Dissertations / Theses on the topic 'Channelrhodopsin'

Channelrhodopsin-2 ist ein lichtaktivierter Kationenkanal, der zur nichtinvasiven Steuerung neuronaler Aktivität verwendet wird. Einige grundlegende Eigenschaften dieses Proteins sind bereits bekannt, aber die molekularen Mechanismen des Ionentransports und der Aktivierung liegen noch weitgehend im Dunkeln. Ziel dieser Studie war es, anhand von Mutationsstudien die Funktion einzelner Aminosäuren zu bestimmen. Dazu habe ich gezielt potentiell wichtige Reste substituiert und die Channelrhodopsin-2-Varianten elektrophysiologisch untersucht. Um die aufgetretenen Änderungen beim Ionentransport und den Kanalkinetiken zu erklären, habe ich verschiedene mathematische Modelle an die experimentellen Daten angepasst. Dabei stellte sich heraus, dass die Reste H134 und E90 Schlüsselpositionen für den Protonentransport sind. Außerdem haben auch die Reste E235 und D253 einen großen Einfluss auf den Ladungstransport. Dagegen wird die Kanalöffnung von C128 und D156 kontrolliert. Des Weiteren kontrolliert E123 die Übergänge zwischen leitenden und nichtleitenden Zuständen von Channelrhodopsin-2. Aus der zielgerichteten Mutation von Aminosäuren resultierten Varianten, die langsamere oder schnellere Kinetiken hatten oder eine bessere Expression zeigten als der Wildtyp. Das Anwendungspotential der modifizierten Kanäle wurde in Kooperationen mit neurophysiologischen Arbeitsgruppen untersucht. Dadurch konnten drei neue Typen von Channelrhodopsinen in die Neurophysiologie eingeführt werden. Die step-functions opsins führen zu einer anhaltenden Membrandepolarisation, die die Erregbarkeit von Neuronen gegenüber synaptischen Inputs erhöht. ChETA erlaubt das zeitlich präzise Auslösen von Aktionspotentialen auch bei sehr hohen Anregungsfrequenzen. T159C und E123T/T159C ermöglichen durch ihre großen Photoströme und optimierten Kinetiken eine hohe Zuverlässigkeit bei der optischen Steuerung neuronaler Aktivität. Dadurch wird das Anwendungsspektrum von Channelrhodopsin-2 erheblich erweitert.
Channelrhodopsin-2 is a light-activated cation channel which has become a very useful tool in neurophysiology, since it allows the noninvasive control of neural activity. Some of the basic features of this channel are known from previous studies, but the molecular mechanisms of ion translocation and activation are largely unknown. The aim of my thesis is to elucidate the function of single amino acids by mutational studies. I replaced potentially important residues and probed the constructs by electrophysiological measurements under various conditions. Additionally, I fitted the experimental data to several mathematical models in order to explain changes in ion permeabilities and channel kinetics and I assigned particular functions to the mutated residues. Apparently, H134 and E90 are key positions for the proton transportation. Mutations at E235 and D253 also strongly influence ion translocation, whereas C128 and D156 obviously control the channel opening. Moreover, I found that E123 is a key element for the channel activation which controls the transitions between conducting and non-conducting states of Channelrhodopsin-2. The genetically modified Channelrhodopsin-2-variants provide several favorable features, such as, a slower or faster channel opening and closing or an optimized expression. Therefore, we tested the potential of promising constructs for applications in collaboration with neurophysiology laboratories. Finally, we introduced three new tools. First, step-function opsins induce a sustained membrane depolarization which sensitizes neurons to native synaptic inputs. Second, the ChETA variant allows the temporally precise generation of action potentials even at high stimulation frequencies. Third, T159C and E123T/T159C provide large photocurrents and optimized kinetics resulting in an improved performance in the noninvasive control of neural activity. In summary, this significantly broadens the range of application for channelrhodopsin-2.

Maintenance of neuronal function depends on the timely delivery of oxygen and glucose through changes in blood flow that are linked to the level of ongoing neuronal and glial activity, yet the mechanisms underlying this stimulus-dependent control of blood flow remain unclear. Here, using transgenic mice expressing channelrhodopsin-2 in a subset of layer 5b pyramidal neurons, we report that changes in intrinsic optical signals and blood flow can be evoked by activation of channelrhodopsin-2 neurons without direct involvement of other cell types. We have used a combination of imaging and pharmacology to examine the importance of glutamatergic synaptic signaling in neurovascular coupling. In contrast to sensory-evoked responses, we observed that glutamate-dependent neuronal signalling is not essential for the production of channelrhodopsin-evoked hemodynamic responses. Our results rather suggest that ChR2-activated neurons are coupled to the surrounding vasculature through a glutamate-dependent astrocytic pathway mediated by the Group I metabotropic glutamate receptor mGluR5.

The importance of activity during the development of central components of the nervous system such as the visual system has long been recognized (Wiesel & Hubel 1963) and it is beginning to be understood that sensory experience and motor behavior are equally important for neuromuscular development (Brumley et al. 2015; Sharp & Bekoff 2015). The chick embryo model has proven to be especially useful in studying the relationships among motor behavior, sensory experience, and neuromuscular development (Oppenheim et al. 1978; Sharp & Bekoff 2001) due to its accessibility and early onset of movement behavior. Traditionally, neuromuscular blockers have been used to broadly study the role of neural activity and muscle activity during development (Oppenheim et al. 1978; Ding et al. 1983). In order to noninvasively alter neural activity in specific populations of cells, the Sharp lab has developed an optogenetic approach that allows the expression of ChIEF, a variant of channelrhodopsin-2, in the spinal cord of living chick embryos (Sharp & Fromherz 2011). In order to better understand the unique role that muscle activity plays in neuromuscular development, it would be advantageous to directly and noninvasively control muscle activity through light-activation of ChIEF expressed in muscle fibers. Therefore, the primary objective of this thesis research was to achieve ChIEF expression in the plasma membrane of myotubes in living chick embryos. Initial attempts to express ChIEF in chick muscle resulted in low success rates. The CAG promoter in pPB-ChIEF-Tom, the plasmid vector that encodes ChIEF, was likely hindering expression of ChIEF in muscle tissue. Therefore, standard molecular cloning techniques were used to replace the CAG promoter with the myosin light chain promoter which was known to drive transgene expression in chick muscle (Wang et al. 2011). The new DNA construct that resulted from modifying pPB-ChIEF-Tom was identified as pPB-MLC-ChIEF-Tom (mChIEF). ChIEF was successfully expressed in hindlimb muscles of chick embryos via somite electroporation of mChIEF and observed between E7 and E18. Expression patterns corresponded with the current understanding of muscle progenitor contributions of somites to hindlimb muscles (Rees et al. 2003). ChIEF was located in the outer membrane of muscle fibers on E9, E14, and E18 when tissue was histologically examined in conjunction with myosin heavy chain immunofluorescence. Importantly, light-activation of ChIEF in the hindlimb muscle of living chick embryos resulted in muscle contraction and light-evoked hindlimb movements. In addition to demonstrating the functionality of ChIEF expression, an effort was made to characterize the effects of altered parameters of light stimuli on light-evoked movement and determine whether light-evoked muscle contraction could be used to imitate normal, neuronal muscle control. Light intensity was directly related to amplitude and rate of light-evoked movement. Light duration was directly related to amplitude and latency of peak movement. Unfused and fused tetanus were observed when bursts of short duration light pulses with varying interpulse intervals were used to activate ChIEF. This thesis research strongly suggests that light-activation of ChIEF expressed in living, chick embryo hindlimb muscle results in muscle contractions in manner similar to normal, neurally-driven muscle contraction.

The green algae Chlamydomonas reinhardtii senses light through two photosensory proteins, channelrhodopsin-1 (ChR1) and channelrhodopsin-2 (ChR2). The initial discovery of these two photoreceptors introduced a new class of light-gated ion channels. ChR2 is an inwardly-rectified ion channel that is selective for cations of multiple valencies. Similar to microbial-rhodopsin ion pumps, ChR2 has a seven transmembrane domain motif that binds the chromophore all-trans retinal through a protonated Schiff base linkage. Physiologically, ChR2 functions to depolarize the membrane which initiates a signaling cascade triggering phototactic response. This fundamental property has been pivotal in pioneering the field of optogenetics, where excitable cells can be manipulated by light. ChR2 reliably causes neuronal spiking with high spatial and temporal control. Moreover, the recent discovery of new chloride-conducting channelrhodopsins (ChloCs) has further expanded the optogenetic toolbox. Although structurally similar to microbial-rhodopsin ion pumps, ChR2 undergoes more complex conformational rearrangements that lead to ion conductance. Currently, the molecular basis for ChR2 gating remains unresolved. Revealing the specific structural interactions that modulate ChR2 function have important implications in understanding the intricacies of ion transport and molecular differences between ion pumps, channels, and transporters. Here we describe a combined computational and experimental approach to elucidate the mechanism of ion conductance, channel gating, and structure-function relationship of ChR2. Our results have contributed to expanding our understanding of the fundamental properties of ion channels.

