Activation of Dopamine D2 Receptor Is Critical for the Development of Form-Deprivation Myopia in the C57BL/6 Mouse

Abstract

Purpose.: This study used dopamine D2 receptor (D2R) knockout (KO) mice to investigate the role of D2R activity in the development of form-deprivation myopia (FDM). Sulpiride, a D2R antagonist, was administered systemically into wild-type (WT) mice to validate the involvement of D2R in FDM development.

Methods.: The D2R KO and WT C57BL/6 mice were subjected to FDM. Wild-type mice received daily intraperitoneal injections of sulpiride, 8 μg/g body weight, for a period of 4 weeks. The body weight, refraction, corneal radius of curvature, and ocular axial components were measured at week 4 of the experiment. Differences in all ocular parameters between the experimental and control groups were compared statistically.

Results.: Form-deprivation myopia in D2R KO mice (FD-KO) was significantly reduced compared with their WT littermates (interocular difference, −2.12 ± 0.91 diopter [D] in FD-KO versus −5.35 ± 0.83 D in FD-WT, P = 0.014), with a smaller vitreous chamber depth (0.008 ± 0.006 vs. 0.026 ± 0.006 mm, P = 0.044) and axial length (−0.001 ± 0.007 vs. 0.027 ± 0.008 mm, P = 0.007). Furthermore, FDM was attenuated in animals treated with sulpiride (−2.01 ± 0.31 D in FD-sulpiride versus −4.06 ± 0.30 D in FD-DMSO, P < 0.001) compared with those treated with vehicle, with a retardation in growth of vitreous chamber depth (−0.001 ± 0.006 vs. 0.022 ± 0.004 mm, P = 0.003) and axial length (−0.004 ± 0.007 vs. 0.027 ± 0.005 mm, P = 0.001).

Conclusions.: Genetic and pharmacological inactivation of D2R attenuates FDM development in mice, suggesting that dopamine acting on D2R appears to promote the development of FDM in C57BL/6 mice. Further studies are required to confirm these results using animal models in which retinal D2R is selectively blocked.

Introduction

Eye growth in association with refractive development is modulated by different visual stimuli such as lighting, spatial background, form deprivation (FD), and optical defocus. ¹ Dopamine (DA), as a neurotransmitter in the brain and retina, is widely recognized to have a role in ocular development related to ambient lighting intensity and duration. ² Retinal DA is synthesized by amacrine cells in the retinal inner nuclear layer. ³ Its release in chickens ⁴ and mice ⁵ is greater during daytime or in a bright environment compared with nighttime or darkness. In humans ⁶ and animal models, ⁷ extended outdoor activities are protective against myopia compared with an extended indoor stay, probably due to approximately a 100-fold larger illuminance exposure during daytime. ⁶  

Dopamine receptors are classified into five subtypes, and in the mouse retina D1R, DA D2 receptor (D2R), and D4R have been identified. ⁸ D1R is mainly localized in the horizontal and/or amacrine cells, whereas D2R is more widely distributed in dopaminergic cells across the entire retina. ⁹ One D2R subtype functions as a presynaptic autoreceptor, whose activation inhibits DA release. ³ D4R is expressed at a high level in photoreceptors, inner nuclear layer neurons, and ganglion cells. ¹⁰  

In chickens ¹¹ and monkeys, ¹² the retinal DA levels and its main metabolite 3,4-dihydrophenylacetic acid (DOPAC) are reduced when the eye is form deprived but regain their baseline levels after the eyes are reexposed to normal vision. ¹³ Similar to form-deprivation myopia (FDM), in lens-induced myopia retinal DA and DOPAC levels fall. ¹⁴ In contrast, DA measurements after lens-induced hyperopia have produced mixed results with either a decrease (McCarthy S, et al. IOVS 2008;49:ARVO E-Abstract 1733), an increase, ¹⁴ or no change ¹⁵ in retinal DA and DOPAC levels. However, the retinal DA and DOPAC pools only reflect medium-term (rather than acute-term) changes in DA levels, and distinguishing between DA release and DA synthesis may not be possible. ¹⁶ On the other hand, vitreal DOPAC levels appear to be more meaningful as an index of changes in retinal DA release caused by variations in imposed visual signals (McCarthy S, et al. IOVS 2008;49:ARVO E-Abstract 1733). ¹⁷ The nature of acute changes in retinal DA level during myopia development remains to be determined. 

