The Effects of Monochromatic Illumination on Early Eye Development in Rhesus Monkeys

Abstract

Purpose.: Influence of longitudinal chromatic aberration (LCA) on emmetropization during early eye development has not been studied in primates. We investigated the effects of quasi-monochromatic lighting on refractive development and eye growth in rhesus monkeys.

Methods.: Infant rhesus monkeys were raised under one of three lighting conditions for 51 weeks: quasi-monochromatic blue light (peak 455 nm), red light (peak 610 nm), and white light (color temperature 5000 K). All animals underwent biometric measurements using cycloplegic streak retinoscope, keratometry, and A-scan ultrasonography for refraction, corneal power, and axial components, respectively, at designated time points.

Results.: At the 51st week, the mean difference in refraction of the white light and blue light groups, compared with that of the red light group, reached 1.71 diopters (D) and 1.43 D, respectively (both P < 0.001). Two monkeys in the red light group developed myopia at the 16th week, whereas the other seven remained hyperopic throughout the experiment. No significant difference in mean refraction was observed between the blue light group and white light group.

Conclusions.: Illumination from long-wavelength light during early life could be a risk factor for the development of myopia in a small proportion of rhesus monkeys that are sensitive to L-cone stimulation.

Introduction

For most infant eyes, the refractive power of the eye does not match its axial length, and the image of a distant object is focused behind the retina, thus exhibiting a hyperopic refractive error. This early refractive error is usually eliminated as the eye develops from neonate to adolescent in a normal visual environment, and thus the eye reaches an emmetropic refractive status. This postnatal developmental process for eliminating infantile refractive error is called emmetropization, and it is an active and visual experience–dependent process. The active emmetropization process during early eye development has been reported in fishes, chickens, guinea pigs, and monkeys. ^1–9  

The refractive power of the eye varies with the wavelength of light. For an emmetropic human eye looking at a distant object, the focal image for each wavelength is usually formed at a different location, with short wavelength (blue light) focused in front of the retina, long wavelength (red light) behind the retina, and middle wavelength (green or yellow light) focused at the retina. In the human eye, the image focal planes for the whole visual spectrum vary over a range of approximately 2 diopters (D). ¹⁰ Previous studies have investigated the influence of longitudinal chromatic aberration (LCA), on emmetropization in fishes and chickens and it was found that both eye growth and refraction development depended on the wavelength of the illumination lights. ^2,11 The difference in eye size and refraction matched the difference in focal length determined by LCA for different lighting conditions, suggesting that the emmetropization mechanism was able to detect and compensate for the LCA-related differences in the focal plane positions. 

In addition, to make a shift in focal plane, different wavelengths also produce different color appearances. For normal color vision, the color appearance changes from red to blue when the wavelength varies from long to short. Based on the findings of previous animal studies examining the emmetropization process, it is possible that the chromatic mechanism does not have a major effect on emmetropization; eye growth progressed normally no matter what color illumination the animals were exposed to. ^2,12,13 In guinea pigs, the hyperopia that is present at birth progresses to emmetropia by 3 weeks of age. ¹⁴ In a prior study, we reported that guinea pig eyes overcompensated in response to narrowband light, resulting in inaccurate refractive growth. ¹⁵ The LCA between medium-wavelength green light and short-wavelength blue light was approximately 1.5 D, but the difference in refractive errors between those animals exposed to blue light versus green light was 4.5 D (at 12 weeks of rearing). 

Therefore, if this change were simply the effect of LCA compensation, one would not have expected the difference to be so large. Therefore, the results suggest a potential overcompensation. The researchers interpreted these data as evidence of disruption of emmetropization under these monochromatic conditions. The relative abundance of short-wavelength (S) cones in the retina of the guinea pig suggests that it has good spatial acuity in short-wavelength light. ^16,17 Previous studies have hinted that the S-cones prefer myopically defocused images and L- or M-cones (long- and middle-wavelength) prefer hyperopically defocused images. ^12,18,19 We hypothesize that the guinea pig might over-accommodate in short-wavelength light and under-accommodate in middle-wavelength light. This would accentuate the effects of LCA, which could result in larger refractive changes under monochromatic light. To date, no studies similar to those described above have been reported for primates. Interestingly, it has been shown that there is a lower incidence of myopia in schoolchildren with color vision deficiency than with normal color vision. ²⁰  

Thus, the purpose of this study was to investigate the effects of quasi-monochromatic lighting on refractive development and eye growth in healthy infant rhesus monkeys so as to understand the influence of LCA on eye development in primates and also to test the effect of the chromatic mechanism on emmetropization. 

