Project 1
The consequences of HIF1 inactivation in cones
Marijana Samardzija

Activation of HIF1 in photoreceptors leads to retinal degeneration and may be involved in the development of AMD. To define HIF1 as a therapeutic target, it is essential to investigate whether HIF1 can be safely inactivated in adult photoreceptor cells.

In this project we use cone-specific Cre recombinase to inactivate Hif1a in cones of the normal mouse retina and in cones of the all-cone retina of the R91W;Nrl-/- mouse. With these experiments we aim to ultimately define Hif1a as a target for gene therapy in patients.

Project 2
Molecular mechanisms of cone degeneration
Marijana Samardzija

Human vision depends largely on cones, which are responsible for high acuity and color vision. Degeneration of cones (e.g. macular degeneration) leads to severe visual impairment and blindness. Research is hampered by the lack of suitable models to study physiology and pathophysiology of cones. Wild type mice, for example, express 97% rods and 3% cones which are dispersed throughout the retina. A mouse developing only cones has recently been generated by knocking out the NRL transcription factor. However, the all-cone retina of the Nrl-/- mouse has severe morphological disturbances that may hamper detailed investigation of the cone pathophysiology.

We and others showed that abnormal cone photoreceptor layering in
Nrl-/- mice depends on high levels of 11-cis-retinal, the chromophore for cone and rod visual pigments. To prevent rosette formation in the Nrl-/- mouse, we lowered the levels of 11-cis-retinal by combining the Nrl knockout with a mutation (R91W) in the Rpe65 gene, which has been engineered previously in our lab. Due to the hypomorphic function of RPE65 the 11-cis-retinal levels are reduced in R91W;Nrl-/- mice. This leads to rescue of the retinal phenotype - the double mutant mouse has an all-cone retina with sustainable visual function and normal retinal layering.

Using different means to induce cone cell death - such as toxic insults, light exposure, or genetic mutations - we aim to analyze in detail cone pathophysiology in the absence of rods, and recapitulate different aspects of complex macular diseases in the
R91W;Nrl-/- mouse. In parallel cone pathophysiology is monitored also in mouse models having rod-dominant retinas.
Uncovering and analysis of molecules/signaling pathways that are differentially activated in healthy vs. degenerative retinas should lead to the identification of potential neuroprotective factors with ultimate goal to develop treatment strategies to prevent cone loss.

Project 3
Targeting HIF1A and HIF2A for the treatment of dry AMD
Lynn Ebner

Vision loss in the dry form of age-related macular degeneration (AMD) is caused by the degeneration of retinal pigment epithelium (RPE) and photoreceptor cells. Although various exogenous, systemic and genetic risk factors have been identified for disease development or progression, a therapy for dry AMD is still an unmet medical need.

During normal ageing, the posterior ocular tissue undergoes several changes including accumulation of autofluorescent material in RPE cells, formation and accumulation of Drusen deposits, thickening of Bruch’s membrane and reduction of choroidal blood flow. These changes reduce the oxygen availability in RPE and photoreceptor cells and may lead to mild but chronic tissue hypoxia imposing cellular stress. Since photoreceptors have an extraordinarily high energy demand, and maintenance of photoreceptor homeostasis and function requires sufficient tissue oxygenation, chronic hypoxia may lead to disease development when crossing critical threshold levels. Thus, it seems plausible that some of the tissue changes discussed above may be more pronounced in eyes with dry AMD.

Reduced tissue oxygenation leads to a well defined molecular response. Key factors in this response are hypoxia-inducible factors (HIFs), which are transcription factors activated at low oxygen concentrations. HIFs control expression of a large number of genes required for adaptation to low oxygen conditions. Activation of HIF1 and HIF2 is controlled by the von Hippel Lindau (VHL) protein complex that targets hydroxylated HIF-alpha subunits to the proteasomal degradation pathway in normoxic conditions. In hypoxia, HIF-alpha subunits are not hydroxylated and thus not recognised by VHL. Therefore, HIF-alpha proteins are stabilised and can function in gene regulation.

Evidence suggests that chronic activation of HIF1 in photoreceptors and of HIF2 in the RPE leads to a change in cellular metabolism, RPE atrophy and photoreceptor degeneration. Thus, we develop a gene therapy-based RNA interference approach to preserve cellular integrity and function in such conditions. This approach will deliver anti-HIF1a shRNA to photoreceptors and anti-HIF2a shRNA to the RPE using a single AAV vector injected into the subretinal space. Cell type-specific expression of the shRNAs will be achieved through specific Pol II promoters and embedding of the shRNAs into a miR-scaffold.

