December 2015
Volume 56, Issue 13
Free
Retina  |   December 2015
Electrophysiological ON and OFF Responses in Autosomal Dominant Optic Atrophy
Author Affiliations & Notes
  • Enyam Komla A. Morny
    School of Optometry and Vision Sciences, Cardiff University, Cardiff, United Kingdom
  • Tom H. Margrain
    School of Optometry and Vision Sciences, Cardiff University, Cardiff, United Kingdom
  • Alison M. Binns
    Division of Optometry and Visual Sciences, City University, London, United Kingdom
  • Marcela Votruba
    School of Optometry and Vision Sciences, Cardiff University, Cardiff, United Kingdom
    Eye Unit, University Hospital of Wales, Cardiff, United Kingdom
  • Correspondence: Marcela Votruba, School of Optometry and Vision Sciences, Cardiff University, Maindy Road, Cardiff CF24 4HQ, UK; votrubaM@cardiff.ac.uk
Investigative Ophthalmology & Visual Science December 2015, Vol.56, 7629-7637. doi:10.1167/iovs.15-17951
  • Views
  • PDF
  • Share
  • Tools
    • Alerts
      ×
      This feature is available to Subscribers Only
      Sign In or Create an Account ×
    • Get Citation

      Enyam Komla A. Morny, Tom H. Margrain, Alison M. Binns, Marcela Votruba; Electrophysiological ON and OFF Responses in Autosomal Dominant Optic Atrophy. Invest. Ophthalmol. Vis. Sci. 2015;56(13):7629-7637. doi: 10.1167/iovs.15-17951.

      Download citation file:


      © ARVO (1962-2015); The Authors (2016-present)

      ×
  • Supplements
Abstract

Purpose: To assess the effect of autosomal dominant optic atrophy (ADOA) on ON and OFF retinal ganglion cell (RGC) function by evaluating the ON and OFF components of the photopic negative response (PhNR).

Methods: Twelve participants from six families with OPA1 ADOA and 16 age-matched controls were recruited. Electrophysiological assessment involved pattern ERGs (PERGs), focal (20°) and full-field long-duration (250 ms) flash ERGs using a red light-emitting diode flash on a rod-saturating blue background, and full-field brief (300 μs) xenon flash ERGs using a red filter over a continuous rod saturating blue background. Amplitudes and implicit times of the ERG components were analyzed and the diagnostic potential of each electrophysiological technique was determined by generating receiver operating characteristic (ROC) curves.

Results: Mean amplitudes of the N95 and all PhNRs, except the full-field PhNRON, were significantly reduced in participants with ADOA (P < 0.01). Subtraction of the group-averaged focal ERG of ADOA participants from that of controls showed an equal loss in the focal PhNRON and PhNROFF components, whereas in the full-field ERG the loss in the PhNROFF was greater than that in the PhNRON component. The areas under the ROC curve (AUC) for the focal PhNRON (0.92), focal PhNROFF (0.95), and full-field PhNROFF (0.83), were not significantly different from that of the PERG N95 (0.99).

Conclusions: In patients with ADOA, the PhNRON and PhNROFF components are nearly symmetrically reduced in the long-duration ERG, suggesting that ON- and OFF-RGC pathways may be equally affected.

Autosomal dominant optic atrophy (ADOA) is a hereditary optic neuropathy characterized by variable bilateral loss of vision in early childhood, optic nerve pallor, centrocecal visual field scotoma, and color vision defects.15 It is the commonest hereditary optic neuropathy with a prevalence between 1 in 50,000 and 1 in 8000.611 
Autosomal dominant optic atrophy is caused primarily by mutations in the autosomal nuclear gene, OPA1,9,1215 a key player in mitochondrial dynamics, controlling mitochondrial fusion, among other key roles. Histopathologic studies in humans16,17 and mouse models1821 show that ADOA is principally characterized by the degeneration of the RGCs. 
In a mouse model of ADOA, generated in our laboratory,21 the defect is first evident as a dendritic pruning of RGCs in B6:C3-Opa1Q285STOP Opa1 mutant mouse, which appears to be ON-center specific.18,22 This selective vulnerability of ON-center RGCs may reflect their higher energy demands in comparison with their OFF-center counterparts, as OPA1 mutations are thought to curtail mitochondrial energy output.18,22 This new finding has, however, not been investigated in humans with ADOA. 
The functional integrity of RGCs can be evaluated by assessing the photopic negative response (PhNR) of the flash ERG. The PhNR is a negative potential seen after the b-wave in a photopic ERG elicited by a brief flash. The PhNR is believed to primarily originate from spiking activity in RGCs and their axons with contributions from amacrine cells and possible involvement of associated glial cells/astrocytes of the retina.2326 When a long-duration flash is used to evoke the ERG, the PhNR is seen once after the b-wave (PhNRON) and again as a negative going potential after the d-wave (PhNROFF). Furthermore, it has been demonstrated that the ERG obtained in response to a long-duration red flash of moderate intensity provides optimal delineation of the PhNRON and PhNROFF components.23,26,27 
The brief-flash PhNR is attenuated in patients with ADOA28 and in the Opa1Q285STOP mutant mouse.29 In the mouse model, the defect is seen before any changes in visual acuity on optokinetic drum testing and before morphologic changes on retinal histology. This suggests that retinal connectivity may be affected before RGC somal loss affects RGC function.22 Thus, the PhNR deficit could serve as a marker for early disease. These early changes in RGC function may be reversible and need to be defined as markers for targeted therapies in any forthcoming therapeutic trials. 
Miyata et al.28 and Barnard et al.29 highlight the diagnostic potential of the PhNR in ADOA; however, the investigators used a brief white flash (broadband stimulus) to evoke the PhNR, which provides a poor signal-to-noise ratio compared with monochromatic stimuli,26 and cannot distinguish ON and OFF components. Furthermore, the studies elicited full-field (global) PhNRs that, in contrast to the focal PhNR, are less sensitive in detecting focal retinal lesions, such as those seen in early to moderate glaucoma.30,31 As ADOA results in localized centrocecal visual field defects,1 it might be expected that a focal stimulus presented to this region would enhance the sensitivity of the PhNR to early disease-related changes. 
The aim of this study was to assess the relative effect of ADOA on the PhNRON and PhNROFF components elicited using focal and full-field long-duration red flashes on a rod-suppressing blue background. An additional aim was to compare the diagnostic potential of the long-duration PhNRs to responses that have previously been shown to be affected by ADOA; the full-field brief-flash PhNR28,29 and the N95 amplitude of the pattern ERG (PERG),32 which also reflects spiking activity of the RGCs.33 
Methods
Participants
Twelve participants (aged 18–61 years) from six families with documented OPA1 mutations and 16 healthy age-matched controls (aged 19–61 years) were recruited for the study (see Table 1 for characteristics of all 12 participants). Detailed information about the clinical characteristics of nine of the participants have been reported elsewhere.1 The study conformed to the Declaration of Helsinki and was approved by the National Health Service Research Ethics Committee for Wales, as well as the ethics committees of the School of Optometry and Vision Sciences, Cardiff University, and the Division of Optometry and Visual Science, City University, London. All participants provided their written consent after receiving a participant information sheet and having the opportunity to ask questions. Nine participants with ADOA (ID numbers 1010–1017, 1021) were examined in Cardiff and the rest at City University London by the same investigator. 
Table 1
 