Local injection of DA, levodopa (a DA precursor), or nonselective DA agonists such as apomorphine and 2-amino-6,7-dihydroxy-1,2,3,4-tetrahydronaphthalene hydrobromide can inhibit FDM development in various animal models. ^18–21 These results indicate thaThese results indicate that a deficiency in retinal DA may induce excessive axial growth of the eye, resulting in axial myopia. However, depletion of retinal DA by intravitreal injection of reserpine ²² and 6-hydroxy DA (6-OHDA) ²³ can also inhibit FDM in chickens. It should be noted that reserpine inhibits the vesicular monoamine transporter and therefore depletes neural transmitters of DA, norepinephrine, and serotonin, ²⁴ whereas 6-OHDA is toxic to ocular tissue. ²⁵ Therefore, myopia inhibited by these two agents could instead be caused by retardation of the eye growth due to multifunctional inhibition or ocular toxicity rather than only being caused by DA receptor inhibition. The nonselective DA antagonist methylergonovine reportedly has no effect on axial growth of the form-deprived eye in chickens, ¹⁸ probably due to its dual effects on DA receptors by acting as a DA receptor partial agonist or antagonist. ²⁶  

In a 2007 study ¹⁸ in which both the selective D2R agonist quinpirole and the selective D1R agonist SKF-38393 were tested separately in form-deprived eyes in chickens, only quinpirole was effective in inhibiting FDM. Furthermore, inhibition of FDM by apomorphine was abolished by spiperone (a D2R antagonist) but not by SCH 23390 (a D1R antagonist) in the same animal model. ²⁷ Therefore, it appears that D2R activation has a major role in the DA-mediated axial growth of the eye. The effect of D2R antagonists on axial growth of the eye appears to be dose dependent because sulpiride in chickens enhances FDM at a dose between 100 and 400 μg but suppresses FDM at a dose higher than 400 μg. ²² Another selective D2R antagonist, haloperidol, also causes dose-dependent inhibition of FDM development in chickens with a much lower dose range than sulpiride (i.e., 3–300 ng). ¹¹ However, spiperone, which acts as a D2R antagonist, was found to not affect axial growth of the form-deprived eyes, ¹⁸ probably due to only a single dose being applied in the study. All of these D2R antagonists also appear to interact with other neurotransmitter receptors and at multiple sites in reducing depression and psychosis in humans. ²⁸ These mixed results in relation to FDM likely reflect the complexity of D2R actions in the retina, depending on different basal DA levels and the partial specificity of many D2R drugs. Therefore, the exact role of D2R in the development of FDM remains to be critically evaluated. 

D2R knockout (KO) mice provide an alternative approach to D2R antagonists to test the role of D2R in refractive development of the eye and FDM. Furthermore, we complemented the D2R KO model with relatively selective D2R antagonists to address possible developmental confounding effects of D2R gene deletion and to validate our finding regarding D2R involvement in FDM development in mice. 

Methods

Experimental Design

D2R KO Mice.

Ninety-nine D2R KO mice (4 weeks old) were randomly divided into two groups, FD-KO (n = 51) and KO-only (n = 48). Seventy-three wild-type (WT) littermates were randomly divided into two control groups, FD-WT (n = 37) and WT-only (WT mice with no additional treatment, n = 36). These experimental animals were produced as follows: heterozygous D2R KO mice (±) derived from the C57BL/6 background were bred to generate D2R KO (−/−) and their WT littermates (+/+). ²⁹ The genotype of the mice was determined by PCR analysis of nail DNA as described previously. ^29,30 Wild-type mice displayed a single 500-bp band, whereas homozygous D2R KO mice showed a single 300-bp band. Heterozygous D2R KO mice showed both 300-bp and 500-bp bands. 

WT Drug-Treated Mice.

One hundred seventy-two C57BL/6 mice (4 weeks old) were randomly assigned to six groups. These included FD-sulpiride (n = 28), FD–dimethyl sulfoxide (FD-DMSO) (n = 28, DMSO was used as a solvent for injection of sulpiride), FD-only (n = 31), sulpiride (n = 26), DMSO (n = 28), and untreated controls (i.e., free of any injection, n = 31). 

FD and Ethics

Form deprivation was produced by gluing a translucent lens to cover the right eye of the animals for 4 weeks in the relevant groups. ³¹ Body weight, refraction, corneal radius of curvature, and axial components of the eye were measured at week 4 of the experiment in all groups. The treatment and care of animals were conducted according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research, and the protocol for handling animals was approved by the animal care and ethics committee at Wenzhou Medical University (Wenzhou, China). 

Preparation for Drug Injection

All agents were intraperitoneally injected daily using a microliter syringe attached to a 29-gauge needle. The injection site was the lower right or left quadrant of the abdomen. Sulpiride (Tocris, Glasgow, UK) (8 μg/g body weight) was injected after dissolving it in DMSO (Sigma, Buchs, Switzerland) at 2.67 μg/μL (8 μg sulpiride in 3 μL DMSO, 1:10 saturation). ³² The injection volume in the DMSO groups was 3 μL/g body weight of the animal. 

Biometric Measurements

Refraction.

