Dark-reared rd10 mice experience rapid photoreceptor degeneration with short exposure to room-light during in vivo retinal imaging

Abstract

Inherited retinal diseases (IRDs) are a collection of rare genetic conditions, which can lead to complete blindness. A large number of causative genes have been identified for IRDs and while some success has been achieved with gene therapies, they are limited in scope to each individual gene and/or the specific mutation harbored by each patient with an IRD. Multiple studies are underway to elucidate common underlying mechanisms contributing to photoreceptor (PR) loss and to design gene-agnostic, pan-disease therapeutics. The rd10 mouse, which recapitulates slow degeneration of PRs, is an in vivo IRD model used commonly by vision researchers. Light deprivation by rearing animals in complete darkness significantly delays PR death in rd10 mice, subsequently increasing the time window for in vivo studies investigating neuroprotective strategies. Longitudinal in vivo retinal imaging following the same rd10 mice over time is a potential solution for reducing the number of animals required to complete a study. We describe a previously unreported phenotype in the dark-reared rd10 model that is characterized by dramatic PR degeneration following brief exposure to low-intensity light. This exquisite light sensitivity precludes the use of longitudinal studies employing in vivo imaging or other functional assessment requiring room light in rd10 mice and highlights the importance of closely following animal models of IRD to determine any deviations from the expected degeneration curve during routine experimentation.

Short Communication

Inherited retinal diseases (IRDs) encompass a large group of disorders, which are generally characterized by dysfunction of photoreceptors (PRs), leading to retinal degeneration and vision loss (Cremers et al., 2018). The etiology of IRDs is complex with at least 27 different forms being described, and over 300 causative genes or loci having been identified to date (Retnet, https://sph.uth.edu/retnet/, accessed 06/15/2021). Retinitis pigmentosa (RP) is the most common form of IRD, with over 80 genes and loci shown to be causative for isolated RP (Verbakel et al., 2018). An estimated 1 in 4000 people are affected world-wide by IRD, leading to significant societal and economic burden (Burton et al., 2021). Novel neuroprotective strategies that delay or prevent the PR loss underlying IRD will be of tremendous value to preserve vision in patients with IRD.

The rd10 mouse is a widely used, in vivo, IRD model in vision research (Table 1). This spontaneous mutant mouse line carries an autosomal recessive missense mutation in the Pde6b gene (c1678C>T, p.Arg560Cys) (Chang et al., 2007). When raised in 12/12-hour light/dark cycles, mice homozygous for this mutation demonstrate PR degeneration starting at approximately post-natal day 18 (p18) with near complete loss of rods by p30, and complete secondary cone degeneration by p60 (Chang et al., 2007; Hasegawa et al., 2016). The loss of most PRs by about the 5^th week of life limits the therapeutic window for screening neuroprotective interventions in rd10 mice. Interestingly, rd10 animals raised in 24-hour darkness display a reduced rate of PR degeneration and maintain a significant number of rods up to p60 (Cronin et al., 2012). The slower rate of degeneration as a result of dark-rearing increases the potential feasibility of performing longitudinal, non-terminal experiments, such as in vivo retinal imaging, to determine PR survival, which can reduce the number of experimental animals required.