Clinical Characteristics of Participants With ADOA
Table 1
 
Clinical Characteristics of Participants With ADOA
Electroretinograms
All ERGs were recorded monocularly using a DTL fiber active electrode (Unimed Electrode Supplies, Ltd., Surrey, UK) and a contralateral reference. The DTL fiber was placed in the lower fornix to maximize stability during recording and the loose end fastened using medical tape at the inner canthus (Blenderm; Viasys Healthcare, Ltd., Warwick, UK). A silver–silver chloride 10-mm diameter touch-proof skin electrode (Unimed Electrode Supplies, Ltd.), placed at the midfrontal forehead position was used as ground electrode. 
Electroretinogram responses were obtained using an evoked potential monitoring system (Medelec EP; Oxford Instruments PLC, Surrey, UK [Cardiff site]; Espion; Diagnosys LLC, Cambridge, UK [City University site]). Responses were bandpass filtered from 1 to 100 Hz and digitally averaged. Signals were recorded in blocks of 10 to 20 responses, with a total of 40 to 60 averaged per trace. Between four and six traces were obtained for each stimulus condition. The traces were superimposed to confirm signal repeatability and averaged off-line into a single averaged trace containing 160 to 300 responses. An automatic artifact rejection system removed signals contaminated by large eye movements and blinks. 
Transient PERG stimuli (four reversals per second; check size = 1°) were generated on a computer monitor at 98% contrast. The screen was masked with a black opaque cardboard with a 13 × 13-cm square cut-out at the center so that it produced a 20° × 20° field at a viewing distance of 36 cm. 
Long-duration ERGs were recorded using a red flash stimulus (peak output 660 nm, 250-ms duration, 3.33 log phot td, 2 Hz) on a rod-saturating blue background (peak output 469 nm, 3.49 scot log td) produced by a handheld miniature Ganzfeld light-emitting diode (LED) stimulator (CH Electronics, Kent, UK). Focal stimulation was produced by mounting the miniature Ganzfeld LED tube into the middle of a light box (44 × 44 × 10 cm) such that the circular stimulus subtended 20° diameter at a viewing distance of 15.6 cm. The 20° stimulus size was chosen to encompass as much of the central field as possible while avoiding the optic disc, which starts approximately 12° to 15° nasal to the fovea. To minimize the effect of stray light stimulating the peripheral retina (i.e., the area outside the stimulus area), the light box contained a strip of white LEDs (color temperature > 7000K) passed through a blue filter (Lee Filter 068 Sky Blue; Lee Filters, Hampshire, UK) to produce a desensitizing blue surround of 3.73 scot log td (field size = 109° × 109° field). Cross hairs centered in the middle of the stimulus served as the fixation target. Full-field ERGs were recorded by holding the stimulator head, fitted with a diffusing cap, directly to the eye. 
Full-field brief (flash) ERGs were elicited by a Ganzfeld stimulator (GS2000; LACE Elettronica, Rome, Italy) presenting a xenon flash stimulus (1.76 log td.s, 300 μs maximum flash duration, 4 Hz). Filters were used to obtain a red stimulus (Lee Filter “Terry Red”; Lee Filters, Hampshire, UK, transmittance < 5% at wavelengths shorter than 575 nm, and above 85% from 625–700 nm) over a continuous rod-saturating 3.39 scot log td blue background (Schott Glass filter BG28; Schott AG, Mainz, Germany, peak transmittance 454 nm). All stimulus backgrounds were of sufficient scotopic illuminance to saturate the rods.34 
All ERGs were recorded by the same investigator using the same protocol at both sites. Long-duration ERGs (focal and full-field) were generated by the same miniature Ganzfeld LED stimulator at both sites. Pattern ERG and full-field brief flash data were obtained only from participants attending Cardiff University, so as to ensure consistency. All stimuli were calibrated using an ILT 1700 radiometer with SED033/Y/R luminance detector (Able Instruments and Controls, Reading, UK) assuming a 7-mm pupil with no correction for the Stiles-Crawford effect. The wavelengths of the light sources were measured using a Specbos 1201 spectro-radiometer (Horiba Jobin Yvon Ltd., Middlesex, UK). 
Procedures
All participants underwent a comprehensive ophthalmic examination that included best corrected visual acuity (ETDRS), contrast sensitivity (Pelli-Robson), visual field assessment (24-2 SITA-FAST, Humphrey Visual Field Analyzer, Carl Zeiss Meditec, Inc., Dublin, CA, USA), slit lamp biomicroscopy, optical coherence tomography (OCT; Topcon 3D-OCT 1000; Topcon Medical Systems, Inc., Tokyo, Japan), fundus photography, color vision (D-15 desaturated test) and auto-refraction. To target earlier-stage ADOA, the eye with the better visual field mean deviation score was selected for ERG recording, with the dominant eye chosen in the case of equal scores between the two eyes. 
Pattern ERGs were always recorded first with natural pupils and near refractive correction when necessary. Pupils were then dilated using 1% tropicamide to a minimum of 7 mm and flash ERGs were recorded in the following order: focal long-duration, full-field long-duration, and full-field brief-flash ERG. 
Signal Analysis
Pattern ERGs and focal ERGs were Fourier analyzed to remove high-frequency noise above 30 Hz and 50 Hz, respectively. The method for measuring the amplitude of the various subcomponents is described in Figure 1. The PhNRON (PhNR for brief-flash ERG) and PhNROFF amplitudes were measured from the prestimulus baseline and voltage at stimulus offset, respectively, to a fixed time point in their respective troughs. When determining the most appropriate fixed time point at which to measure the PhNRON and PhNROFF responses, the group-averaged ERG of ADOA participants was subtracted from the group-averaged ERG of the controls to obtain a difference ERG. The implicit time of the greatest discrepancy between the two was identified for the PhNRON and PhNROFF responses and was used as the fixed time point for all measurements. The fixed times at which the PhNR amplitudes were measured were as follows: focal PhNRON at 95 ms after onset, focal PhNROFF at 97 ms after offset, full-field PhNRON at 83 ms after onset, full-field PhNROFF at 102 ms after offset and full-field brief PhNR at 72 ms after onset. The identification of all peaks and troughs was determined objectively using Microsoft Excel (Microsoft Corp., Redmond, WA, USA) (i.e., as the minimum/maximum voltage within a fixed time window). 
Figure 1
 
