.01) but not in AM. Scleral fibril diameters were smaller in myopic animals, both albino and pigmented.

Conclusions.: Deprivation myopia could not be induced in spontaneously myopic but only in hyperopic albino guinea pigs, where it was even higher than in pigmented animals. The distinct effects of albinism on emmetropization will help to elucidate the mechanisms underlying the emmetropization.

While vertebrate eyes grow considerably from birth to adulthood, based on their genetic programs (for instance: the human eye grows from 17 mm length at birth to 24 mm), the match between focal length and eye length is not optimal at the beginning. These variables are tuned to each other during postnatal development by a process called emmetropization and it is well established that visual cues are used to achieve the best match. Accordingly, variability in refractive state is high at birth, both in humans ¹ and chickens, ² but declines with age. 

There is ample evidence that emmetropization is controlled by a visually-guided feedback loop ^3,4 which is modulated by genetic factors. ^5,6 Myopia can arise either as a result of abnormal function of the visual feedback loop that controls axial eye growth, or as a result of inappropriate visual input. In addition, based on genetic factors, globe dimensions can deviate from normal such that the refractive state is outside the range over which emmetropization can operate, and high pathologic forms of myopia may result. 

The initial experimental evidence for a visually-guided feedback loop for emmetropization came from animal models in which deprivation of a sharp retinal image was found to trigger exaggerated eye growth. Furthermore, it was found that imposing optical defocus by spectacle lenses can either enhance (negative lenses) or inhibit (positive lenses) axial eye growth. ^{7 –9} The importance of visual feedback in the process of emmetropization in humans is also suggested by the fact that myopia can be induced by too much near-work. ¹⁰ As mentioned above, genetic factors can modulate the visually-guided feedback loops. For example, different chicken strains have been shown to respond differently to equivalent experimental manipulations. ^11,12 Human twin studies have demonstrated the importance of genetic factors in refractive error development. ^13,14  

Currently, the guinea pig represents an important new mammalian model to study myopia for the following reasons. (1) It is precocial, born with a well-developed visual system. ¹⁵ (2) Its eye growth responds readily to manipulations of visual experience, such as treatment with diffusers or lenses. ^16,17 (3) Both albino and pigmented guinea pigs have good spatial vision (approximately 2.4 cyc/deg, shown in an optomotor paradigm), ¹⁸ which is better than pigmented rats (approximately 1.5 cyc/deg) ¹⁹ or mice (0.5 cyc/deg), ²⁰ but similar to the tree shrew (1.2–2.4 cyc/deg). ²¹ Good visual acuity is a prerequisite for proper emmetropization. (4) The guinea pig is easy to breed and to handle, and very cooperative. (5) It has larger eyes than mice ²² and rats. ²³ (6) It is a mammalian model and has therefore more similarities to primates than the common avian myopia model, the chicken. In fact, recent work has demonstrated clear differences between guinea pig and chicken: compensation of defocus imposed by positive lenses required an intact optic nerve in guinea pigs, but not in the chick. ²⁴ (7) Guinea pigs are available as pigmented and albino strains, making it possible to study the effects of albinism on emmetropization. Human albinos exhibit both high myopia or hyperopia, as well as impaired emmetropization. ²⁵ Furthermore, Wildsoet and colleagues (Wildsoet CF, et al. IOVS 2002;43:ARVO E-Abstract 2926) ²⁶ have shown that emmetropization is abnormal in an albino chicken model. Finally, (8) pigmented guinea pigs have been recently found to display spontaneous myopia, ²⁷ and the present study shows that the same is true for albino animals. Spontaneous myopia is potentially of high interest because this type of myopia may be more similar to human myopia than experimentally-induced versions. 

In the present study, emmetropization was studied in albino and wild type pigmented guinea pigs to find out how albinism might affect emmetropization in this animal model. Furthermore, it was studied how emmetropization might be different in guinea pigs with spontaneous myopia. Three experiments were performed:

•  
  Experiment 1—refractive error development was followed in spontaneously myopic albino and pigmented animals.
•  
  Experiment 2—refractive error and eye growth were measured in response to deprivation of sharp vision, a condition which is known to induce deprivation myopia in other animal models and guinea pigs. ^16,28,29
•  
  Experiment 3—to better understand the involvement of the scleral tissue in myopia development in the albino guinea pigs, fibril diameters were examined by transmission electron microscopy.

