Idiopathic Bilateral Optic Atrophy in the Rhesus Macaque

Abstract

purpose. To document the existence of idiopathic bilateral optic atrophy (BOA) in rhesus macaque monkeys and to characterize the structural and functional consequences of this condition.

methods. In vivo assessment of retinal and optic nerve structure included fundus biomicroscopy and stereophotography. Functional analyses included transient pattern-reversal electroretinography (PERG) and full-field flash ERG, with both white flashes while dark adapted and red flashes on a blue background used to assess the photopic negative response (PhNR). Also measured were visual evoked cortical potentials (VEPs) and multifocal (mf)ERGs, with both a standard fast and slowed (7F) stimulation sequence. Post mortem histologic evaluation was performed on a subset of five animals with BOA and compared with data from 22 healthy normal animals. Blood tests, including vitamin E, B₁₂, folate, lead, and complete blood cell count with differential were obtained on the four animals that remained alive.

results. Animals with BOA showed temporal pallor of the optic nerve head and thinning of the retinal nerve fiber layer (RNFL) between the temporal vascular arcades (i.e., of the papillomacular bundle). Severity of optic atrophy and RNFL loss varied between animals from mild to severe, but was similar in the two eyes of each animal. Functional changes included greater loss of the PERG N95, compared with the P50 component and substantial reduction of mfERG high-frequency components. The mfERG low-frequency components were slightly larger than normal. None of the full-field flash ERG amplitudes (a-wave, b-wave, oscillatory potentials, or PhNR) was significantly different from normal. There were no consistent abnormalities found in the results of any blood test. Histologic findings included axonal loss and gliosis limited to the temporal optic nerve, reduction of nuclei within the retinal ganglion cell layer, and thinning of the temporal retinal RNFL.

conclusions. The existence of BOA in nonhuman primates warrants caution on the part of investigators who use these animals in experimental models of ophthalmic disease.

The purpose of this article is to report the existence of idiopathic bilateral optic atrophy (BOA) in rhesus macaque monkeys (Macaca mulatta) and to characterize the structural and functional consequences of the condition. Initially, this form of optic atrophy was an incidental finding in a single animal, but further investigation and careful evaluation has demonstrated a significant prevalence of BOA among nonhuman primate from several different sources. Thus, BOA should be investigated and characterized so that it can be readily identified in primates that are used in vision research. Identification of BOA is important, not least because it may confound interpretation of scientific results. An extensive search of the National Library of Medicine database (PubMed; National Institutes of Health, Bethesda, MD), using keywords such as optic neuropathy, optic atrophy, optic nerve, monkey, and rhesus macaque, suggests that this entity has never before been described. 

Methods

Animals

All experimental methods and animal care procedures adhered to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by the local Institutional Animal Care and Use Committee (IACUC). In total, data for 38 adult rhesus monkeys (Macaca mulatta) are presented in this study. Nine animals were found to have BOA; their demographic data are presented in Table 1 . Six of these were isolated from a population at one of our laboratories (Discoveries in Sight, Devers Eye Institute) after the initial subject was identified and confirmed by histopathology (BOA4). The other three monkeys with BOA were identified from a separate cohort located at the LSU Eye Center. Data for 29 other animals were used as the control. The control population consisted of normal healthy adults with an age range of 4 to 19 years. The control data were collected over the course of several years at Discoveries in Sight as a part of other IACUC-approved protocols. Intraocular pressure (IOP) was measured with a handheld tonometer (Tonopen; Oculab, Inc., Glendale, CA) with monkeys under ketamine or isoflurane anesthesia. The average of three readings each on two separate days is recorded in Table 1 . 

Anesthesia

All procedures were performed with animals under general anesthesia. In all cases, anesthesia was induced with 15 mg/kg intramuscular ketamine (Fort Dodge Animal Health, Fort Dodge, IA) and 1.5 mg/kg intramuscular xylazine (Phoenix Scientific Inc. St Joseph, MO), along with a single subcutaneous injection of atropine sulfate (0.05 mg/kg, Phoenix Scientific Inc.). Animals were intubated and breathed 100% oxygen. Heart and pulse rates and arterial oxyhemoglobin saturation were monitored continuously (Propaq Encore model 206EL; Protocol Systems, Inc., Beaverton, OR). Body temperature was maintained with a warm-water heating pad set at 37°C. Pupils were fully dilated (≥7 mm) with 1.0% tropicamide and 2.5% phenylephrine (Alcon Laboratories Inc., Fort Worth, TX). For all electroretinography (ERG) and visual evoked potential (VEP) sessions, anesthesia was maintained with intravenous ketamine (5 mg/kg per hour) and intramuscular xylazine (0.8 mg/kg per hour IM). Topical corneal anesthesia was provided with 0.5% proparacaine (Alcon Laboratories Inc.), and an ocular lubricating agent (Celluvisc; Allergan, Irvine, CA) was periodically applied. Recording sessions lasted approximately 2 hours. During all stereo optic disc photography sessions and clinical retinal examinations, anesthesia was maintained with 2% to 3% isoflurane (Baxter, Deerfield, IL). 

In Vivo Clinical Optic Nerve Assessment

The optic nerve and retinal nerve fiber layer (RNFL) were evaluated by clinical fundus examination, including binocular biomicroscopy using a 90-D lens (Volk Optical Inc., Mentor, OH) and slit lamp biomicroscope (model 900; Haag Streit AG, Köniz, Switzerland), as well as by simultaneous stereoscopic optic disc and retinal photography (3-Dx; Nidek Co., Ltd., Aichi, Japan). Four masked observers (GAC, BF, LW, and JD) graded all the optic nerves by scoring the stereoscopic optic disc photographs on a five-point scale (Table 1 ; grade 5, normal; grade 1, severe optic atrophy). Examples of optic disc photographs (half of the stereo pair for each eye) are shown in Figure 1 . 

Functional Evaluation: ERG and VEPs

Pattern-Reversal ERGs.

Traditional large-field, transient, pattern-reversal ERGs (PERGs) were recorded (UTAS-E3000 system; LKC Technologies, Gaithersburg, MD), as previously described. ¹ Custom-designed Burian-Allen contact lens electrodes (10 mm diameter, +3.0 D; Hansen Ophthalmics, Iowa City, IA) were used for all ERGs. The corneal ring on the stimulated eye served as the active electrode, whereas the corneal ring of the unstimulated (patched) contralateral eye served as the reference electrode. Both electrodes were referenced to a subcutaneous ground electrode placed in the thigh. Electrode impedance was accepted if <5 kΩ. The PERG stimulus was a checkerboard pattern (check size, 1°), reversing at 2.5 Hz (5 reversals/sec). The stimulus subtended 32° × 24° at the 50-cm test distance. Stimulus luminance was 75 cd/m², and contrast was >90%. The position of the foveal projection determined during mfERG testing was used for central alignment of the PERG stimulus. Residual refractive error was measured for the test distance and corrected to the nearest half diopter. Signals were band-pass filtered 1 to 500 Hz and sampled at 2 kHz. Two records were obtained for each eye and then averaged. Each single record was an average of 200 sweeps. Eye position was monitored continuously and remained stable, with sufficient depth of anesthesia. Peak-to-trough amplitudes were measured for the primary features commonly known as P50 and N95, as shown in Figure 2A . The slope of the leading edge of the N95 (same as the trailing edge of the P50) was also calculated as slope = N95 amplitude/(N95 implicit time − P50 implicit time). 