Representative ERG traces of the (A) PERG, (B) full-field long-duration ERG, and (C) full-field brief-flash ERG showing their components and how their amplitudes were measured (double-headed arrows). The amplitudes of the P50, N95 (A), a-wave (a) and b-wave (b) (B, C) were measured as recommended by the International Society of Clinical Electrophysiology of Vision.35 The d-wave (d) amplitude was measured from the point of light offset to the peak of the d-wave. The PhNRON (PhNR in brief flash) and PhNROFF amplitudes were measured from the prestimulus baseline and voltage at stimulus offset respectively to a fixed time point in their respective troughs (see main text for details). The focal long-duration ERG had the same profile as the full-field long duration except that in the focal ERG there was only one prominent positive peak after light offset, the d-wave.
Figure 1
 
Representative ERG traces of the (A) PERG, (B) full-field long-duration ERG, and (C) full-field brief-flash ERG showing their components and how their amplitudes were measured (double-headed arrows). The amplitudes of the P50, N95 (A), a-wave (a) and b-wave (b) (B, C) were measured as recommended by the International Society of Clinical Electrophysiology of Vision.35 The d-wave (d) amplitude was measured from the point of light offset to the peak of the d-wave. The PhNRON (PhNR in brief flash) and PhNROFF amplitudes were measured from the prestimulus baseline and voltage at stimulus offset respectively to a fixed time point in their respective troughs (see main text for details). The focal long-duration ERG had the same profile as the full-field long duration except that in the focal ERG there was only one prominent positive peak after light offset, the d-wave.
Statistical Analysis
Data expressed on a logarithmic scale (i.e., visual acuity, contrast sensitivity, and visual field mean deviation) were converted (antilogged) into a linear scale to calculate mean and SD values. The mean and SD values were then converted back to log units. The distribution of the ERG data was checked for normality using the Shapiro-Wilk test. Where data were normally distributed, independent sample t-tests (2-tailed) were used to compare controls and participants with ADOA; the Mann-Whitney U test was used where data were non-normally distributed. To minimize Type 1 errors due to the number of comparisons made (n = 35), we applied a Bonferroni adjustment to the α level (0.05) and report observations as significant when P < 0.0014. Receiver operating characteristic (ROC) curve analysis was used to calculate the area under the curve (AUC) to assess the diagnostic potential of the various ERG components. The comparison between AUCs was made using the method described by Hanley and McNeil.36 
Results
The clinical characteristics of all 12 ADOA participants from six families are shown in Table 1. The means and SDs for visual acuity, contrast sensitivity, and mean deviation were 1.10 ± 1.07 logMAR, 1.30 ± 1.26 log units, and −7.39 ± 7.09 dB, respectively. The visual field defects were mostly central or centrocecal and color vision defects were variable, but participants from the same family had similar defects. More details regarding the relationship between the clinical characteristics and ERG data in ADOA participants is to be the subject of a future manuscript. 
Pattern ERGs
Pattern ERGs recorded from nine ADOA participants are shown superimposed on the group-averaged trace of 16 controls in the left column of Figure 2A. It shows that the negative N95 component is reduced in amplitude for all participants with ADOA, beyond the 95% confidence intervals (CIs) for the control data. The P50 amplitudes in ADOA participants were also below the lower 95% CIs except for one participant. The middle column of Figure 2A and the data in Table 2 demonstrate that the mean P50 and N95 amplitudes were significantly reduced in ADOA participants compared with controls. The mean N95:P50 ratio in ADOA participants of 1.05 was significantly reduced compared with 1.73 in controls (Table 2). Although there was evidence of P50 and N95 loss in people with ADOA, the difference plot in the right column of Figure 2A is dominated by a negative going signal corresponding to the loss of the N95 component. 
Figure 2
 