Methods

Animals

In total, 448 guinea pigs (Cavia porcellus, “the English short hair stock,” tricolor strain and albino strain) were involved in this study. Albino and pigmented guinea pigs were reared under a 12-hour light/12-hour dark cycle in the animal facilities, as previously described, with normal visual exposure to the cage environment. ³⁰ The room temperature was kept at 25°C. All guinea pigs had free access to standard food and water, and fresh vegetables were provided twice daily. Procedures were in agreement with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and the animal research was approved by the Animal Care and Ethics Committee at Wenzhou Medical College, Wenzhou, China. 

Biometric Measurements

Refractive errors were measured in the vertical pupil meridian by an eccentric infrared photorefractor, as previously described ²⁷ at a camera distance of 1 m. Alert guinea pigs are inherently cooperative, and it was easy to align their heads by hand until the pupil was clearly visible in the video frame. Room light was dimmed to approximately 5 lux ambient illuminance although both albino and pigmented guinea pigs showed a striking absence of a flash-induced pupillary light response. For each eye, three readings of the refractive error in the vertical meridian were recorded, and averaged data were used for further analyses. 

Corneal curvature was measured in alert guinea pigs with a modified keratometer (OM-4, Topcon, Tokyo, Japan) mounted with a +8.0 diopters (D) lens, as previously described. ³⁰ In brief, the corneal radii of two perpendicular meridians of each eye were measured three times, and then averaged. Corneal refractive power was calculated using F = r/(n _cornea − n _air), where n _cornea is the refractive index of the cornea (assumed 1.336 from Howlett's schematic guinea pig eye ¹⁶ ), r is corneal radius of curvature (m), and F the refractive power in diopters. Both eyes were measured but only data of right eye were calculated and reported in Table 1. 

Table 1.

 

Corneal Radius of Curvature and Astigmatism in the Right Eyes in Albino and Pigmented Guinea Pigs

Age (wk)	Radius (mm)*		Astigmatism (D)†
	Albino	Pigmented	Albino	Pigmented
2	3.337 ± 0.068	3.319 ± 0.048	3.92 ± 1.65‡	1.11 ± 1.00
4	3.406 ± 0.052	3.442 ± 0.068	2.47 ± 1.18‡	0.86 ± 0.88
6	3.498 ± 0.048‡	3.567 ± 0.061	1.75 ± 0.83‡	0.88 ± 1.21
8	3.622 ± 0.047	3.658 ± 0.091	1.63 ± 1.00‡	0.60 ± 0.99
10	3.687 ± 0.056‡	3.754 ± 0.052	1.48 ± 1.05‡	0.34 ± 0.99

 Data are presented as mean ± SD.

*  Averages of the two radii of corneal curvature in right eyes in the vertical and horizontal meridian.

†  Astigmatism of right eye refers to the difference in the calculated powers of the two measured corneal meridians.

‡  Indicates that there was a significant difference between albino and pigmented guinea pigs at the respective time point (P < 0.05).

Axial dimensions of the eyes were measured in alert animals with A-scan ultrasound (AVISO Echograph Class I-Type Bat, Quantel Medical, Clermont-Ferrand, France) on the same day as the refractive error. The cornea was topically anesthetized with one drop of 0.5% proparacaine hydrochloride (Alcon, Puurs, Belgium). Velocities of sound were assumed to be 1534 m/s for the aqueous and vitreous humor, and 1774 m/s for the lens. A stand-off rubber tube was attached to the probe tip, as described by Schaeffel and Howland. ³¹ Recorded parameters were the principal ocular components, i.e., anterior chamber depth, lens thickness, vitreous chamber depth, and axial length. Each eye was measured at least eight times, and the averages of these measurements were used for further analyses. 

Experimental Protocol

Albino (n = 214) and pigmented guinea pigs (n = 234) were refracted in both eyes at 2 weeks of age. To be able to study emmetropization separately in myopic and hyperopic animals, both the albino and the pigmented group were divided into two subgroups, based on the refractive errors of their right eyes (albino: AM and AH; pigmented: PM and PH, respectively). Animals were assigned to either group just by the sign of the photorefractor readings although it is clear that their true refractive errors were likely somewhat more myopic because the small eye artifact was ignored. ³² Refractive state was treated as a continuous variable with no special consideration of the value of zero. Due to the small eye artifact, it was unknown which of the measured refractive errors were equivalent to emmetropia—probably not zero but some amount of hyperopia. 