Project 4
The retinal response to normobaric, hypobaric and acute hypoxia
Lynn Ebner

Oxygen is very important for retinal cells to function. Chronic hypoxia may contribute to the development of retinal diseases such as AMD. Acute hypoxia, however, activates a response that protects photoreceptors against a toxic light insult. Since the cellular response to hypoxia is regulated mainly by the activity of hypoxia-inducible transcription factors and transcriptional adaptation, we will determine the hypoxic response of the retina under various conditions. Data may provide valuable information about the connection of hypoxia to disease.
In a first project, we will compare the retinal transcriptome of mice that were kept for several weeks at normobaric hypoxia to the retinal transcriptome of mice that spent several weeks at high altitude (3’500 m). We will also investigate the transcriptome in mice that spent only 2 days at high altitude to determine potential adaptation processes after prolonged hypobaric hypoxia. The retinal transcriptome will be determined by RNA-Seq.
In a second project, we will investigate the transcriptome of individual retinal cells after acute hypoxia using DropSeq. Single-cell sequencing will reveal the response of the individual cell types to reduced oxygenation of the retina. Data will be validated and interesting changes will be followed up by various approaches.

Project 5
The influence of hypoxia on energy metabolism in photoreceptors visualized by two-photon imaging
Vyara Todorova

The retina is one of the metabolically most active tissues of the body requiring large amounts of oxygen for energy production, especially in photoreceptor cells (Yu and Cringle 2001). Any disturbances in oxygen availability or energy metabolism may affect survival of photoreceptors and other retinal cells (Osborne and del Olmo-Aguado 2013; Jaiswal et al. 2015). Several ophthalmologic diseases including diabetic retinopathy, glaucoma and age related macular degeneration (AMD) are increasingly recognized to have a ‘mitochondrial component’ such as impaired mitochondrial function, leading to reduced ATP levels (Barot et al. 2011). Interestingly, all of these diseases may also have a hypoxic component (Grimm and Willmann 2012) that may link the molecular hypoxic response to energy production, and thus to disease onset and / or progression.

Using advanced two-photon microscopy we will address the link between energy metabolism and hypoxia in photoreceptors in physiological and pathophysiological conditions. Our investigations of energy metabolism in the retina will not only advance the knowledge of basic biological principles in retinal neurons but may also significantly deepen the understanding of disease mechanisms.

Project 6
Single-cell transcriptomic profiles in the degenerating mouse retina
Duygu Karademir

The retina is a highly complex, heterogeneous tissue harboring various cell subpopulations. The characterization of cell populations in the retina under healthy and degenerative conditions has been highly difficult, as whole-retina approaches conceal large gene expression changes in small numbers of cells. Therefore, we are utilizing droplet sequencing (DropSeq) to elucidate transcriptomic profiles of single-cells in a model of retinitis pigmentosa, the rd10 mouse, at the peak of rod photoreceptor degeneration.

We have previously characterized LIF as a key neuroprotective factor in the mouse retina, which is expressed by a subpopulation of Müller cells upon retinal damage, with no essential function in healthy retinas. LIF also influences the pluripotency of murine stem cells in vitro (Hirai et al., Biochem J. 2011, 438, 11-23) and is involved in the maintenance of adult neural stem cells in the mouse forebrain (Shimazaki et a., J. Neurosci. 2001, 21, 7642-53). Others have shown that transgenic overexpression of LIF during development suppresses differentiation of all retinal cell types including Müller cells (Sherry et al., J. Neurosci. 2005, 82, 316-32). In teleost fish like zebrafish, and to some extent in birds, retinal Müller cells can assume progenitor properties upon photoreceptor injury and function as retinal stem cells effectively replacing lost neurons to restore visual function (Goldmann, Nat Rev Neurosci., 2014, 15, 431-42). This suggests that the LIF expressing subpopulation of Müller cells may have special functions in retinal protection and repair, making the analysis and characterization of these cells highly relevant. Thus, by characterizing single-cell transcriptomes at the peak of injury in the rd10 retina, where the Lif-expressing Müller cell subpopulation is active, we aim to characterize Lif(pos) and Lif(neg) Müller cells and explore their function in the diseased retina.

Project 7
The transcriptomic profile of retinal Müller cells in the presence and absence of leukemia inhibitory factor (LIF)
Duygu Karademir

LIF is a survival factor in the retina and is produced in a subset of Müller glia cells in response to photoreceptor injury. Lif knockout mice are highly susceptible to degeneration and display altered signaling mechanisms in the retina. It is also of interest that LIF is required to maintain pluripotency of mouse embryonic stem cells in vitro and that Müller glia cells are thought to have some stem cell potential.
Here we will FACS sort fluorescently labeled Müller glia cells from wild type and Lif knockout mice to investigate how absence of LIF affects Müller cells. This information may help to elucidate the function of Müller glia in retinal homeostasis and neuroprotection.