Ganzfeld ERGs.

Using the same system (UTAS-E3000; LKC Technologies), we obtained photopic, full-field, flash ERGs (fERGs) after 5 minutes of light adaptation to a rod-saturating blue background (30 scotopic cd/m²; Wratten filter 78; Eastman Kodak Co., Rochester, NY). Red stimulus flashes (Wratten filter 29; Eastman Kodak Co.) with an intensity of 0.42 log photopic cd-s/m² were presented monocularly at 0.5 Hz by a Ganzfeld integrating sphere. Stimulus and background intensities were measured with a calibrated photometer (Spectra Pritchard PR-1980A; Photograph Research, Chatsworth, CA). Signals were band-pass filtered 0.3 to 500 Hz and sampled at 2 kHz. Two records were obtained and then averaged. Each single record was an average of 6 to 10 sweeps. 

Peak-to-trough amplitudes were measured for three features: the a-wave, b-wave, and photopic negative response (PhNR), as shown in Figure 2B . Oscillatory potential (OP) isolation was achieved by post hoc filtering with a Blackman filter (−3 dB at 70 and 280 Hz). The amplitude of the OP complex was quantified by calculating the root mean square [RMS] of the filtered waveform over the epoch beginning at the a-wave trough and ending after OP4, as described previously. ² Flash VEPs were recorded simultaneously (during acquisition of the flash ERG) in four animals with optic atrophy (BOA6 to -9; Table 1 ) and in four control subjects, by subdermal platinum electrodes placed in the occipital scalp 1 cm to the right and left of the vertical midline and referenced to a midfrontal electrode. 

Scotopic ERGs were recorded from both eyes simultaneously after 20 minutes of dark adaptation in the same four animals (BOA6 to -9) and four control animals. Burian-Allen contact lens electrodes were used in their normal bipolar arrangement (corneal ring referenced to ipsilateral speculum). Single responses were recorded for stimulus flash intensities ranging from −1.6 to 4.4 log scot td sec. There was a 10-second pause between successive intensities at the lower end of the range and a 2-minute pause at the upper end. The scotopic P3 was modeled ³ using the responses from 1.6 to 4.4 log scot td sec to obtain two parameters of photoreceptor function, the maximum response amplitude (R _m) and sensitivity (S), as previously described (see Fig. 1and the Appendix in Ref. ⁴ ). It should be noted that no attempt was made to isolate the dark-adapted cone portion of the leading edge of the scotopic a-wave. After subtraction of the modeled P3 and band-pass filtered OPs (35–275 Hz) from the raw ERG, the amplitude of the isolated scotopic P2 was measured, plotted against stimulus intensity, and fitted with the Naka-Rushton equation, ⁵ to obtain two parameters of bipolar cell function ^{6 7 8} : the maximum response voltage (V _m) and the semisaturation constant (K), a measure of sensitivity. 

Multifocal ERGs were also recorded (VERIS; ver. 4; EDI, San Mateo, CA). Pupils were fully dilated (≥7 mm), corneal anesthesia and lubrication were provided periodically throughout the session, and the active Burian-Allen corneal electrode was referenced to the contralateral cornea, as described earlier. 

Residual refractive error was measured by retinoscopy for the test distance (25 cm) and corrected to the nearest 0.5 D. The mfERG stimulus was presented on a 21-in. monochrome monitor with a 75-Hz refresh rate. Before the actual recording session, an initial set of brief recordings (2 minutes each) were used to center foveal responses within the response array and to position the blind-spot responses appropriately. 

The mfERG stimulus consisted of 103 unscaled hexagonal elements subtending a total field size of ∼55° (Fig. 3A) . The luminance of each hexagon was independently modulated between dark (1 cd/m²) and light (200 cd/m²), according to a predetermined pseudorandom, binary m-sequence with a base interval of 13.3 ms, providing local contrasts of ∼99%. Stimulus luminance was measured with a calibrated spot photometer (SpectraScan PR-650; Photo Research, Inc., Chatsworth, CA). Each recording was ∼8 minutes in length (usually obtained in eight 60-second segments). Signals were amplified (gain, 100,000), band-pass filtered (10–300 Hz; with an additional 60-Hz line filter), sampled at 1.2 kHz (i.e., sampling interval, 0.83 ms), and digitally stored for subsequent off-line analyses. In a subset of four of the monkeys with optic atrophy and 15 control animals, mfERGs were also recorded with a slow stimulation sequence that had seven dark frames inserted into each m-step (7F). ^{9 10 11 12}  

Multifocal ERG records were exported for further analyses. Figure 3Ashows the stimulus locations corresponding to these records. The responses from the central (C) stimulus element, and the two surrounding concentric rings were evaluated. Locations were numbered 1 to 6 around the first ring and 1 to 12 around the second ring. These central locations were evaluated because they contain the largest relative contributions from ganglion cells. ^{11 12}  

The frequency content of these local responses was evaluated by fast Fourier transform (FFT) analysis on computer (Excel; Microsoft, Redmond, WA). Each local mfERG response was then band-passed filtered (85–300 Hz) to extract the high-frequency components (HFCs). The low-frequency component (LFC) of each response was represented as the raw response minus the HFC. The amplitude of the HFC was calculated as the RMS for the epoch between 0 to 80 ms of each filtered record. For reference, the mean amplitude of the noise was calculated using the RMS for an identically filtered 80 ms epoch taken from the eighth slice of the first-order kernel (where it is assumed that no signal is present) from the same 19 locations in each normal animal. 

Peak amplitudes for LFC features were quantified as follows. The first negative feature (N1) was calculated as the maximum negative excursion from baseline in the epoch up to 30 ms. The amplitude of the first positivity (P1) was calculated as the voltage difference between the maximum peak and the N1 trough. The second negativity (N2) was calculated as the difference between baseline and the minima from 30 to 50 ms, whereas the P2 amplitude was calculated as the difference between the maximum voltage from 50 to 70 ms, minus the N2 trough (Fig. 4B) . 