Refractive error, corneal curvature, and ocular dimensions (anterior chamber depth, lens thickness, vitreous chamber depth, axial length) of both eyes were measured every other week. To study normal refractive error development, 20 albino guinea pigs (10 hyperopic and 10 myopic) and 20 pigmented guinea pigs (10 hyperopic and 10 myopic) were randomly selected at 2 weeks of age and followed until 10 weeks of age (Experiment 1). Thirty-six albino guinea pigs (18 hyperopic and 18 myopic) and 36 pigmented guinea pigs (18 hyperopic and 18 myopic) were randomly selected, and deprivation myopia was induced in half of the sample from each group by treating them with a face mask as previously described. ²⁹ In brief, a rubber balloon of suitable size was adapted to fit over the head. Five holes were cut into it, one for neck, one for mouth, one for contralateral eye to receive normal vision, and two for ears to stabilize the position of the face mask; the experimental eyes remained covered by the rubber tissue with approximately 2 log units of light attenuation, as measured with a spectrophotometer (U-4100; HITACHI, Tokyo, Japan). Treatment started at 2 weeks of age and continued until 6 weeks of age (Experiment 2). For analysis of scleral tissue by electron microscopy, three additional animals from each subgroup (AH, AM, PH, PM) which were not included in Experiment 1, were killed at the age of 2 weeks. Another three animals from each subgroup (AH, AM, PH, PM) from Experiment 1 were killed for examination of the sclera tissue at the age of 10 weeks. 

Electron Microscopy of Scleral Tissue

The animals were killed by a lethal dose of pentobarbital sodium (130 mg/kg). A small mark was made at the temporal corneal limbus, using an indelible ink marker pen, to allow orientation of the eye cup after enucleation. Eyes were immersed into 0.05 M cacodylic acid sodium and 2.5% glutaraldehyde with phosphate buffer saline (pH 7.4) for 2 hours. At the end of this period, cornea and lens were dissected away, leaving the mark in the limbal region. Furthermore, a cornea trephine of 6 mm diameter was used to punch out a posterior tissue button (containing retina, choroid, and sclera). Two 2 × 1 mm strips of sclera tissue were excised from the tissue button near the nasal region, 1 mm away from the optic nerve with razor blades. The strips were immersed in 0.05 M cacodylic acid sodium and 2.5% glutaraldehyde for 24 hours. Tissue samples were postfixed in 1% osmium tetroxide with phosphate buffer saline (pH 7.4) for 1 hour at 4°C, stained with 1% uranyl acetate with double distilled water for 2 hours, then rinsed and dehydrated in graded acetones before embedding in Araldite. Micrographs of histologic cross-sections were collected using transmission electron microscope (Hitachi 7650, Tokyo, Japan). The sclera tissue button was divided into outer, middle, and inner layers and 4 to 6 electron micrographs were taken from the middle layer of the sclera for each eye at magnification ×10,000. Scleral fibril diameter was analyzed by the method as previously described. ³³ In brief, the diameters of fibrils were measured using image analysis software (Photoshop 10.0, Adobe Systems, San Jose, CA). All scleral fibrils visible in a section were measured without selection. However, because the numbers of visible fibrils varied in different sections, the numbers that were counted ranged approximately between 100 and 300. For statistical analyses, the median was calculated for the respective sample and analyzed. 

Statistical Analyses

In Experiment 1, only the data of right eyes from albino and pigmented guinea pigs were analyzed. Comparisons of the data between albino myopia and pigmented myopia, or comparisons between albino hyperopia and pigmented hyperopia were analyzed using two-way, repeated measures ANOVA. Differences between albino and pigmented guinea pigs at specific time points were analyzed by the Bonferroni post hoc test. In Experiment 2, data were compared among three levels: deprived eyes (right eyes of treated animals), fellow eyes (left eyes of treated animals), and right eyes of control animals. In detail, comparisons of deprived eyes and fellow eyes involved paired t-tests whereas comparisons to eyes of control animal with normal vision involved unpaired t-tests. Comparisons between deprived eyes in AH and PH were performed by unpaired t-tests, as were comparisons between deprived eyes of AM and PM groups. For statistical analyses of collagen fiber diameters (Experiment 3), differences among different groups were tested by unpaired t-tests. A P value of <0.05 was considered significant. 

