The critical period for darkness-induced recovery of the vision of the amblyopic eye following early monocular deprivation

Abstract

Exposure of kittens to complete darkness for 10 days has been shown (Duffy & Mitchell, 2013) to reverse the loss of visual acuity that follows a prior period of monocular deprivation (MD). In that study, recovery of acuity in the previously deprived eye was fast despite the fact that darkness was imposed 2 months after the period of MD when kittens were 3 months old. In a later study (Holman, Duffy, & Mitchell, 2018), it was demonstrated that the same period of darkness was ineffective when it was imposed on cats about 1 year old, suggesting that dark exposure may only promote recovery when applied within an early critical period. To determine the profile of this critical period, the identical period of darkness (10 days) was imposed on kittens at various ages that had all received the same 7-day period of MD from postnatal day 30 (P30). Recovery of the acuity of the deprived eye as measured by use of a jumping stand was complete when darkness was imposed prior to P186 days, but thereafter, darkness induced progressively smaller acuity improvements and was ineffective in kittens when it began at or beyond P191 days of age. These data indicate a critical period for darkness-induced recovery with an abrupt end over a 5-day period.

Introduction

Amblyopia, a common neurodevelopmental visual disorder, has long been modeled in experimental animals by monocular deprivation (MD), which disrupts patterned visual input to one eye in early postnatal life (Daw, 2006; Hubel & Wiesel, 2005). It has long been known that MD produces long-standing visual deficits in this eye as well as large anatomical and physiological changes in the mammalian central visual pathways (Daw, 2006; Hensch, 2005; D. E. Mitchell & Timney, 1984; Movshon & Van Sluyters, 1981). Although originally demonstrated in cats (Wiesel & Hubel, 1963) and monkeys (Baker, Grigg, & Von Noorden, 1974; Hubel, Wiesel, & Levay, 1977), similar but somewhat less extensive consequences of MD have been demonstrated more recently in rats (Fagiolini, Pizzorusso, Berardi, Domenici, & Maffei, 1994), mice (Dräger, 1978), and ferrets (Issa, Trachtenberg, Chapman, Zahs, & Stryker, 1999). In cats and monkeys as well as the species studied later, the visual cortex was shown to be vulnerable to MD only within an early critical period with a timing that was species dependent (Daw, 2006). Consistent with the concept of critical periods in development during which neural connections are malleable, the consequences of an early extended period of MD were thought to be largely intractable to subsequent experiential or other interventions initiated in adulthood. However, two studies conducted on Long Evans rats provided a serious challenge to this opinion by demonstrating a remarkable capacity for recovery from both the physiological (He, Ray, Dennis, & Quinlan, 2007) and visual (He, Hodos, & Quinlan, 2006) changes induced by an early period of MD following exposure to a 10-day period of complete darkness. An important feature of these studies is that darkness was imposed on mature rats at 100 days of age, a time considered well beyond the critical period for the consequences of MD, which is believed to end by about 7 weeks of age for this species (Fagiolini et al., 1994). The remarkable ability of darkness to reverse the consequences of an early period of MD at 100 days of age raises the possibility that it may be effective either throughout life or over a prolonged span of time in comparison to the short critical period of vulnerability to the physiological and behavioral consequences of an early period of MD. Strong support for this contention was provided by the findings from a study on mice (Stodieck, Greifzu, Goetze, Schmidt, & Löwel, 2014) that reported rescue of ocular dominance plasticity in the visual cortex following a 14-day period of darkness imposed at 535 days of age. 

Recent studies of kittens have demonstrated that 10 days of darkness can reverse both the severe anatomical consequences of an early period of MD in the dorsal lateral geniculate nucleus (Duffy, Lingley, Holman, & Mitchell, 2016) as well as the profound loss of spatial vision in the deprived eye (Duffy & Mitchell, 2013; D. E. Mitchell, MacNeill, Crowder, Holman, & Duffy, 2016). A key observation from the latter studies was the fast and seemingly complete recovery of visual acuity in the deprived eye of kittens in which darkness was imposed 8 weeks after termination of a 7-day period of MD that began at postnatal day 30 (P30). Darkness was imposed at around 3 months of age within the waning stages of the period of vulnerability of the kitten visual cortex to MD (Olson & Freeman, 1980). However, the possibility that 10 days of darkness may promote recovery from the visual consequences of MD throughout life was eliminated by the subsequent demonstration that it was ineffective when imposed on cats at 1 year of age (Holman et al., 2018). In the present study, we report the results of a systematic study of the effectiveness of a constant 10-day period of darkness imposed on kittens at progressively later ages to promote recovery of visual acuity in the deprived eye of kittens that received the same 7-day period of MD at P30. The results indicate that, for the latter short but constant length of MD, 10 days of total darkness can promote recovery of the acuity of the deprived eye of kittens only when imposed within a critical period that ends sharply over a 5-day interval from 186 days of age. Although the timing of the critical period may vary with the length of MD or darkness or both, the clear indications of a restricted time window for the benefits of darkness in kittens provides evidence of species differences with respect to both rats and mice for which darkness appears efficacious throughout life. 

Materials and methods

Animals and rearing conditions

Data were collected from 21 kittens that included seven animals from five separate litters (Table 1, litters A–E) that participated in three previously completed investigations, namely four animals from the earliest published study (Duffy & Mitchell, 2013) and three examined more recently (C304 from D. E. Mitchell et al., 2016, and C422 and C439 from D. E. Mitchell, Aronitz, Bobbie-Ansah, Crowder, & Duffy, 2019). The 14 new animals reared for the current study were derived from five litters (litters F–J) from three queens and four unrelated studs that were bred and reared in a closed animal colony at Dalhousie University over a 3-year period from September 2013. The animal colony and all animal procedures followed protocols approved by the Dalhousie University committee on laboratory animals and conformed to the guidelines of the Canadian Council on Animal Care. All 21 animals received 7 days of MD at about P30 followed by a 10-day period of total darkness that began at progressively later ages. Because of the variable and generally small litter sizes, kittens were assigned to the various exposure conditions without consideration of gender but with kittens of each litter assigned to a variety of different exposure conditions. Table 1 displays the rearing history of all animals as well as their gender and litter of origin designated by a letter. The small size of the breeding colony dictated that all animals were assigned to experimental groups. Comparisons of the recovery of the acuity of the deprived eye were made with the acuity of the fellow eye and to acuity norms established over the last decade on age-matched normal kittens (D. E. Mitchell et al., 2016). 

