Morphometrics of corneal growth in chicks raised in constant light

Abstract

In this study we wish to augment our understanding of the effect of environment on corneal growth and morphology. To understand how corneal development of chicks raised in constant light differs from that of ‘normal’ eyes exposed to cyclic periods of light and dark, white Leghorn chicks were raised under either constant light (approximately 700 lux at cage top) or in 12 h light/12 h dark conditions for up to 12 weeks after hatching. To determine whether corneal expansion is uniform, some birds from each group received corneal tattoos for periodic photographic assessment. By 16 days of age, constant light corneas weighed less than light/dark regimen corneas [7.39 ± 0.35 mg (SE) vs. 8.47 mg ± 0.26 mg SE wet weight, P ≤ 0.05], and corresponding differences were seen in corneal dry weights. Spatial expansion of the corneal surface was uniform in both groups, but the rate of expansion was slower in constant light chicks [0.0327 ± 0.009 (SE) vs. 0.144 ± 0.018 (SE) mm² day⁻¹ for normal chicks, P ≤ 0.001]. At 1 day of age, there were 422 ± 12.5 (SE) stromal cells 0.01 mm⁻² in the central cornea and 393 ± 21.5 (SE) stromal cells 0.01 mm⁻²peripherally. Although this difference is not statistically significant, the cell densities in the central cornea were always larger than those of the peripheral cornea in all eight measurements over a 10.5-week period, and this difference is significant (P ≤ 0.008, binomial test). Light/dark regimen birds show no such consistent difference in cell densities between central and peripheral corneas. Thus, the density distribution of corneal stromal cells of chicks grown in constant light differs from that of normal chicks. Taken together, all these observations suggest that diurnal cycles of light and darkness are necessary for normal corneal growth.

Introduction

We have very little understanding of the direct effect of environment on corneal growth and development, although the environment may well influence healing processes following surgical, as well as accidental, incursions on corneal integrity. The common perception is that the lens and cornea display fixed patterns of development that are independent of non-visual environmental influence. This study documents the effect of constant light exposure on the growth and development of the chick cornea.

The regulation of eye growth and development of ametropia (persistent defocus of the eye) is frequently studied in the chick because its eyes are responsive to manipulations of its visual experience (Wallman & Turkel, 1978). Refractive errors (myopia, or nearsightedness, hyperopia, or farsightedness) have been induced in chick eyes using constant darkness (CD, Gottlieb et al. 1987), and constant light (CL, Lauber et al. 1970; Li et al. 1995). Raising chicks in CL produces corneal flattening and hyperopia within 3 weeks (Lauber et al. 1970; Lauber, 1987; Li et al. 1995). Flattening of the cornea is accompanied by vitreous chamber elongation in Cornell K strain chicks maturing in CL; however, the birds remain hyperopic (Li et al. 1995). Long-term CL produces shallow anterior chambers, corneal thickening, lenticular thinning, cataracts, and damage to the retina, pigment epithelium, and choroid (Li et al. 1995).

From these studies, it is clear that corneal shaping during growth is influenced by ambient light. However, the physiology underlying this phenomenon is not known. The corneal epithelium and adjacent conjunctivum undergo a circadian pattern of mitosis in the rat (Haskjold et al. 1989), quail (Sasaki et al. 1995) and chicken (Oishi, 1984). Mitotic rate in the corneal epithelium declines towards the center of the cornea in the rat (Haskjold et al. 1989) but is highest in the central cornea of the chicken (Oishi, 1984). The mitotic rate of lens epithelial cells declines towards the central zone of the rat (von Sallmann et al. 1962) and human (Kleiman & Worgul, 1994), although these central zone cells have been shown to respond to mitotic stimuli, including hormones (Kleiman & Worgul, 1994). The mitotic rate in the rat and human epithelium is highest between midnight and early morning, and lowest between noon and 18:00 hours.

Cyclic variations in the axial length of the chick eye occur throughout the day under normal light conditions (Weiss & Schaeffel, 1993; Nickla et al. 1998). The autonomic nervous system may regulate this cyclic change (Schmid et al. 1999). Although cyclic effects on eye dimensions and mitotic rates occur in both growing and adult animals, it is only in the growing eye that significant morphological changes result when these cyclic patterns are disrupted. It is not known whether the changes are due to overall differences in cell proliferation, matrix production, or a combination of both.

Matrix production has been shown to follow cyclic patterns in sclera (Nickla et al. 1999) as well as in other connective tissues (Simmons, 1992), thus it is reasonable to suppose that the corneal stroma is laid down in a similarly rhythmic fashion. The normal, rhythmic increase in numbers of mitotic corneal epithelial cells during darkness is eliminated from the overall mitotic rate in CL (von Sallmann et al. 1962). Over time, this means that fewer cells are born within the epithelium in conditions of CL.