The eyespot of the single-celled alga Chlamydomonas aids the cell in detecting the direction of light in the environment. The complex assembly and asymmetric placement of the eyespot provides a model to ask questions about assembly and asymmetric placement of organelles. Understanding the mechanisms that underlie assembly and asymmetric placement of the eyespot can be applied more broadly to their functions in other eukaryotic organisms. This study sought to understand the role of a key protein in those processes, Channelrhodopsin-1 (ChR1). ChR1 was found to localize along the entire length of the D4 rootlet from the region around the daughter basal body to the eyespot. ChR1 was found to primarily localize to the plasma membrane side of the D4, suggesting that ChR1 was being pulled through the plasma membrane from the region around the basal bodies to the eyespot. Further, ChR1 was found to be able to localize to the eyespot even with the truncation of the large cytoplasmic C-terminal domain, suggesting that ChR1 is able to complex with another protein that is being trafficked to the eyespot. One such protein was thought to be ChR2, the other light-activated ion channel localized to the eyespot. Efforts to isolate a mutation in ChR2 were unsuccessful. Initial efforts were made in this dissertation to perform proteomic studies of ChR1 and identify its interacting partners. ChR1 is not the master regulator of either placement or assembly of the eyespot, but work in this study lays the groundwork to further investigate transport of ChR1 and interacting proteins to the eyespot and their role in assembly of the eyespot.

Cellular lipid membranes can – and often do – support a transmembrane electric field, serving as biological capacitors that maintain a voltage difference between their two sides. It isn't hard to see why these voltage gradients matter; the electrical spiking of neurons gives rise to our thoughts and actions, and the voltage dynamics of cardiomyocytes keep our hearts beating. Studies of bioelectricity have historically relied on electrode-based techniques to perturb and measure membrane potential, but these techniques have inherent limitations. I present new optogenetic methods of studying membrane potential that will broaden the scope of electrophysiological investigations by complementing traditional approaches. I introduce the microbial rhodopsin Archaerhodopsin-3 (Arch), a transmembrane protein from Halorubrum sodomense. The fluorescence of Arch is a function of membrane potential, allowing it to serve as an optical voltage reporter. We use time-dependent pump-probe spectroscopy to interrogate the light- and voltage- dependent conformational dynamics of this protein, to elucidate the mechanism of voltage-dependent fluorescence in Arch. I then present two new methods for imaging voltage using engineered variants of Arch. Both techniques take advantage of the unique photophysical properties of Arch(D95X) mutants. The first method, Flash Memory, records a photochemical imprint of the activity state -- firing or not firing -- of a neuron at a user-selected moment in time. The Flash Memory technique decouples the recording of neural activity from its readout, and can potentially allow us to take large-scale snapshots of voltage (e.g. maps of activity in a whole mouse brain). The second method allows for the quantitative optical measurement of membrane potential. This technique overcomes the problems that typically hinder intensity-based measurements by encoding a measurement of voltage in the time domain. Finally, I present a method to visualize cellular responses to changes in membrane potential. I engineer mutants of Channelrhodopsin-2 (ChR2), a light-gated cation channel from Chlamydomonas reinhardtii that is used for optical control of neural activity, and use these optogenetic actuators in conjunction with GFP-based sensors to study the activity-dependent behavior of cultured neurons.

Seit mehr als 10 Jahren kann biologische Aktivität durch eine Vielzahl photosensorischer Proteine beeinflusst werden. In diesem als Optogenetik bezeichneten Forschungsgebiet, werden Kationen leitende Kanalrhodopsine (CCRs) als lichtinduzierte neuronale Aktivatoren eingesetzt. Diese Arbeit soll zur Vervollständigung von optogenetischen Werkzeugen durch die Entwicklung Anionen leitender Kanalrhodopsine (ACRs) dienen, um die bestehenden Nachteile mikrobieller lichtgetriebener Ionenpumpen zu überwinden, die bislang zur neuronale Inhibition genutzt wurden. Der Austausch von E90 in C. reinhardtii Kanalrhodopsin 2 (CrChR2) durch positiv geladene Aminosäuren führte zu Entwicklung Chlorid leitender ChRs (ChloCs), die jedoch eine Restkationen-permeabilität aufwiesen. Durch Substitution zweier weiterer negativen Ladungen innerhalb des Ionenpermeationsweges, konnte die Kationenleitung vollständig aufgehoben werden. Parallel wurde durch A. Berndt et al. ein inhibitorisches C1C2 (iC1C2), basierend auf der CrChR1/2 Chimäre entwickelt. Wie auch bei den ChloCs, zeigte iC1C2 verbesserungswürdige biophysikalische Eigenschaften. Mutagenesestudien des Ionenpermeationsweges führten zur Entwicklung der verbesserten Nachfolgervariante iC++. Um ausgehend von weiteren CCRs neuartige ACRs zu entwickeln (eACRs), wurden die zuvor angewandten Mutagenesestrategien auf weitere CCRs übertragen. Zwei neue eACRs, Phobos und Aurora, mit jeweils blau- und rotverschobenen Aktionsspektrum konnten generiert werden. Bistabile eACRs wurden erzeugt, die ein lichtgesteuertes Schalten zwischen offenen und geschlossenen Zuständen ermöglichen. Schlussendlich wurde ein natürlich vorkommendes ACR (nACR) aus Proteomonas sulcata (PsACR1) identifiziert und charakterisiert. Die Maximalaktivität von PsACR1 zählt mit 540 nm zu den am stärksten rotverschobenen unter den nACRs. Elektrophysiologische und spektroskopische Untersuchungen ergaben, dass sich der Photozyklus von PsACR1 signifikant von jenen der CCRs unterscheidet.
For more than 10 years, photosensory proteins have developed as powerful tools to manipulate biological activity. In this research field termed optogenetics, cation-conducting channelrhodopsins (CCRs) mainly are utilized as light-induced neural activators. This study aimed at a complementation of the optogenetic tool box by engineering anion-conducting channelrhodopsins (ACRs) to overcome the existing drawbacks of microbial light-driven ion pumps utilized for neural inhibition so far. Replacement of E90 in the cation-conducting C. reinhardtii channelrhodopsin 2 (CrChR2) with positively charged residues reversed the ion selectivity and yielded chloride-conducting ChRs (ChloCs). Applied in neuronal cell culture, ChloCs showed residual cation permeability occasionally leading to excitation instead of the desired inhibition. Further charge elimination within the ion permeation pathway completely abolished cation conduction. In parallel, an inhibitory C1C2 (iC1C2) was developed by A. Berndt et al. based on a CrChR1/2 chimera. Though, iC1C2 displayed unsatisfactory biophysical properties as well. Further mutational modifications of the ion permeation pathway led to the development of the improved successor variant iC++. A systematic transfer of both conversion strategies to other CCRs was conducted to create engineered ACRs (eACRs) with distinct biophysical properties. Two novel eACRs, Phobos and Aurora, with blue- and red-shifted action were obtained. Additionally, step-function mutations greatly enhanced the operational light sensitivity and enabled temporally precise toggling between open and closed states using two different light colors. Finally, a natural ACR (nACR) originating from Proteomonas sulcata (PsACR1) was identified and characterized. With a maximum activation at 540 nm it is one of most red-shifted nACRs. Single turnover electrophysiological measurements and spectroscopic investigations revealed an unusual photocycle compared to that of CCRs.

Kanalrhodopsine (ChRs), lichtgesteuerte Ionenkanäle, vermitteln phototaktische Reaktionen in beweglichen Algen und sind als optogenetische Werkzeuge zur Manipulation der Zellaktivität mittels Lichts weit verbreitet. Viele Kationen- und Anionen-leitende ChRs (CCRs und ACRs) wurden aus kultivierbaren Chlorophyten- und Cryptophytenarten identifiziert. Die meisten mikrobiellen Organismen kann jedoch nicht kultiviert werden, was zu einem unvollständigen Bild der ChR-Vielfalt führt. Die Metagenomik öffnet die Tür für Erkenntnisse über die Verteilung von ChRs in unkultivierten Organismen. Diese Arbeit beschreibt die biophysikalische Charakterisierung von zwei Gruppen metagenomisch identifizierter ChRs. Die MerMAIDs (Metagenomically discovered marine, anion-conducting, and intensely desensitizing ChRs) sind eine neue ChR-Familie und zeigen nahezu komplette Photostrom-Inaktivierung unter Dauerlicht. Die Photoströme lassen sich durch einen Photozyklus erklären, der zur Akkumulation eines langlebigen und nicht-leitenden Photointermediats führt. Ein konserviertes Cystein ist für dieses Phänomen entscheidend, da seine Substitution zu einer stark reduzierten Inaktivierung führt. Die Prasinophyten ChRs, die große carboxyterminale Domänen aufweisen, wurden in großen, marinen Viren identifiziert, die sie von ihren beweglichen und einzelligen Grünalgen-Wirten durch lateralen Gentransfer übernommen haben. Heterolog exprimiert, sind die viralen ChRs nur nach Ergänzung von Transportsequenzen und carboxyterminaler Kürzung funktional. Die Grünalgen- und viralen ChRs sind Anionen-leitend mit nicht-inaktivierenden Photoströmen, wenn sie in Säugetierzellen exprimiert werden, obwohl die viralen Vertreter weniger leitfähig und zytotoxisch sind. Nichtsdestotrotz repräsentiert diese ChR-Gruppe die ersten Grünalgen- und Virus-ACRs. Diese Arbeit zeigt eine breite Verteilung der ACRs unter marinen mikrobiellen Organismen und die Bedeutung der Funktionsmetagenomik bei der Entdeckung neuer ChRs.
Channelrhodopsins (ChRs) are light-gated ion channels mediating phototactic responses in motile algae and widely used as optogenetic tools to manipulate cellular activity using light. Many cation- and anion-conducting ChRs (CCRs and ACRs) have been identified from culturable chlorophyte and cryptophyte species. However, most microbial organisms cannot be cultured, resulting in an incomplete view of the diversity of ChRs. Metagenomics opens the door to gather insights on the distribution of ChRs in uncultured organisms. Here, the biophysical characterization of two groups of metagenomically identified ChRs is described. The MerMAIDs (Metagenomically discovered marine, anion-conducting, and intensely desensitizing ChRs) represent a new ChR family with near-complete photocurrent desensitization under continuous illumination. The photocurrents can be explained by a single photocycle leading to the accumulation of a long-lived and non-conducting photointermediate. A conserved cysteine is critical for this phenomenon, as its substitution results in a strongly reduced desensitization. The prasinophyte ChRs, harboring large carboxy-terminal extensions, were identified in marine giant viruses that acquired them from their motile and unicellular green algal hosts via lateral gene transfer. Expressed in cell culture, the viral ChRs are only functional upon the addition of trafficking sequences and carboxy-terminal truncation. The green algal and viral ChRs are anion-conducting and display non-desensitizing photocurrents when expressed in mammalian cells, though the viral representatives are less conductive and cytotoxic. Nonetheless, this group of ChRs represents the first green algal and viral ACRs. This thesis highlights a broad distribution of ACRs among marine microbial organisms and the importance of functional metagenomics in discovering new ChRs.