Histopathology

After completion of the in vivo studies, five of the animals were euthanatized with an intravenous injection (Euthasol; Diamond Animal Health, Inc., Des Moines, IA) after am intravenous bolus injection of heparin (∼5000 IU). Perfusion fixation was completed immediately with approximately 1 L of 4% buffered paraformaldehyde injected bilaterally into the precannulated carotid arteries. The perfusion lasted approximately 30 to 45 minutes, and the eyes were enucleated. For the purpose of comparison, 22 normal eyes from 22 monkeys were processed with the same techniques. These animals were part of a normal control group from several other protocols. 

Retrobulbar optic nerves were sampled approximately 2 mm posterior to the globe, from all 10 eyes of the five BOA animals killed and from all normal eyes, and were fixed in 4% formaldehyde for an additional 2 to 3 hours. A 0.5-mm-thick transverse section was then obtained from the nerve and fixed in 5% glutaraldehyde in phosphate buffer (pH 7.4) for 1 hour. After a thorough wash in phosphate-buffered saline (PBS, 5 minutes × 3), the tissue was postfixed in 2% osmium tetroxide for 3.5 hours. The tissue was rinsed, dehydrated in a graded ethanol-acetone series, and embedded in Epon 812. Semithin sections (1 μm) were cut and mounted onto glass slides and stained with 1% toluidine blue in phosphate buffer (0.01 M, pH 7.0–7.4) for 3 minutes followed by a few drops of Sörensen’s buffer for another 2 minutes. The slides were rinsed with distilled water and air-dried. The cross-sectional area of each retrobulbar optic nerve section was measured under a microscope by image analysis software (Bioquant; R&M Biometrics, Inc., Nashville, TN). 

Retinal histology was performed in one eye from each of four animals with BOA. In these cases, the eyes were hemisectioned along a horizontal plane located just above the optic disc. The tissue was processed for paraffin-embedded sections and stained by using a standard hematoxylin-eosin (H-E) method. In one of the BOA animals (the monkey with the lowest clinical ON grade by stereophotographic evaluation, i.e., the most severe optic atrophy), the retinal tissue from one eye was also evaluated with specific immunohistochemical labeling for astrocytes and axonal neurofilament (NF). 

Immunohistochemical Labeling.

Antiserum of monoclonal mouse anti-human glial fibrillary acidic protein (GFAP, 1:100; Novocastra Laboratories Ltd., Newcastle-upon-Tyne, UK) and NF (200 kDa; Novocastra Laboratories Ltd.) were used to label the astrocytes and axonal NF, respectively, with the avidin-biotin method for the paraffin-embedded sections. GFAP was used as a marker of astrocytes, whereas NF was used to identify axons within the nerve fiber layer and optic nerve. 

The sections were dewaxed and placed in 0.3% H₂O₂ in methanol for 30 minutes for antigen retrieval. The sections were then incubated with a mixture of 1% serum corresponding to the host species of secondary antibodies and 1% bovine serum albumin for 30 minutes. Primary antibodies of GFAP (1:200) or NF (1:50) were applied, and slides were incubated at room temperature for 90 minutes, or overnight at 4°C. After three 10-minute rinses in 0.01 M PBS, corresponding biotinylated secondary antibody (1:100, Vectastain Elite ABC kit; Vector Laboratories, Inc. Burlingame, CA) was applied for 30 minutes. This was followed by the avidin-biotin complex, which was applied for another 30 minutes, 3,3′-Diaminobenzidine (DAB Kit; Vector Laboratories, Inc.) was used for substrate chromogen staining for 2 to 10 minutes. The sections were counterstained with 0.1% Mayer’s hematoxylin (Sigma Diagnostics, St. Louis, MO) and mounted for microscopy. Negative control experiments for immunohistochemistry stains were performed with the omission of the corresponding primary antibody from the solution. 

Blood Work

From the four living BOA animals, a complete blood cell count (CBC) plus differential was obtained, as well as several other basic blood tests, including levels of vitamin E, B₁₂, and folate (Table 2 ; IDEXX Laboratories, Sacramento, CA). Normative ranges for B₁₂ and folate were determined by the same laboratory by using blood samples from 22 normal monkeys, whereas in all other tests the normal range was obtained from the available literature. Table 2also lists normal ranges for human blood. 

Statistics

Functional parameters for individual eyes with BOA were compared with the normal ranges obtained from the control group. Normal ranges (photopic flash ERG and PERG, n = 26 eyes of 26 animals; mfERG, n = 29 eyes of 29 animals) are presented as ordinary box-and-whisker plots in which the lower and upper whiskers indicate the 2.5 and 97.5 percentiles, respectively. Comparisons between normal and BOA group means were tested with one-way analysis of variance (ANOVA) and the Tukey multiple-comparison post hoc test (Prism 3.0; GraphPad Software, Inc., San Diego, CA). A conservative criterion for significance was adopted (α = 0.01) because multiple comparisons were made for each ERG technique (e.g., three to five parameters each, Table 3 ). In all cases, the assumption of equal variance was examined using Bartlett’s test and found to be valid. (Note that Bartlett’s test is also very sensitive to deviations from a Gaussian distribution.) 

Results

Study Population Demographics

The demographic data for the nine rhesus monkeys with BOA are summarized in Table 1 . All nine of these animals were born in China and arrived in the United States between 3 and 15 years after birth. They were acquired from the four sources listed in Table 1 . The five nonhuman primates in the original study group, which had complete histology and clinical measures, ranged in age from 5 to 17 years at the time of euthanasia. All these animals were female. Their IOPs were within a normal range: between 12 and 19 mm Hg at each examination time point. All animals appeared to be generally healthy, with an average weight of 4.9 kg (range, 4.4–5.8 kg). Clinical optic nerve grades were also obtained in a masked fashion, as described in the Methods section. Four of the five animals had severe bilateral temporal optic nerve atrophy, whereas one had only mildly abnormal (if not normal)-appearing optic nerve heads by stereo optic disc photography. Three of the four living animals were male and ranged in age from 4 to 19 years (the age of one animal was not known). The IOP was normal in three of the four and borderline elevated in one of the animals. The clinical optic nerve grades ranged from severe to moderate temporal optic atrophy. 

Of the five animals that had histopathologic studies completed (BOA1–5; Table 1 ), three were acquired through the Oregon Regional Primate Research Center (ORPRC), the other two through Sierra Biomedical; all five were obtained by Discoveries in Sight. The three ORPRC animals were bred and reared in captivity in China (Yunnan National Laboratory Primate Center of China) and were research-naïve when they arrived at ORPRC 4 to 5 years after birth. During quarantine, they received prophylactic medical therapy: ivermectin and/or valbazen (antiparasitics) and ofloxacin or cefazolin (antibacterials). All three were cleared and then assigned to studies on reproductive biology. They received one or more of the following: human chorionic gonadotrophin, luteinizing hormone, follicle-stimulating hormone, and gonadotropin-releasing hormone. All animals were also exposed to anesthetic agents (ketamine, isoflurane) before each regularly scheduled routine examination and/or during uncomplicated laparoscopic surgical procedures, which included oophorectomy (BOA3, BOA4), ovarian follicle aspiration (BOA4), and intraovarian catheterization (BOA5). After surgery, these animals also received pain control medication (buprenorphine). 

