289

Naoyuki Tanimoto (ed.), Mouse Retinal Phenotyping: Methods and Protocols, Methods in Molecular Biology, vol. 1753,https://doi.org/10.1007/978-1-4939-7720-8_20, © Springer Science+Business Media, LLC, part of Springer Nature 2018

Chapter 20

Functional Assessment of Melanopsin-Driven Light Responses in the Mouse: Multielectrode Array Recordings

Shi-Jun Weng, Jordan M. Renna, Wei-Yi Chen, and Xiong-Li Yang

Abstract

Intrinsically photosensitive retinal ganglion cells (ipRGCs) are a special subset of retinal output neurons capable of detecting and responding to light via a unique photopigment called melanopsin. Melanopsin activation is essential to a wide array of physiological functions, especially to those related to non-image- forming vision. Since ipRGCs only constitute a very small proportion of retinal ganglion cells, targeted recording of melanopsin-driven responses used to be a big challenge to vision researchers. Multielectrode array (MEA) recording provides a noninvasive, high throughput method to monitor melanopsin-driven responses. When synaptic inputs from rod/cone photoreceptors are silenced with glutamatergic blockers, extracellular electric signals derived from melanopsin activation can be recorded from multiple ipRGCs simultaneously by tens of microelectrodes aligned in an array. In this chapter we describe how our labs have approached MEA recording of melanopsin-driven light responses in adult mouse retinas. Instruments, tools and chemical reagents routinely used for setting up a successful MEA recording are listed, and a standard experimental procedure is provided. The implementation of this technique offers a useful paradigm that can be used to conduct functional assessments of ipRGCs and NIF vision.

Key words Intrinsically photosensitive retinal ganglion cells, Melanopsin, Multielectrode array, Mouse, Retina

1 Introduction

Intrinsically photosensitive retinal ganglion cells (ipRGCs) send their axons to non-image-forming (NIF) visual centers [1, 2] and drive numerous biological activities associated with ambient light detection, such as circadian rhythms, pupillary light reflex and nocturnal suppression of pineal melatonin release [3, 4]. Distinguished from conventional retinal ganglion cells, ipRGCs express a novel photopigment called melanopsin, thus being able to signal light directly [5, 6]. There is evidence that knocking out the melanopsin protein in mice with rod-cone dysfunction results in complete loss of NIF visual responses [7, 8]. Furthermore, ani-mals deficient in melanopsin have been reported to exhibit abnor-malities in various functions, including diurnal variation of

290

cone-based electroretinography [9], sleep homeostasis [10, 11], light modulation of retinal wave burst duration [12], and light- induced circadian phase shifting [13]. All these evidence suggest that melanopsin-driven responses play an important role in a wide variety of ipRGC-mediated functions. Assessing responses driven by this photopigment will aid in the exploration of the both ipRGC function and NIF vision.

Since ipRGCs only constitute a small fraction of the total gan-glion cell population (approximately 10%), a focused functional study on these neurons could require targeted recording tech-niques. Single-cell patch clamp recording on genetically or retro-gradely labeled ipRGCs is a common method for monitoring melanopsin- driven responses [14, 15]. While detailed intracellular response properties can be gleaned from this type of recording, the laborious nature of this technique significantly decreases the number of cells that can be examined in a single experiment. Additionally, patch clamp recordings are more invasive, which usually induce response run down over longer recordings (30–60 min). The tech-nique of calcium imaging provides a relatively noninvasive alterna-tive with the added benefit of recording multiple ipRGCs simultaneously [16]. Technical considerations regarding lower temporal resolution and the inability to directly measure ganglion cell action potentials, the critical component for signaling between the eye and brain, still remain to be fully resolved.