In this study, we attempted to determine whether there are differences in the pattern of corneal growth between chicks raised in CL and those raised in normal light conditions (N, raised in 12 h light/12 h darkness). Specific observations comparing CL and N birds in this study include: (1) a comparison of eye weights and wet and dry corneal weights, (2) measurement of corneal thicknesses and corneal diameters, (3) spatial dynamics of corneal expansion, and (4) stromal cell densities at different ages, in peripheral and central regions of the cornea.

Materials and methods

Animal husbandry

In all, 135 White Leghorn chicks (Cornell-K strain) were used in this study. At 1 day of age, 12 birds were sacrificed for baseline measurements. The remainder were randomly divided into control N (n = 59) or CL (n = 64) groups. Except for illumination regimes, they were maintained under identical conditions. The ambient illumination level in the aviary averaged 700 lux during the light-on period. Illumination was supplied by fluorescent lamps (Sylvania 40 W, Cool White). Both groups were raised for the first 4–6 weeks in temperature-controlled brooders (30 °C). They were later maintained in large cages at room temperature (21 °C). Food (Agway), crop gravel, and water were provided ad libitum. All experiments were carried out under the supervision of the Cornell Institutional Animal Care and Use Committee, and adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research.

Morphology: eye weights, corneal wet and dry weights, and corneal diameters

Birds were sacrificed at 1, 3, 5, 8, 16, 28, 32, 42, 63, 74, and 84 days of age and measurements were made on the excised left eyes.

Some of the right eyes of these birds were used for histology (see below). Eye wet weight, corneal wet and dry weights, and corneal diameters were measured on the left eyes of all birds up to and including day 42. Due to the pressure of time, corneal diameters were not measured on day 63 and corneal dry weights were not measured after day 42. Corneas from the left eye were removed using iridectomy scissors and then weighed. The wet corneas were then placed in a 60 °C oven for 2 days and the dry weights were recorded. The number of CL and N birds used in each group is given in Table 1.

Table 1.

Numbers of chicks collected at indicated ages (see Fig. 3)

Spatial growth of the cornea

The corneas of four CL-treated birds and three N birds were tattooed by first anesthetizing the cornea with 0.5% proparacaine HCl and then introducing small quantities of undiluted India ink (Pelikan, black drawing ink A17) into the corneal stroma. The tattoo was created by lightly pricking the corneal surface with a fine dissection needle through an ink droplet (Fig. 1A). The quantities injected were just sufficient to make a visible spot. The spots did not spread noticeably over the course of the experiment. Those in Fig. 1A are typical of spots at the end of the experiment.

Fig. 1.

(A) Photo of tattooed cornea. Arrows indicate tattoos. Orientation: N = nasal, D = dorsal, V = ventral, and T = temporal. (B) Diagram of triangles described by the tattoos in Fig. 1A. Area of each spherical triangle was calculated and summed for an indication of overall corneal expansion with age.

To compute the actual areas of the corneas bounded by the tattoo marks, we first measured the distances between the five marks in prints of video frames of the corneas. The frames were taken with a 105-mm f/2.8 Micro Nikkor lens with a small depth of field. We calibrated these measurements by measuring the photographic images of a ruler. We then measured the offset of the central tattoo mark from the center of the cornea, and entered this data, together with the radii of the corneas at the appropriate ages, into a basic computer program written by one of the authors (H.C.H.). This program computed the location of the rest of the tattoo marks on the photograph using the lengths of the lines connecting them (Fig. 1B), solving the triangles with plane trigonometry. Then, using the known constants of the video camera set-up and the corneal radii, we computed the locations of the tattoo marks on the cornea by tracing rays from the image plane of the video camera through the center of the camera lens to the spherical surface of the cornea. We then solved the spherical triangles using equations from Vilhelm (1969) to compute the triangular (segmental) and total areas subsumed by the tattoos. The program was tested by entering artificial input data for which the corresponding spherical areas were known.

Corneal curvature measurements

Corneal curvature measurements were made by taking video images of reflections from an array of eight infra-red light emitting diodes arranged in a 30-cm circle around a video camera at a distance of 137 mm from the animal. Measurements were made in four orthogonal meridians and averaged. The distance between opposed LEDs is inversely proportional to the dioptric power of the cornea, and the apparatus was calibrated using ball bearings of known diameter. This technique is more fully described in Glasser et al. (1994).