The first part of this thesis aims to develop new tools to explore mitochondrial Ca2+ dynamic in vivo by the improvement of a mitochondria targeting Cameleon probes and by creating a mitochondria targeted Channelrhodopsin. Genetically encoded calcium indicators (GECIs) allow quantitative Ca2+ measurements in different experimental models. Organelle-specific targeting signals are fused with the GECI’s sequence, achieving selective targeting to a specific organelle or cytoplasmic domain. Moreover, GECI’s coding sequences can be placed under the control of tissue specific or inducible promoters, allowing spatial and temporal control of their expression. Different types of GECIs have been created: in our laboratory we use a class of FRET-based Ca2+ sensors, called Cameleons. FRET (Förster resonance energy transfer) microscopy detects the direct transfer of energy from a donor to an acceptor fluorescence protein (FP) in a living cell. Cameleon structure consists of two Ca2+-responsive elements that alter the efficiency of FRET between two FPs: a cyan fluorescent protein (CFP), the donor, and a yellow fluorescent protein (cpV), the acceptor. The two Ca2+-responsive elements are Calmodulin (CaM) and the CaM-binding domain of myosin light chain kinase. As reported in the beginning, the aim of this study is to develop novel molecular sensors and new methodologies to express the probes in vivo in intact tissues as well as in organisms in order to explore organelle Ca2+ dynamics in vivo (in particularly in brain and heart). One of the limitations of Cameleon probes, especially critical for in vivo applications, is the low fluorescence of CFP, reducing the maximal obtainable signal-to-noise ratio, and a multi-exponential lifetime, indicating the presence of multiple excited-state decay pathways. Recently, a brighter and more stable FP, compared to CFP, has been developed: mCerulean3. It has high fluorescence quantum yield and high photostability, making this protein a good donor candidate in Cameleon probes. For this reason, CFP has been replaced with mCerulean3 in two different Cameleons: the cytosolic D3cpv and the mitochondria-targeted probe named 4mtD3cpv. The new probes have been tested in different cell types: HeLa, neonatal rat cardiomyocytes and neonatal mouse neurons. The brightness, the photostability, the pH-sensitivity and the dissociation constant (Kd) of the new probes have also been measured in situ and the data show a clear improvement in brightness and in photostability, compared to the original Cameleons, in both cytosolic and mitochondrial probes. The only drawback of the new probes is a reduced amplitude (about 20-30%) in the maximum change in the fluorescence emission ratio due to Ca2+ binding (dynamic range). In order to extend the dynamic range different approaches have been used. The addition of 16 glycines between the two Ca2+ responsive elements allowed us to generate a new cytosolic probe (D3mCerulean3+16) with an increased dynamic range (+ 20%). Similarly, we have modified the mitochondrial probe, starting from the targeting sequence, each N-terminal sequence of COX-VIII (subunit VIII of human cytochrome C oxidase) was elongated of 5 aminoacids. Indeed, 24 h after transfection a significant mis-localization in the cytosol was observed with the original probe, while the mis-targeting of the modified targeting sequence containing probe was decreased. Moreover, the reduction in the dynamic range due to the mCerulean3 presence was almost fully recovered adding the 16 glycines linker. We than analysed the effect of pH, observing a general stability of all the probe variants in the pH range tested. Finally, we evaluated the Ca2+ affinity of all the mitochondrial variants generated so far, obtaining a Kd of 0.06 µM and one of 10.84 µM in 4mtD3cpV, a Kd of 6,5 µM in 4mtD3mCerulean3 and a Kd of 0.03 µM and one of 10.7 µM in 4mtD3mCerulean3+16. To better characterize the new cytosolic and mitochondrial Cameleon probes, we purified the proteins expressed in E.Coli in order to measure also in vitro all the parameters described above. Up to now only preliminary data have been obtained, however a comparison of in situ and in vitro data suggests that the DR calculated for cytosolic probes is about 3, while that for a mitochondria like environment is about 2.5. Concerning the Ca2+ affinity, just few [Ca2+] concentrations have been tested so far, therefore all the mitochondrial and cytosolic probes display a single Kd. However, further experiments are required to confirm the in vitro calculated Kd. In parallel, we are also assessing another method to measure Ca2+ with FRET-GECIs, as an alternative of the classical intensity-based method. We are indeed employing FLIM technique that measures the lifetime of FPs (the time the molecule spend in the exited state), which is totally independent from phenomena such as probe photobleaching, expression level, image shading. Preliminary results in cells expressing all the Camelon probes described so far, suggest the presence of homo-FRET phenomena and mCerulean3-based Ca2+ sensors seem to be less affected by them. Finally, to express the new probes in vivo we are testing three strategies: i) adeno-associated virus (serotype 9) with cardiac tropism viral vectors; ii) adeno-associated virus (serotype 9) with neuron-specific promoters (synapsin promoter) for intracranial injection; iii) transgenic mice. These strategies will allow us to perform in vivo Ca2+ measurements in different tissues and subcellular compartments. We have just generated the AAV9 for the expression of the D3mCerulean3 and 4mD3mCerulean3 under the control of CMV promoter. A good expression levels in heart was obtained through intraperiotneal injection of neonatal mice. In conclusion, the results obtained so far make the mCerulean3-based Cameleons an attractive choice for in vivo experiments. To study the role of mitochondria in situ and in vivo, we also took advantage of an optogenetic tool: Channel Rhodopsins (ChRs). ChRs represent the only type of ion channels directly gated by light. When excited by light, ChRs open and depolarize the plasma membrane. For activation, opsins require binding of retinal, a vitamin A–related organic cofactor. In collaboration with Prof. Sekler’s group (Ben-Gurion University-Israel), we have developed a new mitochondria-targeted ChR, called mitochondrial Stabilized Step Function Opsin (4mt-SSFO). This ChR variant was modified to stabilize the conducting state of the channel: SSFO deactivation occurs in 30 minutes after a brief pulse of activating blue light (460-480 nm). It has also been reported the possibility to terminate the photocurrents with red light. To create a non-functional control, a truncated form Chr2(TR) lacking retinal binding site was generated. The targeting of proteins to the inner mitochondrial membrane (IMM) was obtained fusing four signal sequences, derived from human cytochrome C oxidase subunit VIII sequence, to the N-terminus of ChR. To verify the correct topology of the probe, confocal microscopy, elecrtophisiological recordings and fluorescence quenching experiments were carried out. When 4mtSSFO-YFP is properly inserted in the IMM, its C-terminal YFP tag should face the mitochondrial matrix. Indeed, the application of proteinase K to Digitonin permeabilized cells did not cause a significant reduction of 4mtChR2B-YFP signal, while that of N33D3cpv (a probe in which the YFP is located on the cytoplasmic surface of the OMM) is completely abolished. Trypan blue addition, that is permeable across the OMM, but not the IMM, did not affect the fluorescence of 4mtSSFO-YFP or that of the matrix located Cameleon 4mtD3cpv. The 4mtD3cpv fluorescence however was totally lost after alamethicin application that permeabilizes both mitochondrial membranes and releases all matrix proteins into the medium, while the fluorescence of 4mtSSFO-YFP was not affected by the latter treatment, as expected for a membrane bound protein. Taken together these data are consistent with a proper IMM localization. The new mitochondrial 4mtSSFO-YFP constructs were then tested in situ for their ability to change the mitochondrial membrane potential in response to light, resulting in a significant organelle depolarization in cells expressing 4mtSSFO-YFP, while no effect was induced by blue light in control cells. We then analysed in more details the effects of mitochondrial depolarization on mitochondrial Ca2+ uptake using two genetically encoded probes, 4mtD3cpv and mt-aequorin. Both approaches demonstrate that the depolarization of mitochondria triggered by photo-activation of the channel causes, as predicted, a significant reduction in the amplitude of the mitochondrial Ca2+ rise observed upon stimulation of HeLa cells with an IP3-generating agonist. Thus, we generated a tool able to modulate diverse mitochondria activities in a temporary-controlled, reversible (at least in principle) and cell-specific manner, offering an approach to quantitatively investigate mitochondrial role in a large variety of critical cellular processes. In the second part of the thesis, the investigation was focused on the role of Presenilins in Ca2+dyshomeostasis associated to Alzheimer's Disease (AD) using a single cell analysis approach and focusing on the role of AD Presenilin 1 and Presenilin 2 mutations in the modulation of Capacitative Calcium Entry (CCE). AD is the most frequent form of dementia. A small percentage of cases is inherited (Familial AD, FAD) and is due to dominant mutations on three genes, coding for Amyloid Precursor Protein (APP), Presenilin-1 (PS1) and Presenilin-2 (PS2). Mutations in these genes cause alterations in the cleavage of APP by a PS1- or PS2- containing enzyme, named y-secretase complex, thus leading to an increase in the ratio between Abeta42 and Abeta40, the two main peptides eventually derived from APP maturation. This event, in turn, leads to an increased deposition of amyloid plaques, one of the main histopathological feature of AD. To date, the generation of A42 peptides, its oligomers and amyloid plaques is the core of the most widely accepted pathogenic hypothesis for AD, the “Amyloid Cascade Hypothesis”. PS1 and PS2 are ubiquitous 9 transmembrane domains homologous proteins localized mainly in intracellular membranes (Endoplasmic Reticulum, ER, Golgi apparatus, and endosomes) and plasma membrane. Despite being the catalytic core of y-secretase complex, PSs display also some specialized, y-secretase independent activities. On this line, numerous studies reported a role for FAD-linked PS mutations in cellular Ca2+ alterations. Ca2+ is a key second messenger in living cells and it regulates a multitude of cell functions; thus, alterations in its signalling cascade can be detrimental for cell fate. Ca2+ mishandling has been proposed as a causative mechanism for different neurodegenerative diseases and in particular for AD. Although supported by several groups for many years, the Ca2+ hypothesis for AD pathogenesis has never been undisputedly accepted, since some data were clearly conflicting, especially those considering FAD-PS2 mutations. One of the Ca2+ pathway reported to be modulated by different FAD-PSs mutation is the so called CCE or Store Operated Calcium Entry (SOCE). CCE is the mechanism responsible for Ca2+ entry in response to ER Ca2+ depletion. The key molecules responsible for this Ca2+ entry have been identified only recently: STIM and Orai. Basically, Orai forms the channels located in the PM, while STIM is the protein that can “sense” the [Ca2+] in the ER lumen. Upon store depletion, STIM1 changes its distribution from diffuse to clusterized “puncta” and interacts with plasma membrane-located Orai1. Employing a cytosolic Cameleon probe (D3cpV), the CCE variation in SH-SY5Y cells overexpressing PSs or in human FAD and control fibroblasts was investigated. In particular, measuring the effect of PS1-A246, PS2-T122R (in overexpression) and PS1-A246 and PS2N141I (in fibroblasts) FAD-linked mutations, a decrease in both peak and rate of CCE was observed. This phenomenon could be explained by a decrease in STIM1 protein levels, while Orai1 level was not analysed because no antibody sufficiently specific is available. In the over-expression system of SH-SY5Y cells also wild type forms of PS1 and PS2 cause a decrease in CCE and this could be due to the accumulation of the full-length form of the proteins that is thought to be the mediator of the effect