Electroretinogram traces of the (A) PERG, (B) focal long-duration ERG, (C) full-field long-duration ERG, and (D) full-field brief-flash ERG recorded from participants in this study. Left column: Individual traces of participants with ADOA (thin lines) superimposed on the group averaged ERG of 16 controls (thick lines) for each type of ERG recorded. The number of participants with ADOA are 9 (A), 12 (B), 12 (C), and 7 (D). Dotted lines represent 95% CIs. Middle column: Comparison between group-averaged traces of controls (thick black line) and ADOA participants (thick red line). Right column: Difference plots generated by subtracting the group-averaged ADOA ERG from the control ERG.
Figure 2
 
Electroretinogram traces of the (A) PERG, (B) focal long-duration ERG, (C) full-field long-duration ERG, and (D) full-field brief-flash ERG recorded from participants in this study. Left column: Individual traces of participants with ADOA (thin lines) superimposed on the group averaged ERG of 16 controls (thick lines) for each type of ERG recorded. The number of participants with ADOA are 9 (A), 12 (B), 12 (C), and 7 (D). Dotted lines represent 95% CIs. Middle column: Comparison between group-averaged traces of controls (thick black line) and ADOA participants (thick red line). Right column: Difference plots generated by subtracting the group-averaged ADOA ERG from the control ERG.
Table 2
 
Means of Amplitudes and Implicit Times in Controls and Participants With ADOA
Table 2
 
Means of Amplitudes and Implicit Times in Controls and Participants With ADOA
Focal Long-Duration Cone ERGs
Focal long-duration cone ERGs recorded from 12 participants with ADOA are shown superimposed on the group-averaged trace of 16 controls in the left column of Figure 2B. The typical ERG responses were characterized by the a-wave, b-wave, PhNRON, d-wave, and PhNROFF . The PhNRON was reduced in amplitude below the lower 95% confidence limit of the control data in almost all ADOA participants except one. Notably, the waveform after stimulus offset varied considerably among ADOA participants. For instance, in participants with ADOA, the most prominent positive peak after stimulus offset, assumed to be the d-wave, was delayed and had a broad peak whose maximum amplitude occurred at highly variable times (Fig. 2B, right and middle columns). In comparison, this prominent peak was highly consistent among control participants with respect to implicit time and was reflected in the much smaller SD of the d-wave implicit time in controls than in participants with ADOA (Table 2). 
The difference plot (Fig. 2B, right) was dominated by two negative going waves representing the PhNRON and PhNROFF components affected by ADOA. The difference plots of the ON and OFF components had similar profiles and amplitudes of 2.80 μV and 2.88 μV, respectively. 
Long-Duration Full-Field Cone ERG
The long duration full-field cone ERGs in Figure 2C were recorded from the same ADOA participants (thin lines) and controls (group-averaged thick black line) as the focal cone ERGs in Figure 2B. The form of the long-duration ERG was similar under focal and full-field conditions with one exception. There were two positive peaks immediately after light offset in the full-field ERG; the first being the d-wave.23,37 The mean amplitude of the PhNROFF, but not the PhNRON, was significantly reduced in participants with ADOA (Table 2). On the difference plot (Fig. 2C, right) the amplitude of the PhNROFF difference (8.76 μV) was more than twice the amplitude of the PhNRON difference (3.42 μV) when measured. 
Once again, the OFF components showed greater variability than ON components for participants with ADOA. In fact, in at least six participants with ADOA, there was a third positive peak (3PP) after light offset not seen in controls (Fig. 2C, left). There was no obvious pattern to the presence or absence of the 3PP in ADOA participants. The amplitude and implicit time of the 3PP measured from the ADOA group-averaged trace was 13.34 μV and 75 ms after light offset, respectively. Comparatively, none of the control traces displayed the 3PP prominently, although on close visual inspection, a kink corresponding in time with the 3PP was observed in some individual control traces. 
Comparison of Focal and Full-Field Long-Duration ERGs
The waveform of the focal and full-field long-duration ERGs was further compared by normalizing the group-averaged ERGs to their respective b-wave amplitudes (Fig. 3). The focal and full-field ERGs of controls (Fig. 3A) and participants with ADOA (Fig. 3B) had similar profiles, although implicit times of the b-wave, PhNRON, and d-wave were significantly delayed in the focal ERG (P ≤ 0.01, data not shown). The most prominent positive peak of the focal ERG after light offset coincided with the 2PP of the full-field ERG in the control traces, whereas in the ADOA group, the broad peak of the focal ERG after offset described a curve that roughly matched the profile of the 2PP and 3PP of the full-field ERG. 
Figure 3
 
A comparison of the long-duration focal (dashed lines) and full-field (solid lines) group-averaged ERGs for (A) controls, (B) participants with ADOA, and (C) difference plots. Electroretinograms have been normalized to the b-wave amplitude of their respective control group–averaged ERG.
Figure 3
 
A comparison of the long-duration focal (dashed lines) and full-field (solid lines) group-averaged ERGs for (A) controls, (B) participants with ADOA, and (C) difference plots. Electroretinograms have been normalized to the b-wave amplitude of their respective control group–averaged ERG.
In controls, the PhNRs are proportionally greater in the focal ERG than in the full-field ERG (Fig. 3A). The losses in amplitudes of the PhNRs were also greater in the focal ERG than the full-field ERG in participants with ADOA (Figs. 3B, 3C). 
The Brief Full-Field ERG
Brief full-field ERGs recorded from seven ADOA participants are shown in Figure 2D. Typical ERG responses had a-wave, b-wave, i-wave, and PhNR components. The PhNR amplitude was reduced significantly in people with ADOA compared with controls (Fig. 2D; Table 2). The difference plot in the right column of Figure 2D indicates that the greatest deficit in ADOA corresponds to the timing of the b-wave and the PhNR. An i-wave was recorded for all participants (controls and ADOA). Although it appeared more prominent in ADOA participants, there was no statistical difference in amplitude or implicit time between control and ADOA participants (Table 2). 
Specificity and Sensitivity of the Different ERGs
Receiver operating characteristic curves were used to determine the effectiveness of the N95 and long-duration focal and full-field PhNRs at discriminating participants with ADOA from controls for nine participants with ADOA and 16 controls for whom PERG and long-duration focal and full-field ERG data were available (Fig. 4). The AUC, sensitivity, specificity, and cutoff value, which produced an optimal sensitivity while maintaining minimum specificity of approximately 90%, are shown in Table 3. The N95 amplitude had the greatest diagnostic power. However, a comparison of the AUCs of the focal and full-field PhNRs with the N95 amplitude, using the method described by Hanley and McNeil36 showed that the N95 amplitude was only significantly more sensitive than the full-field PhNRON amplitude (z = 2.12). Therefore, considered in terms of their diagnostic ability, the focal PhNRs and N95 component were not significantly different. 
Figure 4
 