Methods

Animals

The animal research was approved by the Animal Care and Ethics Committee at the Eye and Ear, Nose, and Throat Hospital of Fudan University, Shanghai, China. The monkeys were purchased from the SuZhou XiShan Zhongke Laboratory Animals Co., Ltd. (Jiangsu, China), and were held at the Animal Facility of the Shanghai Public Health Center (Shanghai, China). The treatment and care of animals were conducted according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. A total of 24 male and female healthy infant rhesus monkeys ( Macaca mulatta ), at 50 to 60 days of age, were randomly assigned into three groups: two experimental groups of red lighting (RL, n = 9) and blue lighting (BL, n = 7), and a control group of white lighting (WL, n = 8). All monkeys were examined to confirm clear media and no abnormities. Refractive status and biometric parameters were determined at 0, 2, 4, 6, 8, 10, 12, 16, 22, 29, 34, 41, and 51 weeks. The follow-up examination of one monkey in each group was not completed due to injury during play, and the experiment was discontinued for these animals at the 41st, 22nd, and 12th week for monkeys in the RL, BL, and WL groups, respectively. 

Infant monkeys were separated from mother monkeys at approximately 50 days after birth, and we carried out the adaptive training so that they could adapt to the artificial feeding regimen. Infant monkeys were acclimated for approximately 1 week after purchase before they were raised in standard cages with specialized lighting as described below. Monkeys were kept under a 12-hour light/dark cycle (light: 8:00 AM–8:00 PM), temperature of 22 to 26°C, and relative humidity of 55% to 65%. Briefly, the monkeys were initially bottle-fed formula used for human infants once every 4 hours, and the feeding frequency and amount of formula were adjusted with increasing age. After approximately 6 months, supplementary food, including bread and fruit, was gradually introduced. When the monkeys were older, they were fed a fruit-based diet supplemented by cow's milk. At an early stage, some toys were placed in the cages, such as a small ball and metal chain; however, the beneficial effects of having these items were not obvious and the toys were removed due to safety issues. The handler and researchers were the only persons allowed to interact with or watch the monkeys. The animals were killed, after general anesthesia was administered, using an intracardiac paraformaldehyde perfusion, which is a customary international practice. 

The LCAs were assessed when the monkeys were 2 months of age (the time when they were first included into each experimental group). The mean value of 0.96 D in the current study is similar to that reported in the literature for monkeys: 1 D (656 nm–486 nm). ²¹  

The monkeys were raised in special cages with light-emitting diode (LED) tubes installed on the walls, ceiling, and bottom of the cage to obtain maximum light intensity and homogeneous illumination. The dimensions of the cages were 100 cm long, 70 cm wide, and 90 cm high, and there were two or three monkeys in each cage. The spectral distribution of the red and blue LEDs peaked at 610 nm (half bandwidth of 20 nm) and 455 nm (half bandwidth of 25 nm), respectively, and the color temperature of the white LEDs was 5000 K. Intensity of the LED lights was controlled by modulating the voltage. A cage requires approximately 30 LED lamps, and each LED lamp includes 150 LEDs. 

The spectral sensitivity function assay of rhesus monkeys proposed by Jacobs and Deegan ²² was used to determine the irradiance of the lights for keeping similar brightness to the monkey's eyes. Irradiances of the LED lights, calibrated with an IL-1700 Research Radiometer (International Light Technologies, Inc., Peabody, MA), were 0.043 mW/cm², 0.14 mW/cm², and 0.024 mW/cm² for the red, blue, and white lights, respectively. 