Morphology: corneal cell numbers

Some right eyes (n= 39) from birds of various ages and both treatments were collected for histology. They were sliced open just posterior to the ora serrata to ensure rapid fixative penetration, and immersed in 4% phosphate-buffered paraformaldehyde, pH 7.4. The eyes were prepared for paraffin embedding and serial sectioning at 6 µm, followed by conventional hematoxylin and eosin staining. The tissue was oriented during dissection with a transverse cut across the dorsal margin of the iris. The position of this cut, the position of the optic nerve stump, and left/right eye information were used as landmarks to obtain naso-temporal (horizontal) sections. Alternate sections from each of a minimum of 150 serial sections located between iris margins were analyzed for corneal cell densities, including two to three birds per age and treatment group. These counts were divided into ‘peripheral corneal’ and ‘central corneal’ locations. The peripheral cornea was defined as that part of the cornea situated over the iris, and the central cornea was defined as that part of the cornea situated over the pupil. Results were analyzed using statview statistical analysis software (SAS Institute, Cary, NC, USA).

Results

Rates of eye growth

As the ratio of eye weight/body weight did not significantly differ between N and CL chicks, we simply measured eye weight vs. age. By day 42, CL eyes were significantly heavier than N eyes (Fig. 2A; P < 0.001).

Fig. 2.

Measurements of the left eyes of 135 chicks raised in normal and constant light conditions. Data points on each day of all the graphs refer to the same set of chicks sacrificed on that day whose numbers are given in Table 1. Error bars indicate ± 1 SEM (in many cases they are smaller than the plotted points). For practical reasons it was impossible to make all measurements on each group. (A) Mean wet weight of the left eye of CL and N chicks with growth. (B) Mean wet weights of the left corneas of CL and N chicks with growth. (C) Mean dry weights of the left corneas of CL and N chicks with growth (D) Corneal diameters of left eyes with growth in CL and N birds.

Corneal wet/dry weights

N corneas were heavier than CL corneas by 16 days of age and thereafter [mean values: 16 days, wet weights; N = 8.47 + 0.30 mg (SE); CL = 7.40 ± 0.35 mg (SE); P = 0.03] (Fig. 2B). Dry N corneas were also heavier than dry CL corneas, and this difference was consistently significant at 16 days and thereafter [N = 1.81 ± 0.066 mg (SE), CL = 1.54 ± 0.065 mg (SE); P = 0.015] (Fig. 2C).

Corneal diameters

Corneal diameters for birds aged 1–84 days are plotted in Fig. 2D. CL corneal diameters are significantly smaller (P = 0.019) than N corneas beginning at 16 days of age. By 84 days of age, CL corneal diameters average 7.23 ± 0.105 mm (SE), and N corneal diameters 8.385 ± 0.29 mm (SE), P = 0.023.

Corneal curvatures

As expected, the chicks raised in constant light had flatter corneas, as seen from the side, than those raised under normal conditions (Fig. 3; see also Li et al. 1995). The regression statistics for corneal curvatures (which determine the optical power of the cornea) were: Normal condition, corneal radius (mm) = 2.65 ± 0.308 (± 0.022 SD) * Age (days) (r² = 0.90, P < 0.0001); CL conditions, corneal radius (mm) = 2.42 ± 0.606 (± 0.043 SD) * Age (days) (r² = 0.93, P < 0.0001). It will be noted from the standard deviations of the regression slopes that they are significantly different from each other.

Fig. 3.

Corneal flattening is grossly perceptible in living profile, as seen when comparing frontal photographs of 12-week-old constant light (A) and normal (B) chicks.

Spatial growth of the cornea

We examined corneal growth by computing regression slopes both for the mean total areas as a function of age for the two treatments, and for the growth of individual regions of the cornea for each tattooed bird.

The mean regressions (Fig. 4) are as follows: for normal birds raised in 12/12: mean corneal area (mm²) = 6.70 + 0.144 (± 0.018 SE) * age (days) (r² = 0.928, P < 0.0005); for birds raised in constant light: mean corneal area (mm²) = 6.27 + 0.0327 (± 0.009 SE) * age (days) (r² = 0.719, P < 0.016). It will be noted that the regression slope for CL birds is very small, indicating that the growth of the CL corneas in area was negligible.

Fig. 4.

Total mean corneal areas as a function of age for 3 normal (N) and 4 constant light (CL) raised chicks. For days 42 and 49 we had data for only two of the three normal chicks. The two regression slopes are significantly different. See text for further explanation.

We examined the slopes of corneal segmental areas vs. age for all of the tattooed birds. Three of the four CL birds and one of the N birds did not have segmental slopes of area vs. age that were significantly different from zero. The remaining three birds all had segmental growth slopes which, within the same bird, did not differ significantly from each other. To test for effects of differing initial segmental size, we also tested the slopes of log segmental area vs. age for the same birds and again found no significant difference between the slopes. We conclude that none of the corneas that we examined appeared to grow at different rates in different quadrants.