Multielectrode array (MEA) recordings on isolated whole- mount retinal preparations (Note: “MEA” is also an abbreviation for “microelectrode array”; in most circumstances in retinal stud-ies, these two terms are interchangeable) provides a noninvasive, high- throughput method for the assessment of ipRGC physiologi-cal response properties with high temporal resolution. When syn-aptic inputs from rod/cone photoreceptors are silenced with glutamatergic blockers melanopsin-mediated ipRGC action poten-tials (spikes) can be isolated extracellularly and recorded simultane-ously in a large scale from tens or even hundreds of electrodes integrated in a single array. This technique does not require geneti-cally labeling ipRGCs with a fluorescent reporter for visually tar-geted recordings and it avoids the use of epifluorescence light prior to data collection, two common issues with the patch clamp method of recording from ipRGCs. Using this method, some functional characteristics of ipRGCs, such as its heterogeneity in response characteristics [17], circadian modulation [18], and synaptic influence from the primary rod pathway [19] have been demon-strated. Here we describe an MEA protocol designed and implemented in the authors’ laboratories with high reliability for recording light-evoked, melanopsin-driven ipRGCs responses in an isolated adult mouse retina.

Shi-Jun Weng et al.

291

2 Materials

See Fig. 1 for a detailed schematic demonstrating the MEA recording setup described in this protocol.

C57BL/6 mice 4–10 weeks of age were used (see Note 1). They were raised in a 12-h light/dark cycle (light on at 8:00 am) with ad libitum access to food and water. Illumination was provided by cool white fluorescent bulbs, which produced an ambient illumina-tion of approximately 200 lux. Use and handling of animals were strictly in accordance with the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research and were approved by Institutional Animal Care and Use Committees of Fudan University and the University of Akron.

2.1 Animals

Fig. 1 A schematic demonstrating the setup for MEA recordings. (a) MEA1060-Inv-BC preamplifier (headstage, MCS); (b) FA60SBC filter amplifier (MCS); (c) USB-ME64 main unit (A-D converter, MCS); (d) PS40W power supply (MCS); (e) 60MEA200/30iR-ITO-gr MEA chip (MCS); (f) MPIOH-T magnetic perfusion port holder (ALA); (g) MPIOH-S magnetic perfusion port holder (ALA); (h) MMP coated steel magnetic mounting plate (ALA); (i) Ames’ medium container (Sigma); (j) Ismatec peristaltic pump (Cole-Parmer); (k) SH-27B in-line solution heater (Warner); (l) TC-324B perfusion heat controller (Warner); (m) TA-29 thermistor (Warner); (n) mechanical micromanipulator (Narishige); (o) ScopeLED LED illuminator (DiCon); (p) FO-6000FILT inline filter holder adapter (WPI); (q) 6593T7 Fiber-Optic cable (McMater-Carr); (r) SYS-A310 pulse generator/stimulator (WPI); (s) Dell personal computer

MEA Recording of Melanopsin-Based Photoresponses

292

1. The MEA recording system used in this protocol is a commercial available, USB-MEA60-Inv-BC-System (Multi Channel Systems MCS GmbH, Reutlingen, BW, Germany) (see Note 2). This system includes four major components:

(a) USB-ME64 main unit, which contains the data acquisition and analog-digital converter board.

(b) MEA1060-Inv-BC preamplifier (headstage). (c) FA60SBC filter amplifier. (d) PS40W external power supply. 2. The type of MEA chip used in this protocol is 60MEA200/30iR-

ITO- gr (MCS) (see Notes 3 and 4). This “standard” type has 60 TiN (titanium nitride) electrodes of 30 μm in diameter aligned in an 8 × 8 grid layout, with interelectrode distances of 200 μm. It provides an internal reference (iR) electrode (elec-trode 15), a glass ring (gr) in height of 6 mm to constitute a recording chamber, with transparent contact pads and tracks made of indium tin oxide (ITO).

1. Ismatec 78023-00 peristaltic pump (Cole-Parmer, Vernon Hills, IL, USA) equipped with one 2-stop 07616-40 tubing (inside diameter 1.85 mm, for solution inflow) and one 2-stop 07616-48 tubing (inside diameter 2.79 mm, for solution outflow).

2. MPIOH-T magnetic perfusion port holder with flexible Teflon tubing (for inflow, ALA Scientific, Farmingdale, NY, USA).

3. MPIOH-S magnetic perfusion port holder with stainless steel cannula (for outflow, ALA).

4. MMP coated steel magnetic mounting plate for positioning MPIOH-T and MPIOH-S (ALA), which is mounted on the top of the lid of the MEA preamplifier.