Refraction of the eye was measured in a darkened room using an eccentric infrared photorefractor (provided by Schaeffel ³¹ ) without anesthesia. ³¹ Briefly, the mouse was gently restrained by grabbing its tail and placed on a small platform in front of the photorefractor. The position of the platform was adjusted until a clear first Purkinje image occurred in the center of the pupil, indicating an on-axis measurement. Each measurement was repeated at least three times, with the mean recorded using a system designed by Schaeffel et al. ³¹  

Keratometry.

A keratometer (OM-4; Topcon Corporation, Dongguan, Japan) was modified by mounting a +20.0-diopter (D) aspherical lens. A group of stainless-steel ball bearings with diameters ranging from 2.0 to 3.98 mm were used for calibration. The corneal radius of curvature measured with the keratometer was then deduced from the readings on the balls with known radii. ³³ Each eye was measured three times to obtain a mean value. 

Axial Components of the Eye.

Anterior chamber depth (ACD), lens thickness (LT), vitreous chamber depth (VCD), and axial length (AL) were measured by real-time optical coherence tomography (a custom-made OCT). ³⁴ The ACD was defined as the distance from the posterior surface of the cornea to the anterior surface of the lens. The AL was defined as the distance between the anterior surface of the cornea and the vitreous-retina interface. After being anesthetized, a mouse was placed in a cylindrical holder mounted on a positioning table in front of an optical scanning probe (Model 6215H; Cambridge Technology, Hartwell Avenue, Lexington, MA, USA). A video system (LifeCam Cinema 720p HD Webcam; Microsoft, Redmond, WA, USA) measured animal orientation. The optical axis of the eye was aligned with the axis of the probe by detecting the specular reflex on the corneal apex and posterior surface of the lens in two-dimensional OCT images. Axial components of the eye were determined by measuring the traveling distance between two adjacent ocular interfaces. The optical path was recorded and converted into physical length using a refractive index appropriate for each component of the eye. ³⁴ Each measurement was determined based on the mean of three OCT images recorded. 

All measurements were performed by a research optometrist, with help from an assistant. The identities of the different groups were masked. 

Statistical Analysis

Due to the difference in baseline development (body weight and ocular dimensions) between D2R KO mice and their WT littermates, the interocular difference (right eye minus left eye) was adopted to investigate the effects of FD in the D2R KO study, while both the raw data and interocular differences were compared among the groups for the pharmacological study. One-way ANOVA with Bonferroni correction was used to compare biometric results among different groups (experimental versus control groups). A difference was defined as being significant at P < 0.05 and highly significant at P < 0.01 (using SPSS version 16.0; SPSS, Inc., Chicago, IL, USA). 

Results

Developmental Differences Between WT and D2R KO Mice

At week 4 of experiments, D2R KO mice had a smaller body weight compared with WT mice (20.14 ± 0.43 g in KO-only versus 21.78 ± 0.48 g WT-only: P = 0.013, one-way ANOVA) (Table 1), with a shorter ACD (KO-only versus WT-only, P ≤ 0.033) and a thinner LT (KO-only versus WT-only, P ≤ 0.003), indicating that the physical growth in the KO-mice was slower. Therefore, a comparison of raw data between WT-only and KO-only eyes would provide potentially misleading results due to the different developmental baselines between these two groups. There was no significant difference in all biometric results between the fellow eyes of each FD group and their respective control eyes (FD-WT versus WT-only, P > 0.286 and FD-KO versus KO-only, P > 0.074, one-way ANOVA) except for LT (1.665 ± 0.005 mm in FD-WT versus 1.689 ± 0.005 mm in WT-only, P = 0.002 and 1.643 ± 0.004 mm in FD-KO versus 1.665 ± 0.005 mm in KO-only, P = 0.001). This agreement indicates that the fellow eyes of two FD groups (FD-WT and FD-KO) could be used as a control to assess biometric changes in the deprived eyes. Therefore, to minimize the confounding effect of the developmental difference in body weight, interocular differences (right eye minus left eye) of the animals rather than raw data were used for result analysis for these groups. 