Biometric Measurements

The refractive status, corneal curvature, and axial dimensions of the eyes were measured at the onset of each experiment and the scheduled follow-up time points. Before each measurement, the monkeys were anesthetized with an intramuscular injection of ketamine hydrochloride (15–20 mg/kg) and acepromazine maleate (0.15–0.20 mg/kg), and topically administered 0.4% oxybuprocaine hydrochloride. Cycloplegia was achieved by topical administration of 1 drop of 1% cyclopentolate hydrochloride every 5 minutes for a total of four drops. 

Both the spherical and cylindrical components were measured along the pupillary axis by two experienced investigators (using a streak retinoscope in a dark room), and the refractive error was defined as the mean spherical-equivalent, spectacle-plane refractive correction. Examiners were blinded to the lighting conditions of the animals, and to the results of the other examiner. For each examination, the average of two measurements from the two investigators was used as the estimate of refraction. The mean unsigned difference in all of the refraction measurements between the two investigators was 0.26 ± 0.2 D (range, 0.0–0.5 D). 

Chromatic defocus between the 455-nm and 610-nm lights was evaluated for the right eyes of the 24 completely cycloplegic monkeys with a streak retinoscope by using blue and red streak lights that were controlled by 455-nm and 610-nm interference filters (half bandwidth of 5 nm). Using this method, the average chromatic focal difference was 0.96 ± 0.30 D, similar to the LCA between the two lights for monkey and human eyes as reported by other authors. ^10,21,23  

The anterior radius of curvature of the cornea was assessed using keratometry (Topcon OM-4; Topcon Corporation, Tokyo, Japan), and the corneal power (CP) was calculated using a refractive index of 1.3375 for the aqueous. The axial dimensions of the eye, including anterior chamber depth (ACD), lens thickness (LT), vitreous chamber depth (VCD), and the sum of these measures, the axial length, were measured using A-scan ultrasonography (Cinescan A/B; Quantel Medical, Inc., Clermont-Ferrand, France). An 11-MHz transducer was placed in direct contact with the cornea while care was taken not to indent the cornea. The intraocular distances were determined using ultrasound velocities for the normal human eye, and the reported data represent the average of 10 readings. 

Data Analysis

In this study, data of two eyes were collected for every monkey. Data from biometric measurements were summarized as mean ± SD. The differences in baseline biometric measurements among the three groups were analyzed by repeated measures ANOVA because data of two eyes of the same monkey were correlated. The differences in biometric measurements (other than baseline) were compared at each time point among the three groups using a linear mixed model (a linear mixed model is appropriate for normally or non-normally distributed data). A χ² analysis was not performed because the data (two eyes from the same monkey) were not independent. Bonferroni corrections were applied to control the type I error rate during multiple comparisons. The relationships between refraction and other biometric measurements were also investigated using a linear mixed model. The statistical analyses were performed with SAS software version 9.2 (SAS Institute, Inc., Cary, NC). A two-tailed P value less than 0.05 indicated statistical significance. 

Results

Baseline Biometric Measurements

Before exposure to different illumination, the mean spherical refraction was comparable among the three groups (WL: 2.55 ± 0.66 D; RL: 2.46 ± 0.77 D; BL: 2.13 ± 0.84 D; P = 0.546). There was no significant difference in the VCD, LT, ACD, or CP among the three groups at the beginning of the experiment (all P > 0.05). Comparisons of baseline biometric measurements among the three groups are shown in Supplementary Table S1. 

Longitudinal Change in Refraction

Figure 1 shows the mean (Fig. 1A) and individual measurements (Figs. 1B–D) of spherical equivalent refraction for both eyes over 51 weeks. The mean refraction in the RL group became significantly different from that of the WL group starting at the 12th week and became significantly different from that of the BL group starting at the 22th week (Fig. 1A). The mean difference in refraction of the WL and BL groups, compared with that of the RL group, reached 1.71 D and 1.43 D at the 51st week, respectively (both P < 0.001; Fig. 1A). However, no significant differences in mean refraction were found between the BL and WL groups at all time points (P > 0.05). 

