Microtubule Contribution to the Reflectance of the Retinal Nerve Fiber Layer

Abstract

purpose. The reflectance of the retinal nerve fiber layer (RNFL) arises from light scattering by cylindrical structures oriented parallel to ganglion cell axons. In amphibian retinas, at 440 nm, microtubules (MTs) contribute about one half of RNFL reflectance. In rodent retinas, MTs are the only structure contributing to RNFL birefringence. To increase understanding of the anatomic basis for clinical RNFL measurements, this study was conducted to evaluate the MT contribution to RNFL reflectance in rodent retinas by using the MT depolymerizing agent colchicine.

methods. Reflectance of nerve fiber bundles in isolated rat retinas was measured at 460, 580, and 830 nm with a multispectral imaging reflectometer. Images were taken frequently over an extended period. During baseline, the tissue was perfused with a physiological solution. During a treatment period, the solution was switched either to a control solution or to a solution containing colchicine.

results. Because of the high reflectance of the RNFL, nerve fiber bundles appeared as bright stripes against a darker retina. The reflectance of bundles was relatively stable in control experiments. With colchicine treatment, however, bundle reflectance at first decreased rapidly and then became stable. After 70 minutes of colchicine treatment, RNFL reflectance had declined to approximately 50% below baseline at all wavelengths.

conclusions. MTs contribute to RNFL reflectance at all wavelengths. Unlike RNFL birefringence, however, which totally disappears after colchicine treatment, about one half of RNFL reflectance remained after colchicine treatment. This result suggests that, in addition to MTs, other mechanisms may contribute to RNFL reflectance.

The retinal nerve fiber layer (RNFL) in humans consists of bundles of unmyelinated ganglion cell axons running just under the surface of the retina. ¹ It is damaged in glaucoma and other diseases of the optic nerve. Direct assessment of RNFL structure is attractive for early diagnosis and clinical management of disease because damage often precedes detectable vision loss. ^{2 3 4 5 6 7} Various optical methods, such as optical coherence tomography (OCT) and scanning laser polarimetry (SLP), have been developed for RNFL assessment. ^{8 9 10 11} Identification of the mechanisms responsible for the optical properties of RNFL should enhance the interpretation of clinical measurements. 

RNFL reflectance arises from light scattering by cylindrical structures oriented parallel to ganglion cell axons. ^{12 13 14} Oriented cylinders also cause the RNFL to exhibit birefringence. ^{10 13 15 16 17 18} Electron microscopy reveals the cylindrical structures within nerve fiber bundles, which include microtubules (MTs), neurofilaments and axonal membranes. ^{13 19} Studies to date have focused on the role of MTs in determining the optical properties of the RNFL. 

Axonal MTs are long (10–25 μm), hollow cylinders with outer and inner diameters of approximately 25 and 15 nm, respectively. ²⁰ MTs are polymers of tubulin, protein heterodimers that join end-to-end to maintain the structure of MTs in a state of dynamic equilibrium. The antimitotic agent colchicine irreversibly binds to tubulin and leads to depolymerization of MTs. ^{21 22 23} Colchicine binding is specific to tubulin, ²² and, other than a decrease in the number of MTs in axonal cross sections, colchicine does not appear to change the structural organization of axonal organelles. ²⁴ Suspensions of MTs in vitro become less turbid (scatter less light) after colchicine treatment converts the large MTs into smaller proteins. ^{25 26} We assume, therefore, that if MTs contribute to an optical property of the RNFL, colchicine treatment should decrease the magnitude of that property. 

MTs have a significant role in RNFL reflectance and birefringence. ^{16 18 27} Colchicine treatment of amphibian retinas showed that MTs contribute about half of the RNFL reflectance at 440 nm. ²⁷ In contrast, colchicine treatment of rat retinas showed that depolymerization of MTs causes near total disappearance of RNFL birefringence at wavelengths ranging from 440 to 780 nm, suggesting that MTs may be the only mechanism underlying RNFL birefringence. ¹⁸ The different behaviors of reflectance and birefringence may result from species differences or differences in the underlying mechanisms. Evidence for mechanistic differences comes from spectral reflectance measurements of rat RNFL ¹⁴ ; theoretical modeling of the reflectance spectra suggests both thin and thick cylinder mechanisms, with thin cylinders (perhaps MTs) dominating the reflectance at short wavelengths and thick cylinders dominating at longer wavelengths. ¹⁴  