The other two animals acquired by Discoveries in Sight from Sierra Biomedical were also bred in captivity in China (BOA2 at the Yunnan facility, shipped to Sierra Biomedical at age 3; BOA1 source unknown, as initial import was by HRP, Inc. (Alice TX). BOA1 was involved in a bioavailability study of a small molecule that was given in a single dose, and blood samples were drawn for analysis for 2 days thereafter. All blood chemistry was normal on January 10, 2000, and October 4, 2000, just before delivery to our laboratory. BOA2 was involved in a study of reproductive biology and teratology, in which recombinant human relaxin was given during the first month of her pregnancy to evaluate possible effects on the offspring. Blood chemistry was normal on October 12, 1999, just before delivery to our laboratory. 

All three of the animals acquired by the LSU Eye Center (BOA7–9) were purchased from Three Springs Scientific (TSS; Perkasie, PA) and had been bred and reared in captivity in China (Shared Enterprises, Ltd., Bejing). The breeding center has its own feed mill, enabling monkeys to become adapted to monkey chow and ensuring that the chow is supplemented with vitamins and minerals. The daily diet of these monkeys also included fresh fruits and vegetables with every meal. On arrival in the United States (5, 13, and 17 years after birth, respectively), these three animals were quarantined (at Buckshire Corp., Perkasie, PA) and received prophylactic antiparasitic and antibiotic treatments (ivermectin and ofloxacin). All three were still research-naïve when they arrived at LSU. All eight of the monkeys for which records are available consistently tested negative for tuberculosis, salmonella, shigella, helminths, and ectoparasites. Two of these eight (BOA1 and -2) tested positive for herpes B virus; the other six consistently tested negative. No records are available for the single animal acquired from the University of Michigan (BOA6). 

In Vivo Clinical Optic Nerve Assessment

Figure 1illustrates the optic disc appearance in one representative normal monkey (Fig. 1A) , and in monkeys with mild (Fig. 1B , BOA5), moderate (Fig. 1C , BOA9), and severe (Fig. 1D , BOA4) bilateral temporal optic atrophy. Among the group of nine animals, nearly all had bilateral temporal pallor of the optic nerve head and marked thinning of the temporal RNFL, compared with normal. The typical normal pattern of RNFL striations between the temporal vascular arcades was generally absent from eyes with severe optic atrophy, although in two less-severe cases (e.g., Fig. 1C ), there were only rakelike bundle defects present. 

Functional Findings

Figure 2Apresents the results for transient pattern-reversal ERGs. Representative waveforms for BOA8 (left eye, bold trace) and the left eye of one normal animal (thin trace) are shown in the left panel, followed by the group data for the P50, N95, and slope parameters in the three panels to the right. Eight (44%) of the BOA eyes are below the lower limit of normal for the P50 amplitude, whereas all but two (89%) of the BOA eyes were below the normal limits for both N95 amplitude and the P50–N95 slope. The mean amplitudes of both P50 and N95 and the PERG slope parameter were all significantly reduced in the BOA groups compared with normal (Table 3) . Because the P50 component was also reduced in several of the eyes with BOA, the N95-to-P50 ratio was only below normal in six (44%) of eyes with BOA. PERG implicit times were not significantly different from normal (BOA average P50, 49.3 ± 7.8 ms; N95, 108.8 ± 10.4 ms; control group average P50, 47.5 ± 3.8; N95, 106.6 ± 7.0 ms). However, variability of PERG implicit times was higher in the BOA group than in the control subjects, in part because these measurements are less reliable when the amplitudes approach noise level. 

Figure 2Bsummarizes the findings for photopic, full-field, flash ERGs. As in Figure 2A , waveforms for the same two representative individuals are shown in the left panel (BOA8, left eye, bold trace; and normal animal, thin trace), followed by group data for a-wave, b-wave, PhNR and summed OP amplitudes in the four panels to the right. Only three of the BOA eyes (17%) were below the normal range for a-wave, whereas approximately 25% of the BOA eyes fell below the normal ranges for b-wave, PhNR and summed OP amplitudes. BOA group mean amplitudes were not significantly different from normal for any of these four ERG parameters (Table 3) . 

Figures 3B 3C 3D 3E 3Fdemonstrate the findings for the multifocal ERG. The geometry of the stimulus array is shown in Figure 3A . The array of responses for one representative normal eye and one with optic atrophy are shown in Figures 3B and 3C , respectively. In the left column of Figure 3D , the normal mfERG responses are arranged according to the legend shown in Figure 3A . The response to the central stimulus element (C) is at the top of the column, and the responses to locations within the two concentric rings around the center are aligned down the column in numerical order (1–6 for ring 1 and 1–12 for ring 2). The middle column in Figure 3Ddisplays the results for the left eye of BOA4 in the same manner. Note that the records for BOA4 are more smooth and regular, regardless of their position in the array, whereas the normal records contain more obvious HFCs and appear to vary systematically with position around the two rings. Both of these differences between normal responses and those of the animals with BOA are highlighted in the right column where the difference records are displayed. Note the robust HFCs and systematic nasal–temporal variation as response location changes. 

Figure 3Epresents the average Fourier power spectra (±SEM) for the 19 responses shown in Figure 3Dfor the representative normal eye (dashed line) and for BOA4 (left eye, solid line). The normal responses have greater power throughout the high-frequency range (beyond ∼80 Hz). Figure 3Fshows the average Fourier power spectra (±SEM) in the group of nine BOA animals (n = 18 eyes, circles and solid curve) and for the group of 29 normal animals (n = 29 eyes, triangles and broken curve). Comparison between groups confirmed that the HFCs (area under the curve from 85–300 Hz) were significantly reduced in eyes with BOA (F = 4.8, P = 0.02; ANOVA). There were no significant differences, however, between the normal and BOA groups for either the low-frequency band (0–75 Hz; F = 4.1, NS) or the area under the whole spectral power function (0 to 300 Hz; F = 2.6, NS). 