Table 1 –

Review of literature

#	Paper Title	DOI	First and Last Author	Year	LE	OCT	ERG
1	Two mouse retinal degenerations caused by missense mutations in the beta-subunit of rod cGMP phosphodiesterase gene	10.1016/j.visres.2006.11.020	Chang and Boatright	2007	No	No	Yes
2	AAV-mediated gene therapy for retinal degeneration in the rd10 mouse containing a recessive PDEbeta mutation	10.1167/iovs.07-1622	Pang and Hauswirth	2008	Yes	No	Yes
3	AAV-mediated gene replacement, either alone or in combination with physical and pharmacological agents, results in partial and transient protection from photoreceptor degeneration associated with betaPDE deficiency	10.1167/iovs.10-6269	Allocca and Auricchio	2011	Yes	No	Yes
4	Caspase inhibition with XIAP as an adjunct to AAV vector gene-replacement therapy: improving efficacy and prolonging the treatment window	10.1371/journal.pone.0037197	Yao and Zacks	2012	Yes	No	Yes
5	Knockout of ccr2 alleviates photoreceptor cell death in a model of retinitis pigmentosa	10.1016/j.exer.2012.08.013	Guo and Yoshimura	2012	No	Yes	Yes
6	Cone phosphodiesterase-6alpha' restores rod function and confers distinct physiological properties in the rod phosphodiesterase-6beta-deficient rd10 mouse	10.1523/JNEUROSCI.1536-13.2013	Deng and Hauswirth	2013	Yes	No	Yes
7	Dark-rearing the rd10 mouse: implications for therapy	10.1007/978-1-4614-0631-0_18	Cronin and Bennett	2012	No	No	Yes
8	Reduced phosphoCREB in Müller glia during retinal degeneration in rd10 mice	not given	Dong and Weiss	2017	Yes	No	No
9	Cone Phosphodiesterase-6gamma' Subunit Augments Cone PDE6 Holoenzyme Assembly and Stability in a Mouse Model Lacking Both Rod and Cone PDE6 Catalytic Subunits	10.3389/fnmol.2018.00233	Deng and Hauswirth	2018	NC	No	Yes
10	Dark Rearing Does Not Prevent Rod Oxidative Stress In Vivo in Pde6brd10 Mice	10.1167/iovs.17-22734	Berkowitz and Roberts	2018	No	Yes	No
11	Establishment of Immunodeficient Retinal Degeneration Model Mice and Functional Maturation of Human ESC-Derived Retinal Sheets after Transplantation	10.1016/j.stemcr.2018.01.032	Iraha and Mandai	2018	No	Yes	Yes
12	Light-Dependent OCT Structure Changes in Photoreceptor Degenerative rd 10 Mouse Retina	10.1167/iovs.17-23011	Li and Qian	2018	No	Yes	No
13	The PDE6 mutation in the rd10 retinal degeneration mouse model causes protein mislocalization and instability and promotes cell death through increased ion influx	10.1074/jbc.RA118.004459	Wang and Chen	2018	No	No	Yes
14	Sex-related differences in the progressive retinal degeneration of the rd10 mouse	10.1016/j.exer.2019.107773	Li and Shen	2019	No	No	Yes
15	Rhodopsin signaling mediates light-induced photoreceptor cell death in rd10 mice through a transducin-independent mechanism	10.1093/hmg/ddz299	Sundar and Ramamurthy	2020	No	No	Yes
16	Microglial dynamics in retinitis pigmentosa model: formation of fundus whitening and autofluorescence as an indicator of activity of retinal degeneration	10.1038/s41598-020-71626-2	Makabe and Takahashi	2020	No	No	No

ERG – electroretinogram, LE – Light Exposure, NC – Not Clear, OCT- optical coherence tomography

During our studies of metabolic reprogramming to boost PR survival in the rd10 mouse model, we identified a previously unreported phenotype confounding the experimental results (Wubben et al., 2017; Zhang et al., 2020). Animals raised in 24-hour darkness demonstrated rapid degeneration of the outer retina when subjected to serial OCT imaging. This phenotype was driven primarily by brief exposure to low-intensity room light. Our data show that the dark-reared rd10 mutant mouse is extremely sensitive to low-intensity light and may not be a suitable model for studies that employ longitudinal in vivo retinal imaging or other functional assessment which require normal room lighting.

All animals were treated in accordance with the Association for Research in Vision and Ophthalmology (ARVO) Statement for the Use of Animals in Ophthalmic and Vision Research and all research was performed following the ARRIVE guidelines. The protocol was approved by the University Committee on Use and Care of Animals of the University of Michigan (Protocol number: PRO00009399). Animals were maintained at a constant temperature and in 24-hour darkness. All animal husbandry was conducted under dim red-light illumination (light intensity ≤ 20 lux). The rd10 mouse model was obtained from Jackson Laboratories (Stock # 004297) and was originally characterized by Chang et al. (Chang et al., 2007). Animals were assigned to groups randomly with no consideration to sex. Animal lines are routinely tested to ensure they do not carry the common rd8 allele.