To understand better the contribution of MTs to RNFL reflectance, we measured reflectance at multiple wavelengths in isolated rat retina before and after treatment with colchicine. To enhance comparisons between reflectance and birefringence, we followed the same treatment procedures as have been used in birefringence measurements. ¹⁸  

Materials and Methods

Experimental Preparation

Rat retinas were used in the experiments because, as in humans, the axons in the rat RNFL are unmyelinated, and the effect of colchicine on the birefringence of rat retinas has been well investigated. ¹⁸ The protocol for the use of animals was approved by the Animal Care and Use Committee of the University of Miami, and procedures adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. Tissue preparation followed previously developed procedures. ²⁸ Briefly, the eye of an anesthetized rat was removed and the animal was euthanatized. An eye cup of 5-mm diameter that included the optic nerve was excised with a razor blade and placed in a dish of warm (33–35°C), oxygenated, physiologic solution. Excess vitreous was drained. The retina was dissected from the retinal pigment epithelium and choroid with a fine glass probe and then draped across a slit in a black membrane with the photoreceptor side against the membrane. A second, thinner membrane with a slit matched to the black membrane was put on the RNFL surface to gently stretch the retina and eliminate wrinkles. This procedure was performed with intense white illumination, which thoroughly bleached the visual pigment in the photoreceptors and ensured that the reflectance in this layer remained constant. The mounted retina was placed in a chamber perfused with warm physiologic solution, to maintain the tissue alive. 

An experiment consisted of a baseline period during which the chamber was perfused with a solution containing no colchicine, followed by a treatment period during which the solution was switched either to a solution containing colchicine or a control solution identical with the baseline solution. For the first 10 minutes after switching, the fluid leaving the chamber was shunted out of the circulation system to avoid mixing of the two solutions. The baseline and treatment periods lasted approximately 40 and 70 minutes, respectively. The perfusion fluids were based on a solution containing 110 mM NaCl, 5.0 mM KCl, 30 mM NaHCO₃, 0.8 mM Na₂HPO₄, 1.0 mM MgCl₂, 1.8 mM CaCl₂, 22 mM glucose, and 0.25 mM glutamine. The solutions were bubbled with a humidified gas mixture of 95% O₂ and 5% CO₂ (pH 7.55–7.65). Because electron microscopy has shown that 10 mM colchicine significantly decreases the density of axonal MTs, ^{24 27} this concentration was added to the base solution to form the colchicine solution. To avoid retinal turbidity caused by osmotic change when switching to the colchicine solution, an additional 10 mM glucose was added to the base solution to form the baseline and control solutions. 

Multispectral Imaging Microreflectometer

A multispectral imaging microreflectometer was used to detect the reflectance of the RNFL (Fig. 1) . The device has been described in detail previously. ²⁹ Briefly, light from a tungsten-halogen lamp followed by an interference filter (10 nm full width at half maximum) and lenses L₁ and L₁′ provided monochromatic illumination to a retina. A diaphragm controlled the illumination area. The mounted retina was placed in the chamber at the center of a spherical window. It was imaged by lens L₂ onto a cooled charge-coupled device (CCD) (AT200; Photometrics, Inc., Huntington Beach, CA) that provided a pixel size of approximately 3 μm in an aqueous medium and a full field of view of 1.5 × 1.5 mm on the retina. The chamber was mounted on a precision stage that could be adjusted for translation and rotation. The optical axes of light source and camera both approximately intersected with the center of the spherical window. The light source probe could be moved in both azimuth and elevation. The camera could be moved in azimuth with the elevation fixed at 13° below the equator. 

RNFL reflectance is very directional; the reflection is confined to a conical sheet centered on the axes of the cylindrical structures of the RNFL. ^{12 14} In experiments, bundles were oriented approximately vertical and the azimuths of the camera and light source were adjusted to positions such that there was no specular reflection from the retinal surface and the background reflection was dark and uniform. The elevation of the light source was then moved to a position that gave near maximum reflectance of the nerve fiber bundles. 