La prima parte di questa tesi si propone di sviluppare nuovi strumenti per esplorare l'omeostasi del calcio (Ca2+) mitocondriale in vivo, migliorando la sonda Cameleon indirizzata ai mitocondri, già presente in laboratorio, e creando un canale regolato dalla luce (Channelrhodopsin) indirizzato a sua volta ai mitocondri. Gli indicatori per il Ca2+ geneticamente codificati (Genetically encoded calcium indicators, GECIs), consentono di eseguire misure quantitative di Ca2+ in diversi modelli sperimentali. Segnali di indirizzamento a specifici organelli possono essere fusi con la sequenza codificante i GECIs, consentendo di misurare variazioni della concentrazione di Ca2+ in diversi compartimenti subcellulari. Inoltre, le sequenze codificanti i GECIs possono essere poste sotto il controllo di promotori tessuto-specifici o inducibili, consentendo così il controllo spaziale e temporale della loro espressione. Diversi tipi di GECIs sono stati creati: nel nostro laboratorio usiamo una classe di sensori Ca2+ basati su FRET, chiamati Cameleon. La microscopia FRET (trasferimento di energia di risonanza Förster) si basa sul trasferimento diretto di energia da una proteina fluorescente (FP), detta donatore, a un'altra FP, detta accettore, in una cellula vivente. Strutturalmente, la sonda Cameleon, è composta da due elementi responsivi per il Ca2+, che modificano l'efficienza di FRET tra le due FPs. All'interno di questa sonda sono infatti presenti due FPs: una proteina fluorescente di colore ciano (CFP), il donatore, e una proteina fluorescente gialla (cpV), l'accettore. I due elementi che rispondono a variazioni di Ca2+ sono la Calmodulina (CaM) e il dominio di legame della calmodulina (M13) della chinasi della catena leggera della miosina. Come anticipato, lo scopo del mio studio è lo sviluppo di sensori molecolari e di nuove metodologie per esprimere le sonde in vivo al fine di esplorare le dinamiche di Ca2+ di particolari organelli in vivo (in particolare nel cervello e nel cuore). Un limite delle sonde Cameleon, particolarmente critico per le applicazioni in vivo, è il basso grado di fluorescenza della CFP, che causa un basso rapporto segnale-rumore ed è caratterizzato da un tempo di vita multi-esponenziale, che indica la presenza di percorsi multipli di decadimento dallo stato eccitato. Recentemente è stata sviluppata una FP più luminosa e più stabile, rispetto alla CFP: mCerulean3. Quest'ultima ha elevata resa quantica di fluorescenza e alta fotostabilità, rendendo questa proteina un buon candidato alla sostituzione del donatore originale nelle sonde Cameleon. Per questo motivo, la CFP è stata sostituita con mCerulean3 in due differenti sonde Cameleon: la sonda citosolica D3cpv e la sonda mitocondriale 4mtD3cpv. Le nuove sonde sono state testate in diversi tipi cellulari: HeLa, cardiomiociti di ratto neonato e neuroni da topi neonati. Sono quindi stati misurati in situ vari parametri indicativi della qualità delle nuove sonde create, quali la luminosità, la fotostabilità, la sensibilità al pH e la costante di dissociazione (Kd), mostrando un netto miglioramento della luminosità e fotostabilità, rispetto alle sonde Cameleon originali. L'unico inconveniente delle nuove sonde è una riduzione del range dinamico, di circa il 20-30%, un parametro molto importante che rende conto della massima variazione del rapporto di fluorescenza emessa alle due lunghezze d'onda in seguito a variazioni di Ca2+. Al fine di sopperire a tale svantaggio, sono stati utilizzati diversi approcci, tra cui l'aggiunta di una sequenza di 16 glicine tra i due elementi responsivi al Ca2+ che ha consentito di recuperare il range dinamico della sonda mitocondriale e di aumentare del 20% quello della sonda citosolica, senza modificare i miglioramenti prima descritti, ottenuti con la semplice sostituzione del donatore. Parallelamente, poiché era stata osservata una scarsa localizzazione mitocondriale a 24 ore dalla trasfezione del 4mtD3cpV, anche la sequenza di indirizzamento è stata modificata (allungando ciascuna delle sequenze originali di 5 minoacidi), diminuendo notevolmente la quantità di sonda espressa erroneamente nel citosol. Come anticipato, l'effetto del pH è stato valutato, evidenziando una stabilità generale delle varie delle sonde tra pH 7 e pH 9. Infine, sono state calcolate le costanti di affinità delle varie sonde mitocondriali. Una doppia affinità per il Ca2+ è stata rilevata: la sonda 4mtD3cpV ha una Kd di 0.06 µM e una di 10.84 µM, mentre la sonda 4mtD3mCerulean3+16 ha una Kd di 0.03 µM e una di 10.7 µM. Per caratterizzare meglio le nuove sonde Cameleon, citosolica e mitocondriale, esse sono state prodotte in E.Coli e purificate per misurare anche in vitro tutti i parametri sopra descritti. Ad oggi sono stati ottenui solo dati preliminari che evidenziano una riduzione del range dinamico passando dall’ambiente citosolico (DR=3) a quello mitocodnriale (DR=2.5). Per quanto riguarda l’affinità per il Ca2+, solo poche concentrazioni di questo ione sono state testate, dando come risultato una Kd singola sia per la sonda mitocondriale che per quella citosolica. Tale dato dovrà essere confermato da ulteriori esperimenti. Tipicamente, le variazioni di Ca2+ rilevate da sonde basate su FRET vengono effettuate utilizzando metodi basati sull'intensità delle due proteine fluorescenti, tuttavia tecniche alternative basate sul tempo di vita della proteina sembrano dare risultati più accurati e robusti. Stiamo perciò iniziando a valutare biofisicamente le sonde Cameleon utilizzando la tecnica FLIM, tecnica che misura quanto tempo una molecola fluorescente resta nello stato eccitato. Il FLIM presenta diversi vantaggi in quanto è totalmente indipendente da fenomeni come il photobleaching, i diversi livelli di espressione, i cambi di fuoco. I dati prelimianri raccolti eveidenziano la presenza di fenomeni di homo-FRET nelle sonde testate, fenomeni che sembrano tuttavia essere meno evidenti nelle sonde in cui la CFP è stata sostituita da mCerulean3. Infine, per esprimere questi due nuove sonde in vivo stiamo applicando tre strategie: i) la creazione di un virus adeno-associato (sierotipo 9) con tropismo cardiaco; ii) la generazione di un altro virus adeno-associato (sierotipo 9) dotato di un promotore neurone specifico (sinapsina) per l'iniezione intracranica; iii) la creazione di topi transgenici. Queste strategie ci permetteranno di effettuare misurazioni in vivo di Ca2+ in diversi tessuti e compartimenti subcellulari. Recentemente, la generazione di un AAV9 per l'espressione del D3mCerulean3 e 4mD3mCerulean3 sotto il controllo di un promotore ubiquitario (CMV) ci ha consentito di osservare un buon livello di espressione nel cuore, iniettando il virus per via intraperitoneale in topi neonati. In conclusione, i risultati finora ottenuti rendono le nuove sonde Cameleon, basate su mCerulean3, una scelta interessante per gli esperimenti in vivo. Come anticipato, al fine di studiare il ruolo dei mitocondri in situ e in vivo, abbiamo anche approfittato di uno strumento optogenetico, le Channelrhodopsine (ChRs), canali appartenente alla famiglia delle opsine. I ChRs sono l'unico tipo di canali ionici, direttamente controllati dalla luce, in grado di indurre depolarizzazione della membrana plasmatica, se opportunamente illuminati. Per l'attivazione, le opsine richiedono il legame con un cofattore organico della vitamina A, chiamato retinale. In collaborazione con il gruppo del Prof. Sekler (Ben-Gurion University-Israel), abbiamo fuso una variante del classico ChR2, detta SSFO (Step stabilized function Opsin), con una sequenza di localizzazione mitocondriale, al fine di creare una canale controllato dalla luce in grado di depolarizzare i mitocondri in modo reversibile (4mt-SSFO). Infatti, le SSFOs sono varianti di ChR in grado di aprirsi, se eccitate con luce blu, e di chiudersi, dopo circa 30 minuti dall'attivazione. É possibile tuttavia accelerare la chiusura del canale utilizzando della luce rossa. Per creare un controllo non funzionante di questo 4mt-SSFO, è stata creata una forma tronca di ChR2(TR) mancante del sito legante il retinale. La localizzazione di SSFO alla membrana mitocondriale interna (IMM) è stato ottenuta fondendo al N-terminale del ChR quattro sequenze di indirizzamento mitocondriale derivate da quella presente nella subunità VIII della citocromo C ossidasi umana. Per verificare la corretta topologia della sonda, sono stati effettuati esperimenti in microscopia confocale, di elettrofisiologia e di fluorescenza sfruttando la variante del 4mtSSFO fusa ad una YFP. Utilizzando vari “quenchers” di fluorescenza e diverse proteasi (Trypan Blue, Proteinase K, Alamethicin) abbiamo dimostrato che il nuovo canale localizza correttamente in IMM con il C-terminale protetto nella matrice mitocondriale. Il secondo passo è stata la verifica della funzionalità dei nuovi costrutti 4mtSSFO-YFP in situ, verificando la loro capacità di modificare il potenziale di membrana mitocondriale in risposta alla luce. Confermata la funzionalità, abbiamo poi analizzato più in dettaglio l'effetto della depolarizzazione mitocondriale sulla capacità del mitocondrio di regolare l'ingresso di Ca2+, utilizzando due sonde geneticamente codificate, il 4mtD3cpv e l'aequorina indirizzata ai mitocondri. Entrambi gli approcci dimostrano che la depolarizzazione dei mitocondri, innescata dalla fotoattivazione del canale, induce, come atteso, una significativa riduzione dell’entrata del Ca2+ all'interno del mitocondrio. L'ingresso di Ca2+ in questo organello è stato indotto stimolando cellule HeLa con un agonista producente IP3. Concludendo, i dati ottenuti sino ad ora confermano la generazione di uno strumento molecolare in grado di modulare l'attività dei mitocondri con precisione temporale e in modo reversibile (almeno in linea di principio) e specifico, offrendo così un nuovo approccio per determinare quantitativamente il ruolo mitocondriale in una grande varietà di processi cellulari critici. Nella seconda parte della mia tesi, mi sono concentrata sullo studio del ruolo delle Preseniline nell'alterazione dell'omeostasi del Ca2+ nella malattia di Alzheimer (AD), utilizzando un approccio a singola cellula e concentrandomi sul ruolo delle mutazioni presenti in Presenilina 1 e Presenilina 2 correlate ad AD nella regolazione dell'ingresso capacitativo di calcio (CCE). AD è la forma più frequente di demenza. Una piccola percentuale di casi è ereditaria (Familial AD, FAD) ed è dovuta a mutazioni dominanti in tre geni, che codificano per la proteina precursore dell'amiloide (APP), Presenilina-1 (PS1) e presenilina-2 (PS2). Mutazioni in questi geni provocano alterazioni nel taglio di APP mediato da un enzima, detto complesso y-secretasico, che annovera tra i suoi componenti PS1 e PS2, e che è in grado di causare un aumento del rapporto tra Abeta42 e Abeta40, i due peptidi derivati dalla maturazione di APP. La generazione di tali peptidi, a sua volta, aumenterebbe la deposizione di placche amiloidi, una delle caratteristiche istopatologiche principali dell'AD. Ad oggi, la generazione dei peptidi Abeta42, dei suoi oligomeri e delle placche amiloidi, costituisce il cuore dell’ipotesi patogenetica più accreditata per AD, “l'ipotesi della cascata amiloide". PS1 e PS2 sono proteine omologhe e ubiquitarie formate da 9 domini trans-membrana. Esse sono localizzate prevalentemente nelle membrane interne di reticolo endoplasmatico (ER), apparato di Golgi, endosomi e nella membrana plasmatica (PM). Pur essendo il nucleo catalizzatore della y-secretasi, le PSs svolgono anche ruoli indipendenti dall'attività di questo enzima, tra cui la regolazione dell'omeostasi del Ca2+. Quest'ultimo fenomeno infatti risulta variamente alterato in presenza di mutazioni FAD nelle PSs. Il Ca2+ è un secondo messaggero intracellulare chiave nelle cellule viventi e regola una moltitudine di funzioni cellulari, pertanto alterazioni nelle sue vie di segnale possono essere dannose per il destino della cellula. Alterazioni nell'omeostasi del Ca2+ sono state proposte come meccanismo causale per diverse malattie neurodegenerative e in particolare per l'AD. Sebbene supportata da diversi lavori, l'ipotesi del Ca2+ per la patogenesi dell'AD è stata a lungo messa in discussione, a causa dello scarso consenso della comunità scientifica sull'effetto di mutazioni in PSs nella regolazione di questo ione. Uno dei meccanismi che sembra essere modulato da diverse mutazioni FAD associate a PSs, è il cosiddetto CCE, o Store Operated Calcium Entry (SOCE). Il CCE è il meccanismo responsabile dell'influsso di Ca2+ all'interno della cellula in risposta allo svuotamento dell'ER. Le molecole chiave responsabili del CCE sono state identificate solo di recente: STIM e Orai. Brevemente, Orai1 forma i canali situati in PM, mentre STIM1 è la proteina che "sente" la [Ca2+] nel lume dell'ER. Quando l'ER si svuota, STIM1 cambia la sua distribuzione diffusa, oligomerizza in punte (puncta) discrete che, a questo punto, sono in grado di interagire con Orai1 a livello della PM. Utilizzando una sonda citosolica Cameleon (D3cpV), ho valutato le variazione di CCE in cellule SH-SY5Y sovra-esprimenti PSs e in fibroblasti derivati da pazienti FAD o da controlli sani. In particolare, le mutazioni FAD, PS1-A246 e PS2-T122R (in sovra-espressione) e PS1-A246 e PS2-N141I (in fibroblasti) causano una diminuzione del CCE, sia della sua entità che della sua velocità di attivazione. Questo fenomeno può essere spiegato da un diminuito livello di STIM1 nei campioni FAD rispetto ai controlli, mentre non è stato possibile valutare i livelli proteici di Orai, poiché non è disponibile un anticorpo abbastanza specifico. Negli esperimenti di sovra-espressione, anche la forma wild type di PS1 e PS2 causa una diminuzione del CCE, fenomeno probabilmente dovuto all'accumulo della forma lunga (non tagliata) delle PSs, forma che sembra essere responsabile degli effetti mediati dalle PSs sull'omeostasi del Ca2+