The Fourier analysis does not permit determination of specific waveform features that may have been affected by BOA (e.g., an increase in P2 offset by a decrease in N2). To address this question, we band-pass filtered all mfERG responses, as described in the Methods section, to extract the HFCs and LFCs for further analyses. Results are presented for the same individual normal and BOA eyes in Figure 4 . The extracted HFCs are shown in Figure 4Aand the residual LFCs are shown in Figure 4B . As expected, the HFCs were substantially larger in the normal eye and exhibited more nasal–temporal variation than those from the eye with BOA. The LFCs from the normal eye also showed greater nasal–temporal variation than those from the eye with BOA, which had nearly none. The major waveform features of the LFCs are identified on the top record of the second column. Some of the LFC features are actually larger in the eye with BOA, which was expected, given the raw data (Figs. 3B 3C 3D)and results of the Fourier analysis (Fig. 3E)for this pair of eyes. The results of the analyses by group for each feature’s peak-to-trough amplitude and for the RMS amplitude of the HFCs are presented in Figure 5 . 

The five panels in Figure 5present the normal ranges and individual data for BOA eyes for each of the five mfERG response components. Seventeen (94%) of the BOA eyes were at or below the lower limit of normal for the HFCs, whereas none of the BOA eyes fell below the normal range for any of the LFC features. In fact, there was a tendency for the LFC features to have larger than normal amplitudes. The N2 amplitude was significantly larger in both right- and left-eye optic atrophy groups compared with the control group (Table 3) . In contrast, the HFC amplitudes were much smaller in eyes with optic atrophy than those in the control eyes, and this finding was statistically significant for both the right and left eyes (Table 3) . 

Figure 6Ashows an example of the mfERG response array to slow stimulation (7F) for a representative normal eye (left) and for the typical animal with optic atrophy (right, BOA9 OS). The individual waveforms from the central retina (as in Fig. 3 ) are shown below each response array. The HFCs that are so prominent in the normal responses are profoundly reduced in eyes with optic atrophy without substantial change in the LFC amplitudes. This was a uniform finding in all eight of the eyes (4 animals) with optic atrophy for which 7F recordings were obtained. 

Figure 7demonstrates severe bilateral reduction of the full-field flash VEP in three of the four animals with BOA compared with the prototypical control responses. The VEPs in each eye of the fourth animal with BOA (BOA9) were delayed by ∼20 ms but had normal amplitudes. Clinically, this animal had only mild dropout of the RNFL, showing bundles of atrophy within the vascular arcades (Fig. 1C) . However, the mfERGs were severely affected in both eyes of this animal (e.g., see Fig. 6for 7F mfERG OS). 

Scotopic full-field ERG responses from both eyes of one representative animal with BOA are shown in Figure 8A . The scotopic b-wave increased normally with stimulus intensity, and the a-wave appeared at its normal intensity threshold. Scotopic ERG waveforms were normal in all respects, including the OPs. There were no differences between control and BOA animals in either the Rm_P3 or S parameters of photoreceptor function (Rm_P3 control = −128.7 ± 58.6 μV, BOA = −153.3 ± 12.4 μV, P = 0.26; S control = 21.7 ± 4.5 s⁻² · td s⁻¹, BOA = 24.0 ± 5.2 s⁻² · td s⁻¹, P = 0.35. Note, the time delay parameter t _d was fixed to 0.0035 s for all fits). Similarly, there were no differences between control and BOA animals in either the Vm_P2 or K parameters of bipolar cell function (Vm_P2 control = 324.8 ± 48.1 μV, BOA = 367.9 ± 11.6 μV, P = 0.40; K controls = 0.30 ± 0.2 td s, BOA = 0.33 ± 0.4 td s, P = 0.52. Note, the slope parameter n was fixed to 1.0 for all fits). Figure 8Bshows the amplitude of the isolated scotopic P2 against stimulus intensity for the control and BOA animals, as well as the Naka-Rushton functions defined by the average best-fit parameters from each group. 

Blood Work

Table 3lists the results of blood analyses. Blood levels of vitamin E, B₁₂, folate, and lead were within normal limits. There were no consistent abnormalities revealed by CBC with differential. According to the laboratory technicians, the glucose levels were low in two samples because of a storage problem that allowed glucose consumption to continue after the samples were drawn. 

Histopathologic Findings

Cross-sections of retrobulbar optic nerves from nonhuman primates with BOA revealed degenerative signs, including axonal loss and glial cell (astrocyte) proliferation, that were limited to the temporal sector in each nerve, but varied in severity across individual animals. The severity of histologic signs of degeneration was similar between the two eyes of each animal with BOA. Figure 9illustrates the general findings in one animal with mild (BOA5, left) and another with severe temporal optic nerve atrophy (BOA4, middle). A typical normal eye is shown for comparison (right). Within the region of degeneration, there was gliosis and a profound reduction of retinal ganglion cell axon density. The overall cross-sectional area of retrobulbar optic nerve tissue was significantly smaller in animals with BOA compared with a group of 20 normal eyes (3.85 ± 0.61 mm² vs. 5.84 ± 0.94 mm², respectively; P < 0.001, unpaired t-test). This is largely due to degeneration of the temporal portion of the optic nerve, as is apparent by the relatively temporal position that the central retinal artery and vein occupy in the atrophic optic nerves. 

Longitudinal sections through the anterior optic nerve of animals with BOA also demonstrated clear differences between the temporal and nasal sectors (Fig. 10) . In optic nerves with severe degeneration, obvious demarcation lines (Fig. 10 , arrowheads) became apparent between the gliotic temporal side and the relatively normal nasal side of each optic nerve. In addition, the RNFL was severely attenuated on the temporal side of the optic nerve (i.e., the papillomacular bundle, Fig. 10 ). In one of the BOA eyes, vacuoles or cysts were present within the inner nuclear layer (INL) of the temporal retina (Fig. 10A , asterisks). These cysts were not observed in any of the sections obtained from the other three eyes (from three other animals, see e.g., Fig. 10B ). They are unlikely to represent tissue processing or histologic artifact (because they were localized similarly in every section from that eye only). But as yet, the origin or content of these vacuoles is unknown. Evidently, the functional correlates (or consequences) of these cysts were relatively mild, given that the photopic b-wave was only slightly below the lower limit of normal and the mfERG LFC component amplitudes were within normal limits for this eye (scotopic ERGs were not obtained for this animal). 

Immunohistochemical labeling with antibodies to GFAP showed an increased immunoreactivity within the temporal portion of the optic nerve in animals with BOA (Figs. 11A12) . Conversely, immunohistochemical labeling of NF protein (NF, 200 kDa) was reduced within the same regions of the temporal optic nerve in animals with BOA, but was present normally in the nasal optic nerve and nerve fiber layer of these same animals (Figs. 11B 12) . The areas of increased GFAP immunoreactivity corresponded well with the areas of axonal loss and degeneration evidenced by decreased immunoreactivity to NF protein (e.g., compare Figs. 11Awith 11B ; insets in Fig. 12 ). 