Table 1

 

Animals and rearing conditions. Notes: Timing (postnatal days of age) of the initial period of MD and the subsequent period of darkness for all 21 kittens including the seven animals that participated in three previous studies (C151, C152, C15, C157 from Duffy & Mitchell, 2013; C304 from D. E. Mitchell et al., 2016; and C422 and C439 from D. E. Mitchell et al., 2019). Animals are listed in order of their age when exposed to darkness. The C number of the animals defines the order of their birth. Also shown are litter identifications, designated by a letter to identify littermates as well as the gender (male, M or female, F) of each kitten.

 

For all but the period spent in darkness, animals were housed in colony rooms that were illuminated on a 12:12 hr light/dark cycle that was changed on occasions to as high as a 14:10 cycle in some rooms to promote breeding. During the day, animals ran free in the colony rooms that contained many toys and purpose-built stands for climbing and at night were housed in large interconnected cages within the colony rooms. The behavioral measurements of acuity that we made did not require any reduction in the amount or nature of their daily food. For the 10-day period of total darkness, animals were moved to a large darkroom (3.8 × 3.5 m) that was part of a dark-rearing facility that is described in detail elsewhere (D. E. Mitchell, 2013). The facility contained two adjacent darkrooms that were accessible through several small anterooms and doors that ensured that the two darkrooms were light tight. To entrain an activity cycle, a radio in the darkroom was automatically turned on and off at times that corresponded to the lighting cycle of the colony rooms. In the darkroom, animals were kept in a large cage (1.5 × 0.7 × 0.9 m) with a 24-cm ledge running the length of the cage. As the animals were at least 2.5 months old at the time they entered the darkroom, they had all been weaned and so were held in the darkroom without their mother. Usually only one or two kittens were held in the darkroom at the same time, and they were held in separate large cages. 

As a possible consequence of the activity cycle entrained by the radio onset, regular feeding, and cage cleaning times as well as petting by experimenters and technicians, the animals appeared to adjust quickly to the darkroom as evidenced by purring and a calm demeanor. Upon removal from the darkroom, they also appeared to adjust rapidly to an illuminated environment to the extent that it was possible to initiate measurements of acuity within 2 hr. Thereafter, measurements of the acuity of the deprived eye were made daily until the vision appeared to reach a stable level. 

Surgical procedures

Monocular deprivation by eyelid suture of the left eye was achieved by use of a procedure developed (Murphy & Mitchell, 1987) to maintain a secure eyelid closure while allowing a fast recovery of a normal patent palpebral aperture upon its termination to facilitate the behavioral assessment of the vision of this eye after the eyelids were opened. Detailed descriptions of these procedures have been provided in recent papers (Duffy, Fong, Mitchell, & Bear, 2018), so only a brief summary is provided here. All surgical procedures were performed under gaseous isoflurane anesthesia (2%–3% in oxygen), and a subcutaneous injection of Anafen for postprocedure analgesia was administered once the animals were anesthetized. Local anesthesia was administered with Alcaine sterile ophthalmic solution (proparacaine hydrochloride). The first stage of the procedure was to dissect the palpebral conjunctivae free from the upper and lower eyelids and to suture them together with 6-O vicryl suture thread. A broad-spectrum topical antibiotic (chloromycetin 1%) was then applied to the sutured conjunctivae. The second stage of the surgical procedure was to oppose and suture the exposed tissue on the underside of the eyelids together with 6-O or 5-O silk. 

To open the eyelids after the 7-day period of MD, animals were anesthetized with isoflurane, and any remaining suture material was removed. The eyelids and underlying conjunctivae were then gently cut and pulled apart to achieve a normal palpebral opening. A broad-spectrum topical antibiotic (chloromycetin 1%) was applied to the cornea and surrounding conjunctivae. 

Behavioral measurements of visual acuity

Measurements of the visual acuity for square-wave gratings were made by use of a jumping stand and procedures developed and refined over the last four decades in this (D. E. Mitchell, 2013; D. E. Mitchell, Giffin, & Timney, 1977; Murphy & Mitchell, 1987) and other laboratories (Murphy, Roumeliotis, Williams, Beston, & Jones, 2015; Williams, Balsor, Beshara, Beston, & Jones, 2015). The training and testing procedures that are used currently and employed for this study have been described in detail recently (D. E. Mitchell, 2013) and so are summarized only briefly here. Some modifications to the usual procedure were necessary to accommodate the long duration of testing, which, for some animals, extended to 8 months of age. As in the past, the kittens were trained initially when between 4 and 5 weeks old. However, unlike animals in most of our previous studies, tests of acuity had to be made over a long period of time and at ages at which the animals were large and more easily distracted. All the tests of acuity were made with two of the authors seated on either side of the jumping platform with one person providing the food and social reward after each response while the other person recorded the response and prepared the stimuli for the next trial. The stimuli were adjacent, large (19 × 19 cm) horizontal and vertical square-wave gratings of the same period (and, hence, spatial frequency), a Michelson contrast of 1.0 and with a luminance of 80 cd/m². The stimuli were printed by an ink-jet printer, pasted onto thick cardboard, and coated with a clear protective matte finish (Krylon, Sherwin-Williams). 