5. PVC tubing of 1/16 in. inside diameter and 1/8 in. outside diameter (8000-0004, Thermo Fisher Scientific, Waltham, MA, USA).

6. Three-port manifold (30600-41, Cole-Parmer). 7. Luer valve assortment kit (14011, World Precision Instruments,

Sarasota, FL, USA), which provides all the connectors neces-sary to fit the tubing/cannula described in this protocol.

8. Perfusion heat controller (TC-324B, Warner Instruments, Hamden, CT, USA).

9. Rapid flow inline solution heater (SH-27B, Warner). 10. Thermistor with BNC end compatible with TC-324B (TA-29,

Warner).

2.2 MEA System

2.3 Perfusion System

Shi-Jun Weng et al.

293

1. White-light LED illuminator (ScopeLED, DiCon Fiberoptics, Richmond, CA, USA).

2. Fiber-Optic cable (6593T7, McMater-Carr, Elmhurst, IL, USA). 3. 480 nm narrow band interference filter (OS095524, Chroma

Technology, Bellows Falls, VT, USA). 4. Neutral density filter kit (66155, Edmund Optics Inc.,

Barrington, NJ, USA). 5. Inline filter holder adapter (FO-6000FILT, World Precision

Instruments) for coupling (items 1–4). 6. Mechanical micromanipulator (Narishige, Amityville, NY,

USA) for holding and adjusting the position of the fiber-Optic cable output end.

7. Pulse generator/stimulator (SYS-A310, World Precision Instruments) for triggering the illuminator and sending syn-chronized signals to the MEA system.

1. Ames’ medium (A1420, Sigma, St. Louis, MO, USA). It should be freshly made and always saturated in 95% O2 + 5% CO2.

2. Glutamatergic synaptic blockers (All three reagents are freshly prepared from 1: 1000 aqueous stock solutions by dilution with Ames’ medium.):

(a) L-(+)-2-amino-4-phosphonobutyric acid (L-AP4, 0103, Tocris Bioscience, Bristol, UK) at 50 μM.

(b) 6,7-dinitroquinoxaline-2,3-dione disodium salt (DNQX, 2312, Tocris) at 40 μM.

(c) D-(−)-2-amino-5-phosphonopentanoic acid (D-AP5, 0106, Tocris) at 30 μM.

1. Stereomicroscope (SZ61TRC-SET, Olympus Corporation, Shinjuku, Tokyo, Japan).

2. Surgical spring scissors (15025-10, Fine Science Tools, Foster City, CA, USA).

3. Surgical “#55” fine forceps (11255-20, Fine Science Tools). 4. Filter membrane carriers (Anodisc 25, GE Healthcare Bio-

Sciences, Pittsburgh, PA, USA). 5. Stainless steel ring (a “washer,” inner diameter 6 mm, outer

diameter 12 mm, height 1 mm, weight 1 g). 6. Red headlamp LED (XL-2003; Xuelang Illumination, Ningbo,

Zhejiang, China) or narrow band (660 nm) LED (LED660N-03, Roithner Lasertechnik GmbH, Vienna, Austria) that provides dim red illumination.

2.4 Light Stimulation System

2.5 Chemicals

2.6 Dissecting/Mounting Retinas

MEA Recording of Melanopsin-Based Photoresponses

294

1. MC_Rack (Multi Channel Systems). 2. MEA_Select (Multi Channel Systems) (see Note 5).

1. Offline Sorter (Plexon Inc., Dallas, TX, USA). 2. MC_DataTool (Multi Channel Systems). (see Note 6) 3. Clampfit (Molecular Devices, Sunnyvale, CA, USA). 4. NeuroExplorer (Nex Technologies, Madison, AL, USA). 5. OriginPro 2015 (OriginLab Corp., Northampton, MA, USA). 6. Microsoft Excel (Microsoft Corporation, Redmond, WA,

USA).

3 Methods

1. One day before the experiment dark adapt the mouse over-night (>12 h) (see Note 7). Meanwhile, check the perfusion system and the light stimulation system to make sure they are functioning properly. Doing a “practice” the day before allows you to trouble shoot electrical noise and check flow rates with-out valuable tissue in the recording chamber.