Optical coherence tomography (OCT) was performed as previously described (Weh et al., 2020). Briefly, animals were anesthetized via intraperitoneal injection of a mixture of 90mg/kg Ketamine and 10mg/kg Xylazine. Eyes were dilated using 2.5% Phenylephrine Hydrochloride and 1% Tropicamide ophthalmic solution drops. Systane® eye drops were used to wet the surface of the cornea before obtaining images. In vivo retinal thickness was assessed using the Envisu-R SD-OCT imager (Leica Microsystems Inc., Buffalo Grove, IL, USA). A 1.4mmx1.4mm rectangular volume scan was performed with 1000 A-scans by 30 B-scans with 6 frames obtained at each B-scan location. The 6 frames were registered and averaged using the built-in software. Average thickness of various retinal layers was assessed using Diver 1.0 (Leica Microsystems Inc). A 9x9 spider-plot consisting of 17 points (9 along the nasal/temporal axis, and 9 along the superior/inferior axis, with the center point being shared at the optic nerve head) along the superior/inferior and nasal/temporal axes, spaced approximately 140μm from each other and centered on the optic nerve head, were placed to measure layer thickness and the results were average to obtain neural retinal and outer nuclear layer (ONL) thickness. The point at the optic nerve head was omitted as no retinal layers are present, totaling 16 points used to obtain average thickness. The relative luminance on the retina by the OCT laser, assuming an 840nm wavelength and a power of 700μW, would be equivalent to approximately 0 cd/m² (or 0 lux). Room light luminance was measured using a Digital Lux Meter (Model # LX1010B, Dr. Meter) placed in the bottom of a standard mouse cage in the approximate location where light exposure was conducted. The room lights were left on for 15 minutes before assessing relative luminance. The same conditions were used when exposing animals to room light. Statistical analysis was performed using GraphPad Prism 7.0.

Our observations, which prompted this study, showed that multiple litters homozygous for the rd10 mutation were equally, and robustly, affected by short exposure to room light (data not shown). In order to rule out any novel mutations acquired in our facilities, a fresh strain of rd10 animals was obtained from The Jackson Laboratory. A litter of 10 animals was separated into 3 cohorts: 3 animals were imaged with OCT initially at p28 and a follow-up at p35 (Cohort #1, plotted as, “p28 Baseline” and “p35 Follow-up”), 4 animals exposed to 15 minutes of room light only (~650lux, eyes were dilated, and the animals were placed in a clear cage with only bedding material and no top) at p28, returned to the dark for 7 days, and then imaged at p35 (Cohort #2, plotted as, “15 mins Light at p28”), and 3 animals imaged at p35 (otherwise remained in 24-hour darkness) (Cohort #3, plotted as, “p35 Naïve”). As seen in Figure 1A&C, OCT obtained at p28 and p35 with no previous light exposure show retention of ONL thickness with a small, but non-significant thinning of the ONL (p=0.0814) whereas significant thinning of the neural retinal (p=0.036) was observed at p35 compared to p28, which is typical for dark-reared rd10 animals (Chang et al., 2007; Cronin et al., 2012). The cohort of animals which were imaged with OCT at p28 (approximately 10-15 minutes of room light exposure), and then again at p35, showed significant retinal degeneration compared to their p35 littermates (ONL & Neural Retina thickness – p <0.04), which were neither previously exposed to light nor had undergone a previous imaging session. p28 mice exposed to only 15 minutes of room light without OCT imaging show complete degeneration of the ONL layer by p35 (Fig. 1A&C). These eyes also display characteristic retinal separations, as reported previously, with 100% of eyes being affected (Fig. 1A) (Hasegawa et al., 2016). We also investigated if performing OCT under near-dark conditions with only dim-red light for illumination could prevent PR degeneration. We initiated a fourth cohort treated identically to Cohort #1, except all imaging was performed in near darkness (Cohort #4. N=4 animals). These animals were raised in 24/7 dark until p28, imaged with OCT under near-dark conditions and plotted as “p28 Dark”. The animals were returned to 24/7 darkness, and then imaged again at p35 under near-dark conditions, plotted as, “p35 Dark Follow-up”. We did not observe any significant thinning of the ONL layer (p=0.1492) (Fig. 1 D&E), similar to what was seen between p28 Baseline and p35 Naïve (Fig. 1C). Cohort #4 did not display any evidence of the severe PR degeneration associated with short exposure to room light, which we observed in Cohorts #1 and #2, suggesting that the near-infrared light used by OCT imaging is not sufficient to induce PR degeneration (Fig. 1D&E). We did not observe any differences in degeneration in the nasal/temporal or superior/inferior quadrants in Cohort #2 and thus the data above are presented as an average across the entire retina for all Cohorts. We also did not investigate for differences between sexes, but our observations which prompted this study demonstrated both sexes were equally affected by room light. Our data demonstrate that even short room light exposure can drive PR degeneration in rd10 animals; however, careful planning may allow for longitudinal experiments if exposure to room lighting can be eliminated.