Acuity was measured in the morning between the hours of 9 and 11 a.m. by use of a descending method of limits with jumps to the vertical (positive) stimulus rewarded by petting that was supplemented during part of the session by small amounts of wet food (paté) until the animal became satiated. Errors resulted in a denial of these rewards. The presence of people on both sides of the jumping platform reduced the frequency of side preferences by the animals as can occur when the discrimination becomes difficult and also discouraged attempts to escape when the animals became large. Because the changes in spatial frequency between blocks of trials were very small and equated on a logarithmic scale with as many as 12 steps/octave changes in spatial frequency, only a single trial was provided at the lowest spatial frequencies until an error was made. At this point, the animal had to make five consecutively correct responses or a minimum of seven correct responses out of a maximum of 10 trials provided at any spatial frequency before the spatial frequency was increased. Within about five steps of spatial frequency of threshold, the minimum number of trials was increased to five. Threshold, defined as the highest spatial frequency for which the animal performed at a level of 70% or better, was typically sharp so that performance fell from flawless to chance within three step changes of spatial frequency. Kittens exhibited a number of stereotypical behaviors near threshold that included a drastic increase in latency to respond, crying, looks toward one or both of the people involved in testing as well as attempts to back away from the edge of the jumping platform. These behavioral patterns in addition to further formal tests provided evidence that the thresholds reflected the animal's visual acuity (i.e., the angular dimensions of the stimuli) and not by a lack of motivation. Among the additional tests were the use of stimuli of higher spatial frequency at the beginning of the testing session, thereby reducing the number of steps to the same angular dimension at threshold or a drastic change in the distance of the jumping platform to the stimuli that would result in a proportional increase or decrease of the period of the grating at threshold if it was dictated by the angular size of the stimulus. 

Once animals were trained, which usually occurred in the fifth and sixth week, measurements of thresholds were made two to three times a week and more frequently in the 2 weeks before they were placed in the darkroom. With only a few exceptions, the testing of the acuity was conducted in the morning at about the same time for each individual animal. Tests of the acuity of the deprived eye were made with a hard opaque contact lens occluder placed in the other eye. Six occluders of different base curvatures selected to match the mean corneal curvatures of young kittens of various ages (Freeman, 1980) were used as the animals matured. To mitigate against any possible pain, a drop of a local ophthalmic anesthetic (proparacaine hydrochloride 1%) was administered to the eye to be occluded immediately prior to insertion of the contact lens. No signs of any discomfort were evident in the 20 min of occlusion of the fellow eye that was typically required for the measurement of the acuity of the deprived eye. Measurement of binocular grating acuity was used as a substitute for monocular measurement of the grating acuity of the nondeprived eye as in the past they have been demonstrated to be identical. An additional advantage of this practice was it enabled measurement to be made of the acuity of the deprived eye in the immediate aftermath of the binocular measurement. Monocular measurements of the acuity of the nondeprived eye were made on nearly all animals at the conclusion of testing, and without exception these matched the binocular measurements. Because the contact lens occluders have been shown to cause a temporary distortion of the cornea, it was not possible to obtain monocular measurements of the acuity of the two eyes on a single day (Dzioba, Murphy, Horne, & Mitchell, 1986). Following the period of MD, the acuity of the deprived eye improved gradually to reach a stable level in about 3–4 weeks and as many as 6 months prior to the period of darkness. During the time when the acuity of the deprived eye was stable, the starting spatial frequency for each acuity measurement was altered to ensure that the threshold reflected a true visual barrier irrespective of the number of trials or length of the testing session. 

Curve fits to recovery data

The change in the grating acuity of the deprived eye over time was fit to a three-part piecewise linear function using the least squares method:  