Discussion

Presently, nine cases of BOA have been documented in a population of rhesus macaques. In each case, the clinical findings consisted of bilateral, symmetric thinning of the RNFL, predominantly within the papillomacular bundle, and corresponding temporal pallor of the optic nerve. Histologic analysis revealed degeneration of ganglion cell axons and gliosis within the temporal sector of the anterior optic nerve, as well as shrunken fascicles and overall reduction of optic nerve cross-sectional area. Retinal histology showed significant thinning of the temporal RNFL and loss of ganglion cells from the macula. The functional data were indicative of retinal ganglion cell disease. 

These results are all consistent with pathologic findings in human optic atrophy, in particular, conditions such as toxic/nutritional optic neuropathy, Leber hereditary optic neuropathy (LHON), and dominant optic atrophy (DOA). ^{13 14 15 16 17} It has been suggested that all these causes of optic atrophy share important aspects of pathophysiology and a final common pathway of mitochondrial dysfunction that ultimately lead to optic nerve degeneration. ¹⁸ The smallest retinal ganglion cells and their axons that make up the papillomacular bundle are the most susceptible in these conditions. ^{17 18 19} Their degeneration leads to characteristic clinical findings such as loss of the papillomacular bundle of the RNFL and temporal optic atrophy, primarily manifested as temporal pallor of the optic disc. ^{14 17 18 19} The equally distinctive histologic findings in these conditions are similar to those reported herein for monkeys with BOA. ^{17 18} It should be noted that the pattern of clinical and histologic findings are not like that in human glaucoma or experimental glaucoma based on chronically elevated IOP in nonhuman primates. ^{20 21} Specifically, all these animals had normal IOP, and only one or two of the 18 optic nerves had a clinical appearance similar to glaucomatous excavation (excavation and cupping). Moreover, histologic signs of degeneration were predominant within the temporal portion of the retrobulbar optic nerve in monkeys with BOA, in contrast to the upper and lower poles most typically observed in experimental and human glaucoma. ^{13 20 21 22 23} The features of spontaneous glaucoma in the rhesus monkey are more similar to those of human glaucoma than to the features of BOA described herein. ^{24 25}  

The functional data are also consistent with retinal ganglion cell disease. The VEP was reduced to near noise levels in most BOA animals tested. Full-field scotopic ERGs were normal in the animals with BOA. There were no differences between BOA and control animals in the maximum response amplitude or the sensitivity parameters of either dark-adapted photoreceptor function (P3) or bipolar cell function (P2). Full-field photopic ERG parameters, including a-wave, b-wave, PhNR, and OP amplitudes were unaffected in this group of animals on the whole, although some of the individual eyes were below normal limits for certain parameters, particularly for the b-wave and PhNR. Reduced photopic ERG b-waves have been reported in some patients with longstanding LHON, ^{26 27} but more typically, the full-field ERG is normal (reviewed by Sherman and Kleiner ²⁸ ). Loss of function as measured with the PERG is also consistent with ganglion cell/optic nerve disease, given the larger reduction of the N95 amplitude and the N95–P50 slope parameters, relative to the P50 amplitude. ²⁹  

The changes observed for mfERG responses from the central retina of the BOA animals, specifically loss of the HFCs from these responses, are also a marker of ganglion cell dysfunction (or death). ^{11 12 30 31} The findings of intact low frequency components, including the N1, P1, N2, and P2 provide additional evidence that cone photoreceptor and bipolar cell function remain normal throughout the central retina in these animals. ^{12 32} Taken together, these electrophysiological results are consistent with loss of only ganglion cells from central retina, with relative preservation of both cone and rod photoreceptor and bipolar cell function throughout the retina. ¹¹  

None of the authors nor any member of the veterinary staff has been able to discern overt behavioral abnormalities that would be indicative of central vision loss in the animals with BOA. However, the captive environment in which these animals reside imposes few demands on central vision or on behavior guided exclusively by central vision—unlike a more natural setting in which activities such as foraging, feeding, grooming, and navigating might all more readily reveal the manifestations of central vision impairment. Four of the surviving animals, including two who appeared to have a relatively early stage of this disease, are currently being observed to determine whether BOA is progressive. 

All nine of these animals were bred and reared in captivity in China and subsequently imported by one of several different sources (see Table 1 ). The eight monkeys for which records are available were research-naïve when they arrived in the United States. Before their delivery to either of our laboratories for ophthalmic research, five of those eight were involved in research, whereas the other three remained research-naïve at the time BOA was first documented. Of the five monkeys previously involved in research, four were in studies on reproductive biology and the other in a short-duration bioavailability study. The former received hormone treatments and had uncomplicated laparoscopic surgical procedures as part of IACUC-approved protocols. 

Collectively, the records indicate that none of these eight monkeys were exposed to drugs or neurotoxins known to cause optic atrophy in humans. ^{16 18} Although there is one report of a complicated case in which high-dose ciprofloxacin may have caused toxic optic neuropathy in a human patient, ³³ none of these monkeys with BOA were given high-dose ciprofloxacin; rather, fewer than half (four of nine) received routine prophylactic doses of the related drug ofloxacin. Similarly, there is one report of neurologic signs of ivermectin toxicity, including possible blindness, in a rhesus macaque that was inadvertently given 40 times the recommended dose. ³⁴ All the animals in the current report had received routine prophylactic antiparasitic doses of ivermectin in the past; however, according to the records none displayed signs of neurologic toxicity and all received doses that are routinely administered. Thus, it is unlikely that BOA represents toxic reactions to routine doses of these drugs. Nonetheless, it is difficult to ascertain perinatal nutritional status, and to definitively rule out potential past exposure to environmental toxins. To date, all toxicology and blood analyses have returned negative results. However, this does not exclude the possibility of previous toxicity or nutritional insufficiency. ^{35 36} Vitamin B₁₂ deficiency is known to cause optic atrophy in monkeys, ³⁷ but none of the animals presented here had abnormal B₁₂ levels nor any other neurodegenerative signs (e.g., spastic paralysis) of B₁₂ deficiency. ³⁸ Reversible blindness due to accidental lead poisoning has been documented in monkeys, ³⁹ but seems to occur only when blood levels of lead are repeatedly above 200 μg/dL in older monkeys, and is then also accompanied by severe systemic signs. ⁴⁰  

Thus, it is possible that BOA in these monkeys was due to some environmental cause, although the fact they originated in at least two separate breeding centers and that three were still research-naïve when BOA was first documented, makes this less likely. Thus, it is possible that BOA in monkeys is inherited, perhaps like LHON or DOA is in humans. Birth records of the four monkeys bred at the Yunnan facility establish that none of them are first-degree relatives. If other records become available, it may be possible to determine whether more distant genetic relationships exist in those four, and whether the other animals are closely related. Three of the surviving animals were male and one was female, whereas all five from the original group were female. This pattern at least serves to rule out sex-linked inheritance. 