Figure 1 – Brief, low-intensity exposure to light causes rapid and complete degeneration of photoreceptors in rd10 animals reared in 24/7 darkness.

(A) Representative B-scans showing rapid loss of ONL thickness after OCT imaging or exposure to low-intensity light. Arrows indicate retinal separations. (B) Schematic of retinal layers measured and quantified in (C). (C) Quantification of ONL and Neural Retina thickness. All animals were raised in 24-hour darkness until p28. The first group of animals (Cohort #1, 3 total, individual eyes were counted as separate data points) were imaged at p28 (p28 Baseline) then returned to 24-hour darkness before being imaged again at p35 (p35 Follow-up). The second group of animals (Cohort #2, 4 total, individual eyes were counted as separate data points) were exposed to 15 minutes of room-light at p28 without OCT imaging (15 mins Light at p28), returned to 24-hour darkness, then imaged at p35. The third group of animals (Cohort #3, 3 total, individual eyes were counted as separate data points) were exposed to 24-hour darkness only, then were imaged at p35 (p35 Naïve). Individual p-values are shown on the graphs. (D) Representative B-scans of animals (Cohort #4) undergoing OCT imaging in near-darkness at p28 and again at p35 showing minor decreases in ONL thickness. (E) Quantification of ONL and Neural Retina thickness from (D). For each mouse, both eyes were averaged and the resulting value was used as an independent data point for statistical analysis using a t-test. IPL – inner plexiform layer, ONL – Outer Nuclear Layer, OPL – outer plexiform layer, RNFL – retinal nerve fiber layer, RPE – retinal pigment epithelium

We report that rd10 animals raised in the dark are extremely sensitive to short periods of relatively low-intensity light. To our knowledge, this phenotype has not been described previously despite the widespread use of this mutant line as a model of IRD. Anecdotal evidence of similar light-induced degeneration in dark-reared rd10 mice was received during the preparation of this report (Bo Chang (Chang et al., 2002; Chang et al., 2007), The Jackson Laboratory, personal communication).

A literature review was conducted to identify any published reports of this extreme sensitivity to light in dark-reared rd10 mice. We searched the PubMed database using the search terms “rd10” and “mouse” and identified 236 publications as of May 2021. We scanned each manuscript for the following keywords to determine if animals were dark-reared: “dark”, “reared”, “cyclic”. Of the 236 studies, only 16 included rd10 animals raised in 24/7 darkness for at least part of the study (Table 1) (Allocca et al., 2011; Berkowitz et al., 2018; Chang et al., 2007; Cronin et al., 2012; Deng et al., 2018; Deng et al., 2013; Dong et al., 2017; Guo et al., 2012; Iraha et al., 2018; Li et al., 2019; Li et al., 2018; Makabe et al., 2020; Pang et al., 2008; Sundar et al., 2020; Wang et al., 2018; Yao et al., 2012). The report by Pang and colleagues is likely the first study to include data showing rapid loss of PRs in rd10 mice after being exposed to cyclic room light (Pang et al., 2008). Two additional studies reported rapid PR degeneration within 4-7 days of being moved to 12/12 cyclic light after dark-rearing (Dong et al., 2017; Yao et al., 2012). Five of the 16 studies presented in Table 1 reported exposing dark-reared animals to room lighting before examination of animals. To this end, the untreated animals in these manuscripts show significant PR degeneration in contrast to the large number of PR cells dark-reared animals should still have at the stated age (Allocca et al., 2011; Deng et al., 2013; Dong et al., 2017; Pang et al., 2008; Yao et al., 2012). In another one of the studies, the methodology did not include enough detail to definitively determine if animals had been exposed to light before examination, although the data presented in the manuscript suggest that this occurred due to the advanced PR degeneration as stated above (Deng et al., 2018). Based on the review of published literature, our data is the first to demonstrate extreme light sensitivity of rd10 mice after a very brief exposure to room light.