2. Cut the Anodisc filter membrane into squares of approxi-mately 8 mm × 8 mm using a razor blade.

3. Under dim red light provided by a headlamp LED, sacrifice the animal (see Note 8).

4. Enucleate one eye immediately (see Note 9), make a small cut on the cornea using a 26 gauge needle, and quickly transfer the eye into a 10 cm Petri dish filled with Ames’ medium, which should be equilibrated with 95% O2 and 5% CO2.

5. Under a stereo microscope, remove the cornea and the lens using a pair of spring scissor. The illuminator of the stereo microscope should be covered by a red filter membrane to generate red light, and its output should be minimized.

6. Using two pairs of #55 fine forceps, dissect the retina from the eye cup, and remove as much vitreous as possible (see Note 10).

7. To help flattening the retina, make 4 very small (0.5–1 mm) radial cuts in the isolated retina.

8. Mount and flatten the retina with the photoreceptor side down, on a small piece of Anodisc filter membrane prepared in [2].

9. Rivet the retina onto the filter membrane by pressing the edges of the retina with the forceps.

10. Dry the retina-free part of the filter membrane with a small piece of Kimwipes wiper (see Note 11).

2.7 Software for Data Recording

2.8 Software for Off Line Spike Sorting and Data Analysis

3.1 Preparing Retinal Wholemounts

Shi-Jun Weng et al.

295

1. Transfer the retinal wholemount, which is now attached to a filter membrane, to the recording chamber on an MEA chip, by grabbing a corner of the filter membrane with forceps. Make sure that the ganglion cell side is not only facing the chamber but touching the central area of the chamber where the electrode array is located.

2. Anchor the retinal preparation with a stainless steel ring. Make sure that the ring only covers the retina-free part of the prepara-tion so that the retina is exposed (inside the aperture of the ring).

3. Fill the recording chamber with fresh, oxygenated Ames’ medium.

1. Place the MEA chip inside the trough of the base of the MEA preamplifier. Double check whether the MEA chip is correctly oriented: the wedge-shaped mark for internal reference elec-trode (electrode 15) should be on the left (see Note 12).

2. Place the lid of the MEA preamplifier onto the base and close it gently. This should be done very carefully. Before pressing the lid to close, make sure that all four small rods on the four corners of the lid are now inside the four holes on the base, otherwise, the MEA chip or contact pins of the preamplifier might be damaged.

1. Place the two magnetic perfusion port holders, one holding the inflow tubing and the other holding the suction cannula, onto the magnetic mounting plate of the MEA preamplifier.

2. Adjust the magnetic perfusion port holders, so that the open-ing of the inflow tubing is at the edge and on the bottom of the chamber, and the opening of the suction cannula is at the edge (but at far side relative to the inflow tubing) and right on the dam of the chamber.

3. Turn on the peristaltic pump and set the perfusion speed at approximately 4 mL/min.

4. Turn on the heat controller. 5. Place the thermistor probe inside the recording chamber. 6. Maintain the temperature of the solution at 32 ± 2 °C (at the

level of the tissue).

1. Start the MC-Rack software. 2. Add a Recorder to the virtual rack. 3. To monitor spike activities, add a Filter to the Recorder.

Choose High Pass with a Cutoff of 200–300 Hz. 4. Add a Data Display to the virtual rack. Consider adding a

Longterm Data Display if the recording would last for minutes. 5. Select the option Peak Detection in the Display.

3.2 Transferring and Anchoring Retinal Preparations to MEA Chips

3.3 Mounting the MEA Chips Inside the Preamplifier

3.4 Switching on the Perfusion System

3.5 Setting Up Recording Software

MEA Recording of Melanopsin-Based Photoresponses

296

6. Set sampling frequency to 25 kHz (can be reduced to 10 kHz in case of long-term recording).

7. Set the display X-axis to 1000 ms and Y-axis to 50 μV. The Y-axis can be set to a larger value later when the spike ampli-tude is increased.