Future studies could include analysis of possible genetic mutations, such as those known to be associated with LHON and DOA. Similarly, a screening study would be useful to estimate the prevalence of BOA in larger populations. However, it should be noted that BOA may be isolated to a relatively small population of monkeys with Chinese origin, as a survey of other investigators revealed that BOA has, in their collective experience, never previously been observed (Quigley H, Kaufman P, Harwerth R, Neuringer M, Dawson B, Horton J, personal communication, May 2005). 

In summary, the structural and functional consequences of BOA have been thoroughly characterized in nine rhesus monkeys. Investigators should carefully screen all nonhuman primates for this condition, using the techniques described in this and other studies, before inclusion in vision and ophthalmology research. Identification of BOA is important, not least because it may confound interpretation of scientific results. The presence of this entity could lead to erroneous recommendations being applied to humans. As this form of optic atrophy is predominantly localized to the temporal optic nerve, nonhuman primates with BOA may need to be eliminated from investigations that employ experimental models of retinal and optic nerve disease. However, if the basis of the disease can be more fully characterized in the future, it may also serve as an experimental model for human diseases such as LHON or DOA. 

 Supported by National Eye Institute Grants RO1-EY05231 (GAC); RO1-EY11610 (CFB); and unrestricted funds from Allergan, Inc., Irvine, California.

 Submitted for publication September 30, 2004; revised March 24 and June 13, 2005; accepted September 1, 2005.

 Disclosure: B. Fortune, None; L. Wang, None; B.V. Bui, None; C.F. Burgoyne, None; G.A. Cioffi, 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.

 

Table 1.

 

Demographics of Monkeys with BOA

ID (Sex)	Symbol	Age (y)	Weight (kg)	IOP* (mm Hg)	Clinical ON Grade, †	Histologic Signs of Damage	Source
BOA1 (f)	•	5	4.8	16.3/16.0	1.75/1.5	Moderate OU	SB/CR
BOA2 (f)	○	17	5.8	12.8/12.2	1.0/1.25	Severe OU	SB/CR
BOA3 (f)	▾	7	4.4	16.0/16.0	1.25/1.0	Severe OU	ONPRC
BOA4 (f)	▿	16	4.6	18.7/13.2	1.0/1.0	Severe OU	ONPRC
BOA5 (f)	▪	7	5.2	14.3/13.3	4.0/4.13	Mild OU	ONPRC
BOA6 (f)	□	-	7.5	21.0/22.0	3.75/3.0	Animal living	U-M
BOA7 (m)	⧫	19	8.8	18.0/19.0	1.5/1.75	Animal living	TSS
BOA8 (m)	◊	15	11.2	16.0/17.0	2.0/2.25	Animal living	TSS
BOA9 (m)	▴	4	2.3	11.0/10.0	3.0/3.0	Animal living	TSS

 SB/CR, Sierra Biomedical, Sparks, NV (Division of Charles River Laboratories, Inc.); ONPRC, Oregon National Primate Research Center, Beaverton, OR; U-M, University of Michigan, Ann Arbor, MI; Three Springs Scientific, Inc,, Perkasie, PA.

*  Average intraocular pressure.

†  Average grade for stereoscopic disk photographs OD/OS.

Figure 1.

 

Optic disc photographs for a representative normal monkey (A), and for one each with mild (B), moderate (C), or severe (D) BOA. Left: Right eyes; right: left eyes.

Figure 2.

 

Representative examples of transient PERG responses (A) and full-field flash ERG responses (B) from an individual normal eye (thin trace) and an eye with optic atrophy (bold trace). The distributions of normal amplitudes for each response component are shown by the box plots in the labeled panels to the right (horizontal hash mark: median; box: 25th and 75th percentiles; whiskers: 5th and 95th percentiles). Symbols: response amplitudes for individual eyes with optic atrophy (optic atrophy right [OAR] and left [OAL] eyes; see Table 1for identification of each symbol).

Figure 3.

 

Multifocal ERG stimulus array (A). Locations with responses that were exported for analyses are indicated: C, central element; 1 to 6, first eccentricity ring; 1 to 12, second eccentricity ring. Representative individual response array to standard mfERG stimulus for a normal left eye (B) and one with optic atrophy (C). Responses from locations marked in (A), for the same eyes shown in (B) and (C), but at higher magnification (D). Average (±SEM) frequency spectrum (E) for responses shown in (D); dashed curve: normal eye, solid curve: eye with optic atrophy. Group average (±SEM) frequency spectrum (F) for the same 19 response locations for 29 normal eyes (triangles, dashed curve) and 18 optic atrophy eyes (circles, solid curve).

Figure 4.

 

HFCs of mfERG responses to the standard stimulus (A) for a representative individual normal eye (left) and an eye with optic atrophy (right). LFCs from same two eyes (B).

Table 2.

 

Blood Chemical Assays in Four of the Animals with BOA and Normal Ranges

	Unit	Normal Range			BOA Monkey
				Human	Monkey	6	7	8	9	Mean	SD
Vitamin E	gm/mL	5–20	N/A		5.97	7.15	6.64	6.6	0.6
Vitamin B12	pg/mL	200–900	>1200*	2000	2000	2000	2000	2000
Folate	ng/mL	2.7–17	10.3 ± 4.1*	13.7	10	13.5	15.9	13.3	2.4
ALK	IU/L		118	71				71.0
ALT	IU/L	8–40	40 ± 12	16	56	36	48	39.0	17.4
AST	IU/L	8–35	41 ± 11	37	48	35	42	40.5	5.8
Creatine kinase	IU/L	22–198	442 ± 231	697				697.0
GGT	IU/L		41 ± 11		77	52	48	59.0	15.7
Albumin	g/dL	30–50	4.4	4.2	3.9	3.4	4.4	4.0	0.4
Total bilirubin	mg/dL	0.1–0.8	0.1	0.2	0.2	0.2	0.1	0.2	0.1
Total protein	g/dL	60–80	7.8	7.7	6.9	5.3	6.6	6.6	1.0
Cholesterol	mg/dL	116–210	180	142	161	187	168	164.5	18.6
Calcium	mg/dL	8.5–10.5	10.2	9.1				9.1
Glucose	mg/dL	80–120	74	57	99	84	27	66.8	31.7
Phosphate	mEq/L	3–4.5	3.5	3.6				3.6
Chloride	mEq/L	98–106	116	109				109.0
Potassium	mEq/L	3.5–5	4	3.6				3.6
WBC	k/μL	4–11	7.6–10.6	5.4	3.1	4.3	2	3.7	1.5
RBC	m/μL	4.5–6.5	5.5	5.59	4.25	4.88	5.14	5.0	0.6
Hb	g/dL	11.5–18.0	13.5	14.3	10.6	11.7	11.8	12.1	1.6
MCV	fL	80–98	74	74	78	73	68	73.3	4.1
MCH	pg	26–34	24	25.5	24.9	24	23	24.4	1.1
MCHC	g/dL	32–36	33	34.6	32.1	32.8	33.7	33.3	1.1
Lymphocyte	%		60	65	62	38	62	56.8	12.6
Monocyte	%		3	5	3	2	5	3.8	1.5
Eosinophil	%		4	2	2	3	1	2.0	0.8
Neutrophil	%		38	28	33	56	32	37.3	12.7
Lead	μg/dL	<10	<10	1	2.5	2.5	4.7	2.7	1.5

*  Averages from 22 normal monkeys (judged by clinical and sterophotographic examination of the optic nerve and RNFL).