Results

Distributions of Refractive State, and Refractive Error Development in Albino and Pigmented Guinea Pigs with Normal Vision (Experiment 1)

At 2 weeks of age, both albino and pigmented guinea pigs showed double-peaked refractive error distributions (Fig. 1). In detail, in the albino group, 64 (29.9%) were hyperopic (averages and SDs, AH: 4.79 ± 2.00 D), and 150 (70.1%) myopic (AM: −6.13 ± 2.70 D); in the pigmented group, 167 (71.4%) were found hyperopic (PH: 5.14 ± 2.08 D), and 67 (28.6%) myopic (PM: −5.38 ± 2.65 D). Percentages differed significantly between the albino and pigmented group (χ² test: P < 0.01). Also the average refractive errors of albino (−2.86 ± 5.60 D) and pigmented guinea pigs (2.13 ± 5.27 D) in the right eye were significantly different (t = 9.712; P < 0.001, unpaired t-test). There was no difference between the distributions of right and left eyes in albino (average refractive errors of left eyes: −3.15 ± 5.68 D, t = 0.284; P = 0.776, paired t-test) or pigmented (average refractive errors of left eyes: 1.99 ± 5.34; t = 0.541; P = 0.589, paired t-test) guinea pigs. 

Figure 1.

 

Distributions of refractive errors in both eyes in 234 pigmented guinea pigs (A1, A2), and in 214 albino guinea pigs (B1, B2) at 2 weeks of age, as measured with an infrared photorefractor as previously described. ²⁷ A striking bimodality of the distributions was found which was not previously described in other animal models. The distributions of refractive errors in left and right eyes were similar.

During the observation period of 8 weeks, hyperopia declined similarly in both albino and pigmented guinea pigs by 1.51 ± 0.50 D in PH and by 1.35 ± 1.85 D in AH (Fig. 2A). The decline can partially be attributed to the decrease of the small eye artifact because the ratio of retinal thickness to axial length declines during global eye growth. On the other hand, myopic albino and pigmented guinea pigs showed clearly different developmental trends with age. The most striking observation here was that spontaneous myopia in pigmented myopic guinea pigs declined over time but progressed in albino myopic guinea pigs. As a result, there was a significant difference between the two myopic groups (F = 40.704; P < 0.001, two-way repeated measures ANOVA) at the end of the observation period. This difference was not yet present at the beginning of the experiment at 2 weeks of age, but developed over time and was significant from 4 to 10 weeks of age (P < 0.01, two-way ANOVA, followed by post-hoc Bonferroni test). 

Figure 2.

 

(A) Refractive error, (B) axial length, (C) anterior chamber depth, (D) lens thickness, and (E) vitreous chamber depth in the right eyes of hyperopic and myopic albino guinea pigs and hyperopic and myopic pigmented guinea pigs, as measured over a period from 2 to 10 weeks of age. Data points denote the means. Error bars are standard errors of the means.

All groups showed an increase in axial length with age, and there was a significant difference between albino and pigmented hyperopic guinea pigs (F = 4.849; P = 0.041), but no difference between albino and pigmented myopic guinea pigs (F = 0.015; P = 0.905; Fig. 2B). Additionally, there was a significant difference in growth rate between AM and PM (0.10 mm/wk versus 0.07mm/wk, t = 3.925; P < 0.001), and in hyperopic ones (0.09 mm/wk versus 0.10 mm/wk, t = 2.572; P = 0.011). Anterior chamber depth and lens thickness increased with age in both albino and pigmented guinea pigs. However, the anterior chamber depth of albino guinea pigs was consistently shallower than in pigmented animals (PH versus AH, F = 8.455; P = 0.009; PM versus AM, F = 50.269; P < 0.001; Fig. 2C). Lens thickness was not different in albino and pigmented guinea pigs (AH versus PH, F = 0.513; P = 0.483; AM versus PM, F = 1.149; P = 0.298; Fig. 2D). Vitreous chamber depth increased with age in both AH and AM, as well as in PH, but not in pigmented myopic guinea pigs (Fig. 2E). There was no significant difference in vitreous chamber depth between AH and PH at any time point of observation (F = 0.001; P = 0.980), but there was significant difference between AM and PM (F = 5.757; P = 0.027). Regression analyses demonstrated that refractive errors were correlated with axial lengths in albino and pigmented guinea pigs at 2 weeks of age (albino, refractive error = −21.134 × axial length + 165.88; r = 0.667; pigmented, refractive error = −17.345 × axial length + 137.59; r = 0.723) and 10 weeks of age (albino, refractive error = −26.711 × axial length + 227.68; r = 0.766; pigmented, refractive error = −10.446 × axial length + 92.15; r = 0.583; Figs. 3A, 3B). Detailed inspection of the changes of the individual data points between Figure 3A and Figure 3B reveals that the clustering of the refraction data into two clouds that was seen at the age of 2 weeks had changed into a more homogenously scattered distribution, but with more myopic refractions among the albino animals. 

