All experimental procedures were conducted according to the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes, set out by the National Health and Medical Research Council in Australia. Ethics clearance was obtained from the University of Melbourne, Science Faculty, Animal Ethics Committee (approval number 0911322.1).

1. Pre-implantation of Chronic VEP Electrodes

Note: If concurrent ERG and VEP signals are to be collected animals must be surgically implanted with VEP electrodes at least 1 week prior to signal collection.

1. Sterilize the surgical bench prior to experimentation by cleaning with chlorhexidine (0.5% in 70% ethanol). Autoclave all surgical equipment before use. Cover the animal with a sterilized surgical drape. Ensure all experimenters wear surgical masks, gowns and sterilized gloves.
2. Induce anesthesia with 3 – 3.5% isoflurane with O₂ at a flow rate of 3 L/min. Maintain anesthesia at 1.5% and 2 L/min throughout the surgery. Ensure sufficient depth of anesthesia by the absence of paw pinch reflex.
3. Apply 1% carboxymethylcellulose sodium on the cornea to prevent drying of the eyes.
4. Shave a 30 mm x 30 mm area over the forehead, posterior to the eyes and anterior to the ears.
5. Place animal on a heat pad (37 ˚C) to maintain body temperature and stabilize animals' head with a stereotaxic frame.
6. Disinfect the shaved area with 10% povidone-iodine three times. Avoid the use of alcohol-based antiseptics for area near the eye, being consistent with the Standard of Practice set out by the Association of Surgical Technologists.
7. Make a median-sagittal incision on the head with a scalpel and from this excise a ~ 20 mm diameter circle of dermal tissue to expose the cranial bone.
8. Remove underlying periosteum by scraping and drying with gauze to expose the coronal and sagittal cranial sutures.
9. Using a dental burr attached to a drill, trepan two holes (0.7 mm diameter, depth ~ 1 mm) through the skull on both hemispheres at the stereotaxic co-ordinates: 7 mm caudal to bregma, 3 mm lateral to midline.
10. Screw in stainless steel screws (diameter 0.7 mm, length 3 mm, sterilized with chlorhexidine) into the two pre-made holes up to a depth of ~ 1 mm (2 mm of screw exposed) to allow firm anchorage. This contacts the dura without damaging the underlying cortical tissue.
11. Prepare surgical area for dental amalgam by drying the cranial bone with gauze, and retracting loose skin with two 3 – 0 sutures at ~ 4 and 8 o'clock.
12. Spread dental amalgam over the exposed skull to secure the screw electrodes (stainless steel screws described in step 1.10) in place. Ensure ~ 1.5 mm of the screws remain exposed for recording.
13. Remove the retraction sutures.
14. Inject 0.5% carprofen subcutaneously (5 mg/kg) for analgesia and saline (sodium chloride 0.9%, 1.5 ml) subcutaneously for fluid replacement.
15. Allow animal to recover in separate cages. Do not leave animal unattended until it has regained sufficient consciousness to maintain sternal recumbency.
16. Do not return animal to the company of other animals until it has fully recovered from surgery (minimum 5 days).
17. Continue to administer 0.5% carprofen subcutaneously for analgesia (5 mg/kg) once a day for 4 days.
18. Record ERG and VEP 1 week following surgery.