\(\def\upalpha{\unicode[Times]{x3B1}}\)\(\def\upbeta{\unicode[Times]{x3B2}}\)\(\def\upgamma{\unicode[Times]{x3B3}}\)\(\def\updelta{\unicode[Times]{x3B4}}\)\(\def\upvarepsilon{\unicode[Times]{x3B5}}\)\(\def\upzeta{\unicode[Times]{x3B6}}\)\(\def\upeta{\unicode[Times]{x3B7}}\)\(\def\uptheta{\unicode[Times]{x3B8}}\)\(\def\upiota{\unicode[Times]{x3B9}}\)\(\def\upkappa{\unicode[Times]{x3BA}}\)\(\def\uplambda{\unicode[Times]{x3BB}}\)\(\def\upmu{\unicode[Times]{x3BC}}\)\(\def\upnu{\unicode[Times]{x3BD}}\)\(\def\upxi{\unicode[Times]{x3BE}}\)\(\def\upomicron{\unicode[Times]{x3BF}}\)\(\def\uppi{\unicode[Times]{x3C0}}\)\(\def\uprho{\unicode[Times]{x3C1}}\)\(\def\upsigma{\unicode[Times]{x3C3}}\)\(\def\uptau{\unicode[Times]{x3C4}}\)\(\def\upupsilon{\unicode[Times]{x3C5}}\)\(\def\upphi{\unicode[Times]{x3C6}}\)\(\def\upchi{\unicode[Times]{x3C7}}\)\(\def\uppsy{\unicode[Times]{x3C8}}\)\(\def\upomega{\unicode[Times]{x3C9}}\)\(\def\bialpha{\boldsymbol{\alpha}}\)\(\def\bibeta{\boldsymbol{\beta}}\)\(\def\bigamma{\boldsymbol{\gamma}}\)\(\def\bidelta{\boldsymbol{\delta}}\)\(\def\bivarepsilon{\boldsymbol{\varepsilon}}\)\(\def\bizeta{\boldsymbol{\zeta}}\)\(\def\bieta{\boldsymbol{\eta}}\)\(\def\bitheta{\boldsymbol{\theta}}\)\(\def\biiota{\boldsymbol{\iota}}\)\(\def\bikappa{\boldsymbol{\kappa}}\)\(\def\bilambda{\boldsymbol{\lambda}}\)\(\def\bimu{\boldsymbol{\mu}}\)\(\def\binu{\boldsymbol{\nu}}\)\(\def\bixi{\boldsymbol{\xi}}\)\(\def\biomicron{\boldsymbol{\micron}}\)\(\def\bipi{\boldsymbol{\pi}}\)\(\def\birho{\boldsymbol{\rho}}\)\(\def\bisigma{\boldsymbol{\sigma}}\)\(\def\bitau{\boldsymbol{\tau}}\)\(\def\biupsilon{\boldsymbol{\upsilon}}\)\(\def\biphi{\boldsymbol{\phi}}\)\(\def\bichi{\boldsymbol{\chi}}\)\(\def\bipsy{\boldsymbol{\psy}}\)\(\def\biomega{\boldsymbol{\omega}}\)\(\def\bupalpha{\unicode[Times]{x1D6C2}}\)\(\def\bupbeta{\unicode[Times]{x1D6C3}}\)\(\def\bupgamma{\unicode[Times]{x1D6C4}}\)\(\def\bupdelta{\unicode[Times]{x1D6C5}}\)\(\def\bupepsilon{\unicode[Times]{x1D6C6}}\)\(\def\bupvarepsilon{\unicode[Times]{x1D6DC}}\)\(\def\bupzeta{\unicode[Times]{x1D6C7}}\)\(\def\bupeta{\unicode[Times]{x1D6C8}}\)\(\def\buptheta{\unicode[Times]{x1D6C9}}\)\(\def\bupiota{\unicode[Times]{x1D6CA}}\)\(\def\bupkappa{\unicode[Times]{x1D6CB}}\)\(\def\buplambda{\unicode[Times]{x1D6CC}}\)\(\def\bupmu{\unicode[Times]{x1D6CD}}\)\(\def\bupnu{\unicode[Times]{x1D6CE}}\)\(\def\bupxi{\unicode[Times]{x1D6CF}}\)\(\def\bupomicron{\unicode[Times]{x1D6D0}}\)\(\def\buppi{\unicode[Times]{x1D6D1}}\)\(\def\buprho{\unicode[Times]{x1D6D2}}\)\(\def\bupsigma{\unicode[Times]{x1D6D4}}\)\(\def\buptau{\unicode[Times]{x1D6D5}}\)\(\def\bupupsilon{\unicode[Times]{x1D6D6}}\)\(\def\bupphi{\unicode[Times]{x1D6D7}}\)\(\def\bupchi{\unicode[Times]{x1D6D8}}\)\(\def\buppsy{\unicode[Times]{x1D6D9}}\)\(\def\bupomega{\unicode[Times]{x1D6DA}}\)\(\def\bupvartheta{\unicode[Times]{x1D6DD}}\)\(\def\bGamma{\bf{\Gamma}}\)\(\def\bDelta{\bf{\Delta}}\)\(\def\bTheta{\bf{\Theta}}\)\(\def\bLambda{\bf{\Lambda}}\)\(\def\bXi{\bf{\Xi}}\)\(\def\bPi{\bf{\Pi}}\)\(\def\bSigma{\bf{\Sigma}}\)\(\def\bUpsilon{\bf{\Upsilon}}\)\(\def\bPhi{\bf{\Phi}}\)\(\def\bPsi{\bf{\Psi}}\)\(\def\bOmega{\bf{\Omega}}\)\(\def\iGamma{\unicode[Times]{x1D6E4}}\)\(\def\iDelta{\unicode[Times]{x1D6E5}}\)\(\def\iTheta{\unicode[Times]{x1D6E9}}\)\(\def\iLambda{\unicode[Times]{x1D6EC}}\)\(\def\iXi{\unicode[Times]{x1D6EF}}\)\(\def\iPi{\unicode[Times]{x1D6F1}}\)\(\def\iSigma{\unicode[Times]{x1D6F4}}\)\(\def\iUpsilon{\unicode[Times]{x1D6F6}}\)\(\def\iPhi{\unicode[Times]{x1D6F7}}\)\(\def\iPsi{\unicode[Times]{x1D6F9}}\)\(\def\iOmega{\unicode[Times]{x1D6FA}}\)\(\def\biGamma{\unicode[Times]{x1D71E}}\)\(\def\biDelta{\unicode[Times]{x1D71F}}\)\(\def\biTheta{\unicode[Times]{x1D723}}\)\(\def\biLambda{\unicode[Times]{x1D726}}\)\(\def\biXi{\unicode[Times]{x1D729}}\)\(\def\biPi{\unicode[Times]{x1D72B}}\)\(\def\biSigma{\unicode[Times]{x1D72E}}\)\(\def\biUpsilon{\unicode[Times]{x1D730}}\)\(\def\biPhi{\unicode[Times]{x1D731}}\)\(\def\biPsi{\unicode[Times]{x1D733}}\)\(\def\biOmega{\unicode[Times]{x1D734}}\)\begin{equation}{\rm{A}}\left( t \right) = \left\{ {\matrix{ {o + {s_2} \times t + \left[ {\left( {{s_2} - {s_1}} \right) \times \left( {{b_1} - t} \right)],\ {\rm{if}\ }t \ \lt \ {b_1}} \right.} \cr {o + {s_2} \times t,\ {\rm if\ }{b_1} \le t \le {b_2}} \cr {o + {s_2} \times t + \left[ {\left( {{s_3} - {s_2}} \right) \times \left( {t - \left( {{b_1} + {b_2}} \right)} \right)} \right],\ {\rm if\ }t \ \gt \ {b_2}} \cr } } \right.\end{equation}

where A(t) is the deprived eye's grating acuity at time t. The first, second, and third slopes are s₁, s₂, and s₃, respectively. The first and second break points are b₁ and b₂, respectively. Finally, o is an additive term that shifts the function up or down. We favored a piecewise linear function over a logistic function because the speed of visual recovery captured by the s₂ term of the piecewise fit had a more intuitive interpretation than the exponential term in logistic functions that determines slope. However, to ensure our piecewise linear fits were not objectively worse, we also fit our data to a sigmoidal logistic function using the least squares method:  