Figure 3.

 

Refractive errors plotted against axial lengths at 2 weeks (A) and 10 weeks of age (B) in 20 albino and 20 pigmented guinea pigs. At 2 weeks of age, refractive errors appear clustered into two clouds but merge into a more homogenous distribution at 10 weeks of age, both in pigmented and albino animals.

Because there was no difference in corneal radius of curvature between AH and AM (P = 0.813), and no difference between PH and PM (P = 0.250), we combined the data of all ametropic albino guinea pigs into one data set (n = 20). The same was done for the ametropic pigmented animals (n = 20). After pooling, significant differences appeared between albino and pigmented animals in the average radius of corneal curvature (F = 8.170; P = 0.007). Furthermore, consistent with previous reports, ^25,26 albino guinea pigs had higher amounts of corneal astigmatism at all time points of observation (mostly against-the-rule), compared with pigmented animals (F = 32.025; P < 0.001). Corneal astigmatism in albino guinea pigs decreased from 3.92 D at 2 weeks of age to 1.49 D at 10 weeks of age, as detailed in Table 1. 

Effects of Diffusers on the Development of the Eyes in Albino and Pigmented Guinea Pigs (Experiment 2)

At the start of the experiment, the refractive errors for AH and PH were similar. The same was true for AM and PM (refractive error: AH: 4.81 ± 1.40 D in eyes that were later treated with diffusers, and 4.97 ± 1.12 D in fellow eyes; PH: 5.76 ± 1.21 in eyes that were later treated with diffusers, and 5.95 ± 1.58 D in fellow eyes; AM: −5.68 ± 1.67 in eyes that were later treated with diffusers, and −5.75 ± 1.99 D in fellow eyes; PM: −4.97 ± 1.52 in eyes that were later treated with diffusers, and −5.39 ± 1.66 D in fellow eyes). There were some initial differences in anterior chamber depth and corneal radius of curvature at the 2 weeks of age. These differences likely trace back to differences in genotype in albino and pigmented animals. 

As described in a previous report, ³⁴ the translucent face mask used in this study successfully induced deprivation myopia in PH (Fig. 4A). At 2 and 4 weeks, the eyes treated with such diffusers were more myopic than their uncovered fellow eyes (difference between eyes at 2 weeks: −4.48 ± 1.46 D, t = 9.725, P < 0.01; 4 weeks: −8.28 ± 2.13 D, t = 12.29, P < 0.01, paired t-tests). There was no significant difference between the fellow eyes of treated animals and the right eyes of control animals in PHs. 

Figure 4.

 

Refractive error (A, D), vitreous chamber depth (B, E), and axial length (C, F) in the deprived eyes, fellow eyes, and eyes of untreated albino and pigmented hyperopic (A–C) and myopic (D–F) guinea pigs, as observed after a period of 4 weeks of deprivation. Data are expressed as means ± standard errors.

The translucent face mask induced deprivation myopia in AH, and even more so than in PH (difference between eyes for AH at 2 weeks: −7.61 ± 2.71 D, t = 8.416, P < 0.01; 4 weeks: −11.17 ± 2.55 D, t = 13.140, P < 0.01, paired t-tests). Again, there was no significant difference between the fellow eyes of treated animals and the right eyes of control animals in AHs. Deprivation myopia developed much more rapidly in albino guinea pigs than in pigmented guinea pigs (2 weeks: t = 3.712, P < 0.01; 4 weeks: t = 3.506, P < 0.01, unpaired t-test). In fact, the degree of myopia in AH after 2 weeks was similar to that produced in PH after 4 weeks. Potential yoking of diffuser-treated and fellow eyes was considered but statistical analyses did not reveal any significant effects (Table 2). 

Table 2.