Table 1

 

Body Weight, Refraction, and Ocular Biometry at Week 4 of the Experiment in D2R KO Groups (Mice Aged 8 Weeks)

Group	Weight, g	Eye	Refraction, D	ACD, mm	LT, mm	VCD, mm	AL, mm	CRC, mm
WT-only, n = 36	21.78 ± 0.48	OD	5.29 ± 1.03	0.386 ± 0.004	1.681 ± 0.006	0.639 ± 0.005	3.020 ± 0.030	1.401 ± 0.023
		OS	5.41 ± 0.99	0.381 ± 0.004	1.689 ± 0.005	0.636 ± 0.004	3.024 ± 0.030	1.431 ± 0.012
FD-WT, n = 37	21.10 ± 0.38	OD	0.76 ± 0.85	0.375 ± 0.004	1.666 ± 0.006	0.657 ± 0.006	3.025 ± 0.030	1.409 ± 0.013
		OS	6.11 ± 0.93	0.376 ± 0.004	1.665 ± 0.005	0.631 ± 0.005	2.998 ± 0.029	1.410 ± 0.015
KO-only, n = 48	20.14 ± 0.43	OD	3.18 ± 0.87	0.375 ± 0.004	1.658 ± 0.005	0.641 ± 0.004	2.996 ± 0.027	1.384 ± 0.015
		OS	3.14 ± 0.65	0.369 ± 0.004	1.665 ± 0.005	0.638 ± 0.004	2.994 ± 0.028	1.402 ± 0.012
FD-KO, n = 51	17.09 ± 0.37	OD	1.09 ± 0.75	0.357 ± 0.004	1.640 ± 0.004	0.653 ± 0.006	2.955 ± 0.026	1.383 ± 0.012
		OS	3.22 ± 1.00	0.359 ± 0.003	1.643 ± 0.004	0.645 ± 0.005	2.956 ± 0.025	1.383 ± 0.012

 Data are expressed as the mean ± standard error. CRC, corneal radius of curvature.

FD in D2R KO Mice

The interocular difference in refraction for the FD-WT group was approximately −5.35 ± 0.83 D compared with −0.12 ± 0.46 D for the WT-only group (P < 0.001, one-way ANOVA) at week 4 of FD. In contrast, the interocular difference in refraction for the FD-KO group was −2.12 ± 0.91 D compared with 0.04 ± 0.73 D for the KO-only group (P = 0.069). Therefore, the myopia induced in the FD-KO group was approximately 60% lower than that in the FD-WT group (−2.12 ± 0.91 D in FD-KO versus −5.35 ± 0.83 D in FD-WT, P = 0.014) (Table 1; Fig. 1A). 

Figure 1

 

The myopia induced (deprived eye minus fellow eye) in the FD-KO group was reduced by 60% compared with the FD-WT group, (A), with a reduction in both VCD (B) and AL (C). The corneal radius of curvature (D), ACD (E), and LT (F) were not affected by FD or genetic KO (*P < 0.05 and **P < 0.01, one-way ANOVA).

The interocular difference in VCD for the FD-WT group increased by 0.026 ± 0.006 mm compared with 0.001 ± 0.006 mm for the WT-only group (P = 0.002, one-way ANOVA) at week 4 of the experiment, with an increased interocular difference in AL (0.027 ± 0.008 mm in FD-WT versus −0.004 ± 0.008 mm in WT-only, P = 0.005). In contrast, the FD-KO group showed no significant difference compared with the KO-only group in the interocular difference in VCD (0.008 ± 0.006 mm in FD-KO versus 0.003 ± 0.005 mm in KO-only, P = 0.539) or AL (−0.001 ± 0.007 mm in FD-KO versus 0.002 ± 0.007 mm in KO-only, P = 0.730). The elongation of both VCD and AL in the FD-KO group was reduced by at least 60% compared with the FD-WT group: the values for VCD were 0.026 ± 0.006 mm in FD-WT versus 0.008 ± 0.006 mm in FD-KO (P = 0.044), and the values for AL were 0.027 ± 0.008 mm in FD-WT versus −0.001 ± 0.007 mm in FD-KO (P = 0.007) (Table 1; Fig. 1). 

The corneal radius of curvature, anterior chamber depth, and LT in the FD-WT and FD-KO groups were similar to their respective control groups at week 4 of treatments. The P value for both FD-WT versus WT-only and FD-KO versus KO-only was P > 0.05 (Table 1; Figs. 1D–F). 

FD in Pharmacologically Treated Mice

There was also no significant difference in eyes between any two groups for all baseline results. The P value was P > 0.05 for all by one-way ANOVA (Table 2). 

Table 2

 

Baseline Measurements of Body Weight, Refraction, and Ocular Biometry for Pharmacological Groups (Mice Aged 4 Weeks)