\begin{equation}{{\ A}}\left( {{t}} \right) = {{{{{A}}_{{{max}}}}{\rm{\ }} \times {\rm{\ }}{{{t}}^{{n}}}} \over {{{{t}}^{{n}}} + {\rm{\ }}{{{t}}_{50}}^{{n}}{\rm{\ }}}} + o{\rm {,}}\end{equation}

where n is the exponent that determines the steepness of the curve, A_max is the maximum elevation in acuity above the baseline, and t₅₀ is the time when an acuity of half A_max occurs. Goodness of fit to each function was compared by calculating an F statistic (Motulsky & Ransnas, 1987):  

\begin{equation}F = {{\left( {S{S_{sig}} - S{S_{3pl}}} \right)/\left( {d{f_{sig}} - d{f_{3pl}}} \right)} \over {\left( {S{S_{3pl}}/d{f_{3pl}}} \right)}}{\rm {,}}\end{equation}

where SS_sig and SS_3pl are the sum of squared residuals between the deprived eye acuity and sigmoidal or three-part piecewise linear fits, respectively. Degrees of freedom were calculated as the number of data points minus the number of free parameters. We then obtained p values for each fit from an F table. For individual animals, the piecewise linear function fit the data significantly better (α = 0.025) more often (12/21 animals) than the sigmoid function (8/21 animals), but across our sample, there was no evidence of a difference between the sum of squared residuals from each function (p = 0.14, Wilcoxon rank sum). Thus, we were confident that the three-part piecewise linear function was adequate for our analysis.  

Results

Results from all the animals reared since the original paper (Duffy & Mitchell, 2013) are displayed in four figures (Figures 1 to 4) in which each figure shows the results from three animals in order of the age at imposition of darkness. Each figure displays the results of daily measurements of the grating acuity of the deprived eye immediately prior to (plotted at −11 on the abscissa) and following the 10-day period of darkness. The data shown immediately prior to dark exposure represents the best amblyopic eye acuity achieved and usually represents multiple identical values achieved on consecutive days. Although measurements of the binocular acuity were used to assess the acuity of the nondeprived eye for most of this interval, monocular measurements of the acuity of this eye were made on nearly all the animals once the acuity of the deprived eye had stabilized. In agreement with prior studies from this laboratory, the monocular measured acuity of the nondeprived eye matched that obtained with both eyes open. For animals displayed in Figure 1, darkness was either imposed at the same age (C304 at P94) or 1 week later than the age of the kittens from the original study (Duffy & Mitchell, 2013). Although the speed of acuity recovery of the acuity of the deprived eye was fast for all three animals, it was also quite variable. The largest discrepancy in the speed of recovery of the acuity of the deprived eye to match that of the other eye was observed in two animals (C422, 9 days and C439, 13 days) that were placed in darkness at very similar ages (P101 or P102). The speed of recovery of the third animal (C304, 10 days) that was placed in darkness earliest at P94 matched that observed for one of the three animals (C151) of the original study that were placed in darkness at P93 but was slower than the 5 days required by the other two (C157, C152). Although the recovery speed from the three animals of the original study were also variable, they were generally faster than that for C304, a finding that may in part be due to the better acuity of the deprived eye of the former animals at the time they were placed in the darkroom. 

Figure 1

 

The grating acuity of the deprived eye of three female animals (C304, C422, and C439) immediately before and in the days that followed 10 days of darkness imposed at, respectively, 94, 101, and 102 days of age. The kittens were reared from three separate litters from two different queens and three studs. Solid symbols depict measurements of the binocular visual acuity, open symbols the acuity of the deprived eye, and half-shaded symbols the acuity of the nondeprived eye. Results for the three individual animals are shown by symbols of different shapes and colors as follows: C304 orange squares, C422 green circles, and C439 blue triangles. The brackets to the right indicate the range of values for the monocular acuity of normal kittens of comparable ages and tested in a similar manner (D. E. Mitchell et al., 2016). Darkness promoted recovery of the acuity of the deprived eye of all three animals to match that of the fellow eye.

Figure 2

 

Grating acuity of the deprived eye of three male animals (C282, C283, and C292) immediately before and in the days that followed 10 days of darkness imposed at, respectively, 122, 157, and 169 days of age. The kittens were obtained from two litters with separate queens and studs. Solid symbols depict measurements of the binocular visual acuity, open symbols the acuity of the deprived eye, and half-shaded symbols the acuity of the nondeprived eye. Results for the three individual animals are shown by symbols of different shape and colors as follows: C282 orange squares, C283 green circles, and C292 blue triangles. The brackets to the right indicate the range of values for the monocular acuity of normal kittens of comparable ages and tested in a similar manner. Darkness promoted recovery of the acuity of the deprived eye of all three animals to match that of the fellow eye.

Figure 3

 

Grating acuity of the deprived eye of three male animals (C427, C291, and C284) immediately before and in the days that followed 10 days of darkness imposed at, respectively, 185, 186, and 188 days of age. The kittens were obtained from three separate litters that had different queens and studs. Solid symbols depict measurements of the binocular visual acuity, open symbols the acuity of the deprived eye, and half-shaded symbols the acuity of the nondeprived eye. The results for the three individual animals are shown by symbols of different shapes and colors as follows: C427 orange squares, C291 green circles, and C284 blue triangles. The brackets to the right indicate the range of values for the monocular acuity of normal kittens of comparable ages and tested in a similar manner. Whereas darkness promoted recovery of the acuity of the deprived eye of C427 and C291 to match that of the fellow eye, the acuity of the deprived eye of C284 recovered to a level just short of this level.