2. ERG and VEP Recording 

1. Data collection preparation
   1. Use computer software to simultaneously trigger stimulus and acquire data² according to the settings recommended below.
      1. Amplify the signals³ (ERG: ×1,000, VEP: x10,000) with the gain internally set by an isolated pre-amplifier and amplifier, and with both eyes matched for impedance.
      2. Set sampling rate for the ERG to 4 kHz over a 650 msec recording window (2,560 points), To do this, click the tab for "time base" in the data acquisition software (for name and version of software see Table for Materials), select "2,560" for Samples, and "500 ms" for time which will return a 650 msec recording window.
         1. Use the same method to set the sampling rate for the VEP to 10 kHz over a 250 msec epoch. Allow a 10 msec pre-stimulus baseline for both ERG and VEP recordings. To do this, click the "Setup" tab; select "Stimulator" to bring up a new dialogue window; in that window select "pulse" from the drop down list for "Mode"; and set the value for "delay" to "10 ms".
      3. Set ERG band-pass filtering to 0.3 – 1,000 Hz (- 3 dB). This is done by clicking "Bio Amplifier" in the data acquisition software. Then set the value for "High Pass" to "0.3Hz", and the value for "low pass" to "1 kHz".
      4. Using the abovementioned method in 2.1.1.3, set VEP band-pass settings to 0.1 – 100 Hz (- 3 dB) as recommended by the International Society for Clinical Electrophysiology of Vision (ISCEV) for human VEP recordings⁶.
2. Electrode preparation
   1. Custom-make the ERG active/inactive and VEP active/inactive electrodes by attaching silver wire or an alligator clip to an electrode lead, respectively². Commercially obtain ground electrode.
   2. For the 4 custom-made electrodes, cut the male end from the electrode lead extension. Remove 1 cm of the outer polytetrafluoroethylene insulation coating with a scalpel blade ensuring the inner wire is not damaged.
   3. Pre-fashion the ERG inactive electrodes by cutting a 70 mm length of silver wire (0.3 mm thickness) and forming a loop ~ 8 mm in diameter to encircle the rat eye. Prepare a uniform circle by shaping the loop on a 1 ml pipette tip.
   4. Pre-fashion VEP inactive electrodes by cutting a 70 mm length of silver and forming an ellipse ~ 8 mm in length-wise diameter to hook onto rat incisors.
   5. Pre-fashion the ERG active electrodes by cutting a 30 mm length of silver wire and forming a small loop to gently contact the rat cornea (~ 1-2 mm in diameter)
   6. Securely attach electrodes (2 ERG active, 2 ERG inactive, 1 VEP inactive) to the electrode lead by entwining the silver with the exposed inner wire.
   7. Insulate excess exposed metal with masking tape to reduce photovoltaic artifacts.
   8. On the ERG inactive electrodes stick a small piece of hook-and-loop fastener (~ 5 mm × 20 mm) to the masking tape to enable stable attachment to the rodent neck strap.
   9. Attach alligator clip to the inner wire of the electrode leads to make the VEP active electrodes.
   10. Prior to recordings, electroplate the exposed surfaces of the silver wires (i.e., the inactive ring and active tip) with chloride using a 9 V DC source for 20 sec to improve signal conduction.
       1. To do this, immerse the silver tip of the ERG electrode wire (acting as the anode of a primary cell) into normal saline; connect the other end of this electrode wire to the positive terminal of a 9 V battery.
       2. Connect another wire (the cathode) to the negative terminal of the battery, and immerse the other end into saline as well. Disconnect after 20 sec and observe the sliver tip of the ERG electrode wire to be coated evenly in white color.
          Note: Prepare new ERG electrodes for each experimental session (~ up to 8 hr) to ensure patency of the chloride coating.
3. Animal preparation
   1. Dark-adapt the animals overnight (≥ 8 hr) prior to recordings in a light tight room. Ensure maximal dark adaptation by turning off room lights, closing all doors and blinds. Minimize light leakage by placing light-proof materials around junctions of doors/windows and placing computer screens outside thick black curtains.
   2. Conduct animal preparation in a dark room with the aid of dim red light-emitting diode (LED; 17.4 cd.m^-2, λ_max = 600 nm) to sustain rod sensitivity.
   3. Anesthetize rat by injecting ketamine/xylazine (60 : 5 mg/kg) intramuscularly. Confirm sufficient depth of anesthesia by the absence of a paw pinch reflex.
   4. To maintain sedation, administer a further dose of anesthesia (50% of initial dose) after 50 min if necessary.
   5. For additional topical anesthesia apply one drop of 0.5% proxymetacaine to each eye, and blink off excess fluid.
   6. For pupil dilation apply one drop of 0.5% tropicamide to each eye, then dry off excess fluid.
4. ERG and VEP electrode positioning
   1. Place animal on the ERG platform in front of the Ganzfeld bowl situated in the Faraday cage. Avoid using an electrical heating pad, as it can introduce electrical noise into the electrophysiological recordings. Note: The platform is attached to a circulated heated water platform to maintain body temperature.
   2. Secure animal to platform with a strip of hook-and-loop fastener placed firmly but not tightly around the nape.
   3. Hook the inactive VEP electrode around bottom incisors of anesthetized rat.
   4. Position the ERG inactive electrodes by encircling the scleral ring non-invasively around the eye's equator. Stabilize this by attaching electrodes to the hook-and-loop fastener strip around the nape. Repeat for the contralateral eye.
   5. Fasten VEP active electrodes by attaching alligator clips to stainless-steel screws pre-implanted on the skull.
   6. Place a small drop of 1% carboxymethylcellulose sodium on cornea prior to placement of the ERG active electrode to improve signal quality. Note: The viscose fluid also helps maintain corneal hydration throughout experimentation to minimize the formation of desiccation-type cataract in rodents⁷.
   7. Place a small drop of 1% carboxymethylcellulose sodium on the lower incisors to improve contact of the VEP inactive electrode and thus signal quality.
   8. Position the ERG active electrodes to lightly touch the central corneal surface using a micromanipulator attached to a custom-built stereotaxic arm.
   9. Insert 2 – 5 mm of the ground needle electrode (stainless steel) subcutaneously into the tail.
   10. If necessary dry any excess fluid from the inferior eyelid prior to recording to improve signal quality.
   11. Slide platform closer to the Ganzfeld bowl ensuring the animal's eyes align with the opening of the bowl to enable even illumination of both retinas (see step 2.4.1).
   12. Close the Faraday cage to reduce extraneous noise.
5. Data collection
   1. Use a dim test-flash (- 0.52 log cd.s.m^-2) to assess whether electrode placement is satisfactory². Under control conditions this would result in an ERG amplitude of ~ 800 μV and an inter-eye variability of no greater than 10%. If required reposition electrodes.
   2. Following the test-flash allow animals to dark-adapt for 10 min in complete darkness prior to recording.
   3. Present flashes of light stimuli using a Ganzfeld bowl whilst collecting ERG and VEP signals simultaneously over a ~ 500 msec time-window. Progress from dimmer to brighter light levels in order to maintain sufficient dark-adaptation for particular waveforms.
   4. Collect signals over a range of luminous energies to elicit STR, b-wave and a/b-wave waveforms of the ERG. Average more signals at the dimmer light levels (20 repeats) and less at the brighter luminous energies (1 repeat). Gradually lengthen the inter-stimulus interval from 1 to 180 sec from dimmest to the brightest light level. See Table 1 for an example protocol.
   5. To isolate the ERG rod and cone responses, utilize a paired-flash paradigm⁸. Initiate four flashes at 1.52 log cd.s.m^-2 with a 500 msec inter-stimulus interval² in-between. Digitally subtract the cone waveform (3^rd or 4^th flash) from the mixed waveform (1^st flash) to derive the putative rod response.
   6. To record VEP signals, average 20 repeats at the brighter luminous energies (i.e., – 0.52 to 1.52 log cd.s.m^-2, 5 sec inter-stimulus interval). Note that the first flash in this sequence returns the conventional dark-adapted ERG response.
   7. Allow 1 – 3 min for re-adaptation after (20) VEP sweeps before the next brighter ERG step, depending on the luminous energy.
   8. After completion of data collection, euthanize the anesthetized animal with intracardiac injection of pentobarbital sodium (325 mg/ml, 3 ml).