Receiver operating characteristic curves derived using (A) N95 component and N95:P50 ratio of the PERG, (B) focal PhNRON and PhNROFF amplitudes, and (C) full-field PhNRON and PhNROFF amplitudes. Diagonal dashed line is the reference line.
Figure 4
 
Receiver operating characteristic curves derived using (A) N95 component and N95:P50 ratio of the PERG, (B) focal PhNRON and PhNROFF amplitudes, and (C) full-field PhNRON and PhNROFF amplitudes. Diagonal dashed line is the reference line.
Table 3
 
Sensitivity, Specificity, and Area Under Curve of ROC Analysis for ERG Components
Table 3
 
Sensitivity, Specificity, and Area Under Curve of ROC Analysis for ERG Components
Discussion
Effect of ADOA on ON and OFF RGCs
In this study, we sought to determine whether the PhNRON was preferentially affected in ADOA as might be predicted based on the study by Williams et al.18 Our findings, however, showed that in human patients, the PhNRON and PhNROFF amplitudes were equally reduced in the focal ERG, whereas in the full-field ERG, there was a greater reduction in the PhNROFF amplitude than the PhNRON amplitude. What then might explain this apparent contradiction? 
In the study by Williams et al.,18 evidence for the preferential loss of ON-RGCs was based on mouse retinal flat mounts showing significant dendritic pruning of ON- but not OFF-RGCs. The experiment reported here, however, assessed the effect of ADOA on the ON- and OFF-RGCs by evaluating the PhNR amplitude of the human ERG, a functional measure. The role of RGCs as primary originators of the PhNR has been demonstrated in previous studies.2326,33 In experiments using long-duration full-field ERGs, they showed that that PhNRON and PhNROFF components were both reduced or eliminated after experimental glaucoma and intravitreal injection of tetrodotoxin (TTX) (an agent that blocks generation of sodium-dependent spikes in retinal neurons) in macaques, as well as in patients with glaucoma. Although the origins of the PhNRON and PhNROFF have not been conclusively traced to the ON- and OFF-RGCs respectively, Luo and Frishman33 showed that the PhNRON (and b-wave) component but not the PhNROFF (or d-wave) was eliminated after injecting 2-amino-4-phosphonobutyric acid (APB) into the macaque retina to block synaptic transmission from photoreceptors to ON-bipolar cells and hence ON-RGCs. Injecting TTX after APB then removed the PhNROFF but not the d-wave, thereby linking the PhNRON and PhNROFF components (although indirectly) to the ON and OFF pathways, respectively. 
Previous human30,38,39 and animal23,29 studies (including our mouse model) have shown that the PhNR amplitude is very susceptible to RGC damage with severe attenuation of PhNR amplitude recorded even when morphologic and other functional parameters were within normal range (i.e., in early-stage disease). It is possible that the PhNRON pathways may be selectively compromised at an earlier stage of the disease process than that studied here. A similar study in presymptomatic people with the OPA1 mutations or in people with ADOA at a much earlier stage of the disease (e.g., children with ADOA) could provide additional insights. 
Our findings also may be a reflection of the heterogeneous nature of ADOA. There are more than 200 OPA1 mutations14,15 that cause ADOA, with wide phenotypic variations both within and between affected families.1,40 Genotype–phenotype correlations have been difficult to establish in previous studies,2,41 and the number of patients in each family of OPA1 mutations (except family E) was insufficient to reliably explore such correlations. In the mouse model, the mutant mice (>10 months old) were genetically homogeneous and disease severity correlated with age. Participants studied here were from six families, with a different mutation in each family (Table 1), and at different stages of the disease. This may have diluted observations that would have been made from a homogeneous cohort. 
Comparison of Focal and Full-Field PhNRs
The long-duration focal and full-field ERGs in this study were recorded using the same stimulus parameters, which were comparable to the parameters recommended by Kondo et al.42 for eliciting focal responses. Although the waveforms of the focal and full-field ERGs were similar, they were not identical (Figs. 3A, 3B). There was a greater contribution of PhNRON and PhNROFF components to the focal ERG than to the full-field ERG (Fig. 3A), which reflects the decreasing proportion of RGCs to other retinal cells with eccentricity.43 The focal PhNRs were more severely affected than their full-field counterparts by ADOA, and this was reflected in the larger AUCs found for the focal signals. These findings were consistent with the central field defects recorded in ADOA participants in this study and in others.1,5,44 In addition, whereas the focal PhNRs were both significantly reduced (P < 0.001), only the full-field PhNROFF was significantly reduced in the full-field ERG (Table 2). Although the N95 and focal PhNR amplitudes were highly discriminatory for ADOA, it should be noted that the participants in this study had relatively late-stage disease. 
The symmetrical loss in the focal PhNRON and PhNROFF amplitudes (Fig. 2B, right column; Fig. 3C) may reflect the 1:1 ratio of ON- to OFF-RGCs in the macula, whereas the greater loss in the full-field PhNROFF amplitude than the PhNRON amplitude (Fig. 2C, right column; Fig. 3C) may reflect the nearly 1:2 ratio of ON- to OFF-RGCs in the peripheral retina.4547 The broadening of the d-wave peak in the focal ERG and the presence of the 3PP in the full-field long ERG in participants with ADOA may be due to contributions from the cone receptor potential and/or depolarizing OFF-bipolar cell responses after light offset, which were unmasked in the relative absence of the negative going PhNROFF .4850 The 2PP may be the iOFF-wave described by Horn et al.,51 although in contrast to their results, this study did not record a significant difference in amplitude between controls and participants with ADOA. 