Group	Weight, g	Eye	Refraction, D	ACD, mm	LT, mm	VCD, mm	AL, mm	CRC, mm
Untreated control, n = 31	10.83 ± 0.29	OD	−2.73 ± 0.21	0.308 ± 0.002	1.392 ± 0.003	0.715 ± 0.004	2.513 ± 0.008	1.365 ± 0.018
		OS	−2.71 ± 0.20	0.308 ± 0.003	1.392 ± 0.004	0.717 ± 0.006	2.517 ± 0.010	1.353 ± 0.016
DMSO, n = 28	10.78 ± 0.32	OD	−2.60 ± 0.34	0.308 ± 0.004	1.392 ± 0.006	0.716 ± 0.005	2.506 ± 0.010	1.336 ± 0.014
		OS	−2.56 ± 0.34	0.307 ± 0.004	1.387 ± 0.005	0.719 ± 0.004	2.509 ± 0.008	1.332 ± 0.014
Sulpiride, n = 26	11.00 ± 0.19	OD	−2.60 ± 0.22	0.309 ± 0.003	1.407 ± 0.003	0.716 ± 0.006	2.534 ± 0.007	1.325 ± 0.024
		OS	−2.58 ± 0.22	0.306 ± 0.002	1.404 ± 0.003	0.720 ± 0.004	2.538 ± 0.006	1.330 ± 0.020
FD-only, n = 31	10.84 ± 0.25	OD	−2.86 ± 0.31	0.310 ± 0.003	1.389 ± 0.004	0.716 ± 0.004	2.508 ± 0.007	1.349 ± 0.011
		OS	−2.76 ± 0.33	0.308 ± 0.003	1.388 ± 0.004	0.721 ± 0.005	2.508 ± 0.007	1.342 ± 0.015
FD-DMSO, n = 28	10.95 ± 0.26	OD	−2.84 ± 0.29	0.312 ± 0.003	1.396 ± 0.005	0.717 ± 0.004	2.522 ± 0.007	1.339 ± 0.018
		OS	−3.00 ± 0.32	0.310 ± 0.003	1.392 ± 0.005	0.720 ± 0.005	2.526 ± 0.006	1.323 ± 0.015
FD-sulpiride, n = 28	10.96 ± 0.27	OD	−3.45 ± 0.37	0.312 ± 0.003	1.396 ± 0.005	0.716 ± 0.005	2.541 ± 0.008	1.343 ± 0.009
		OS	−3.45 ± 0.37	0.312 ± 0.003	1.399 ± 0.004	0.717 ± 0.004	2.544 ± 0.008	1.334 ± 0.011

At week 4 of the experiment, there were significant differences in refraction (P < 0.001, one-way ANOVA) and AL (P = 0.025, one-way ANOVA) among the fellow eyes (left eyes) of three FD groups (Table 3). The fellow eyes of the FD-sulpiride group were significantly less hyperopic (0.92 ± 0.31 D in FD-sulpiride versus 2.61 ± 0.34 D in FD-DMSO, P < 0.001, one-way ANOVA with Bonferroni correction) and had a longer AL (2.746 ± 0.007 mm in FD-sulpiride versus 2.726 ± 0.005 mm in FD-DMSO, P = 0.023) compared with the FD-DMSO group or the FD-only group. There was also a significant difference in VCD (P = 0.023, one-way ANOVA) among the deprived eyes (right eyes) of three FD groups (Table 3). The deprived eyes of the FD-sulpiride group showed a smaller VCD (0.651 ± 0.004 mm in FD-sulpiride versus 0.665 ± 0.005 mm in FD-DMSO, P = 0.033) compared with the FD-DMSO group or FD-only group at week 4 of the experiment, while other parameters were similar among three FD groups (P > 0.05, one-way ANOVA). There was no significant difference in the refraction, VCD, or AL of the deprived eyes between the FD-only group and the FD-DMSO group (P > 0.05, one-way ANOVA). 

Table 3

 

Body Weight, Refraction, and Ocular Biometry at Week 4 of FD and Pharmacological Treatment in Mice (Aged 8 Weeks)

Group	Weight, g	Eye	Refraction, D	ACD, mm	LT, mm	VCD, mm	AL, mm	CRC, mm
Untreated control, n = 31	23.65 ± 0.19	OD	1.45 ± 0.37	0.392 ± 0.003	1.604 ± 0.003	0.642 ± 0.004	2.743 ± 0.005	1.483 ± 0.011
		OS	1.44 ± 0.35	0.385 ± 0.003	1.603 ± 0.004	0.644 ± 0.004	2.738 ± 0.005	1.485 ± 0.012
DMSO, n = 28	23.44 ± 0.31	OD	0.56 ± 0.33	0.382 ± 0.004	1.604 ± 0.004	0.638 ± 0.003	2.731 ± 0.006	1.474 ± 0.012
		OS	0.67 ± 0.28	0.376 ± 0.004	1.601 ± 0.004	0.643 ± 0.004	2.727 ± 0.006	1.471 ± 0.011
Sulpiride, n = 26	23.02 ± 0.22	OD	0.78 ± 0.25	0.378 ± 0.002	1.606 ± 0.003	0.626 ± 0.004	2.719 ± 0.005	1.505 ± 0.015
		OS	0.77 ± 0.25	0.374 ± 0.003	1.606 ± 0.003	0.629 ± 0.003	2.715 ± 0.006	1.467 ± 0.012
FD-only, n = 31	22.94 ± 0.12	OD	−1.54 ± 0.42	0.387 ± 0.003	1.596 ± 0.004	0.670 ± 0.005	2.756 ± 0.005	1.498 ± 0.015
		OS	2.83 ± 0.35	0.388 ± 0.003	1.593 ± 0.004	0.649 ± 0.005	2.728 ± 0.005	1.490 ± 0.009
FD-DMSO, n = 28	22.97 ± 0.20	OD	−1.45 ± 0.32	0.383 ± 0.004	1.605 ± 0.004	0.665 ± 0.005	2.753 ± 0.006	1.479 ± 0.014
		OS	2.61 ± 0.34	0.384 ± 0.003	1.600 ± 0.003	0.643 ± 0.004	2.726 ± 0.005	1.478 ± 0.011
FD-sulpiride, n = 28	22.49 ± 0.23	OD	−1.09 ± 0.31	0.383 ± 0.003	1.605 ± 0.003	0.651 ± 0.004*	2.742 ± 0.007	1.485 ± 0.009
		OS	0.92 ± 0.31**	0.382 ± 0.002	1.607 ± 0.004	0.652 ± 0.004	2.746 ± 0.007*	1.506 ± 0.011