Assessment of the RNFL Reflectance

Images of the RNFL were collected by the CCD camera approximately every 2 minutes at wavelengths of 460, 580, and 830 nm. Black images taken with the same exposure duration, but with the light source off, were subtracted from each image to compensate for the dark current and bias level of the CCD. The resultant pixel values were directly proportional to the reflected intensity. Pixel values were converted to relative diffuse reflectance using images of a diffuse white reflector (6080 White Reflectance Coating; Eastman Kodak Company, Rochester, NY). For comparison of measurements of the same bundles over time, the entire set of images was registered by horizontal and vertical translation. 

To measure the reflectance of nerve fiber bundles, rectangular areas were chosen both on bundles and on nearby gaps between bundles and the reflectances of the pixels in each area were averaged. (Examples of bundle and gap areas are shown in Figs. 2A and 3 .) Area selection was necessarily subjective, but was made on pretreatment images without reference to posttreatment images. Areas for any one bundle were sufficiently numerous that exact area placement did not materially affect results. Care was taken to avoid blood vessels, which were recognized by any of three criteria: first, although the brightness of a blood vessel varies with the direction of the incident beam, the vessel does not disappear for angles far from peak reflectance, as do nerve fiber bundles; second, blood vessels often contain residual red blood cells, which are visible as dark spots at short wavelengths; and third, blood vessels show a characteristic branching into smaller vessels that often cross nerve fiber bundles. Only bundles that were oriented near vertical and had approximately uniform surrounding gaps were chosen for analysis. Reflectance measured on bundle areas included light reflected from the RNFL and from underlying tissue. Because the weak scattering of the RNFL caused little attenuation to an incident beam, we assumed that the reflectance from deep layers was approximately the same as that from nearby gap areas. The average reflectance of gap areas, therefore, was subtracted from the total reflectance measured on the bundle areas to get an estimate of the bundle reflectance alone. The reflectances of several areas on the same bundle were then averaged. 

The 10 mM colchicine solution appears yellowish due to absorption at short wavelengths. The effect of this absorption on the bundle reflectance measured after switching to colchicine solution was corrected as follows. A piece of Teflon sheet (duPont, Wilmington, DE) was mounted as a specimen in the chamber with its surface at the spherical window center to achieve a path length in the solution approximately the same as in retinal measurements. The relative reflectance of the Teflon sheet was measured with the chamber filled with the control and treatment solutions, respectively. The ratio of control and treatment reflectances provided transmission coefficients at each measured wavelength that were applied to bundle reflectance. The measured coefficients were 0.858, 0.968, and 0.991 at 460, 580, and 830 nm, respectively. 

To quantify reflectance change over time, the percentage change of reflectance was calculated from the average reflectance in the 20-minute period before the switch to the control or colchicine solution and the 20-minute period before the end of the experiments. Data analysis was implemented in a commercial software (MatLab; The MathWorks, Inc. Natick, MA). 

Results

In the absence of colchicine, the RNFL reflectance was stable, as demonstrated by control experiments in which the perfusion solution was switched to an identical baseline solution (Fig. 2) . The image in Figure 2Awas taken approximately 1.6 hours after the isolated retina was placed in the measurement chamber. Because in the direction of maximum reflectance the RNFL reflects much more light than the retina, the nerve fiber bundles appear as bright stripes against a darker background. The time course of the reflectance of one bundle shows little change at the three measured wavelengths (Fig. 2B) . To display temporal change in multiple bundles, the reflectance was normalized by the average of the values in the 20 minutes just before the solution was switched. Figure 2Cdisplays the average of eight normalized curves at 580 nm (four bundles in each of two control retinas). The normalized time courses at 460 and 830 nm (not shown) were similar. The average percentage decrease of reflectance was 2.80% ± 10.1%, 8.2% ± 8.5%, and 8.0% ± 11.0% (mean ± SD) at 460, 580, and 830 nm, respectively. 

In contrast with control experiments, bundle reflectance decreased immediately after a switch to colchicine solution, rapidly at first, then slowly and finally leveling off at a lower value after about 1 hour. Change of reflectance over time was similar at different wavelengths. Figures 3 and 4show an example of these experiments. Nerve fiber bundles that were bright during the baseline measurements (Fig. 3 , left column) faded in intensity but were still visible at all wavelengths after 70 minutes in colchicine solution (Fig. 3 , right column). Although colchicine treatment caused a small increase in background reflectance, this was removed from the calculated bundle reflectance as described in the Methods section. Figure 4shows the temporal change of normalized reflectance for the bundle marked in Figure 3 . Data collected immediately after the switch were omitted because of incomplete exchange of the baseline and colchicine solutions. The reflectance was high before the solution switch, decreased rapidly in the first 40 to 50 minutes after the switch, and then remained at a lower value for the rest of the measurements. 

