Evaluation of focal photoreceptor degeneration using a multi-focal ERG

Background Photoreceptor degenerative diseases such as retinitis pigmentosa and age-related macular degeneration gradually proceed to photoreceptor degeneration. In animal models of photoreceptor degeneration, it is difficult to evaluate the loss of function in the retina, especially in a focal lesion. Methods To induce local lesion in rat retina, one microliter of N-Methyl-N-Nitrosourea (MNU) solution was injected subretinally into a temporal site of the retina. A week after the injection, multifocal electroretinogram (mfERG) was recorded to presume the degenerated site. Based on the data from mfERG, optical coherence tomography (OCT) was performed. After the OCT, eyes were fixed with 4% paraformaldehyde and the flat-mounted retinas were stained with peanut agglutinin (PNA) and wheat germ agglutinin (WGA). Following observation under a fluorescence microscope, the retina was embedded into an OCT compound and cryo-sections were obtained. The were detected Most the The OCT at the recorded a lower response of the data from and showed the severe destruction however, a structure was decreased number and degeneration in a rat by subretinal injection of N-methyl-N-nitrosourea (MNU) and tried to evaluate the focal retinal dysfunction by using an OCT and a multifocal ERG (mfERG) system. The recorded amplitudes of mf-ERGs in the temporal side


Abstract Background
Photoreceptor degenerative diseases such as retinitis pigmentosa and age-related macular degeneration gradually proceed to photoreceptor degeneration. In animal models of photoreceptor degeneration, it is difficult to evaluate the loss of function in the retina, especially in a focal lesion.

Methods
To induce local lesion in rat retina, one microliter of N-Methyl-N-Nitrosourea (MNU) solution was injected subretinally into a temporal site of the retina. A week after the injection, multifocal electroretinogram (mfERG) was recorded to presume the degenerated site. Based on the data from mfERG, optical coherence tomography (OCT) was performed. After the OCT, eyes were fixed with 4% paraformaldehyde and the flat-mounted retinas were stained with peanut agglutinin (PNA) and wheat germ agglutinin (WGA). Following observation under a fluorescence microscope, the retina was embedded into an OCT compound and cryo-sections were obtained.

Results
The decreased responses in some divisions of the retina were detected in the recording of mf-ERGs.
Most of the decreased responses were localized in the temporal part of the retina corresponding to the MNU injection site. The OCT images at the area recorded a lower response of the data from mfERG and showed the severe destruction of photoreceptors; however, a well-organized retinal structure was observed in other parts of the retina. PNA and WGA staining also showed a decreased number of cones and rods in the degenerated area.

Conclusions
Subretinal injection of MNU induced the local photoreceptor degeneration without effects on other retinal layers. mfERGs could detect the focal lesion, and it might be a very helpful indication to perform OCT. These methods would help the evaluation of the effects of regenerative medicine in rodents.

Background
Retinitis pigmentosa (RP) and age-related macular degeneration (AMD) are among the major retinal disorders that cause severe visual dysfunction (https://www.nei.nih.gov/learn-about-eye-health/eyeconditions-and-diseases). In RP, many gene mutations, most of them related to the phototransduction 3 pathways (https://sph.uth.edu/retnet/), have been identified. Generally, in the early phase of RP, it starts with the degeneration of rod cells and followed by gradual degeneration of cone cells. 1, 2 On the other hand, in AMD, cone-rod dystrophy and macular dystrophy, starts with the degeneration of In this study, we induced local photoreceptor degeneration in a rat by subretinal injection of Nmethyl-N-nitrosourea (MNU) and tried to evaluate the focal retinal dysfunction by using an OCT and a multifocal ERG (mfERG) system. The recorded amplitudes of mf-ERGs in the temporal side corresponding to the injection site were decreased, and the local photoreceptor degeneration was clearly shown by the OCT. These methodologies would be useful for evaluating the photoreceptor degeneration in animal models.

Methods Animals
All animal experiments were conducted following the guidelines of the Animal Experiment Committee of Iwate University, Japan. Wistar rats were obtained from CLEA Japan (Tokyo) and they were housed in a 12-h/12-h light/dark cycle with access to water ad libitum. Male 6-12 months old wistar rats (n = 16) were used in this study.

Optical Coherence Tomography (OCT)
Under the anesthetized condition, their pupils were dilated with tropicamide (Midrin-P, Santen Co., Ltd., Osaka, Japan). A purified sodium hyaluronate (Hyalein Mini ophthalmic solution Santen Co., Ltd., Osaka, Japan) was applied to the eye, to avoid dryness during the procedure of OCT. Image 5 acquisition of 1.1 mm length of the rat retina including the optic disk was performed using the line scan mode on an OCT imaging device equipped with a special ordered lens (RS-3000, NIDEK Co., Ltd., Aichi, Japan).