Figure 1

 

Spherical equivalent refraction during a 51-week treatment period (measured at 0, 2, 4, 6, 8, 10, 12, 16, 22, 29, 34, 41, and 51 weeks). (A) Both eyes in the three groups (n = 8 in the white light group, one discontinued at the 16th week; n = 9 in the red light group, one discontinued at the 51st week; n = 7 in the blue light group, one discontinued at the 29th week). (B) Right eyes and left eyes of monkeys in the white light group. (C) Right eyes and left eyes of monkeys in the red light group. (D) Right eyes and left eyes of monkeys in the blue light group. Data presented as mean ± SD.

Two monkeys in the RL group developed myopia at the 16th week, whereas the other seven remained hyperopic throughout the experiment (Fig. 1C). At the 51st week, the mean refraction of both eyes was −2.75 ± 0.35 D and 1.44 ± 0.78 D for myopic and hyperopic monkeys, respectively. The biometric measurements of myopic and hyperopic monkeys in the RL group at all time points are shown in Supplementary Table S2. 

Longitudinal Change in Axial Components

Figure 2 shows the mean (Fig. 2A) and individual measurements (Figs. 2B–D) of VCD for both eyes over 51 weeks. No intergroup differences in mean VCD were found at any time point studied (P > 0.05; Fig. 2A). Figure 3 shows the mean (Fig. 3A) and individual measurements (Figs. 3B–D) of LT for both eyes over 51 weeks. A significant difference in mean LT between the RL and WL groups was found only at the second week, and significant differences between the BL and WL groups were found only at the 22nd and 29th week (all P < 0.05; Fig. 3A). No significant differences were found between the RL and BL groups at any time points (Fig. 3A). 

Figure 2

 

Vitreous chamber depth during 51-week treatment period (measured at 0, 2, 4, 6, 8, 10, 12, 16, 22, 29, 34, 41, and 51 weeks). (A) Both eyes in the three groups (n = 8 in the white light group, one discontinued at the 16th week; n = 9 in the red light group, one discontinued at the 51st week; n = 7 in the blue light group, one discontinued at the 29th week). (B) Right eyes and left eyes of monkeys in the white light group. (C) Right eyes and left eyes of monkeys in the red light group. (D) Right eyes and left eyes of monkeys in the blue light group. Data presented as mean ± SD.

Figure 3

 

Lens thickness during 51-week treatment period (measured at 0, 2, 4, 6, 8, 10, 12, 16, 22, 29, 34, 41, and 51 weeks). (A) Both eyes in the three groups (n = 8 in the white light group, one discontinued at the 16th week; n = 9 in the red light group, one discontinued at the 51st week; n = 7 in the blue light group, one discontinued at the 29th week). (B) Right eyes and left eyes of monkeys in the white light group. (C) Right eyes and left eyes of monkeys in the red light group. (D) Right eyes and left eyes of monkeys in the blue light group. Data presented as mean ± SD.

Figure 4 shows the mean (Fig. 4A) and individual measurements (Figs. 4B–D) of ACD for both eyes over 51 weeks. The differences in mean ACD between the BL group and WL group were significant only at the 41st and 51st week (P < 0.05; Fig. 4A), and between the RL group and BL group only at the 41st week (both P < 0.05; Fig. 4A). No significant differences in mean ACD between the RL group and WL group were found at any time points (Fig. 4A). 

Figure 4

 

Anterior chamber depth during 51-week treatment period (measured at 0, 2, 4, 6, 8, 10, 12, 16, 22, 29, 34, 41, and 51 weeks). (A) Both eyes in the three groups (n = 8 in the white light group, one discontinued at the 16th week; n = 9 in the red light group, one discontinued at the 51st week; n = 7 in the blue light group, one discontinued at the 29th week). (B) Right eyes and left eyes of monkeys in the white light group. (C) Right eyes and left eyes of monkeys in the red light group. (D) Right eyes and left eyes of monkeys in the blue light group. Data presented as mean ± SD.

Longitudinal Change in Corneal Power

Figure 5 shows the mean (Fig. 5A) and individual measurements (Figs. 5B–D) of CP for both eyes over 51 weeks. No intergroup differences in mean CP at any time points were noted (Fig. 5A). 