Figure 4

 

Grating acuity of the deprived eye of three animals (C428, C318, and C319; two female, one male) immediately before and in the days that followed 10 days of darkness imposed at, respectively, 190, 195, and 195 days of age. The animals were obtained from two litters with different queens and studs. Solid symbols depict measurements of the binocular visual acuity, open symbols the acuity of the deprived eye, and half-shaded symbols the acuity of the nondeprived eye. The results for the three individual animals are shown by symbols of different shapes and colors as follows: C428 orange squares, C318 green circles, and C319 blue triangles. The brackets to the right indicate the range of values for the monocular acuity of normal kittens of comparable ages and tested in a similar manner. Darkness promoted little (C428) or no (C318 and C319) recovery of the acuity of the deprived eye.

For the next group of animals (Figure 2), darkness was imposed 3–6 weeks later from P122 to P169 days. Recovery of the acuity of the deprived eye was slower than for the animals exposed to darkness a month earlier (Figure 1) revealing a progressive increase in recovery time from 14 to 20 days with age. Further delay in the age at which darkness was imposed to P185 (C427) or P186 (C291) did not appear to appreciably alter the eventual outcome as the acuity of the deprived eye recovered to match that of the fellow eye without further change in the speed with which it occurred (Figure 3). However, acuity of the deprived eye of the animal (C284) for which darkness was imposed latest at P188, although it did show substantial improvement, did not fully recover to match that of the fellow eye. Further delay in the age of imposition of darkness to P190 or P195 days of age (Figure 4) or later (Table 1) promoted little improvement in the acuity of the deprived eye beyond its value before the animal was placed in the darkroom. 

Two main trends are evident in the recovery data for the animals displayed in Figures 1 through 4. First, the acuity of the deprived eye began to improve soon after the period of darkness and continued at a rate that slowed only gradually with the age of the animal when darkness was imposed up until it failed to promote a significant benefit. And second, there was an upper limit to the age at which a 10-day period of darkness promoted recovery of the acuity of the deprived eye. In order to capture key features of the recovery promoted by darkness, the data were fit by three-part piecewise linear functions (see Methods). A representative fit to the data for one animal placed in darkness at day P188 is shown in Figure 5. The five key parameters of the fits to the data are indicated with arrows in Figure 5. The first break point (b₁) provides a measure of the time of the initial change in the acuity of the deprived eye following the period of darkness. A remarkable feature of the timing of this break point is that, in situations in which darkness was effective, it changed very little with the age of the animal when darkness was imposed: Recovery of the acuity of the deprived eye began at the same time after darkness. This point is emphasized further in Figure 6A in which the time of the first break point is provided for all the animals as a function of the age of imposition of darkness. The linear regression fit to the data indicated by the solid line (95% confidence intervals, dashed lines) reveals that recovery begins within ∼2 days irrespective of the animal's age. Although there was little change in the starting point of recovery with age, the subsequent rate of recovery of acuity gradually slowed. The latter point is emphasized by the data of Figure 6B in which the slope of recovery (s₂) of the linear fits to the data of all the animals had a strong negative correlation with the age at which darkness was imposed. 

Figure 5

 

Grating acuity of the deprived eye of C284 before and following 10 days of darkness imposed at P188 and fit with three-part piecewise linear function. As in earlier figures, the shaded area depicts the period spent in darkness. Empty and solid green circles indicate deprived eye acuity and binocular acuity, respectively. The solid line shows the three-part piecewise linear fit, and the five parameters from the fit are indicated with arrows (see Methods).

Figure 6

 

(A) The time after darkness (days) of the first break point (b₁; ordinate) of the piece-wise linear curves fitted to the recovery of the acuity of the deprived eye of MD kittens following 10 days of darkness imposed at different ages. The solid line shows a linear regression fit to the data (r² = 0.0006; p < 0.92), and dashed lines indicate 95% confidence intervals for this regression. Recovery, as reflected by this break point, occurred ∼2 days after animals were removed from the darkroom irrespective of the time when they entered. (B) A plot of the slope (s₂) of the recovery of the acuity of the deprived eye as a function of the age of imposition of darkness. The solid line shows a linear regression fit to the data (r² = 0.78; p < 0.00001), and dashed lines indicate 95% confidence intervals for this regression. Visual recovery was generally faster when darkness was imposed earlier.

The second point that emerged from the recovery data of Figures 1 through 4 was the existence of an upper limit to the age at which recovery occurred. This upper limit appears to be quite sharp as 10 days of darkness imposed on animals beyond 188 days of age did not promote any improvement in the acuity of the deprived eye. To illustrate this point, a recovery index was calculated for each animal to provide a measure of the extent of recovery. We calculated the recovery index (RI) to reflect the proportional difference between the final acuity of the deprived eye and the fellow eye (and/or binocular acuity) as 

RI = 1 − ([Binoc. Acuity – Dep. Eye acuity]/[Binoc. Acuity + Dep. Eye acuity]). 

The sharp drop from complete to incomplete recovery is highlighted in Figure 7, which displays the RI calculated for each animal as a function of the age at which darkness was imposed. A three-part piecewise linear fit to this data showed that the recuperative effect of darkness began to rapidly wane if applied later than day 188 (at a rate of ∼0.29 RI units per day) to become completely ineffective if applied later than day 191. 

Figure 7

 

The age at imposition of darkness (abscissa) is plotted against the RI (ordinate) as described in the Results section. Early dark exposure was associated with complete recovery and high RI scores, whereas later dark exposure was associated with incomplete recovery and lower RI scores. The solid line shows a three-part piecewise linear fit that modeled changes in RI over time rather than the deprived eye's grating acuity over time. In this case, s₂ gave an estimate of how rapidly 10 days of darkness went from promoting full recovery (high RI scores) to little or no recovery (low RI scores).