The effect of the colchicine treatment on the reflectance of 17 nerve fiber bundles in five retinas was studied. The initial reflectance of individual bundles was different across and among retinas but, in all bundles, reflectance dropped after the switch to colchicine solution and was stable at a lower value after 1 hour in colchicine solution. To combine the results from the 17 bundles, the data were normalized and averaged as described for the control experiments displayed in Figure 2C . Figure 4Bshows the result at 580 nm; the average percentage decrease of reflectance in all 17 bundles was 51% ± 12%. The averaged time courses at 460 and 830 nm were similar, with average percentage decreases of 46% ± 12% and 54% ± 16%, respectively. 

Discussion

The axons of retinal ganglion cells possess oriented cylindrical structures, including MTs, neurofilaments, and axonal membranes, ¹⁹ which are plausible sources for the optical properties of the RNFL. In a recent study, the investigators found that MTs may be the only structure responsible for RNFL birefringence. ¹⁸ The present study followed the same experimental procedure to evaluate the role of MTs in RNFL reflectance. Control experiments demonstrated that RNFL reflectance was relatively stable, declining, on average, approximately 3% to 8%. In contrast, colchicine treatment caused RNFL reflectance to decrease rapidly before stabilizing at lower values. Nerve fiber bundles faded but were still visible after at least 65 minutes of treatment. The temporal change in reflectance was similar at three wavelengths ranging from 460 to 830 nm, with a decrease of bundle reflectance of approximately 50% at all wavelengths. Although colchicine treatment increased the reflectance of retinal tissue, a background subtraction algorithm removed the effect of underlying tissue, and the observed decline in bundle reflectance is assumed to be specific to ganglion cell axons, where colchicine is expected to reduce the density of axonal MTs. ^{24 27} These results imply that MTs contribute significantly to RNFL reflectance. 

The time courses of the colchicine-induced decline of RNFL reflectance and birefringence were grossly similar, ¹⁸ which may indicate that changes in these properties are correlated and associated with the rate of MT depolymerization. A significant difference between RNFL reflectance and birefringence, however, is that colchicine causes birefringence to approach 0, ¹⁸ whereas colchicine did not cause total disappearance of reflectance. Instead, RNFL reflectance approached approximately 50% of its baseline value. This finding suggests that, although MTs may be the only structure contributing to the RNFL birefringence, light scattering from MTs is not the only mechanism responsible for the RNFL reflectance. 

Although its mechanism is unclear, some comments can be made about the RNFL reflectance that remained after colchicine treatment. The directional reflectance of the RNFL was not systematically measured in this study, but both before and after treatment, all nerve fiber bundles consistently displayed the highly directional reflectance characteristic of light scattering by cylinders. ^{12 14 30 31} Cylinders other than MTs are candidates for the remaining mechanism. Axonal membranes, for example, when modeled as thin lipid sheets arranged into parallel arrays, are calculated to scatter more light than actually measured. ¹³ Thus, axonal membranes, perhaps in association with adjacent proteins to reduce scattering strength, provide a candidate reflecting structure that would be resistant to colchicine treatment. Other candidates include neurofilaments and mitochondria within axons and glial processes between axons. Another possibility is that not all MTs disappear with colchicine treatment, ²⁴ but then one must seek an arrangement of the remaining MTs that does not exhibit birefringence. 

An important finding of this study is that the decline and final plateau of reflectance after colchicine treatment was similar across a wide wavelength range. The average decreases across 17 bundles were 46%, 51%, and 54% at 460, 580, and 830 nm, respectively. In an earlier study, we proposed a two-mechanism model to describe the spectral reflectance of the RNFL ¹⁴ : a model that consisted of a thin cylinder mechanism dominating at short wavelengths and a thick cylinder mechanism dominating at long wavelengths. On the assumptions that the thin cylinder mechanism is light-scattering by MTs and that MTs are completely eliminated by colchicine, Figure 7 of Knighton and Huang ¹⁴ predicts declines of 68%, 58%, and 26% at the same three wavelengths. Clearly the present data do not support this simple scheme. Further characterization of the remaining reflectance, including detailed measurements of its spectrum, ¹⁴ may elucidate the mechanisms involved. 