 Data are expressed as the mean ± standard error. CRC, corneal radius of curvature.

*   P < 0.05 and ** P < 0.01 compared with the FD-DMSO and FD-only groups, one-way ANOVA.

At week 4 of the experiment, the mean interocular difference in refraction for the FD-only group was approximately −4.37 ± 0.30 D compared with 0.01 ± 0.13 D for the untreated control group (P < 0.001, one-way ANOVA) (Fig. 2A). The myopic shift in the FD-sulpiride group was 50% lower than that recorded in the FD-DMSO and FD-only groups (interocular difference, −2.01 ± 0.31 D in FD-sulpiride versus −4.06 ± 0.30 D in FD-DMSO versus −4.37 ± 0.30 D in FD-only, P < 0.001, one-way ANOVA). In contrast, there was no significant difference in refraction between the FD-only and FD-DMSO groups (−4.37 ± 0.30 D in FD-only versus −4.06 ± 0.30 D in FD-DMSO, P = 1.000, one-way ANOVA) (Table 3). 

Figure 2

 

Eyes treated with FD-sulpiride showed a significantly lower myopia (interocular difference) (A), with a smaller VCD (B) and AL (C), compared with the FD-DMSO and FD-only groups (*P < 0.05 and **P < 0.01, one-way ANOVA). Results in corneal radius of curvature (D), ACD (E), and LT (F) were not affected by FD or sulpiride.

The mean interocular difference in VCD for the FD-only group increased by 0.020 ± 0.004 mm compared with −0.002 ± 0.004 mm for the untreated control group (P < 0.001, one-way ANOVA) (Fig. 2) at week 4 of the experiment, with an increased interocular difference in AL (0.027 ± 0.005 mm in FD-only versus 0.005 ± 0.005 mm in untreated controls, P = 0.002). Both the FD-only and FD-DMSO groups had a greater VCD and AL than the FD-sulpiride group at week 4 of the experiment: the interocular difference for VCD was −0.001 ± 0.006 mm in FD-sulpiride versus 0.022 ± 0.004 mm in FD-DMSO (P = 0.003) and for AL was −0.004 ± 0.007 mm in FD-sulpiride versus 0.027 ± 0.005 mm in FD-DMSO (P = 0.001), both by one-way ANOVA. In contrast, there was no significant difference in either the VCD or AL (interocular differences) between the FD-only group and the FD-DMSO group (P = 1.000, one-way ANOVA) (Table 3; Fig. 2). 

No significant difference with respect to either raw data or interocular difference was observed in corneal radius of curvature, anterior chamber depth, or LT among different FD groups (P > 0.05) (Table 3) at week 4 of the experiment. All biometric parameters were similar among the three control groups (sulpiride, DMSO, and normal control) at week 4 of the experiment (P > 0.05, one-way ANOVA) (Table 3). The body weight was not affected by week 4 of the drug treatment (P > 0.05 among three FD groups or three control groups, one-way ANOVA) (Table 3). 

Discussion

This study for the first time to date used a D2R KO model complemented with sulpiride, a selective D2R antagonist, to investigate the role of D2R in FDM development in mice. The myopia induced in the FD-KO group was approximately reduced by 60% compared with the FD-WT group, with a proportionately slower growth of VCD and AL. Similarly, the myopia that developed in the FD-sulpiride group was 50% lower than in the FD-DMSO group. These results indicate that either systemic administration of sulpiride or D2R gene deletion inhibits FDM development in mice by at least 50%. 