Comparison With Other Electrophysiological Studies in ADOA
Miyata et al.28 reported a significant reduction in the full-field brief PhNR, but none in the a- or b-wave amplitude, in ADOA patients using white-on-white stimulus. Similar results were obtained by Barnard et al.29 in the mouse model. In this present study, we show similar results using a red-on-blue stimulus. The flash luminance used in this study was adopted from a previous study in this laboratory52 and was comparable to the flash luminance used by Miyata et al.28 This supports findings that the red-on-blue stimulus is effective for clinical evaluation of RGC function. 
Holder et al.33 reported a significant reduction in N95 amplitude and the N95:P50 ratio of the PERG participants with ADOA. We obtained similar results and showed that the focal PhNRs and N95 amplitude were equally effective at discriminating controls from participants with ADOA. The focal ERG could therefore be used as an alternative to the PERG. 
In this study, as well as in that of Holder et al.,32 the P50 amplitude was significantly reduced. This may indicate that bipolar cell function is compromised in ADOA, as has been put forward by Reis et al.53 However, a reduction in P50 amplitude is also seen when only RGCs are compromised33; therefore, the P50 reduction observed in this study could be due to dysfunction of bipolar cells, RGCs, or both. 
Conclusions
This study showed there was a nearly symmetrical reduction in the PhNRON and PhNROFF amplitudes in participants with ADOA with no evidence of a preferential ON-pathway loss. This suggests that ON- and OFF-RGCs may be equally affected in patients. In addition, in terms of their diagnostic potential, the focal PhNR-ON and -OFF amplitudes were better than their full-field counterparts and were not significantly different from the N95 amplitude of the PERG. 
Acknowledgments
Supported by the Ghana Education Trust Fund (EKAM). We acknowledge Fight for Sight Small Grants Award (MV) for their contribution to genotyping. 
Disclosure: E.K.A. Morny, None; T.H. Margrain, None; A.M. Binns, None; M. Votruba, None 
References
Votruba M, Fitzke FW, Holder GE, Carter A, Bhattacharya SS, Moore AT. Clinical features in affected individuals from 21 pedigrees with dominant optic atrophy. Arch Ophthalmol. 1998; 116: 351–358.
Votruba M, Thiselton D, Bhattacharya SS. Optic disc morphology of patients with OPA1 autosomal dominant optic atrophy. Br J Ophthalmol. 2003; 87: 48–53.
Newman NJ, Biousse V. Hereditary optic neuropathies. Eye. 2004; 18: 1144–1160.
Yu-Wai-Man P, Griffiths PG, Hudson G, Chinnery PF. Inherited mitochondrial optic neuropathies. J Med Genet. 2009; 46: 145–158.
Votruba M, Moore AT, Bhattacharya SS. Clinical features molecular genetics, and pathophysiology of dominant optic atrophy. J Med Genet. 1998; 35: 793–800.
Krill AE, Smith VC, Pokorny J. Similarities between congenital tritan defects and dominant optic-nerve atrophy: coincidence or identity. J Opt Soc Am. 1970; 60: 1132–1139.
Kjer B, Eiberg H, Kjer P, Rosenberg T. Dominant optic atrophy mapped to chromosome 3q region. II. Clinical and epidemiological aspects. Acta Ophthalmol Scand. 1996; 74: 3–7.
Votruba M, Aijaz S, Moore AT. A review of primary hereditary optic neuropathies. J Inherit Metab Dis. 2003; 26: 209–227.
Yu-Wai-Man P, Griffiths PG, Burke A, et al. The prevalence and natural history of dominant optic atrophy due to OPA1 mutations. Ophthalmology. 2010; 117: 1538–1546.e1.
Gallus GN, Cardaioli E, Rufa A, et al. High frequency of OPA1 mutations causing high ADOA prevalence in south-eastern Sicily, Italy. Clin Genet. 2012; 82: 277–282.
Lenaers G, Hamel C, Delettre C, et al. Dominant optic atrophy. Orphanet J Rare Dis. 2012; 7: 46.
Alexander C, Votruba M, Pesch UEA, et al. OPA1, encoding a dynamin-related GTPase, is mutated in autosomal dominant optic atrophy linked to chromosome 3q28. Nat Genet. 2000; 26: 211–215.
Delettre C, Lenaers G, Griffoin JM, et al. Nuclear gene OPA1, encoding a mitochondrial dynamin-related protein, is mutated in dominant optic atrophy. Nat Genet. 2000; 26: 207–210.
Ferre M, Bonneau D, Milea D, et al. Molecular screening of 980 cases of suspected hereditary optic neuropathy with a report on 77 novel OPA1 mutations. Hum Mutat. 2009; 30: E692–E705.
Ferre M, Amati-Bonneau P, Tourmen Y, Malthiery Y, Reynier P. eOPA1: an online database for OPA1 mutations. Hum Mutat. 2005; 25: 423–428.
Johnston PB, Gaster RN, Smith VC, Tripathi RC. A clinicopathologic study of autosomal dominant optic atrophy. Am J Ophthalmol. 1979; 88: 868–875.
Kjer P, Jensen OA, Klinken L. Histopathology of eye optic-nerve and brain in a case of dominant optic atrophy. Acta Ophthalmol (Copenh). 1983; 61: 300–312.
Williams PA, Morgan JE, Votruba M. Opa1 deficiency in a mouse model of dominant optic atrophy leads to retinal ganglion cell dendropathy. Brain. 2010; 133: 2942–2951.
Sarzi E, Angebault C, Seveno M, et al. The human OPA1delTTAG mutation induces premature age-related systemic neurodegeneration in mouse. Brain. 2012; 135: 3599–3613.
Alavi MV, Bette S, Schimpf S, et al. A splice site mutation in the murine OpaI gene features pathology of autosomal dominant optic atrophy. Brain. 2007; 130: 1029–1042.