 

Refractive Errors of Hyperopic and Myopic Guinea Pig Eyes after a Period of Deprivation

	Hyperopic Group				Myopic Group
	Eye	Age (wk)			Eye	Age (wk)
		2	4	6		2	4	6
Pigmented	Exp	5.76 ± 1.21	0.95 ± 1.40	−3.10 ± 2.09	Exp	−4.97 ± 1.52	−4.08 ± 2.25	−6.67 ± 1.41
	Fellow	5.95 ± 1.58	5.43 ± 1.09	5.18 ± 0.69	Fellow	−5.39 ± 1.66	−2.44 ± 2.82	−1.50 ± 2.69
Age-matched	Right	5.86 ± 0.59	5.14 ± 0.80	4.89 ± 1.08	Right	−5.08 ± 1.60	−2.94 ± 2.02	−1.75 ± 2.86
	Left	5.61 ± 0.36	4.94 ± 0.85	4.83 ± 0.95	Left	−5.06 ± 1.41	−3.03 ± 2.20	−1.56 ± 2.66
Albino	Exp	4.81 ± 1.40	−3.33 ± 3.34	−7.33 ± 3.12	Exp	−5.68 ± 1.67	−6.76 ± 2.25	−7.98 ± 2.36
	Fellow	4.97 ± 1.12	4.28 ± 1.28	3.83 ± 1.30	Fellow	−5.75 ± 1.99	−7.33 ± 2.23	−8.28 ± 2.50
Age-matched	Right	5.14 ± 0.67	4.39 ± 0.56	4.17 ± 0.67	Right	−6.02 ± 1.47	−7.08 ± 1.33	−7.84 ± 2.24
	Left	5.08 ± 0.67	4.53 ± 0.57	4.28 ± 0.64	Left	−5.75 ± 1.16	−7.33 ± 1.30	−8.11 ± 2.33

 Mean ± SD are shown for both eyes. Animals were treated with diffuser over one eye (Exp) and the fellow eye was untreated (Fellow). Data for the right and left eyes of untreated age-matched animals (Age-matched) are also shown.

In both AH and PH, vitreous chamber depth (VCD) and axial length were significantly elongated in deprived eyes, compared with their fellow eyes after 4 weeks (VCD, AH: t = 4.086, P < 0.01; PH: t = 5.903, P < 0.01, paired t-test; Figs. 4B, 4C). The increase in VCD in deprived eyes showed up much more quickly in AH than in PH (2 weeks: t = 1.661, P = 0.127; 4 weeks: t = 2.496, P = 0.023, unpaired t-test), and was accompanied by a faster growth rate (0.08 mm/wk versus 0.06 mm/wk in albino and pigmented animals, respectively). In our study, deprivation had no effect on anterior chamber depth or lens thickness of deprived eyes or fellow eyes, in albino and pigmented guinea pigs (4 weeks, in AH: anterior chamber depth, t = 0.555, P = 0.594; lens thickness, t = 2.229, P = 0.056; in AM: anterior chamber depth, t = 1.581, P = 0.148; lens thickness, t = 1.500, P = 0.168). 

Like in PH, the translucent face mask induced relative myopia also in PM (compared with the fellow eyes: 2 weeks: −1.64 ± 1.44 D, t = 3.409, P < 0.01; 4 weeks: −5.17 ± 1.88 D, t = 8.230, P < 0.01, paired t-test). However, in AM, deprivation did not induce any further myopia (Fig. 4D). In pigmented guinea pigs, form deprivation interrupted the recovery from spontaneous myopia and vitreous chamber depth and axial length increased even further. There was significant difference in vitreous chamber depth and axial length between the deprived eyes and their fellow eyes after 4 weeks of deprivation (VCD: t = 6.820, P < 0.001; AL (axial length): t = 7.406, P < 0.001). In AM, deprivation had no further effect on ocular components (Figs. 4E, 4F). 

Scleral Fibril Diameter and Myopia (Experiment 3)

To analyze structural features of myopia, scleral collagen fibril diameters were measured at the age of 2 (Fig. 5) and 10 weeks, respectively. The distributions of the collagen fibril diameters of the sclera were analyzed using a Kolmogorov-Smirnov normality test and found to be skewed. As a result, nonparametric statistics were applied to these data and the median fibril diameter was used as variable. No difference was found between albino and pigmented hyperopic guinea pigs, or between albino and pigmented myopic guinea pigs at 2 weeks of age (Fig. 6) (AM 73.52 ± 8.89 nm versus PM 72.33 ± 7.35 nm, t = 0.327, P = 0.747; AH 80.91 ± 7.54 nm versus PH 84.13 ± 6.49 nm, t = 1.214, P = 0.236). However, at 10 weeks of age, the albino strain showed thinner fibril diameters than the pigmented strain when the animals were myopic (AM 64.82 ± 8.23 nm versus PM 70.64 ± 5.21 nm, t = 2.315, P = 0.028) but not when they were hyperopic (AH 76.93 ± 5.26 nm versus PH 77.40 ± 7.39 nm, t = 0.201, P = 0.842). 