Stromal cell densities

Average central and peripheral stromal cell densities for CL and N corneas counted over all sectioned eyes (a total of 39 birds) are plotted vs. age in Fig. 5A,B. Peripheral cornea cell densities in both treatment groups decrease with age, from 393 ± 21.5 cells 0.01 mm⁻² at 1 day of age, to 112.5 ± 27.5 0.01 mm⁻² (CL periphery) and 90.7 ± 12.4 0.01 mm⁻² (N periphery) at 74 days of age (Fig. 5A). In CL (but not in N) birds, central cell densities are uniformly greater than peripheral cell densities at all ages (two-tailed binomial test; P < 0.016).

Fig. 5.

(A) Peripheral stromal cell densities in 18 CL, 18 N, and 3 untreated birds (day 1). (B) Central corneal stromal cell densities for the same birds. See text for detail of methods.

At 28 days and thereafter, CL birds maintain higher central stromal cell densities relative to N birds (Fig. 5B); however, the numbers of birds measured are insufficient to demonstrate statistical significance. Similar but smaller differences between CL and N birds were observed in the peripheral stroma.

Discussion

It is known that the corneas of chicks raised in constant light are significantly thicker than those raised under normal conditions (Li et al. 1995). In addition, our results show that constant light has an early and persistent effect on diameter, weight, and central stromal cell density of the cornea, as summarized by the sketch in Fig. 6. CL produces a lighter, narrower, flatter, and thicker cornea with higher central stromal cell densities (Fig. 6A, normal and B, constant light). Although CL produces lighter corneas than normal, the eyes of CL birds are heavier (Fig. 2A). All of the various influences of constant light rearing appear to be dysfunctional from the standpoint of emmetropization of the eye.

Fig. 6.

Representation of the results of corneal size and cell density in corneas of chicks raised under normal (12L/12D) conditions (above) and constant light conditions (below). Note that the CL corneas are of smaller diameter and thicker with higher cell densities in the central region of the cornea.

Corneal curvatures

We measured corneal curvatures primarily to ensure that the chicks showed their normal response to constant light conditions and to enable the computation of corneal areas from pictures of the corneal tattoos. As expected, chicks raised in constant light have flatter corneas than those raised under 12 L/12/D illumination. This, of course leads to relative hyperopia of the CL chicks (Li et al. 1995).

Morphometrics of corneal growth in constant light

This study demonstrates that the effect of CL on the expansion of the cornea, which becomes stunted, is opposite to its effect on the expansion of the globe of the CL eye, which is promoted (Lauber et al. 1970; Li et al. 1995). The globe of the CL eye, but not the cornea, also displays an exaggerated asymmetry (Li et al. 1995). Thus, the light-sensitive mechanisms regulating growth of the eye do not appear to act uniformly on the ocular tissues.

Light most probably exerts an indirect effect on corneal growth through humoral regulation (Wahl et al. 2004). A candidate for indirect regulation of growth by light is the cyclically secreted hormone melatonin, which diffuses through the tears and perhaps also the aqueous humor (Li & Howland, 2002), acting directly on the cornea via corneal melatonin receptors in the epithelium and stroma (Rada & Wiechmann, 2006). Humorally mediated regulation of eye shape may occur via modulation of intraocular pressure (IOP) and/or modulation of vascular tone at the iris angle, thereby creating (and altering) mechanical forces to which the stromal cells respond as the cornea grows. This is suggested by data showing that the average IOP of constant light chicks does not fluctuate in a daily rhythm (as in normal chicks; Papastergiou et al. 1998) and is lower than that of normal chicks (Li et al. 2002). Evidence suggests that more than one form of regulation may act during growth. In CL, ciliary ganglionectomized chicks exhibit an exaggerated flattening of the cornea (Li & Howland, 2000), whereas superior cervical ganglionectomy has produced shallow anterior chambers and hyperopia, but normal corneal curvature, in chicks raised under 12 h light/12 h dark conditions (Lauber et al. 1972). Chicks raised in CL but ‘protected’ by 12 h daily covering of the pineal gland show no corneal flattening (Li et al. 2001). These results suggest that both humoral and neural regulation of corneal growth is present.

The long-term morphological and optical consequences of early exposure to CL remain to be determined. We have evidence that the optical properties of the eye may recover to a large extent if the bird is returned to normal light conditions within 3 weeks (Li et al. 2002); however, the eye cannot recover after 2-month exposure to CL (unpublished observation).

This work has shown that the eye's ability to model its shape towards emmetropia is diminished in the absence of periods of light and dark. Particularly sensitive are the stromal cells of the cornea, which show significant changes in density and distribution in CL. CL conditions do not occur in nature where chickens evolved, and this fact may account for the chick's inability to regulate some aspects of eye growth in CL.