The retina contains neural circuits that carry out computations as complex as object motion sensing, pattern recognition, and position anticipation. Models of some of these circuits have been recently discovered. A remarkable outcome of these efforts is that all such models can be constructed out of a limited set of components such as linear filters, instantaneous nonlinearities, and feedback loops. The present study explores the consequences of assuming that these components can be used to construct models for all retinal circuits. I recorded extracellularly from several retinal ganglion cells while stimulating the photoreceptors with a movie rich in temporal and spatial frequencies. Then I wrote a computer program to fit their responses by searching through large spaces of anatomically reasonable models built from a small set of circuit components. The program considers the input and output of the retinal circuit and learns its behavior without over-fitting, as verified by running the final model against previously unseen data. In other words, the program learns how to imitate the behavior of a live neural circuit and predicts its responses to new stimuli. This technique resulted in new models of retinal circuits that outperform all existing ones when run on complex spatially structured stimuli. The fitted models demonstrate, for example, that for most cells the center--surround structure is achieved in two stages, and that for some cells feedback is more accurately described by two feedback loops rather than one. Moreover, the models are able to make predictions about the behavior of cells buried deep within the retina, and such predictions were verified by independent sharp-electrode recordings. I will present these results, together with a brief collection of ideas and methods for furthering these modeling efforts in the future.
Physics