Figure 5

 

Corneal power during 51-week treatment period (measured at 0, 2, 4, 6, 8, 10, 12, 16, 22, 29, 34, 41, and 51 weeks). (A) Both eyes in the three groups (n = 8 in the white light group, one discontinued at the 16th week; n = 9 in the red light group, one discontinued at the 51st week; n = 7 in the blue light group, one discontinued at the 29th week). (B) Right eyes and left eyes of monkeys in the white light group. (C) Right eyes and left eyes of monkeys in the red light group. (D) Right eyes and left eyes of monkeys in the blue light group. Data presented as mean ± SD.

Relationship of Refraction and Other Biometric Measurements

Univariate analysis revealed that VCD was significantly related to refractions in all three groups (WL: β = −0.979, P < 0.0001; RL: β = −1.712, P < 0.0001; BL: β = −0.834, P < 0.0001). In addition, CP was also significantly related to refraction in the BL group (β = −0.100, P = 0.004). Multivariate analysis indicated that VCD was significantly related to refractions in all three groups (WL: β = −1.024, P < 0.0001; RL: β = −1.756, P < 0.0001; BL: β = −1.238, P < 0.0001), and CP was significantly related to refraction in the BL group (β = −0.175, P < 0.0001), after adjusting for other biometric measurements. The relationship of refraction and other biometric measurements is shown in Supplementary Table S3, and the refractive errors and vitreous chamber depths obtained at the end of the rearing period for all animals are shown in Supplementary Table S4. 

Discussion

In this study, we examined refractive development and eye growth in infant rhesus monkeys reared under different chromatic illumination for 51 weeks. In the RL environment, two of nine monkeys developed myopia at the 16th week, whereas the others remained hyperopic with a slightly reduced refraction. The change in refraction of the two myopic monkeys was accompanied by faster elongation of VCD, a typical characteristic of myopia development as reported in previous animal studies. ^{1,8,9,24–27} However, myopia development was not found in monkeys reared under BL, suggesting that monochromatic RL might be a risk factor for the development of myopia in a subset of primates that are presumed to be abnormally sensitive to L-cone stimulation. 

The extent of myopia in the RL group was close to the LCA (approximately 0.96 D) between RL and BL (difference between RL and WL: 1.71 D; difference between RL and BL: 1.43 D). This suggests the LCA-induced difference in focus is involved in the regulation of eye development. Different monochromatic lights exert distinct effects on eye development. The relative level of L- and M-cone activities affect the mean level of accommodation and the gain of the accommodative response. ²⁸ Increased L-cone stimulation has been shown to result in increased accommodation. ^18,29,30 The results of the current study suggest that emmetropization is insensitive to the chromatic difference in focus between blue and white light. 

The refractive development of the eye is regulated by a closed feedback loop. ^6,31 However, in the RL group, a large myopic shift occurred in two monkeys. This may be because those two monkeys were more sensitive to long-wavelength light, ²⁸ leading to excessive elongation of the eye and myopia. It can be seen from the results of this study that when two monkeys with obvious myopia in the RL group were excluded, the diopter in the RL group was not significantly different from that in the white or blue light group. The latter finding suggests that the emmetropization mechanism was insensitive to LCA in most of the monkeys. 

A prior study suggests that color opponency in the color-processing pathway may be involved in the development of myopia. ²⁰ Visual processing consists of two opponent processes: L/M and L+M/S. Long-wavelength RL will excite the L/M-cone pathway. In contrast, S-cones are not sensitive to RL. Under short-wavelength BL, S-cones will be excited, whereas L/M-cones will be minimally stimulated. Thus, the different levels of excitation of L/M- and S-cones under the monochromatic environments of blue or red light, may lead to differences in growth modulatory signals, resulting in altered refractive development. Rucker and Wallman, ³² in a study using chick eyes, reported differential effects of red and blue light on the choroidal and ocular length compensatory responses, suggesting that they are driven by different proportions of cone types. 

In our prior work ¹⁵ using guinea pigs, we found that exposure to narrow-band light resulted in exaggerated and inaccurate refractive development, suggesting that chromatic cues may play a more important role in refractive regulation in guinea pigs than in other species. In another study using newborn guinea pigs, Long et al. ³³ reported that monochromic long-wavelength light promoted ocular elongation and myopic shifts. The difference in findings between guinea pigs and monkeys may be attributed to the species differences in the visual system, vision pathways, and the size of the LCA. 