8. You can save this virtual rack configuration (an .rck file) for future use by click Save As on the File menu.

1. Hit the “Start” button in MC-Rack interface to roughly exam-ine the recording quality based on the pattern of spontaneous spiking activities.

2. If spontaneous spiking activities (see Fig. 2) with good signal–noise ratio (>5:1 judged by visual observation) can be imme-diately observed on >10 channels, it is highly likely that recording quality is ideal, and there probably will be more channels showing spiking activities afterward. However, more frequently, there are only very few or no channels exhibiting spontaneous activities at the very beginning of the experiment. In this case, wait for 15 min.

3. In most circumstances, during this 15-min period, spontane-ous activities will appear in more and more channels and the experiment can proceed to the next stage. If it is not the case, stop the experiment, discard the retina, clean the array, and repeat the process with a new retina.

1. Position the fiber-optic cable by adjusting the Narishige manipulator so that the output end of the cable is directly over the electrode array area with a distance of 1.2 cm.

2. Secure the chamber from any ambient light and allow the ret-ina sample time to completely settle onto the array (roughly 45 min). This increases the signal-to-noise ratio and allows for a more stable recording. During this period, probing the ret-ina with a brief, relatively dim light stimulation will be enough to induce rod/cone-driven light responses on most of the MEA channels with various characteristics (ON and OFF, transient and sustained, see Fig. 3 top panel).

3. Use the manifold to switch the perfusion solution from nor-mal Ames’ medium to Ames’ medium containing glutamater-gic blockers (50 μM L-AP4, 40 μM DNQX, and 30 μM D-AP5) (see Note 13).

4. Apply the blocker cocktail on the retina for 15 min to achieve a complete blockade of glutamatergic transmission. On most MEA channels the spontaneous activity will gradually decline. A few channels may exhibit highly rhythmic discharges which

3.6 Starting MEA Recording

3.7 Pharmacological Isolation of Melanopsin-Driven Responses and Light Stimulation

Shi-Jun Weng et al.

297

correlate to certain OFF-type retinal ganglion cells [20]. When the retina is stimulated with light pulses, although on most channels there will be no light-evoked discharges, slug-gish and persistent light-evoked responses will appear on a few channels (see Fig. 3 bottom panel). These are melanopsin-mediated responses from ipRGCs (see Note 14). It should be noted that in order to elicit these responses a relatively brighter pulse (>1011 photons/cm2/s) will be needed.

Probe the retina with a series of 10 s, 480 nm full-field light flash (2 × 1011–2 × 1015 photons/cm2/s) generated by an LED illuminator and delivered onto the retina by a fiber-optic cable. Adjust the light intensity by introducing neutral density filters into the light path. Stimuli should be presented in a series that is monotonically ascending in intensity and identical for all experiments. Inter-stimulus intervals are advised to increase progressively within the series, ranging from 5 min between dim stimuli to 10 min between the brightest ones, which is designed to allow for substantial recovery from the previous stimulus while reducing the total time needed to complete the series so as to avoid response rundown. For each intensity, the recording lasts for 70 s, including 10 s before (baseline activities) and 60 s after stimulus beginning (see Fig. 4).

3.8 Collecting Data for Building Up the Irradiance–Response (I–R) Curves

Fig. 2 A representative screenshot of MC_Rack during an MEA recording in darkness of spontaneous ganglion cell spiking in an adult mouse retina. The amplitude of each spike is significantly larger than the baseline noise (high signal–noise ratio) on most of the channels. Note that on channel 15 (the internal reference electrode) the baseline trace is much thinner than those on other channels

MEA Recording of Melanopsin-Based Photoresponses

298

1. Set the spike detection threshold for each channel at 3–4 times the standard deviation of the voltage (either negative or posi-tive to the baseline) (see Note 15).

2. Run cluster analysis for all the detected spike waveforms on each channel using the first three principal components, with one of the algorithms provided by the software. In most cir-cumstances, two fully automatic algorithms, T-Dist E-M and Valley Seek, are very efficient and reliable, and are recom-mended to use.