Figure 5.

 

Transmission electron microscopy photographs of collagen fibrils in posterior sclera of albino hyperopic (A), myopic (B), pigmented hyperopic (C), and myopic (D) guinea pigs at 2 weeks of age.

Figure 6.

 

(A–D) Distribution histograms of fibril diameters of guinea pigs at 2 weeks of age for the respective groups.

Discussion

In this longitudinal study, we found that there are inherent differences in the emmetropization mechanism in albino and pigmented guinea pigs. All measurements could be performed in alert guinea pigs because they were very cooperative, and no cycloplegia, general anesthesia, or lid retractors were involved. Therefore, all data reflect normal eye growth with no intervention other than diffusers. An important issue is how well albino guinea pigs can see, compared with the wild type. We tested their optomotor responses in albino and pigmented guinea pigs and found no differences. They all responded similarly to drifting high contrast 2.4 cyc/deg gratings, even though the gain of the optomotor response at higher spatial frequencies was reduced in highly myopic animals. ¹⁸  

Nevertheless, there may be differences in temporal processing between albino and pigmented animals because Bui et al. ³⁵ and Racine et al. ³⁶ had found that albino guinea pigs have significant larger amplitudes and faster implicit times in the ERG a-wave and peak to peak response than did pigmented animals. 

Different Refractive Error Distributions Related to Abnormal Ametropia Development in Albino Guinea Pigs

Because neonatal guinea pigs display adult-like electroretinograms and are well developed by postnatal day 12, ³⁷ it can be assumed that they also undergo emmetropization at 2 weeks of age. It is striking that both albinos and pigmented animals showed a double-peaked distribution of refractive errors which became more Gaussian-shaped with age. The same infrared photorefractor was used in Jiang et al. ²⁷ Because such a double-peaked distribution was not found in a previous study in the same strain, ²⁷ an age effect could account for the difference (2 weeks versus 3 weeks). Alternatively, the difference perhaps could also be due to the small sample size in the previous study, because only some pigmented myopic guinea pigs had started to emmetropize between weeks 2 and 3. It seems unlikely that accommodation was a reason because there is no clear evidence for its presence in the guinea pig. At least in the strains used in this study, accommodation could not be elicited by presenting a target. ²⁷  

Interestingly, spontaneously myopic albino guinea pigs showed no emmetropization which was different from myopic pigmented animals. Furthermore, despite that both myopic albino and pigmented animals were exposed to the same visual environment, their refractive error developments differed. Apparently, their emmetropization feedback loops either had different gains or set-points. The possible differences between pigmented and albino guinea pigs can only be speculated as follows:

1.  
   Levels of dopamine could be reduced in the retina in albino eyes. ³⁸ Many studies have shown that L-dopamine or dopamine, the upstream melanin precursor, plays an important role in the pathogenic mechanism of albinism. ^{39 –41} Dopamine has been implicated in the development of deprivation myopia in chicks. ^{42 –44} A recent study demonstrated that restoration of dopamine but not melanin levels through ectopic expression of tyrosine hydroxylase (TH) in the retinal pigment epithelium (RPE) of albino mice could improve their visual function. ⁴⁵
2.  
   Lack of intraocular melanin could result in an increase of intraocular light scatter which would reduce retinal image contrast ^46,47 and promote the development of deprivation myopia.
3.  
   Other biochemical differences in the signaling cascade from retina to sclera between albino and pigmented animals could also result in different properties of emmetropization.