The field of neuroscience has recently seen optogenetics emerge as a highly utilized and powerful method of non-invasive neural activation and inhibition. This thesis seeks to enhance the optogenetic toolbox through the design, construction, and evaluation of a number of hardware and software modules for research in Caenorhabditis elegans neuroscience. In the first aim, we combine optogenetics, microfluidics, and automated image processing, to create a system capable of high-throughput analysis of synaptic function. In the second aim, we develop a multi-modal illumination system for the manipulation of optogenetic reagents. The system is capable of multi-spectral illumination in definable patterns, with the ability to dynamically alter the intensity, color, and shape of the illumination. The illumination system is controlled by a set of software programs introduced in aim three, and is demonstrated through a set of experiments in aim four where we selectively activate and inhibit specific neural nodes expressing optogenetic reagents in freely moving C. elegans. With the ability to target specific nodes in a freely moving animal, we can correlate specific neural states to behaviors allowing for the dissection of neural circuits. Taken together, the developed technologies for optogenetic researchers will allow for experimentation with previously unattainable speed, precision and flexibility.

The heart is densely innervated by sympathetic neurons (SN) that regulate cardiac function both through chronotropic and inotropic effects. During exercise and stress, SN-released norepinephrine activates cardiac beta adrenergic receptors (beta-ARs) on both the conduction and contractile systems. Increased cardiac sympathetic activity leads to arrhythmias in acquired (e.g. myocardial ischemia) or inherited conditions, including Catecholaminergic Polymorphic Ventricular Tachycardia (CPVT), possibly via development of Ca2+ overload-dependent early- or delayed-afterdepolarizations (EAD, DAD, respectively). The DAD would serve as arrhythmogenic focus, leading to the onset of triggered activity in discrete groups of cardiac cells. Unbalanced sympathetic discharge to different regions of the heart has been identified as a potent arrhythmogenic condition 1. In addition to the direct cardiomyocyte damage, alteration in presynaptic NE reuptake from the autonomic neuron endings, leading to catecholamine spillover in the failing myocardium 2, inducing is an arrhythmic event. These data support a model in which autonomic control of cardiac function relies on specialized sites of direct interaction between the neurons and their target cardiomyocytes (CM). The aims of the project are: 1. To investigate whether specific cell-cell interactions have a role in the dynamics of intercellular signaling between SN and CM, aims to understand how unbalanced SN activity leads to arrhythmic condition. 2. To understand whether the unbalanced SN modulation of a limited group of cardiac cells could be involved in generating arrhythmias in vivo, based on an optogenetic approach 3. To study in vivo, non-invasively, the critical mass of myocardium necessary to generate an arrhythmogenic focus, using optogenetics. In the first part of the project, we used an in vitro model of sympathetic neurons/cardiomyocytes (SN-CM) co-cultures to analyze the dynamics of intercellular signaling. Upon NGF treatment, SNs extend their axons and establish direct contact with CMs. NE-synthesizing terminals developed on SN at the contact site, and beta1-ARs were enriched on the CM membrane in correspondence of the active release areas 3. We performed real-time imaging using the FRET-based biosensors EPAC1-camps and AKAR3 to assess intracellular cAMP and PKA activity, respectively. Stimulation of SN was achieved using KCl or bradykinin. We observed that activation of a specific SN lead to cAMP increase in the interacting CM (ΔR/R0 = 5.6% ± 1% mean ± SEM, n = 8, AKAR3 ΔR/R0= 5.3% ± 1.5%, mean ± SEM, n=6). The cAMP response in cardiomyocytes was not due to NE released in the medium, and was absent in cells not in direct contact with the activated neuron. We showed that in cells without SN coupled the intracellular cAMP and PKA activity were not affected. To estimate the [NE] acting on the CM beta-AR at the contact site, we compared the amplitude of the FRET signal evoked by SN activation (ΔR/R0= 2.6 % ± 0.6%, mean ± SEM, n=13 ) to that elicited by different [NE] administered to the cell bathing solution, and we observed that the increase in the CFP/YFP ratio achieved by SN-released NE is comparable to that obtained with 3.5e-10 M NE to whole cell. Using the competitive beta-antagonist propranolol we determined the effective [NE] in the ‘synaptic’ cleft. Competition antagonism of neuronal stimulation to CM was obtained with [Propranolol] equal to that antagonizing 100 nM of NE, indicating that such concentration is achieved in the ‘synaptic cleft’. Moreover, by calculating the fraction of occupancy of the receptor at different concentration of NE we calculated that the fraction of beta-ARs activated by the SN-released NE is < 1%. 2. In the second part of the project we used an optogenetic-based strategy to modulate cardiac sympathetic neurons activity non invasively in vivo. ChR2 is a light-gated cation channel that becomes permeable mainly to Na+ upon light-stimulation, shown to enable control of neuronal activity both in vitro and in the intact brain. We generated a mouse model expressing ChR2 in SN under the tyrosine hydroxilase (TOH) promoter. Photostimulation of the stellate ganglia neurons (SGN) was obtained in an anesthetized, open-chest model using a fiber optic to locally (1mm) deliver light (470nm) generated from a LED. ECG recording demonstrates a rapid (100-150 ms) increase (40%±6%) in heart rate (HR) upon SGN stimulation. The extremely short activation time of the cardiac response upon ChR2 depolarization of the neurons support a model in which NE acts in a short range, consistent with direct interaction between SN and CM. 3. We used ChR2 to modulate cardiac electrophysiology. We determined in cultured neonatal cardiomyocytes that photostimulation allows triggering action potential (AP). Moreover depending on when the light pulses were given we generated normal AP, early- or delayed-aferdepolarizations (EAD or DAD). We generated a mouse model with cardiac expression of ChR2, driven by the α-MHC promoter. Optical control of cardiomyocyte membrane potential was obtained with a fiber optic, while recording the ECG in the anesthesized mouse. Stimulation was applied to different regions of the heart. Atrial illumination was used to obtain non-invasive atrial pacing resulting in tachycardia with unchanged QRS, indicating as expected that the cardiac activation wave followed the natural conduction system. Ventricular photoactivation, on the contrary, bypassing the natural conduction system gave rise to premature ventricular beats. We provide evidence of the existence of a ‘synaptic’ contact between SN and CM that forms a high agonist concentration, diffusion-restricted space allowing potent activation of a small fraction of beta-ARs on the CM membrane upon neuronal stimulation. SN stimulation leads to a rapid increase of the HR supporting the idea of the existence of the synaptic contact between SN and CM. This close interaction has the potential of fast control of local CM signalling, suggesting that SNs control locally discrete groups of myocardial cells. Stimulation of a small fraction of the cardiac cells (< 200 microm-wide area) induced ectopic beats conducted to the whole heart
Il cuore è densamente innervato dai neuroni del sistema nervoso simpatico che regolano la funzionalità cardiaca attraverso un effetto cronotropo o inotropo positivi. Durante lo stress o l’esercizio, la noradrenalina rilasciata dai neuroni attiva i β recettori cardiaci sia sul sistema di conduzione che sul muscolo contrattile. L’aumento dell’attività del sistema nervoso simpatico cardiaco sia in condizioni normali o in presenza di patologie genetiche, come per esempio la Tachicardia Catecolaminergica Polimorfica Ventricolare, porta ad aritmie presumibilmente attraverso l’insorgere di ‘DADs’. Le ‘DADs’ sono un focus di aritmia che porta a una serie di depolarizzazioni che interessano un piccolo gruppo di cellule cardiache. E’ stato identificato un rilascio di catecolamine non bilanciato in diverese regioni del cuore da parte del sistema nervoso simpatico come possibile causa di aritmia. Inoltre alterazioni del ‘reuptake’ di noradrenalina porta a una concentrazione anomala di NE nello scompenso cardiaco che può essere coinvolto in un evento aritmico. Questi dati supportano un modello in cui il controllo della funzionalità cardiaca da parte del sistema nervoso simpatico avviene attraverso un sito d’interazione diretta e specializzata fra neurone e cardiomiocita accoppiato. Gli scopi del progetto sono quindi: 1. Studiare se l’interazione fra neurone e cardiomiocita ha un ruolo nella trasmissione cardiaca del segnale, per capire come un’attività non bilanciata del sistema nervoso simpatico porta a un evento aritmico. 2. Capire se l’attività non bilanciata del sistema nervoso simpatico modulando l’attività di un piccolo gruppo di cellule cardiache, possa essere coinvolto nella generazione di un’aritmia in vivo. Per verificare quest’ipotesi ci serviremo di un approccio innovativo basato su proteine foto attivabili 3. Studiare in vivo e in maniera non invasiva la massa critica di cellule cardiache necessaria per scatenare un evento aritmico. Anche per questo tipo di studio abbiamo utilizzato una metodologia basata sull’optogenetica. Nella prima parte del progetto, abbiamo creato un modello in vitro costituito da cardiomiociti neonatali e neuroni isolati dal ganglio cervicale superiore. I neuroni in seguito a trattamento con NGF sviluppano assoni che stabiliscono contatti con i cardiomiociti. Sotto terminali che sono in contatto con le cellule cardiache si osserva un maggiore accumulo di β1 recettori [3]. Abbiamo misurato l’attivazione dei β recettori monitorando in tempo reale le variazioni di AMP ciclico e attività di PKA, attraverso l’uso di sensori geneticamente codificati e che si basano sul FRET (EPAC1-camps, che ci permette di monitorare cAMP e AKAR3 che ci permette di monitorare l’attività di PKA). I neuroni del SNS sono stati stimolati con KCl o bradichinina. Abbiamo osservato che stimolando il rilascio di noradrenalina da un neurone, l’AMP ciclico e l’attività di PKA aumentano solo nei cardiomiociti accoppiati a neurone e non nei cardiomiociti senza un contatto (ΔR/R0 = 0.056 ± 0.01 mean ± SEM, n = 8, AKAR3 ΔR/R0=5.3% ± 1.5%, mean ± SEM, n=6). Per stimare la [NE] che agisce sui β recettori nel sito di contatto abbiamo paragonato l’ampiezza del segnale FRET generato dall’attivazione neuronale (ΔR/R0= 0.026 ± SEM) con quello generato da diverse [NE] note aggiunte alla soluzione in cui si trovano le cellule. Abbiamo osservato che l’aumento del rapporto CFP/YFP ottenuto dalla noradrenalina rilasciata dai neuroni e paragonabile a quello ottenuto con 3.5e-10 M di noradrenalina che attiva tutti i recettori. Usando un antagonista competitivo dei β recettori (propranololo) abbiamo determinato la concentrazione di noradrenalina nel cleft sinaptico. La concentrazione di propranolol necessaria per abolire totalmente la risposta indotta dalla noradrenalina rilasciata dai neuroni, e pari a quella necessaria per bloccare la risposta indotta da 100 nM di noradrenalina, suggerendo che la concentrazione nel cleft sinaptico è dell’ordine di 100 nM. Sulla base di questi dati abbiamo quindi calcolato che la frazione recettoriale con cui interagisce la noradrenalina rilasciata dai neuroni che è inferiore all’1% del totale. 1.Nella seconda parte del progetto abbiamo usato una strategia che si basa sull’‘optogenetica’ per modulare l’attività del sistema nervoso simpatico in vivo e in maniera non invasiva. ChR2 è un canale la cui permeabilità è regolata dalla luce. Infatti questo canale diventa permeabile soprattutto al Na+ in seguito a stimolazione con luce blu. Negli ultimi anni è stato largamente utilizzato per il controllo dell’attività neuronale sia in vitro che in vivo [4, 5]. Abbiamo generato un modello murino che esprime ChR2 nei neuroni del sistema nervoso simpatico sotto il promotore tirosina idrossilasi. La foto stimolazione del ganglio stellato è stata ottenuta in un modello a ‘torace aperto’ di topo anestetizzato, usando una fibra ottica per indirizzare in uno specifico punto la luce generata da un LED. L’analisi dell’ECG del topo mostra un rapido (100-150 ms) aumento (40%±6%) nella frequenza di contrazione cardiaca in seguito a ‘fotostimolazione’ del ganglio stellato. Questo rapido aumento nella frequenza cardiaca supporta il modello in cui la noradrenalina agisce in uno spazio piccolo e confinato in cui neurone e cardiomiocita interagisccono direttamente. 3. Abbiamo usato ChR2 anche per modulare l’elettrofisiologia cardiaca. Abbiamo determinato che la fotostimolazione di ChR2 è sufficiente a modulare il potenziale d’azione in cardiomiociti neonatali in cultura. Inoltre a seconda di quando viene dato il pulso di luce siamo in grado di generare un battito normale, una DAD o una EAD. Abbiamo quindi generato un modello di topo che esprima ChR2 nel cuore sotto il promotore α-MHC. Abbiamo controllato tramite stimolazione luminosa il potenziale d’azione di cellule cardiache utilizzando fibre ottiche alimentate da LED, durante l’acquisizione dell’ECG del topo. La stimolazione è stata eseguita in diverse regioni del cuore. La stimolazione atriale ci ha permesso di mimare un pacing atriale sfociato poi una tachicardia. Abbiamo osservato che il QRS non ha variazioni rispetto al normale, indicando che l’onda di depolarizzazione segue il sistema di conduzione cardiaco. La foto attivazione ventricolare invece genera un battito prematuro dato che non segue il sistema di conduzione. Abbiamo qui dimostrato l’esistenza di un contatto sinaptico fra i neuroni e i cardiomiociti che forma un sito a elevata concentrazione di neurotrasmettitore, uno spazio a diffusione limitata permettendo quindi l’attivazione di un ristretto gruppo di recettori β localizzati nella membrana della cellula cardiaca. La stimolazione neuronale genera un rapido aumento nella frequenza cardiaca avvalorando l’ipotesi dell’esistenza di un contatto sinaptico fra neuroni e cardiomiociti. Questa interazione è importante per un controllo rapido del segnale locale dei cardiomiociti, suggerendo che i neuroni controllino un gruppo ristretto di cellule cardiache. La stimolazione di una frazione di cardiomiociti è sufficiente a indurre un battito condotto in tutto il cuore