The decrease in RNFL reflectance caused by colchicine treatment is about the same in rat and toad retinas, ²⁷ which may indicate a similar role for MTs among species. Both toad and rat RNFLs are unmyelinated, as is the human RNFL, and substantial qualitative differences in optical properties are not expected among species. Thus, MTs are probably also a major component of the RNFL reflectance in human retinas. 

Optical measurements can provide direct clinical assessment of the RNFL, and various technologies may reveal different aspects of RNFL structure. In cross-sectional images produced by OCT, which display retinal reflectivity versus depth, the RNFL appears as a reflective layer, the thickness of which can be measured directly. ³² This study suggests that only a portion of the RNFL reflectivity arises from MTs (∼50% if the reflecting structures are similar in human and rat axons). Maps of the RNFL produced by SLP show the distribution of retardance, ¹¹ a polarization property that is the product of tissue birefringence and thickness. ^{17 33} Because birefringence varies with position around the optic nerve head, ^{17 33} SLP retardance profiles and OCT thickness profiles are not the same. ¹⁷ Colchicine treatment causes RNFL birefringence to disappear, ¹⁸ suggesting that MTs are the only structure detected by SLP. When axons die, MTs are expected to disappear and the RNFL to thin, signifying irreversible damage. In early glaucoma, however, it is of great interest to know whether MT loss precedes other ultrastructural or functional change, perhaps indicating a therapeutic window within which damaged axons can be rescued. A comparison of retardance from SLP and thickness from OCT (i.e., a measurement of RNFL birefringence) ^{17 33} may provide such knowledge. 

 Supported by National Eye Institute R01 EY008684, National Eye Institute Core Grant P30-EY014801, and Research to Prevent Blindness.

 Submitted for publication April 20, 2006; revised August 28, 2006; accepted October 10, 2006.

 Disclosure: X.-R. Huang, None; R.W. Knighton, None; L.N. Cavuoto, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

Schematic diagram of the multispectral imaging microreflectometer used to measure the reflectance of retinal nerve fiber bundles. LS, light source; IF, interference filter; OF, optic fiber; D, diaphragm; L1, L1′, and L2, lens; SP, specimen; CB, specimen chamber; W, spherical window; CCD, charge-coupled device. The optical axes of light source and camera intersected at the center of the spherical window. The light source probe could be moved in both azimuth and elevation. The camera could be moved in azimuth with the elevation fixed.

Figure 2.

 

Control experiments demonstrated the stability of bundle reflectance. The switch to control solution occurred at time 0. (A) After 1.6 hours in baseline and control solutions, nerve fiber bundles still appeared as bright stripes (arrows) against the darker retinal background. Black boxes: bundle areas; white boxes: gap areas. Image size: 300 μm wide × 180 μm high. (B) Time courses of the relative reflectance of the average bundle areas (black boxes in A). (C) The mean of the normalized reflectance of eight bundles in two retinas at 580 nm (solid line) and ±1 SD (dashed lines). The normalization level for each bundle was its average relative reflectance over the 20 minutes before the solution switch.

Figure 3.

 

Colchicine caused the RNFL reflectance to decline. Left column: images of RNFL measured before the solution switch, after approximately 40 minutes in the baseline solution. Nerve fiber bundles (black arrows) appeared as bright stripes against a dark background. The bundle and gap areas for the bundles marked with an open arrow provided the data for Figure 4A . Right column: images of the same retinal region after 70 minutes of colchicine treatment. Bundles faded but were still visible at all wavelengths. Images at each wavelength are displayed with the same intensity scale in the two columns. Image size: 480 μm wide × 240 μm high. White arrow: blood vessel.

Figure 4.

 

Time course of RNFL reflectance with colchicine treatment. (A) Normalized reflectance of the bundle areas marked in Figure 3 . (B) The mean normalized reflectance of 17 bundles in five retinas at 580 nm (solid line) and one SD above and below the mean (dashed lines). The means of the reflectances highlighted by the heavy lines before and after colchicine treatment were used to calculate the percentage of reflectance decrease.