Ocular Structures that Determine the Refractive State Are Abnormal in Albino Guinea Pigs

In human albinism studies, corneal astigmatism was found to be with-the-rule (WTR). ²⁵ In the present study, we also found large amounts of corneal astigmatism, but against-the-rule (ATR) rather than with-the-rule as in humans. We could not measure the total astigmatism of the eyes, because the eccentric infrared photorefractor only measured the vertical meridian. In both a study in albino chicks, ²⁶ and in the present study on albino guinea pigs, astigmatism declined with age. Similar to observations in the albino chicks, we found that the anterior chamber of albino guinea pigs was shallower than in pigmented animals. More shallow anterior chambers were also found in dopamine D₂ receptor knockout mice (Xiangtian Zhou, personal communication, 2010). In addition, anterior segment developmental abnormalities were found in albino Cyp1b1 ^−/−B6 mice and these abnormalities could be alleviated by administration of L-dopamine. ⁴⁸ In summary, it is possible that abnormal dopamine signaling could account also for abnormal anterior segment development. 

Changes of Refractive Error Development Are Associated with Scleral Collagen Fibrils

It is well established that the sclera plays an important role in the development of myopia, ⁴⁹ and scleral thinning and reduced collagen fibril diameter were previously detected in myopic eyes. ⁵⁰ To determine whether such changes also occur in spontaneous or induced myopia in the guinea pig, we measured scleral collagen fibril diameters at 2 and 10 weeks of age. Consistent with findings in other mammalian species, ^51,52 we found that they were reduced both in myopic albino and pigmented animals. When myopia progressed, collagen fibril diameters continued to thin and when animals recovered, they became relatively thicker. We also found that they were thinner in myopic albinos than in pigmented animals at 10 weeks of age, suggesting some differences in scleral metabolism between albino and pigmented animals. 

An important question is now whether reduced collagen fibril diameters are a consequence, rather than a cause, of myopia (Kang RN, et al. IOVS 1996;37:ARVO Abstract 1491). ⁵³ We found that collagen fibril diameters had already changed at an early age of 2 weeks in those guinea pigs that were myopic. Because these changes were seen at an early age, it seems less likely that these changes were just only a consequence of myopia rather one of the causes. 

Different Susceptibility to Deprivation Myopia in Albino and Pigmented Guinea Pigs

In previous attempts to induce deprivation myopia in guinea pigs, pigmented hyperopic animals were used and reasonable amounts of myopia could be induced. ^28,54 Here, we have studied four different groups, albino animals with myopia and hyperopia, and pigmented animals with myopia and hyperopia. Diffusers had no effect on refractive error development in albino myopic guinea pigs but could further enhance myopia in spontaneously myopic pigmented guinea pigs—perhaps the first time that deprivation myopia was induced in a spontaneously myopic animal. In myopic albino animals, spontaneous myopia developed so rapidly both without and with diffusers (−0.54 diopters/wk versus −0.58 diopters/wk respectively) that it could be possible that the maximal axial eye growth rates were already reached and diffusers could not further accelerate eye growth. 

Furthermore, it was striking that albino hyperopic guinea pigs were more susceptible to diffuser wear than pigmented hyperopic animals. This is similar to previous studies with Nob mice, ⁵⁵ where an increased susceptibility to deprivation myopia was found, compared with wild type mice. Nob mice also displayed lower retinal dopamine levels than wild type, and it could be that retinal dopamine levels are involved in the gain of the emmetropization feedback loop. In addition, many complex models may exist. For example, unknown genetic factors may also influence the susceptibility of the eye to the form deprivation, especially for the different response between the albino hyperopic and myopic guinea pigs and within the albino guinea pigs. Therefore, further studies are needed to elucidate whether such possibilities exist and such investigations would help us to further understand the mechanisms underlying the emmetropization. 

Conclusions

This study showed distinct abnormalities of emmetropization in albino guinea pigs, compared with pigmented wild type. Most interestingly, albinos showed little emmetropization from spontaneous myopia, about doubled susceptibility to deprivation myopia, and deprivation myopia could not be further enhanced in those albino animals that are already spontaneously myopic—even though it is clear that they had comparable spatial vision as pigmented animals. Finally, the higher myopia in the albinos was accompanied by smaller scleral fibril diameters. 

Footnotes

 Supported by National Natural Science Foundation of China (30600227 and 30973278), Zhejiang Provincial Natural Science Foundation of China (Y207598), National Basic Research Program of China (973 project; 2011CB504602), and Doctoral Program of Institution Higher Education of China (20060343002).

Footnotes

 Disclosure: L. Jiang, None; K. Long, None; F. Schaeffel, None; S. Zhang, None; X. Zhou, None; F. Lu, None; J. Qu, None