Kanalrhodpsine (ChRs) sind lichtaktivierbare Kationenkanäle, welche als primäre Fotorezeptoren in Grünalgen dienen. In der Optogenetik werden ChRs verwendet um neuronale Membranen zu depolarisieren und mit Licht Aktionspotentiale auszulösen. Das mit blauem Licht aktivierte Chlamydomonas Kanalrhodopsin 2 (C2) und effiziente Mutanten wie C2 H134R stellen die am häufigsten genutzten depolarisierenden, optogenetischen Werkzeuge dar. Komplementär zu ChRs werden Protonen- und Chloridpumpen aus Archaebakterien zur neuronalen Inhibierung durch lichtinduzierte Hyperpolarisation verwendet. In der vorliegenden Arbeit untersuchten wir die ChR-Chimäre C1V1, ein grünlichtaktiviertes ChR, das sich durch hervorragende Membranlokalisierung und hohe Fotoströme in HEK-Zellen auszeichnet. C1V1 und C1V1-Mutanten mit feinabgestimmten spektralen und kinetischen Eigenschaften ermöglichen die neuronale Aktivierung mit Wellenlängen bis 620 nm sowie die unabhängige Aktivierung zweier Zellpopulationen in Kombination mit C2. Um die strukturelle Basis von Kanalöffnung und Ionentransport in ChRs zu verstehen, wurden gezielt Mutationen in C2 und C1V1 eingeführt. Die Fotoströme der entsprechenden Mutanten wurden auf Kationenselektivität und kinetische Veränderungen untersucht. Während Aminosäuren, die den Kanal an der zytosolischen Seite begrenzen, die Kationenfreisetzung und Einwärtsgleichrichtung der ChRs bestimmen, spielen zentral im Kanal gelegende Aminosäuren ein entscheidende Rolle für Kationenselektivität und -kompetition. Ein enzymkinetisches Modell ermöglichte außerdem die Zerlegung der Fotoströme in Beiträge der verschiedenen, konkurrierenden Kationen. Im letzten Teil der Arbeit wurde pHoenix, ein optogenetisches Werkzeug zur Ansäuerung synaptischer Vesikel, entwickelt. In Neuronen des Hippocampus wurde pHoenix verwendet, um die treibenden Kräfte für die vesikuläre Neurotransmitteraufnahme sowie den Zusammenhang zwischen Vesikelfüllstand und Freisetzungswahrscheinlichkeit zu analysieren.
Channelrhodopsins (ChRs) are light-activated cation channels functioning as primary photoreceptors in green algae. In the emerging field of optogenetics, ChRs are used to depolarize neuronal membranes, thus allowing for light-induced action-potential firing. The blue light-activated Chlamydomonas channelrhodopsin 2 (C2) and high-efficiency mutants such as C2 H134R represent the most commonly used depolarizing optogenetic tools. Complementary to ChRs, green to yellow light-activated proton and chloride pumps originating from archea enable neuronal inhibition by membrane hyperpolarization. In the present work, we developed the chimeric ChR C1V1, a green-light activated ChR with excellent membrane targeting and high photocurrents in HEK cells. Action spectrum and kinetic properties of C1V1 were further fine-tuned by site-directed mutagenesis. The ensemble of C1V1 variants allows for neuronal activation with wavelengths up to 620 nm and can be used in two-color optogenetic experiments in combination with C2 derivatives. In order to understand the structural motifs involved in channel gating and ion transport, conserved residues in C2 and C1V1 were mutated and photocurrents of the respective mutants were analyzed for kinetic characteristics and cation selectivity. In these experiments, residues of the inner gate region were shown to alter cytosolic cation release and inward rectification, whereas central gate residues determine cation competition and selectivity, as well as the equilibrium between the two open channel conformations. Moreover, an enzyme-kinetic model was used to quantitatively dissect ChR photocurrents into the contribution of different competing cations. Finally, we designed pHoenix, an optogenetic tool enabling green-light induced acidification of synaptic vesicles. In hippocampal neurons, pHoenix was used to study both the energetics of vesicular neurotransmitter uptake and the impact of the vesicular contents on synaptic vesicle release.

