Diana M Mitchell

Microglia rapidly respond to injury, stress, and perturbations to neurons in the brain and retina and perform phagocytosis to clear dying cells and debris. Oxidative stress is a feature of neurodegeneration, and while glia are crucial for managing such stress, microglia may also be dysfunctional in diseased tissue. Here we examine the role of microglia in management of oxidative stress and restoring redox homeostasis following death of rod photoreceptors in the larval zebrafish retina. Using rho:nfsb-eGFP transgenic zebrafish and treatment with the pro-drug metronidazole (MTZ), we coupled the generation of reactive oxygen species (ROS) in dying rods to their ablation. Microglia efficiently engulfed and cleared the ROS-laden rods, effectively undertaking the oxidative load. Despite abundant ROS upon MTZ-mediated cell death, oxidative stress overall was minimal in retinal tissue when microglia were present, indicating that they rapidly and efficiently performed redox functions. In irf8-/- mutants, which are deficient in microglia, retinas with MTZ-induced rod ablation showed widespread ROS that localized, at least in part, to Müller glia. Microglia deficient retinas showed evidence of increased oxidative stress, and increased numbers of "off-target" inner retinal neurons that stained positive for the cell death marker TUNEL. Supplementation with the antioxidant Glutathione modestly reduced the number of off-target TUNEL+ cells detected in microglia-deficient retinas following rod ablation. We also found that microglia may be important for mitigating effects of MTZ alone in the absence of Nfsb enzyme. Our results suggest that microglial redox functions are important in maintaining and restoring homeostasis following acute retinal damage.

Zebrafish can spontaneously regenerate retinal neurons after acute damage. Ablation and regeneration of retinal neurons can be achieved in transgenic zebrafish with cell-selective expression of bacterial nitroreductase (NTR/nfsb) upon exposure to pro-drugs such as metronidazole (Mtz). We found that Mtz exposure of gnat2:nfsb-mCherry adult zebrafish at certain dose/duration could limit death to only a small number of cones, while causing transient stress to the surviving cone photoreceptors, providing insight into glial and immune cell responses to transient neuronal perturbations. Using this approach, we identified molecular changes in multiple retinal cell types following Mtz washout. Cones exhibited evidence of stress including downregulation of phototransduction genes and upregulation of oxidative stress response genes. Even with only limited cone death detected, the Müller glia (MG) response involved features known to occur during retinal regeneration, including proliferating progenitor production. Microglia/myeloid cells showed evidence of two response waves. The first response occurred within a day of drug washout while the second response occurred around 4 days after drug washout and was associated with cone photoreceptor stress recovery. Single cell RNA-seq suggested two subpopulations of microglia/myeloid cells participated in this response. Our results provide insight into triggers of proliferative responses from the MG, microglial/myeloid responses coupled to neuronal stress recovery, and potential molecular changes involved with cone recovery to stress.

Separase regulates multiple aspects of the metaphase-to-anaphase transition. Separase cleaves cohesin to allow chromosome segregation and localizes to vesicles to promote exocytosis. The anaphase-promoting complex/cyclosome (APC/C) activates separase by ubiquitinating its inhibitory chaperone, securin, triggering its degradation. How this pathway controls the exocytic function of separase is unknown. During meiosis I, securin is degraded over several minutes, while separase rapidly relocalizes from kinetochore structures at the spindle and cortex to sites of action on chromosomes and vesicles at anaphase onset. The loss of cohesin coincides with the relocalization of separase to the chromosome midbivalent at anaphase onset. APC/C depletion prevents separase relocalization, while securin depletion causes precocious separase relocalization. Expression of non-degradable securin inhibits chromosome segregation, exocytosis, and separase localization to vesicles but not to the anaphase spindle. We conclude that APC/C-mediated securin degradation controls separase localization. This spatiotemporal regulation will impact the effective local concentration of separase for more precise targeting of substrates in anaphase.

Inflammation and microglia appear to be key factors influencing the outcome of retinal regeneration following acute retinal damage. Despite such findings, direct connection of microglia-specific inflammatory factors as drivers of regenerative responses in the retina are still not defined, and intracellular pathways activated to stimulate such signals from microglia are currently unknown. We became interested in MyD88 regulation in microglia because transcriptomic datasets suggest myd88 could be regulated temporally in zebrafish microglia responding to damage in the central nervous system. MyD88 is an intracellular molecular adaptor that initiates signaling cascades downstream of several innate immune receptors, and probably most well-known for inducing gene expression of pro-inflammatory factors. Using zebrafish, which spontaneously regenerate retinal neurons after acute retinal damage, we studied the effects of overactivation of MyD88 signaling in microglia and macrophages on the Müller glia-mediated regenerative response. Our results indicate that increased MyD88 signaling in microglia/macrophages impacts the initial response of Müller glia entering a regenerative response after acute, neurotoxin-induced retinal damage to inner retinal neurons. In addition, increased MyD88 signaling in microglia/macrophages resulted in reduced survival of inner retinal neurons in regenerated retinas. This work supports the idea that temporal control of inflammatory signaling is a key component in the production of MG-derived progenitors yet further indicates that such control is important for differentiation and survival of regenerated neurons.