Davies VJ, Hollins AJ, Piechota MJ, et al. Opa1 deficiency in a mouse model of autosomal dominant optic atrophy impairs mitochondrial morphology, optic nerve structure and visual function. Hum Mol Genet. 2007; 16: 1307–1318.
Williams PA, Piechota M, von Ruhland C, Taylor E, Morgan JE, Votruba M. Opa1 is essential for retinal ganglion cell synaptic architecture and connectivity. Brain. 2012; 135: 493–505.
Viswanathan S, Frishman LJ, Robson JG, Harwerth RS, Smith EL,III. The photopic negative response of the macaque electroretinogram: reduction by experimental glaucoma. Invest Ophthalmol Vis Sci. 1999; 40: 1124–1136.
Viswanathan S, Frishman LJ, Robson JG. The uniform field and pattern ERG in macaques with experimental glaucoma: removal of spiking activity. Invest Ophthalmol Vis Sci. 2000; 41: 2797–2810.
Viswanathan S, Frishman LJ, Robson JG, Walters JW. The photopic negative response of the flash electroretinogram in primary open angle glaucoma. Invest Ophthalmol Vis Sci. 2001; 42: 514–522.
Rangaswamy NV, Shirato S, Kaneko M, Digby BI, Robson JG, Frishman LJ. Effects of spectral characteristics of ganzfeld stimuli on the photopic negative response (PhNR) of the ERG. Invest Ophthalmol Vis Sci. 2007; 48: 4818–4828.
Sustar M, Hawlina M, Brecelj J. ON-, and OFF-response of the photopic electroretinogram in relation to stimulus characteristics. Doc Ophthalmol. 2006; 113: 43–52.
Miyata K, Nakamura M, Kondo M, et al. Reduction of oscillatory potentials and photopic negative response in patients with autosomal dominant optic atrophy with OPA1 mutations. Invest Ophthalmol Vis Sci. 2007; 48: 820–824.
Barnard AR, Issa PC, Perganta G, et al. Specific deficits in visual electrophysiology in a mouse model of dominant optic atrophy. Exp Eye Res. 2011; 93: 771–777.
Machida S, Tamada K, Oikawa T, et al. Comparison of photopic negative response of full-field and focal electroretinograms in detecting glaucomatous eyes. J Ophthalmol. 2011; 1–11.
Tamada K, Machida S, Yokoyama D, Kurosaka D. Photopic negative response of full-field and focal macular electroretinograms in patients with optic nerve atrophy. Jpn J Ophthalmol. 2009; 53: 608–614.
Holder GE, Votruba M, Carter AC, Bhattacharya SS, Fitzke FW, Moore AT. Electrophysiological findings in dominant optic atrophy (DOA) linking to the OPA1 locus on chromosome 3q 28-qter. Doc Ophthalmol. 1998; 95: 217–228.
Luo X, Frishman LJ. Retinal pathway origins of the pattern electroretinogram (PERG). Invest Ophthalmol Vis Sci. 2011; 52: 8571–8584.
Aguilar M, Stiles WS. Saturation of the rod mechanism of the retina at high levels of stimulation. Opt Acta (Lond). 1954; 1: 59–65.
Marmor MF, Brigell MG, McCulloch DL, Westall CA, Bach M. ISCEV standard for clinical electro-oculography (2010 update). Doc Ophthalmol. 2011; 122: 1–7.
Hanley JA, McNeil BJ. A method of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology. 1983; 148: 839–843.
Evers HU, Gouras P. Three cone mechanisms in the primate electroretinogram: two with, one without off-center bipolar responses. Vision Res. 1986; 26: 245–254.
Nakamura H, Miyamoto K, Yokota S, Ogino K, Yoshimura N. Focal macular photopic negative response in patients with optic neuritis. Eye (Lond). 2011; 25: 358–364.
Gotoh Y, Machida S, Tazawa Y. Selective loss of the photopic negative response in patients with optic nerve atrophy. Arch Ophthalmol. 2004; 122: 341–346.
Granse L, Bergstrand I, Thiselton D, et al. Electrophysiology and ocular blood flow in a family with dominant optic nerve atrophy and a mutation in the OPA1 gene. Ophthalmic Genet. 2003; 24: 233–245.
Yu-Wai-Man P, Griffiths PG, Gorman GS, et al. Multi-system neurological disease is common in patients with OPA1 mutations. Brain. 2010; 133: 771–786.
Kondo M, Kurimoto Y, Sakai T, et al. Recording focal macular photopic negative response (PhNR) from monkeys. Invest Ophthalmol Vis Sci. 2008; 49: 3544–3550.
Curcio CA, Allen KA. Topography of ganglion-cells in human retina. J Comp Neurol. 1990; 300: 5–25.
Fuhrmann N, Schimpf S, Kamenisch Y, et al. Solving a 50 year mystery of a missing OPA1 mutation: more insights from the first family diagnosed with autosomal dominant optic atrophy. Mol Neurodegener. 2010; 5: 25.
Dacey DM, Petersen MR. Dendritic field size and morphology of midget and parasol ganglion-cells of the human retina. Proc Natl Acad Sci U S A. 1992; 89: 9666–9670.
Dacey DM. The mosaic of midget ganglion cells in the human retina. J Neurosci. 1993; 13: 5334–5355.
Drasdo N, Millican CL, Katholi CR, Curcio CA. The length of Henle fibers in the human retina and a model of ganglion receptive field density in the visual field. Vision Res. 2007; 47: 2901–2911.
Bush RA, Sieving PA. A proximal retinal component in the primate photopic ERG a-wave. Invest Ophthalmol Vis Sci. 1994; 35: 635–645.
Ueno S, Kondo M, Ueno M, Miyata K, Terasaki H, Miyake Y. Contribution of retinal neurons to d-wave of primate photopic electroretinograms. Vision Res. 2006; 46: 658–664.
Sieving PA, Murayama K, Naarendorp F. Push-pull model of the primate photopic electroretinogram: a role for hyperpolarizing neurons in shaping the b-wave. Vis Neurosci. 1994; 11: 519–532.
Horn FK, Gottschalk K, Mardin CY, Pangeni G, Junemann AG, Kremers J. On and off responses of the photopic fullfield ERG in normal subjects and glaucoma patients. Doc Ophthalmol. 2011; 122: 53–62.
Mortlock KE, Binns AM, Aldebasi YH, North RV. Inter-subject, inter-ocular and inter-session repeatability of the photopic negative response of the electroretinogram recorded using DTL and skin electrodes. Doc Ophthalmol. 2010; 121: 123–134.
Reis A, Mateus C, Viegas T, et al. Physiological evidence for impairment in autosomal dominant optic atrophy at the pre-ganglion level. Graefes Arch Clin Exp Ophthalmol. 2013; 251: 221–234.
Figure 1
 