 ALK, alkaline phosphotase; ALT, alanine aminotransferase; AST, aspartate aminotransferase; GGT, γ-glutamyltransferase; WBC, total white blood cell count; RBC, total red blood cell count; Hb, hemoglobin; MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; MCHC, mean corpuscular hemoglobin concentration.

Table 3.

 

Statistical Results for Comparisons between Group Means

Parameter	F	P	Normal vs. BOAR q (P)	Normal vs. BOAL q (P)	BOAR vs. BOAL q (P)
ERG
a-wave	1.5	NS	NA	NA	NA
b-wave	1.3	NS	NA	NA	NA
PhNR	1.2	NS	NA	NA	NA
OPs	1.2	NS	NA	NA	NA
PERG
P50	17.0	< 0.0001	6.4 (< 0.001)	6.7 (< 0.001)	0.2 (NS)
N95	34.4	< 0.0001	9.9 (< 0.001)	8.7 (< 0.001)	1.0 (NS)
Slope	24.8	< 0.0001	8.4 (< 0.001)	7.4 (< 0.001)	0.8 (NS)
mfERG
N1	0.9	NS	NA	NA	NA
P1	1.1	NS	NA	NA	NA
N2	8.4	<0.001	4.8 (<0.01)	4.3 (<0.05)	NA
P2	5.0	NS	NA	NA	NA
HFCs	15.8	<0.0001	5.6 (<0.01)	6.9 (<0.001)	1.2 (NS)

 BOAR, bilateral optic atrophy right eye; BOAL, bilateral optic atrophy left eye.

Figure 5.

 

Distribution of normal mfERG response component amplitudes (box plots) and eyes with optic atrophy (symbols). Details are as described in Figure 2 .

Figure 6.

 

Multifocal ERG response arrays (A) for the slow stimulus (7F) from a representative normal eye (left) and an eye with optic atrophy (right). Response waveforms from (A) shown at higher magnification (B) for the same locations as Figure 3 .

Figure 7.

 

Full-field flash VEP responses for typical normal animal (top pair of traces, right eye, OD and left eye, OS). VEP responses from four animals with bilateral optic atrophy (remaining traces). Two response averages are shown for each eye.

Figure 8.

 

Scotopic ERG series from both eyes of one representative animal with BOA (A). Each pair of responses in the intensity series is separated vertically by 150 μV; stimulus intensity indicated by number to left (in log scotopic troland-seconds). Scotopic P2 response amplitude versus stimulus intensity (B); (○) average (±SEM) of four control animals (8 eyes); (•) represent average of four BOA animals (eight eyes; monkeys BOA 6–9). Solid curve: Naka-Rushton function, V = (V _m · I ⁿ )/(I ⁿ + K ⁿ ), defined by the average parameters from control animals; broken curve: average parameters from BOA animals.

Figure 9.

 

Top row: representative optic nerve cross-sections from one monkey with mild (A), and another with severe (B) BOA. A representative normal subject is shown for comparison (C). Axonal degeneration was apparent within the temporal sector (T) of the two abnormal optic nerves; arrowheads: the border between the atrophic area and the relatively normal area of each nerve. Bottom row: higher-magnification (100×) micrographs taken from a point midway between the temporal (T) edge of the nerve and central artery (a) and vein (v). The optic nerve with mild degeneration (A) shows reduction of fascicle size and axon loss, although some intact axons remain within most fascicles. The optic nerve with severe regional degeneration (B) shows dramatic reduction of axonal density, shrunken fascicles and advanced gliosis. The horizontal diameter is also reduced compared with normal. Toluidine blue stain.

Figure 10.

 

Horizontal, longitudinal sections through the middle of the anterior optic nerve from one eye each of two monkeys with optic atrophy (BOA4 OS, A and BOA3 OS, B). Arrowheads: the border between the abnormal temporal (T) side and the relatively normal nasal (N) side of the anterior optic nerve. Advanced gliosis was apparent as well as loss of the normal columnar organization in the temporal side of the nerve. Insets: higher-magnification views of the retinal cross-section sampled from nasal (N) and temporal (T) sides of the optic nerve at similar eccentricities (boxes). On the whole, the temporal retina is approximately 20% thinner than the nasal, owing largely to the diminished RNFL. The density of cell bodies in the retinal ganglion cell (RGC) layer was reduced throughout the temporal portion of the section. Vacuoles ( ) were observed throughout the inner nuclear layer (INL) in the temporal retina of one eye. These were not observed in any of the sections from the other three eyes (from three other animals) evaluated in this manner. H-E stain.

Figure 11.

 

Horizontal, longitudinal sections through the middle of the anterior optic nerve from one eye of a monkey with optic atrophy. The serial sections are stained for GFAP (A) and NF (B) (counterstained with H-E). Arrowheads: the demarcation between relatively normal and degenerated regions. Indicative of glial reactivity, GFAP stain (A) was much stronger in the temporal (T) than the nasal (N) side. In contrast, the NF stain (B) was much weaker in the temporal than nasal side, owing to axonal degeneration. The areas of glial reactivity (A) and axonal degeneration (B) correspond closely. Both sections demonstrate increased density of (presumably glial) cell nuclei throughout the temporal side of the anterior optic nerve.

Figure 12.

 

The photomicrograph in the center is an optic nerve cross-section from an animal with BOA (H-E stain). Dotted lines: the region with degeneration. Within this region, there were an increased number of nuclei stained with H-E. The pair of photomicrographs on the left was taken from the relatively normal nasal region, whereas the pair on the right was taken from the damaged temporal region of the same optic nerve, and stained immunohistochemically for GFAP (top) and NF (bottom). In the nasal side of the optic nerve, glial cells were much smaller and less dense. In contrast, NF was diminished in the damaged regions (bottom right). a: central retinal artery; v: central retinal vein. S, T, I, N: superior, temporal, inferior, and nasal, respectively. Scale bars: 25 μm, unless otherwise indicated.

The authors thank Grant Cull, Jin Dong, Pris Zhou, and J. Crawford Downs for insightful commentary and technical assistance.