Project 8
CRISPR-activation of endogenous Lif expression in Müller glia cells to protect photoreceptor cells in the degenerating retina
Duygu Karademir

Müller cells play key roles in retinal physiology that include structural stabilization, uptake and metabolism of neurotransmitters, ion and water homeostasis, maintaining the blood retinal barrier, modulating immune responses and release of pro-survival factors during pathological situations. Thus, Müller cells are extremely important in both the normal and diseased retina.

Results from our lab show that photoreceptor damage activates a sub-population of Müller cells to express leukemia inhibitory factor (LIF) that acts as a central regulator of a survival pathway for the protection of photoreceptor cells. LIF knockouts exhibit no differences to wildtype retinas under healthy conditions, but the absence of Lif significantly accelerates the course and extent of neurodegeneration in the context of light-induced or inherited retinal degeneration (Joly et al., J. Neurosci. 2008, 28, 13765-13774).

Using an AAV-CRISPRa approach, we will upregulate endogenous Lif expression specifically in Müller glia cells. By inducing endogenous Lif, we will keep the redox-based post-transcriptional regulation of Lif intact and will only produce increased LIF levels in the degenerating retina once dying photoreceptors produce H2O2 as signaling molecule. After validation of the system in wildtype mice, we will apply our CRISPRa approach to various models of induced and inherited retinal degeneration and measure the outcome on a morphological, functional and molecular level.

Project 9
Searching for compounds that increase expression of LIF
Maria Cristina De Liso

Endogenous leukemia inhibitory factor (LIF), produced in the retina by a subset of Müller glia cells upon photoreceptor injury, acts as a survival factor and slows down degenerative processes in photoreceptors. We are generating a Müller cell line that expresses a LIF:mRuby fusion protein under the control of the LIF regulatory elements using CRISPR/Cas9 methodology. This cell line will be treated with a commercially available compound library to identify potential compounds that lead to an increased fluorescence in cells and thus increase LIF expression in vitro. Promising candidates will then be tested in non-engineered cell lines and applied in vivo to eventually investigate their neuroprotective effect in retinal degeneration models.

Project 10
Rescue of vision in cblC deficiency
Eva Kiessling

Methylmalonic aciduria with homocystinuria cobalamin deficiency cblC type (MMACHC cblC), with only 500 patients diagnosed worldwide, is a rare disease caused by mutations in the MMACHC gene. This disorder is accountable for various symptoms, including neurological, ophthalmic, cardiovascular, respiratory and facial abnormalities. Especially the patients with early-onset of this disease frequently show a number of ophthalmic manifestations such as macular degeneration, optic nerve pallor and vascular changes. This condition can progress to full blindness.

Oral or injectable cobalamin substitution can improve some of the non-neurological as well as some of the neurological impairments in these patients. However, the ocular phenotype remains progressive for unknown reasons.

This project aims at the identification of the molecular mechanisms that lead to retinal degeneration in MMACHC-deficiency. For this purpose we generated a Mmachcflox/flox mouse to inactivate Mmachc specifically in retinal cells in order to mimic the eye phenotype observed in humans. In these mice we will study the Mmachc expression profile and determine the ocular function of Mmachc. Ultimately, these findings can be the basis for establishing novel avenues for treatment to rescue vision in cblC patients.

Project 11
Translocator protein (18kDa) in the RPE: its function and connection to retinal degeneration
Katrin Klee

The retinal pigment epithelium (RPE) is a monolayer of pigmented cells located in the very back of the eye. Even though these cells are not directly involved in vision, they are essential for the survival of the underlining photoreceptors. In fact, dysfunctions of the RPE can result in photoreceptor loss and thus blindness.

We identified a protein, Translocator Protein (18 kDa) (TSPO), that is constitutively expressed by the RPE cells. TSPO is mostly studied in steroidogenic cells such as adrenals and gonads, but also in the brain where it gets upregulated by glial cells during brain injury and disease. TSPO ligands are widely used to image injury sites of the brain. Moreover, it was observed that TSPO ligands have beneficial effects in several CNS pathologies.

Not much is known about the role of TSPO in the eye. Recent publications have shown beneficial effects of TSPO modulation in a paradigm of light-induced retinal degeneration. The protection achieved with the use of a TSPO ligand can be partly explained by dampening of microglia activation, but is not clear whether the TSPO expressed by the RPE cells also contributes to the retinal protection.

To study the role of TSPO in the murine eye we generated a Tspo knockout mouse. The mouse will be analyzed on during development and ageing, and in connection to retinal degeneration.