Discussion

Data from this study reveal the existence of an upper age limit for the ability of a 10-day period of darkness to promote improvement of the acuity of the deprived eye in kittens that each received an identical early period of MD. The longitudinal results from individual animals displayed in Figures 1 through 4 show that the fixed 10-day period of darkness promoted recovery of the deprived eye to match that of the fellow eye even when imposed as late as 186 days of age (C291, Figure 3). Whereas the acuity of the deprived eye of an animal (C284, Figure 3) placed in darkness 3 days later at 188 days recovered substantially but not quite to the level of the other eye, the acuity of the deprived eye of an animal (C428, Figure 4) so exposed at P190 recovered very little. A more complete picture of the decline in the effectiveness of a 10-day period of darkness with the age at which it was imposed is provided in terms of an RI calculated for every animal, including those from previous studies as well as five others that were placed in darkness beyond P195 days of age in Figure 7. The transition from full recovery of the acuity of the deprived eye to match that of the fellow eye (RI around one) to no recovery at all occurred in the period between P186 and P191 days. 

The limited size of the breeding colony dictated that animals from each litter were allocated to different rearing conditions (age of exposure to dark) as a first priority with gender as a second consideration. As mentioned in the Methods section, the litters were obtained from multiple queens and studs. Importantly, littermates within each litter were exposed to darkness at different ages. Within the constraints set by the limited number of animals, no gender differences were evident in the decline of efficacy of darkness with age. The general rearing conditions of all the kittens in terms of housing, exercise, and exposure to toys were similar and rich so that the data displayed in Figure 7 reflects the critical timing for darkness uncontaminated by either differences or variations in general rearing conditions as are known to effect cortical plasticity in rodents (Stryker & Löwel, 2018). On the other hand, litter differences of unknown origin were identified in a past determination of profile of the critical period for the effects on visual acuity of a 10-day period of darkness imposed on normal kittens (D. E. Mitchell, Crowder, Holman, Smithen, & Duffy, 2015). Fur and skin color may represent a potential source of difference between litters or littermates as it may introduce variation in light scatter through the closed eyelids and, hence, the efficacy of the constant 7-day period of MD that preceded exposure to darkness. 

As forecast by the results of a previous study (Holman et al., 2018) for which a 10-day period of darkness was found to be ineffective when imposed on cats at between 12 and 14 months of age, the data shown in Figure 7 provide strong evidence for the existence of a critical period for the benefits of darkness in cats. In addition to defining the profile of this critical period, the results indicate that its end point is very abrupt as it occurs over a period of about a week from P186 days (approximately 6 months of age). The evidence for a critical period for the efficacy of darkness as a means to reverse the behavioral consequences of MD in kittens stands in contrast to data from rodents in which darkness can reverse both the physiological and behavioral consequences of MD in adult rats (He et al., 2006; He et al., 2007) and mice (Stodieck et al., 2014). Other differences between the consequences of early selected visual deprivation across different species have been summarized recently (D. Mitchell & Sengpiel, 2018). 

Comparison with the timing of other critical periods in the cat

Although the visual recovery observed in this study was documented in terms of the changes in just a single spatial threshold (grating acuity) promoted in response to a fixed duration of darkness and employing minimal duplication of exposure conditions, these shortcomings are no different than the limitations of previous determinations of the timing of critical periods. Following the original demonstration of a critical period in the central visual pathways of cats (Hubel & Wiesel, 1970) that characterized the timing of vulnerability to the effects of a period of MD on the ocular dominance of neurons in the primary visual cortex, it has been determined that this critical period may be just one of many that define periods of environmental influences on cortical neurons. In particular, it is now recognized that critical periods of vulnerability to a particular experiential manipulation, such as MD, may differ from the timing of effectiveness of interventions, such as reverse occlusion (occlusion of the fellow eye), that can reverse the effects of a prior period of MD. Indeed, dissociation between the neural mechanisms mediating the impairments produced by MD and those promoting recovery from MD have been demonstrated (Cho, Knibnik, Philpot, & Bear, 2009; Kaneko, Stellwagen, Malenka, & Stryker, 2008; Ranson, Cheetham, Fox, & Sengpiel, 2012). Because darkness as used here represents a means to recover function lost by a prior period of MD, the most appropriate class of critical period for comparison are ones that define the period during which interventions can reverse the effects of a prior manipulation that induces a deleterious change or damage. The term “damage” or “disruption” has been used to define critical periods of vulnerability to experiential manipulations and “recovery” for the timing of interventions that attempt to remediate this damage (Daw, 2006; Lewis & Maurer, 2005). 

The most studied critical period in the central visual pathways of cats is the timing of vulnerability of cortical ocular dominance to a period of MD. The first detailed attempt (Hubel & Wiesel, 1970) to define this susceptible period employed periods of MD of different durations, thereby confounding the two variables of age and length of deprivation. Nevertheless, because of the use of some short period of MD on some of the youngest kittens, this study provided a good picture of the rising phase of vulnerability and identified the age of peak susceptibility at 4 weeks of age. Three subsequent electrophysiological studies defined the profile of this period by use of fixed periods of MD on animals of different ages, namely 10 days (Olson & Freeman, 1980), 30 days (Jones, Spear, & Tong, 1984), and 3 months (Daw, Fox, Sato, & Czepita, 1992). Good visualization of the rising phase of vulnerability to a peak at around 4 weeks and the slow subsequent decline was provided by the first of these studies; however, vulnerability to MD was still evident in kittens at the oldest age studied (16 weeks). Use of a longer period of MD revealed that the declining phase of vulnerability was even more protracted, and substantial vulnerability remained at 6 months. Finally, use of a 3-month period of MD revealed that vulnerability extended to between 10 and 12 months of age. By contrast, limited studies of the potential for recovery from the ocular dominance shifts induced by a prior period of MD by subsequent reverse occlusion revealed that the capacity for reversal of cortical ocular dominance declined with age at a much faster rate to vanish completely at about 14 weeks (Blakemore & Van Sluyters, 1974). However, the pace of the declining capacity for recovery may have been exaggerated by the use of a common 9-week period of reverse occlusion imposed on animals that had received progressively longer prior periods of MD that began in all animals at the time of natural eye opening. Although there have been no systematic studies of the critical period for the behavioral consequences of MD, the consequences for the acuity of the deprived eye appear to decline with age very gradually as does the capacity for recovery (Mitchell, 1988). 