It is noteworthy that the fellow eyes of the FD-sulpiride group were significantly less hyperopic and had a longer AL compared with the FD-DMSO group or the FD-only group. In contrast, the deprived eyes of the FD-sulpiride group had only a smaller VCD compared with the FD-DMSO group or the FD-only group, while other parameters were similar among the three FD groups. Thus, systemic administration of sulpiride appears to affect both the form-deprived and fellow eyes in animals undergoing FD. However, sulpiride administered by intraperitoneal injection had no effects on the untreated control eyes in this study. These results suggest that an interocular connection exists in the mechanism modulating eye growth in mice, which is consistent with the results of a previous study ¹³ in which the retinas of untreated fellow eyes of chickens wearing a unilateral goggle displayed altered DOPAC levels compared with the eyes of never-goggled chickens. 

Sulpiride did not influence refractive development and axial growth of the eye in normal visual environments. This finding is similar to a previous study ²⁰ in which apomorphine did not change the axial growth of the guinea pig eye in normal visual environments. The different effects of sulpiride on eyes under FD and normal visual conditions could be due to a higher affinity or susceptibility of the deprived eye to an exogenous D2R antagonist because eyes under deprived conditions have a lower level of retinal DA compared with eyes under normal visual environments. ³⁵ This hypothesis is consistent with a finding indicating that D2R becomes more likely to be activated when the level of retinal DA declines. ³  

The refraction in the 4-week-old WT mice (−2.73 ± 0.21 D) observed in this study was largely consistent with our previous study ³³ in which the refraction was −4.61 ± 2.96 D at postnatal day 25. In the present study, WT mice developed a hyperopic shift, as summarized in Table 2 and Table 3 (from −2.7 D in 4-week-old mice to 1.4 D in 8-week-old mice). Consistent with this notion, the C57BL/6 mice developed myopia during early postnatal development but then progressed to more hyperopia with age and reached a stable refraction of hyperopia. ³³ For the BALB/cJ mice, the refraction measured by streak retinoscopy increased from young to adult mice. ³⁶ The trend toward greater hyperopia after 4 weeks is consistent with other studies. ^31,37 The difference in refraction between two WT groups that were both 8 weeks old (5.29 D in Table 1 of the D2R KO study versus 1.45 D in Table 3 of the pharmacological study) is likely attributable to different genetic backgrounds for these two sets of WT mice (heterozygous D2R KO hybridization versus WT C57BL/6 hybridization). 

The effect of D2R antagonists on axial growth of the eye is known to be dose dependent in chicken models. ²² In this study, systemic administration of sulpiride at 8 μg/g significantly inhibited FDM, indicating that this dose of sulpiride diffusing into the eye through the circulation was still sufficient to block D2R activation. In chicken models, intravitreal injection of sulpiride at a dose between 100 and 400 μg enhances FDM but suppresses myopia at a dose higher than 400 μg. ²² The different effects of sulpiride on the FDM chicken ²² and mouse models (our present results) could be related to differences in (1) drug distribution due to differences in routes of drug administration (intravitreal injection in the study by Shaeffel et al. ²² and intraperitoneal injection in our study), (2) concentrations of drug reaching in the retina in these studies, and (3) species (chickens versus mice). Intravitreal injections can potentially produce significantly higher levels of drug than the levels achieved with intraperitoneal injections. Such different D2R drug concentrations in the retina may result in differential activation of the presynaptic versus postsynaptic D2R and/or differential activation of DA receptor subtypes (D2R versus D1R) due to partial specificity of the D2R drugs. As is known, chickens are basically diurnal animals, with a high dependence on vision in refractive development, ³⁸ and mice are more nocturnal, with a less vision-dependent behavior. ³⁹ This difference between chickens and mice may contribute to the different refractive changes between these two animal models under similar experimental conditions. Consistent with the results found in mice, D2R antagonists also inhibit myopia development in albino guinea pigs (Jiang L, et al. IOVS 2012;53:ARVO E-Abstract 3437) and pigmented guinea pigs (Zhou X, et al. IOVS 2013;54:ARVO E-Abstract 3678). Furthermore, results obtained with D2R KO mice were consistent with those treated with sulpiride, confirming that D2R activation has a critical role in the development of FDM in C57BL/6 mice. In the present study, the reduced myopic shift in the FD-D2R KO mice is inconsistent with the current literature in which an increased myopic shift was found when D2R was inhibited. ²² This difference may be in part due to complex actions of D2R, depending on different DA concentrations, different approaches (D2R KO versus D2R drug), and different routes of drug administration (local versus systemic). 