Modulation of light exposure is well known to either accelerate or delay retinal degeneration in various animal models, whereas exposure to high levels of light is known to induce PR death (Contin et al., 2016; Paskowitz et al., 2006). The VPP mouse model of RP was shown to display a similar delay in PR degeneration when raised in total darkness; however, when exposed to bright light (approximately 3300lux) for 24 hours, retinal degeneration was enhanced (Wang et al., 1997). Similar results were found in rat models of the P23H mutant line (Organisciak et al., 2003; Vaughan et al., 2003). Even more striking is the focal degeneration induced by fundus photography in a canine model of RP (Cideciyan et al., 2005). Another animal model of light-induced retinal degeneration is the Tvrm4 mouse. These animals harbor a mutant allele of the Rho gene and display no retinal degeneration unless exposed to brief periods of intense light (5 minutes, ~12,000 lux), which leads to rapid PR degeneration within 1 week (Budzynski et al., 2010). The effect of light modulation on the progression of PR cell loss has recently been reviewed (Collin et al., 2020). PR death can also be induced through continuous (24/7) exposure to low levels of light in albino animals. A report by Contín and colleagues demonstrated that 7 days of continuous exposure to ~200lux of light was sufficient to induce PR degeneration in rats (Contin et al., 2013). To our knowledge, there are no reports demonstrating PR loss after exposure to low levels of light in normal pigmented animals nor the extreme sensitivity to a very short exposure of room lighting as displayed in this manuscript with dark-reared rd10 animals.

During phototransduction, PDE6 is responsible for cleaving cGMP resulting in the closure of cyclic-nucleotide gated (CNG) Ca²⁺ channels. This closure results in the hyperpolarization of the PR which stops the flow of the neurotransmitter glutamate. Recent evidence has shown that the rd10 mutation results in a hypoactive PDE6β, which is still capable of cleaving cGMP. In fact, cGMP levels are lower in dark-reared rd10 animals than in animals raised in cyclic lighting (Wang et al., 2018). The mechanism(s) underlying the rapid loss of PRs described here is uncertain. Recent work looking into the mechanisms surrounding photoreceptor protection during dark-rearing as well as light-induced PR degeneration in rd10 animals have implicated GNAT1 (transducin) and RPE65. The protective effect of dark-rearing may be mediated by transducin, as PRs in rd10 animals lacking transducin degenerated at the same rate in both 12/12 cyclic lighting as well as 24/7 dark-rearing (Sundar et al., 2020). In addition, loss of RPE65 slowed PR degeneration in rd10 animals raised in cyclic lighting to a similar rate seen in animals raised in 24/7 darkness. Sundar and colleagues suggest a model where cyclic nucleotide gated channels are sensitized to cGMP levels by light which is dependent on non-canonical rhodopsin signaling and independent of transducin signaling. Elucidating the molecular mechanisms that drive rapid and irreversible PR loss described in our report is expected to identify common routes of PR degeneration and reveal selective targets for PR neuroprotection.

Our data demonstrate that dark-reared rd10 animals are exquisitely sensitive to light-induced retinal degeneration. We propose that investigators account for this sensitivity to light when using the rd10 model and consider the possibility of light-induced premature PR degeneration that could potentially mask results of their studies. Based on the above observations, we modified our experimental design and image or evaluate each rd10 mouse at a single time-point when light exposure is unavoidable during common procedures including OCT, electroretinography (ERG), optokinetic response, or intravitreal and subretinal injections.