WGA-And PNA-staining Of Retinal Wholemount And Cryo-section
To evaluate the loss of cones and rods, the staining with peanut agglutinin (PNA), preferentially binds to cone photoreceptors, and wheat germ agglutinin (WGA), binds to rods, were performed. 16 After euthanasia with carbon dioxide, rat eyes were fixed with 4% paraformaldehyde in phosphate-buffered saline (PBS) overnight at 4 °C and retinal whole mounts were made following washing steps with PBS.
After taking photographs, the retinal specimen was washed with PBS, immersed in 10%, 20%, and 30% sucrose in PBS sequentially, and then embedded in an OCT compound (Sakura Finetek, Tokyo).
Ten micrometers of cryo-sections were obtained and observed using the fluorescence microscope.

Statistical analysis
Statistical analyses for in vitro experiments were performed using GraphPad Prism (MDF, Tokyo). The statistical method used was Tukey's multiple comparison test, Dunnett's multiple comparison test, and Student's t-test.

Recordings of mfERGs
Retinal dysfunction induced by the subretinal injection of MNU was clearly shown in the 3-D view in mfERGs ( Fig. 2A). A higher amplitude around the centre of the retina was recorded compared to the peripheral retina. After the injection of MNU, the center of the retina still had a higher amplitude; however, the decreased responses in the temporal side were recorded. When the highest and lowest 6 response from the pre-recorded value and the value of the post-MNU injection were compared, a significant difference was observed in between the lowest responses though the highest responses were not detected (Fig. 2B).

Optical Coherence Tomography
All the OCT images shown in Fig. 3 were of the same eye as known from the vessel travelling. The OCT image of the centre in the retina including the optic nerve clearly showed a normal retinal structure even after the MNU injection (Fig. 3A, B, C). On the other hand, the photoreceptor degeneration was observed in the temporal part of the retina that corresponded to the MNU injection site (Fig. 3D, E, F). Besides, the OCT image near the centre in the same scanning image (Fig. 3E) showed photoreceptors which survived (Fig. 3G). These results indicated that the area of photoreceptor degeneration induced by the subretinal injection of MNU was limited around the injection site. There was no difference in retinal thickness around the optic nerve in between the normal retina without MNU-injection and MNU-injected retina. However, in the MNU-injected site, the retinal thickness was significantly decreased (Fig. 3H).

WGA And PNA-staining Of Retinal Wholemount And Cryo-section
A wholemount retina showed that PNA and WGA-staining uniformly near the optic nerve ( Fig. 4A) but not at the site of subretinal injection of MNU (Fig. 4B). The vertical sections also showed that cones and rods corresponding to PNA and WGA-staining were kept around the optic nerve (Fig. 4C).
However, in the peripheral retina expected as the injection site, PNA and WGA-staining were poor and we also observed the decreased number of the photoreceptor cells (Fig. 4D). Bülach, Switzerland) a photosensitizing dye that is used clinically, without any adverse effects, for patients with subfoveal choroidal neovascularization (CNV), caused the photoreceptor degeneration within the light-exposed area. The latter method is thought to be useful for creating a focal lesion.

Discussion
However, this is difficult to apply in case of rodents as it needs exposure to a focal toxic light.
Recordings of mfERGs showed a decreased response after the subretinal injection of MNU. The 3-D view indicated that the responses of the central retina were higher than those of the peripheral retina ( Fig. 2A). The macula has a lot of cones in humans, does not exist in rodents. However, the population of the cones in rodent's retina is also high in the centre of the retina. 21 The responses of the central retina might include the blind spot brought by the optic nerve because the rat eye was set at the centre of the screen to record the mfERGs. However, we could not detect the blind spot on the 3Dview of mfERG. If an ocular axial length assumes 7 mm in rat, the stimulus area is estimated to be 38.465 mm 2 . This stimulus area is divided into 37 divisions and a division corresponds to 1.04 mm 2 .
The division might be too rough to detect the blind spot. Therefore, the blind spot by the optic nerve might be averaged by the high population of cones around the optic nerve. The strongest point of our mfERG system is to record easily, on the other hand, the weak point is that the recorded area does not always correspond to the same area if you rearrange a rat body setting on the recording tray. 8 After the subretinal injection of MNU, the decreased response at the temporal area was observed ( Fig. 2A). The different part of the retina that corresponding to the higher or lower responses in the mfERG also showed a different pattern of PNA and WGA-staining (Fig. 4) and clearly showed the loss of rods and cones.

Conclusions
The present study suggested a useful method to evaluate the focal lesion in the retina by using a combination of the OCT and newly developed mfERG system and to induce a focal lesion by the

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

9
The authors declare that they have no competing interests  Overview of multifocal ERG system. Flashing of a hexagon is randomly presented on the screen divided into 37 areas. The screen was set at 190 mm of distance from the rat eye and the screen wide was 138 mm. The position of the eye was exactly set at the centre of the screen. The condition corresponded to the visual angle of 45 degrees.