Dopaminergic mechanisms have an essential role in mediating signal transmission from the retina to the visual cortex because DA is a neurotransmitter, with its receptors widely distributed in the retina and throughout the brain. ³ The partial inhibition of FDM by sulpiride in the present study suggests that either myopic shift does not entirely depend on the role of D2R receptors or other signaling pathways are also involved in the control of axial growth of the eye. For instance, apomorphine, a nonselective DA receptor agonist, can inhibit FDM in various animals, including chickens, ^11,18,27 guinea pigs, ²⁰ rabbits, ²¹ and rhesus monkeys. ¹⁹ The findings that both apomorphine (a nonselective DA agonist) and sulpiride (a D2R antagonist) could partially inhibit FDM strongly suggest that D2R and other DA receptor subtypes are involved through different mechanisms in modulating the same type of control. ⁴⁰ This hypothesis is supported by recent observations that myopia development is inhibited by D1R agonists and promoted by D2R agonists in albino guinea pigs (Jiang L, et al. IOVS 2012;53:ARVO E-Abstract 3437) and pigmented guinea pigs (Zhou X, et al. IOVS 2013;54:ARVO E-Abstract 3678). Dopamine acts through D1R and D2R, with opposite physiological effects produced by activation of these two receptor subtypes in various tissues. ^41,42 The reduction in FDM in D2R KO mice further suggests that D2R activation acts to promote FDM, opposite to the effect of D1R activator, which inhibits FDM. ²⁰  

Although the retinal DA release or turnover was not measured in this study, previous investigations have shown that retinal DA and DOPAC decreased during FDM. ² Furthermore, the D2R antagonist sulpiride increases retinal DA levels of form-deprived eyes in chickens, ²² probably due to the antagonism of presynaptic autoinhibitory D2R autoreceptors. An increase in retinal DA may have contributed to inhibition of myopia in the present study. However, D2R KO mice lack a D2R-mediated feedback system, and deletion of D2R could sufficiently suppress autoreceptor-mediated inhibition of DA release both in the striatum and in the shell of the nucleus accumbens. ⁴³ How the dopaminergic activity specifically controls myopia remains to be clarified by future studies. 

Although it might be possible that mice lacking the inhibitory effects of presynaptic D2R autoreceptors would show greater DA release, D2R KO mice have basal and K⁺-evoked DA extracellular concentrations similar to those in WT littermates in striatum. ⁴⁴ In addition, no changes are observed in D2R KO mice in extracellular levels in striatum of DOPAC ⁴⁴ or in the expression of mRNA, protein for tyrosine hydroxylase, and the enzymes responsible for DA synthesis. ⁴⁵ Based on these previous results, the unchanged basal DA levels are due to compensatory changes in DA uptake and metabolism. These compensatory changes, including an increase in adenosine A_2A receptors ⁴⁶ and N-methyl-d-aspartate receptors, have been reported in D2R KO mice. ⁴⁷ Further studies are needed to determine whether such compensatory changes also occur during the development of FDM in D2R KO mice. 

Because D2R KO mice used in the present study were generated by targeted mutagenesis of the D2R gene in embryonic stem cells, ⁴⁸ the D2R gene has been deleted throughout the body during early development. In the present study, mice lacking D2R were smaller and lighter, probably due to a decrease in food intake and/or dysfunction of the gastrointestinal system in the absence of D2R. ⁴⁹ It should be noted that the smaller size of the body and eye in D2R KO mice may be linked, offering a partial nonvisual explanation to the refractive changes in these mice. Despite these limitations, a specific role of D2R in myopia development in these groups of animals could not be excluded. The actual role of retinal D2R in inhibition of FDM remains to be resolved. Further studies, including measurement of vitreal DA levels and selective retinal D2R blockade during FDM development, may clarify these issues. 

In summary, our study demonstrated that DA systems participate in the modulation of visually guided ocular growth and that D2R gene deletion reduces myopia induced by FD in mice during postnatal development. This conclusion is further substantiated by selective blockade of D2R by sulpiride, replicating the D2R KO results. Further studies are required to confirm these results using animal models in which retinal D2R is selectively blocked. 

Acknowledgments

The authors thank Peter Reinach, PhD, School of Ophthalmology and Optometry, Wenzhou Medical University, Wenzhou, China, for editorial support that improved the manuscript. 

Supported by Grant 2011CB504602 from the National Basic Research Program of China (973 Project), Grants 30973278 and 81000386 from the National Natural Science Foundation of China, the Zhejiang Provincial Program for the Cultivation of High-Level Innovative Health Talents, Grant NCET-10-0977 from the Program for New Century Excellent Talents in University of the National Ministry of Education, the National Young Excellent Talents Support Program and Grant Y20100206 from the Wenzhou Science and Technology Plan Projects. 

Disclosure: F. Huang, None; T. Yan, None; F. Shi, None; J. An, None; R. Xie, None; F. Zheng, None; Y. Li, None; J. Chen, None; J. Qu, None; X. Zhou, None