Within the WL group, individual variation in refractive development also was found. Specifically, one monkey became more hyperopic than the others. This abnormal pattern of refractive development under normal WL might be genetically determined, similar to hyperopia development in the human eyes. In the current study, the light intensity in the WL group was similar to that of natural indoor light. It is possible that the monkey that exhibited increased hyperopia had altered light sensitivity. However, we cannot rule out the possibility that genetic or other factors caused this individual to become more hyperopic. 

Twenty of the 24 monkeys, even when exposed to different lighting conditions, similarly progressed toward reduced hyperopia, with moderate elongation of VCD, as is typical of normal emmetropization. Nonetheless a significant difference in refraction between the RL and BL groups at the 51st week was found. Furthermore, a significant inverse relationship between refraction and VCD was found in all three groups; however, the magnitudes of the relationship varied across groups (WL: β = −0.979; RL: β = −1.712; BL: β = −0.834). Therefore, the latter difference in refractions between the RL and BL groups may reflect a difference in VCD between the RL and BL groups. However, this does not necessarily mean that there would be a significant difference in VCD between the RL and BL groups, especially because the magnitude of the relationship between refraction and VCD was not identical among the groups. Thus, the results in this study do not agree with the developmental compensation for chromatic aberration linked with refractive error, as observed in fishes and chickens. ^2,11,12 The normal refractive development of most of the monkeys (20 of 24) under different lighting conditions suggests that the emmetropization process in monkeys is relatively unaffected by exposure to monochromatic environments. 

There are some limitations of the study that should be considered. Because it is the first experiment to observe the effects of quasi-monochromatic lighting on refractive development and eye growth in rhesus monkeys, there are no published studies with which to make comparisons. Monkeys, at age 50 to 60 days, are equivalent to 8-month-old human infants in ocular developmental terms, being, in both cases, a critical period of eye development. ^34,35 Thus, changes in the environment should affect the refractive development. However, we cannot prove whether the monochromic light exposure earlier than 50 days could cause more significant refractive changes. The reason we chose monkeys aged 50 to 60 days was mainly because breastfeeding cannot be performed due to the experimental conditions and thus substitution with a supplemented diet would lead to decreased immunity, poor physical development, and subsequently a higher mortality. As for light exposure causing myopia after 1 year of age, we can only speculate that a monkey at 1 year of age is roughly equivalent to a human 4 years of age, by which time most normal-development eye growth is complete. Only two of nine monkeys developed myopia at the 16th week of treatment, and individual differences cannot be ruled out. However, these findings are not likely due to experimental error, as obvious myopic shifts in refraction occurred in these two monkeys. The results suggest monochromatic light has effects on eye development of monkeys; however, the impact varies among individuals. Species differences in color sensitivities and thus refractive development are also possible, but were not examined in this study. Last, although the sample size of the current study was relatively small, larger numbers are difficult to procure because rhesus monkeys are nonhuman primates and thus are strictly regulated as experimental animals. 

Conclusions

In summary, illumination from long-wavelength light during early life might be a risk factor for the development of myopia in a subset of rhesus monkeys that are abnormally sensitive to L-cone stimulation. Longitudinal chromatic aberration in rhesus monkeys can interfere with emmetropization. 

Supplementary Materials

pdf S1 

Acknowledgments

The authors thank the Ophthalmology Department of Shanghai Public Health Clinical Center for providing the appliance to facilitate this study. We acknowledge Julie Crider, PhD, for medical editing contributions. 

Supported by the National Nature Science Foundation of China (30530770 and 81100689), the Natural Science Foundation of Yunnan Province of China (2010–2012, 2010CD122), and the Science and Technology Commission of Yunnan Province of China (2009–2012, 2009CA008). The authors alone are responsible for the content and writing of the paper. 

Disclosure: R. Liu, None; M. Hu, None; J.C. He, None; X.-T. Zhou, None; J.-H. Dai, None; X.-M. Qu, None; H. Liu, None; R.-Y. Chu, None 

Footnotes

 RL and MH contributed equally to the work present here and should therefore be regarded as equivalent authors.