3. For the resulting clusters, perform manual correction for clus-tering errors. Most importantly, check for evidence of the expected refractory period, and any cluster in which >3% spikes

3.9 Offline Spike Sorting with Offline Sorter Software

Fig. 3 Pharmacological isolation of melanopisn-driven responses. Each recording is derived from one of the electrodes in the MEA. The grey bars represent darkness, while the white bars represent light stimulation. The two panels are from the same retinal preparation. Top, in normal Ames’ medium, with intact synaptic input, relatively dim light stimulation evoked responses on many channels. These were brisk and varied in forms: ON or OFF; transient or sustained. Bottom, when glutamatergic transmission was pharmacologically blocked, activities on most channels was silenced. However, light-evoked responses which were sluggish and persistent could still be recorded on some channels. These were melanopsin-driven responses from ipRGCs. Note that to clearly reveal these responses a brighter light pulse with longer duration is recommended

Shi-Jun Weng et al.

299

had inter-spike intervals <1 ms should be excluded from fur-ther analysis.

4. Use the timestamps of all spikes from each resulting cluster to reconstruct single-unit spike trains.

5. Process the spike trains with software such as NeuroExplorer, OriginPro, and Microsoft Excel to generate raster plots (see Fig. 5) and to further quantify I–R relationships.

4 Notes

1. Although in “Materials” section the age of animals is stated to be 4–10 weeks, this protocol actually also applies to both postnatal (before eye opening) and much older (>12-month- old) mice with slight adaptations. However, it is important that animals used for a set of experiments should always be of the same age as the physiological response properties of ipRGCs change throughout development and into the mature retina [21–23].

2. Between experiments, to avoid contamination/damage of contact pins of the MEA preamplifier, always place it onto a dry, clean, soft area, e.g., a medical adult diaper, with the bot-tom side down.

Fig. 4 Irradiance–response characteristics of melanopsin-driven discharges revealed by MEA recording. The two grey bars represent darkness, while the white bar represents light stimulation (10 s, 480 nm full-field light). The five traces are from the same ipRGC, but with ascending light stimulation intensities (2 × 1011 photons/cm2/s for the first trace, with a 1-log unit increment between each trace). The response amplitude increased as a function of light intensity to the point of saturation. Note that for the lowest light intensity the response was so sluggish that the spiking activities appeared well after the termination of the light stimulus

MEA Recording of Melanopsin-Based Photoresponses

300

3. Cleaning the MEA chips appropriately is vital to achieve a successful recording. Immediately after one recording, rinse the chip, especially the surface where the retina was mounted, with hot tap water for 5 min. If there is still visible residue on the chip, place it in an ultrasonic cleaner for 3–5 min for a deeper cleaning. Sometimes cleaning effect can be improved with Terg-A-Zyme detergent. A detailed protocol of MEA cleaning with Terg-A-Zyme can be found in MCS’s MEA Manual.

4. Prior to recording clean the contact pins of the preamplifier as well as the contact pads of the MEA chip with 100% ethanol using a Q-tip. This will almost completely abolish the record-ing noise caused by the contamination of contact pins and pads.

5. When one experiment is done, always click Change MEA in the MEA_Select program before opening the preamplifier. Moreover, always deactivate the Change MEA mode after the preamplifier is reclosed for another experiment. Otherwise, it might take a long time before the amplifier reaches a stable state and is ready for use.

6. Raw data (.mck files) can also be converted into. ABF files with MC_DataTool software and then viewed with Clampfit

Fig. 5 Representative raster plots of melanopsin-driven responses. Spike timestamps from 10 ipRGCs in the same retina, obtained by offline spike sorting, were further processed to generate this plot. Note that these cells exhibited different response characteristics (especially response latency and persistence), suggesting that different subsets of ipRGCs were recorded

Shi-Jun Weng et al.

301

software, and further processed with other software like OriginPro.

7. Recent lighting history affects light response amplitude and ipRGCs are reported to possess a relatively slower response in dark adaption as compared to rod/cone photoreceptors [24]. Consequently, an overnight dark adaption is preferred when performing quantitative assessment of ipRGC light responses.

8. Melanopsin expression levels and melanopsin-driven light responses are nonlinear through a 24 h cycle but under circadian modulation [18, 25]. Therefore, in one set of experiments all recording should be carried out in a fixed circadian period, e.g., CT12-CT18.