In addition to the recovery from MD promoted by reverse occlusion, there have been many studies of the improvement that occurs upon simply restoring normal visual input to the deprived eye to allow simultaneous visual input to both eyes. Although the acuity of the deprived eye may, at first, occur faster than with reverse occlusion, the eventual acuity attained in the latter situation is superior (D. E. Mitchell, Gingras, & Kind, 2001). In addition, the rate of recovery of vision slows substantially with the age at which MD ends, so much so that it may take months for the deprived eye to recover measureable visual acuity (Mitchell, 1988). 

By contrast, the recovery of the acuity of the deprived eye following dark exposure differs in two very substantial ways from that which occurs following either reverse occlusion or binocular recovery. First, the darkness-induced recovery of visual acuity is both much faster and more complete, and second, there are no negative consequences for the visual acuity of the fellow eye, which remains at normal levels. 

Although the rapid decline with age of the efficacy of darkness to reverse the visual consequences of prior MD bears some resemblance to the decline in the ability of reverse occlusion to promote recovery from the cortical changes induced by MD, the former occurs later and ends much more abruptly. A clear dissociation between the two phenomena was the recent observation in cats (D. E. Mitchell et al., 2019) that the recovery of vision promoted by darkness depended on the fellow eye to the extent that, if this eye was occluded for the first 2 days following dark exposure, the acuity of the deprived eye did not improve at all. In other words, the same manipulation (occlusion of the fellow eye) that promotes recovery after a period of MD does not do so after darkness. Another major difference between reverse occlusion and dark exposure in terms of either physiological recovery in the visual cortex or acuity improvement for the vision of the deprived eye is that reverse occlusion can be a zero-sum game. The shifts of ocular dominance toward the formerly deprived eye promoted by reverse occlusion are accompanied by a shift of ocular dominance away from the fellow eye (Blakemore & Van Sluyters 1974; Movshon, 1976). And although reverse occlusion promotes recovery of the visual acuity of the deprived eye, it does so at the expense of the acuity of the other eye, and moreover, the improved vision of the deprived eye is not always maintained afterward once both eyes are open (D. E. Mitchell, Murphy, & Kaye, 1984; Murphy & Mitchell, 1987). 

Implications for the underlying mechanisms of the recovery promoted by darkness

As summarized in contemporary reviews (Hensch & Quinlan, 2018; Stryker & Löwel, 2018), short periods of darkness represent an environmental intervention that may influence the time course of development of key elements among multiple molecular pathways linked to plasticity in the central visual pathways. For example, early in postnatal life, darkness is known to influence the NMDA receptor subunit composition in the visual cortex (Fox, Daw, Sato, & Czepita,1991; Quinlan, Olstein, & Bear, 1999; Quinlan, Philpot, Huganir, & Bear, 1999) and as well alter the excitatory–inhibitory balance and, hence, the activation of critical period plasticity (Fagiolini & Hensch, 2000) by delay of expression of neurotrophins, such as BDNF (Huang et al., 1999). Later, dark exposure may slow the production of molecules that serve as structural brakes on plasticity, such as chondroitin sulphate proteoglycans and the formation of perineuronal nets (Hensch, 2005). In prior work on kittens (e.g. Duffy & Mitchell, 2013; D. E. Mitchell et al., 2016), it has been proposed that darkness may exert its benefits through action on various molecular pathways in a natural sequence that mirrors events in normal development. 

The decline in the efficacy of darkness with age revealed in Figure 7 has two key elements, a uniform beneficial outcome for the visual acuity of the deprived eye over a long period that lasts until P185, followed by a rapid decline that occurs over the course of just a week. These two components of the curve could be explained on the basis of the dual assumption that, first, functionally effective cortical neural connections with the deprived eye may be mediated via the action of multiple molecular pathways during development and, second, that rescue by darkness of any one of them at any age may support normal acuity. The rapid decline in the effects of darkness at around P185 may reflect the loss of the last of many molecular pathways that are influenced by darkness during development. It is possible that the overall superiority of dark exposure as a therapy may rest upon its ability to impact multiple molecular pathways in a sequence that matches events in normal development. 

Clinical implications

The profile of the efficacy of 10 days of dark exposure indicates that it ends at around 6 months of age in cats. It is possible that longer durations of MD that produce more damage may exhibit different recovery profiles compared to that observed in this study that employed a 7-day period of MD. Estimates of an age in humans equivalent to 6 months in a cat with respect to plasticity in the visual cortex cannot be made with great precision, but on the basis of general maturity and the age at which visual functions, such as acuity, matures in both species, a similar age for humans might lie in the range of 16 to 20 years. To ensure an effective age of intervention, whether it be implemented by dark exposure or binocular pharmacological retinal inactivation (Fong, Mitchell, Duffy, & Bear, 2016), it would probably be wise to initiate treatment in the early teen years. 

Acknowledgments

The research was supported by a Canadian Institutes of Health Research project grant (PJT-153333) to all three of us (K. Duffy PI) and by individual Natural Sciences and Engineering Research Council grants to DEM (2015-03819), NC (2015-06761) and KRD (2015-05320). We thank Mathew Smithen, Kaitlyn Holman, and Rebecca Borchert for care of the animals and occasional assistance with testing of certain cats. 

Commercial relationships: none. 

Corresponding author: Donald E. Mitchell. 

Email: . 

Address: Department of Psychology and Neuroscience, Dalhousie University, Halifax, NS Canada.