Representative ERG traces of the (A) PERG, (B) full-field long-duration ERG, and (C) full-field brief-flash ERG showing their components and how their amplitudes were measured (double-headed arrows). The amplitudes of the P50, N95 (A), a-wave (a) and b-wave (b) (B, C) were measured as recommended by the International Society of Clinical Electrophysiology of Vision.35 The d-wave (d) amplitude was measured from the point of light offset to the peak of the d-wave. The PhNRON (PhNR in brief flash) and PhNROFF amplitudes were measured from the prestimulus baseline and voltage at stimulus offset respectively to a fixed time point in their respective troughs (see main text for details). The focal long-duration ERG had the same profile as the full-field long duration except that in the focal ERG there was only one prominent positive peak after light offset, the d-wave.
Figure 1
 
Representative ERG traces of the (A) PERG, (B) full-field long-duration ERG, and (C) full-field brief-flash ERG showing their components and how their amplitudes were measured (double-headed arrows). The amplitudes of the P50, N95 (A), a-wave (a) and b-wave (b) (B, C) were measured as recommended by the International Society of Clinical Electrophysiology of Vision.35 The d-wave (d) amplitude was measured from the point of light offset to the peak of the d-wave. The PhNRON (PhNR in brief flash) and PhNROFF amplitudes were measured from the prestimulus baseline and voltage at stimulus offset respectively to a fixed time point in their respective troughs (see main text for details). The focal long-duration ERG had the same profile as the full-field long duration except that in the focal ERG there was only one prominent positive peak after light offset, the d-wave.
Figure 2
 
Electroretinogram traces of the (A) PERG, (B) focal long-duration ERG, (C) full-field long-duration ERG, and (D) full-field brief-flash ERG recorded from participants in this study. Left column: Individual traces of participants with ADOA (thin lines) superimposed on the group averaged ERG of 16 controls (thick lines) for each type of ERG recorded. The number of participants with ADOA are 9 (A), 12 (B), 12 (C), and 7 (D). Dotted lines represent 95% CIs. Middle column: Comparison between group-averaged traces of controls (thick black line) and ADOA participants (thick red line). Right column: Difference plots generated by subtracting the group-averaged ADOA ERG from the control ERG.
Figure 2
 
Electroretinogram traces of the (A) PERG, (B) focal long-duration ERG, (C) full-field long-duration ERG, and (D) full-field brief-flash ERG recorded from participants in this study. Left column: Individual traces of participants with ADOA (thin lines) superimposed on the group averaged ERG of 16 controls (thick lines) for each type of ERG recorded. The number of participants with ADOA are 9 (A), 12 (B), 12 (C), and 7 (D). Dotted lines represent 95% CIs. Middle column: Comparison between group-averaged traces of controls (thick black line) and ADOA participants (thick red line). Right column: Difference plots generated by subtracting the group-averaged ADOA ERG from the control ERG.
Figure 3
 
A comparison of the long-duration focal (dashed lines) and full-field (solid lines) group-averaged ERGs for (A) controls, (B) participants with ADOA, and (C) difference plots. Electroretinograms have been normalized to the b-wave amplitude of their respective control group–averaged ERG.
Figure 3
 
A comparison of the long-duration focal (dashed lines) and full-field (solid lines) group-averaged ERGs for (A) controls, (B) participants with ADOA, and (C) difference plots. Electroretinograms have been normalized to the b-wave amplitude of their respective control group–averaged ERG.
Figure 4
 
Receiver operating characteristic curves derived using (A) N95 component and N95:P50 ratio of the PERG, (B) focal PhNRON and PhNROFF amplitudes, and (C) full-field PhNRON and PhNROFF amplitudes. Diagonal dashed line is the reference line.
Figure 4
 
Receiver operating characteristic curves derived using (A) N95 component and N95:P50 ratio of the PERG, (B) focal PhNRON and PhNROFF amplitudes, and (C) full-field PhNRON and PhNROFF amplitudes. Diagonal dashed line is the reference line.
Table 1
 
Clinical Characteristics of Participants With ADOA
Table 1
 
Clinical Characteristics of Participants With ADOA
Table 2
 
Means of Amplitudes and Implicit Times in Controls and Participants With ADOA
Table 2
 
Means of Amplitudes and Implicit Times in Controls and Participants With ADOA
Table 3
 
Sensitivity, Specificity, and Area Under Curve of ROC Analysis for ERG Components
Table 3
 
Sensitivity, Specificity, and Area Under Curve of ROC Analysis for ERG Components
×
×

This PDF is available to Subscribers Only

Sign in or purchase a subscription to access this content. ×

You must be signed into an individual account to use this feature.

×