9. To minimize response run down and maximize consistence among experiments, only one eye from each animal is used for MEA experiments.

10. It is essential to thoroughly remove the vitreous body, other-wise the recording quality will be greatly compromised. Failure to remove the vitreous will significantly decrease signal- to- noise ratio. Additionally, the vitreous act as a barrier between the Ames’ medium and the ganglion cell layer. When retinal ganglion cells fail to get oxygen and nutrients the tissue will run down over the course of the recording. Usually this mani-fests itself as extremely high firing rates on multiple channels.

11. Drying the filter membrane carrier (gently and quickly touch-ing the retina-free side of the filter membrane with a small piece of Kimwipes) before mounting it on the MEA chip is an important step. This can not only help anchoring the retina onto the filter membrane, but also avoid excessive solution drop which may contaminate the contact pads of the MEA chip. However, make sure that the Kimwipes tissue does not directly touch the retina.

12. While using an MEA chip with internal reference electrode, make sure that electrode 15 is connected to ground. This can be done by connecting the ground to the reference electrode socket (lower left corner) with a cable connector provided by MCS.

13. The glutamatergic blocker cocktail should always be bubbled with 95% O2 + 5% CO2, even before it is applied to the retina. The cocktail can be recycled by building up a close loop in the solution perfusion system.

14. There are at least five subtypes of melanopsin-expressing reti-nal ganglion cells (M1–M5) [14, 26, 27]. M1–M3 have higher expression levels of melanopsin, whereas M4 and M5 have lower levels. Physiological response properties of individual spike trains (background firing rate, latency to onset, latency

MEA Recording of Melanopsin-Based Photoresponses

302

to peak response, etc.) may allow us to identify specific ipRGC subtypes on the MEA [28]. For example, three groups of ipRGCs have been identified on the MEA in the developing mouse retina [17].

15. Setting the spike detection threshold to a negative standard deviation value will bias the detection for a biphasic wave form (action potential) originating from a soma, whereas selecting a positive standard deviation value will bias the detection for a triphasic waveform traveling down an axon [29–31].

Acknowledgments

The research of the authors is supported by grants from the National Natural Science Foundation of China (31571072, 31100796, 31571075, 31171005, 31421091, 81790640 and 81430007); the Ministry of Science and Technology of China (2011CB504602 and 2015AA020512); NIH R15 EY026255 and the Karl Kirchgessner Foundation.

References

1. Hattar S, Kumar M, Park A et al (2006) Central projections of melanopsin-expressing retinal ganglion cells in the mouse. J Comp Neurol 497:326–349

2. Gooley JJ, Lu J, Fischer D et al (2003) A broad role for melanopsin in nonvisual photorecep-tion. J Neurosci 23:7093–7106

3. Berson DM (2003) Strange vision: ganglion cells as circadian photoreceptors. Trends Neurosci 26:314–320

4. Schmidt TM, Chen SK, Hattar S (2011) Intrinsically photosensitive retinal ganglion cells: many subtypes, diverse functions. Trends Neurosci 34:572–580

5. Hattar S, Liao HW, Takao M et al (2002) Melanopsin-containing retinal ganglion cells: architecture, projections, and intrinsic photo-sensitivity. Science 295:1065–1070

6. Berson DM, Dunn FA, Takao M (2002) Phototransduction by retinal ganglion cells that set the circadian clock. Science 295:1070–1073

7. Hattar S, Lucas RJ, Mrosovsky N et al (2003) Melanopsin and rod-cone photoreceptive sys-tems account for all major accessory visual functions in mice. Nature 424:76–81

8. Panda S, Provencio I, Tu DC et al (2003) Melanopsin is required for non-image-form-ing photic responses in blind mice. Science 301:525–527

9. Barnard AR, Hattar S, Hankins MW et al (2006) Melanopsin regulates visual processing in the mouse retina. Curr Biol 16:389–395

10. Lupi D, Oster H, Thompson S et al (2008) The acute light-induction of sleep is mediated by OPN4-based photoreception. Nat Neurosci 11:1068–1073