To clarify our understanding of glial phagocytosis in retinal development, we used real-time imaging of larval zebrafish to provide cell-type specific resolution of this process. We show that radial Müller glia frequently participate in microglial phagocytosis while also completing a subset of phagocytic events. Müller glia actively engage with dying cells through initial target cell contact and phagocytic cup formation, after which an exchange of the dying cell from Müller glia to microglia often takes place. In addition, we find evidence that Müller glia cellular material, possibly from the initial Müller cell phagocytic cup, is internalized into microglial compartments. Previously undescribed Müller cell behaviors were seen, including cargo splitting, wrestling for targets and lateral passing of cargo to neighbors. Collectively, our work provides new insight into glial functions and intercellular interactions, which will allow future work to understand these behaviors on a molecular level.

Zebrafish regenerate functional retinal neurons after injury. Regeneration takes place following photic, chemical, mechanical, surgical, or cryogenic lesions, as well as after lesions that selectively target specific neuronal cell populations. An advantage of chemical retinal lesion for studying the process of regeneration is that the lesion is topographically widespread. This results in the loss of visual function as well as a regenerative response that engages nearly all stem cells (Müller glia). Such lesions can therefore be used to further our understanding of the process and mechanisms underlying re-establishment of neuronal wiring patterns, retinal function, and visually mediated behaviors. Widespread chemical lesions also permit the quantitative analysis of gene expression throughout the retina during the period of initial damage and over the duration of regeneration, as well as the study of growth and targeting of axons of regenerated retinal ganglion cells. The neurotoxic Na+/K+ ATPase inhibitor ouabain specifically offers a further advantage over other types of chemical lesions in that it is scalable; the extent of damage can be targeted to include only inner retinal neurons, or all retinal neurons, simply by adjusting the intraocular concentration of ouabain that is used. Here we describe the procedure through which these "selective" vs. "extensive" retinal lesions can be generated.

Adult zebrafish respond to retinal injury with a regenerative response that replaces damaged neurons with Müller glia-derived regenerated neurons. The regenerated neurons are functional, appear to make appropriate synaptic connections, and support visually mediated reflexes and more complex behaviors. Curiously, the electrophysiology of damaged, regenerating, and regenerated zebrafish retina has only recently been examined. In our previous work, we demonstrated that electroretinogram (ERG) recordings of damaged zebrafish retina correlate with the extent of the inflicted damage and that the regenerated retina at 80 days post-injury exhibited ERG waveforms consistent with functional visual processing. In this paper we describe the procedure for obtaining and analyzing ERG recordings from adult zebrafish previously subjected to widespread lesions that destroy inner retinal neurons and engage a regenerative response that restores retinal function, in particular the synaptic connections between photoreceptor axon terminals and the dendritic trees of retinal bipolar neurons.

Detection of the protein PCNA (proliferating cell nuclear antigen) is used to identify cells in the S phase of the cell cycle to indicate cellular proliferation. Here we describe our method to detect PCNA expression by microglia and macrophages in retinal cryosections. We have used this procedure with zebrafish tissue, but this procedure could be applied to cryosections from any organism. Retinal cryosections are subjected to a heat-mediated antigen retrieval step in Citrate Buffer, then immunostained with antibodies to label PCNA and microglia/macrophages, and counterstained for cell nuclei. After fluorescent microscopy, the number of total and PCNA+ microglia/macrophages can be quantified and normalized to compare across samples and groups.

IntroductionZebrafish regenerate their retinas following damage, resulting in restoration of visual function. Here we evaluate recovery of retinal function through qualitative and quantitative analysis of the electroretinogram (ERG) over time following retinal damage, in correlation to histological features of regenerated retinal tissue. MethodsRetinas of adult zebrafish were lesioned by intravitreal injection of 10 mu M (extensive lesion; destroys all neurons) or 2 mu M (selective lesion; spares photoreceptors) ouabain. Unlesioned contralateral retinas served as controls. Function of retinal circuitry was analyzed at selected timepoints using ERG recordings from live zebrafish, and whole eyes were processed for histological analyses immediately thereafter. ResultsQualitative and quantitative assessment of waveforms during retinal regeneration revealed dynamic changes that were heterogeneous on an individual level within each sampling time, but still followed common waveform recovery patterns on a per-fish and population-level basis. Early in the regeneration period (13-30 days post injury; DPI), for both lesion types, b-waves were essentially not detected, and unmasked increased apparent amplitudes, implicit times, and half-widths of a-waves (vs. controls). In control recordings, d-waves were not obviously detected, but apparent d-waves (OFF-bipolar responses) from regenerating retinas of several fish became prominent by 30DPI and dominated the post-photoreceptor response (PPR). Beyond 45DPI, b-waves became detectable, and the ratio of apparent d- to b-wave contributions progressively shifted with most, but not all, fish displaying a b-wave dominated PPR. At the latest timepoints (extensive, 90DPI; selective, 80DPI), recordings with measurable b-waves approached a normal waveform (implicit times and half-widths), but amplitudes were not restored to control levels. Histological analyses of the retinas from which ERGs were recorded showed that as regeneration progressed, PKCa + ON-bipolar terminals and parvalbumin + amacrine cell processes became more stereotypically positioned within the deep sublaminae of the INL over recovery time after each lesion type, consistent with the shift in PPR seen in the ERG recordings. DiscussionTaken together, these data suggest that photoreceptor-OFF-bipolar component/connectivity may functionally recover and mature earlier during regeneration compared to the photoreceptor-ON-bipolar component, though the timeframe in which such recovery happens is heterogeneous on a per-fish basis. Collectively our studies suggest gradual restoration of ON-bipolar functional circuitry during retinal regeneration.