Table 1. ERG and VEP Recording Protocol Using a Range of Stimulus Energy. Stimulus presentations progress from dim (top) to bright (bottom) flashes, with sufficient inter-stimulus interval to ensure dark adaption. At the end of protocol, repetition of four flashes with short interval is presented to elicit the cone mediated response.

3. Analysis of ERG Waveforms 

Note: ERG and VEP analysis has been described in detail previously.^3,9,10 The following sections provide a brief overview.

1. Export signals in digital voltage-time format to a spreadsheet software for data analysis.
2. Rod photoreceptoral function
   1. Model the leading edge of the a-wave PIII with a delayed Gaussian (Equation 1)¹¹.
      PIII (i,t) = Rm_PIII∙ [1 – exp(-i∙ S (t – t_d)²)] for t > t_d (Equation 1)
   2. Optimize the fit over an ensemble of two brightest luminous energies^12,13 (i.e., 1.22 and 1.52 log cd.s.m^-2).
   3. Model up to 90% of the a-wave amplitude to avoid post-receptoral intrusions¹⁴.
      Note: The model returns the saturated amplitude (Rm_PIII, µV), sensitivity (S, m².cd^-1.s^-3) and delay (t_d, msec) of the photoreceptoral response.
3. Rod bipolar cell function
   1. Digitally subtract the PIII model (see above) from the mixed waveforms to return the mixed PII with overlying oscillatory potentials.
   2. To extract the rod PII from the mixed PII, digitally subtract the cone response (3^rd or 4^th flash at 1.52 log cd.s.m^-2) from the mixed PII (1^st flash at 1.52 log cd.s.m^-2).
   3. Then, apply a low-pass filter to the waveform (46.9 Hz, -3 dB, Blackman window) to remove oscillatory potentials. The remaining waveform is the rod PII response¹⁰.
   4. Extract the rod PII peak amplitude and plot it against all stimulus intensities (below -2 log cd.s.m^-2 and rod-isolated PII at 1.52 log cd.s.m^-2 )¹⁰.
   5. Model these data using a hyperbolic function (Equation 2), which provides a measure of inner retinal cell integrity.
      V (_i) = V_max (iⁿ/(iⁿ + kⁿ)) (Equation 2)
      Note: This equation returns maximal PII response (V_max, µV), 1/sensitivity (k,log cd.s.m-2) and the slope of the function (n)¹⁵.
4. Cone bipolar cell function
   Note: As the cone response is taken at a single intensity (1.52 log cd.s.m^-2) the amplitude and timing are retuned at this light level.
   1. Extract maximal cone PII response^2,16.
   2. Extract implicit time to which this maximal response corresponds^2,16.
5. Ganglion cell function
   1. As the STR is a small signal, apply a low pass filter with 50 Hz notch to waveform to eliminate high frequency and line noise (46.9 Hz, -3 dB, Blackman window).
   2. Extract maximal pSTR response^3,17.
   3. Extract implicit time to which this maximal response corresponds^3,17.

4. Analysis of VEP Waveforms

1. Extract maximum and minimum components of the VEP (P1, N1 and P2). For detail see references^3,6.
2. Express amplitudes as trough-to-peak amplitudes from their preceding peak or trough (P1N1 and N1P2)^3,6.
3. Extract implicit time (_it) to which this maximal response corresponds (P1_it, N1_it, P2_it)^3,6.