11. Tsai JW, Hannibal J, Hagiwara G et al (2009) Melanopsin as a sleep modulator: circadian gat-ing of the direct effects of light on sleep and altered sleep homeostasis in Opn4(−/−) mice. PLoS Biol 7:e1000125

12. Renna JM, Weng S, Berson DM (2011) Light acts through melanopsin to alter retinal waves and segregation of retinogeniculate afferents. Nat Neurosci 14:827–829

13. Panda S, Sato TK, Castrucci AM et al (2002) Melanopsin (Opn4) requirement for normal light-induced circadian phase shifting. Science 298:2213–2216

14. Ecker JL, Dumitrescu ON, Wong KY et al (2010) Melanopsin-expressing retinal gan-glion-cell photoreceptors: cellular diversity and role in pattern vision. Neuron 67:49–60

15. Wong KY, Dunn FA, Graham DM et al (2007) Synaptic influences on rat ganglion-cell photo-receptors. J Physiol 582:279–296

16. Sekaran S, Lupi D, Jones SL et al (2005) Melanopsin-dependent photoreception pro-

Shi-Jun Weng et al.

303

vides earliest light detection in the mammalian retina. Curr Biol 15:1099–1107

17. Tu DC, Zhang D, Demas J et al (2005) Physiologic diversity and development of intrinsically photosensitive retinal ganglion cells. Neuron 48:987–999

18. Weng S, Wong KY, Berson DM (2009) Circadian modulation of melanopsin- driven light response in rat ganglion-cell photorecep-tors. J Biol Rhythm 24:391–402

19. Weng S, Estevez ME, Berson DM (2013) Mouse ganglion-cell photoreceptors are driven by the most sensitive rod pathway and by both types of cones. PLoS One 8:e66480

20. Margolis DJ, Detwiler PB (2007) Different mechanisms generate maintained activity in ON and OFF retinal ganglion cells. J Neurosci 27:5994–6005

21. Lax P, Esquiva G, Fuentes-Broto L et al (2016) Age-related changes in photosensitive melanopsin-expressing retinal ganglion cells correlate with circadian rhythm impairments in sighted and blind rats. Chronobiol Int 33:374–391

22. Schmidt TM, Taniguchi K, Kofuji P (2008) Intrinsic and extrinsic light responses in mela-nopsin-expressing ganglion cells during mouse development. J Neurophysiol 100:371–384

23. Sexton TJ, Bleckert A, Turner MH et al (2015) Type I intrinsically photosensitive retinal gan-glion cells of early post-natal development cor-respond to the M4 subtype. Neural Dev 10:17

24. Wong KY, Dunn FA, Berson DM (2005) Photoreceptor adaptation in intrinsically pho-tosensitive retinal ganglion cells. Neuron 48:1001–1010

25. Sakamoto K, Liu C, Tosini G (2004) Classical photoreceptors regulate melanop-sin mRNA levels in the rat retina. J Neurosci 24:9693–9697

26. Estevez ME, Fogerson PM, Ilardi MC et al (2012) Form and function of the M4 cell, an intrinsically photosensitive retinal ganglion cell type contributing to geniculocortical vision. J Neurosci 32:13608–13620

27. Schmidt TM, Alam NM, Chen S et al (2014) A role for melanopsin in alpha retinal gan-glion cells and contrast detection. Neuron 82:781–788

28. Zhao X, Stafford BK, Godin AL et al (2014) Photoresponse diversity among the five types of intrinsically photosensitive retinal ganglion cells. J Physiol 592:1619–1636

29. Henze DA, Borhegyi Z, Csicsvari J et al (2000) Intracellular features predicted by extracel-lular recordings in the hippocampus in vivo. J Neurophysiol 84:390–400

30. Marre O, Amodei D, Deshmukh N et al (2012) Mapping a complete neural population in the retina. J Neurosci 32:14859–14873

31. Segev R, Goodhouse J, Puchalla J et al (2004) Recording spikes from a large fraction of the ganglion cells in a retinal patch. Nat Neurosci 7:1154–1161

MEA Recording of Melanopsin-Based Photoresponses

本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具