Kanalrhodopsine (ChRs) sind lichtgesteuerte Ionenkanäle aus einzelligen Grünalgen, die in der Optogenetik verwendet werden. Photonabsorption führt zur Isomerisierung des Retinal-Kofaktors, was eine Reihe von Reaktionen auslöst, die als Photozyklus bezeichnet werden und die Bildung des leitenden Zustands umfassen. In dieser Arbeit wurde der Photozyklus-Mechanismus ausgewählter ChRs mittels FTIR (Fourier Transform Infrarot)- und UV-Vis-Spektroskopie, sowie Retinalextraktion und HPLC (Hochleistungsflüssigkeitschromatographie)-Analyse untersucht. Photorezeptoren sind dafür optimiert, Lichtenergie zu nutzen, um Konformationsänderungen des Proteins hervorzurufen. Dafür wird ein Teil der Lichtenergie durch eine transiente Verdrillung des Chromophors gespeichert. In dieser Arbeit wird gezeigt, dass der Transfer der gespeicherten Energie zum Protein in ReaChR stark vom Protonierungszustand von Glu163 beeinflusst wird; er wird durch eine erhöhte Rigidität des aktiven Zentrums bei protoniertem Glu163 verlangsamt. In Chrimson hingegen relaxiert der Chromophor nach Photoisomerisierung, was auf einen verdrillten Chromophor im Dunkelzustand hinweist, was vermutlich für die bathochrome Verschiebung von Bedeutung ist. Zusätzlich zur Chromophorgeometrie beeinflusst der Protonierungszustand von Glu163 in ReaChR und dem homologen Glu165 in Chrimson die Stereoselektivität der Photoreaktion. Ein weiterer Faktor der Stereoselektivität ist Asp196 in ReaChR (Asp195 in C1C2), welches im Photozyklus deprotoniert. Die Bildung des leitenden Zustands in C1C2 und ReaChR geht mit einem Wassereinstrom ins Protein einher, welcher den Transport größerer Kationen erleichtert. Die Deprotonierung von Glu130 in ReaChR (Glu129 in C1C2) verändert die Ionenselektivität des Kanals, wie aus elektrophysiologischen Messungen bekannt ist. In Chrimson ist das Ausmaß des Wassereinstroms deutlich reduziert, was – in Übereinstimmung mit elektrophysiologischen Experimenten – den Transport von Protonen begünstigt.
Channelrhodopsins (ChRs) are light-gated ion channels found in single-cell algae and used in optogenetics. Photon absorption leads to isomerization of the retinal cofactor, initiating a number of reactions that are referred to as photocycle and involve formation of the ion-conducting state. In this thesis, the photocycle mechanism of selected ChRs was investigated using FTIR (Fourier Transform Infrared) and UV-Vis spectroscopy, as well as retinal extraction and subsequent HPLC (High Performance Liquid Chromatography) analysis. Photoreceptors are optimized to use photon energy to drive conformational changes of the protein. Therefore, a fraction of the photon energy is stored by a transient distortion of the chromophore. In this thesis, it is shown that in ReaChR the transfer of the stored energy to the protein is largely affected by the protonation state of Glu163, being decelerated by protonated Glu163 due to an enhanced rigidity of the active site. In contrast, the chromophore in Chrimson relaxes upon photoisomerization, hinting at a distorted retinal geometry in the dark state, which is probably essential for its unprecedented bathochromic absorption. In addition to the chromophore geometry, the protonation state of Glu163 in ReaChR and the homologue Glu165 in Chrimson affects the stereoselectivity of the photoreaction. Another factor for stereoselectivity is Asp196 in ReaChR (Asp195 in C1C2) which deprotonates in the photocycle. Formation of the ion-conducting state in C1C2 and ReaChR involves water influx into the protein, facilitating transport of larger cations. Deprotonation of Glu130 in ReaChR (Glu129 in C1C2) alters the ion selectivity of the channel as known from electrophysiological experiments. In Chrimson, the extent of water influx is drastically reduced which favors the conductance of protons in agreement with electrophysiological characterization.

IL’utilizzo della luce per lo studio del cervello ha principalmente riguardato lo sviluppo di tecniche di imaging, la fotonica ed i suoi principi sono state utilizzate negli ultimi anni manipolare, stimolare e studiare l’attività di cellule cerebrali. Lo scopo di questo lavoro di tesi è quello di passare in rassegna in modo critico i più recenti risultati ottenuti nella modulazione attivazione di attività di cellule neurali attraverso approcci e tecniche che prevedono l’utilizzo della luce. Le tecniche di stimolazione o modulazione mediante luce più importanti sono: l’optogenetica, l’uso di molecole fotosensibili e l’uso della radiazione infrarossa. L’optogenetica consiste nell’indurre modifiche genetiche in specifici neuroni tramite microiniezioni di vettori virali contenenti transgeni, i quali, una volta integrati nel genoma dei neuroni, andranno ad esprimere proteine fotosensibili. La tecnica che utilizza molecole fotoattive consiste principalmente nell’iniettare neurotrasmettitori legati a molecole che li rendono attivi solo sotto la stimolazione della luce. La radiazione infrarossa, che attraverso protocolli di stimolazione a differenti frequenze è in grado di indurre depolarizzazione o iperpolarizzazione neuronale, senza la necessità di modificare geneticamente o chimicamente le cellule, ma sfruttando il surriscaldamento transiente che avviene a livello della membrana cellulare Concludendo gli approcci descritti mostrano come la luce possa essere un metodo fisico innovativo non soltanto per l’imaging del cervello ma anche per la sua modulazione con elevata risoluzione spaziale e temporale, efficacia e versatilità e sicurezza. Tuttavia, l’assorbimento della luce puo’ causare effetti di fotodanneggiamento o modulazione incontrollata delle cellule. D’altra parte gli strumenti tecnologici che possano consentire di rilasciare la luce al cervello in modo controllato e guidato sono ancora limitati. In questo senso approcci ibridi di fotonica ed elettronica sono in fase di studio

Kanalrhodopsine (ChRs) sind lichtgesteuerte Ionenkanäle, welche nach Absorption eines Photons durch den Retinal-Cofaktor einen passiven Ionentransport über die Zellmembran katalysieren. Im Zuge von optogenetischen Anwendungen wird diese Reaktion für die Beeinflussung der Ionenhomöostase von verschiedenen Zelltypen und Geweben ausgenutzt. Zu Beginn dieser Arbeit wurden lichtinduzierte Strukturänderungen und Protontransferschritte in einem breiten Zeitbereich (Nanosekunden bis Minuten) in dem grün-absorbierenden ChR ReaChR mithilfe von stationärer und transienter UV-vis- und Fourier-Transform-Infrarot-Spektroskopie (FTIR) untersucht. Auf Basis der experimentellen Daten wurde ein komplexes Photozyklus-Modell konzipiert. Anschließend wurde die IR-aktive, nichtkanonische Aminosäure p-Azido-L-phenylalanin (azF) mittels Stopp-Codon-Suppression ortsspezifisch an mehreren Positionen innerhalb der vermuteten ionenleitenden Kanalpore in ReaChR inkorporiert und mit FTIR untersucht. azF ist sensitiv gegenüber Polaritätsänderungen und absorbiert in einem hochfrequenten Bereich (~2100 cm-1). Aufgrund der großen spektralen Separation zu endogenen Proteinschwingungen (< 1800 cm-1) können globale Konformations- und lokale Hydratisierungsänderungen simultan detektiert werden. Die erhobenen Daten leisten einen wichtigen Beitrag zum Verständnis der Bildung einer temporären Wasserpore in ChRs und demonstrieren zum ersten Mal den erfolgreichen in-vivo-Einbau einer artifiziellen Aminosäure in mikrobielle Rhodopsine und dessen schwingungsspektroskopische Analyse. Die Methode bietet aufgrund ihrer hohen Ortsauflösung ein großes Potential für die Studie von Mikroumgebungen innerhalb komplexer Proteinensemble.
Channelrhodopsins (ChRs) are light-gated ion channels. Upon absorption of a photon, the retinal chromophore isomerizes and drives conformational changes within the protein, which lead to a passive ion transport across the cell membrane. This capability is used for optogenetic applications to manipulate ionic homeostasis of different cell types and entire organisms. Within the work, light-induced structural changes and proton transfer steps were studied in the green-absorbing ChR ReaChR in great detail by steady-state and transient UV-vis and Fourier transform infrared spectroscopy (FTIR). The data were merged into a complex photocycle model. Next, the IR-active, unnatural amino acid p-azido-L-phenylalanine (azF) was site-specifically introduced at several sites of the putative ion pore of ReaChR by stop codon suppression. azF is sensitive to polarity changes and absorbs in a clear spectral window lacking endogenous protein vibrations. Thus, FTIR measurements of labeled mutants report for global conformational changes (< 1800 cm-1) and local hydration changes (~2100 cm-1) simultaneously. The presented findings reveal crucial insights regarding formation of a transient water pore in ChRs and demonstrate the first report of the successful in-vivo incorporation of an artificial amino acid into a microbial rhodopsin and its subsequent spectroscopic investigation. Additionally, the so far unprecedented spatial resolution renders this methodology superior over conventional FTIR methods to study microenvironments within complex protein ensembles.

Die Modifikation von Genen ist in den molekularen Biowissenschaften ein fundamentales Werkzeug, um die Funktion von Genen zu studieren (Reverse Genetik). Diese Arbeit hat erfolgreich Zinkfinger- und CRISPR/Cas9-Nukleasen für die Verwendung in C. reinhardtii etabliert, um Gene im Kerngenom gezielt auszuschalten und präzise zu verändern. Basierend auf vorausgegangener Arbeit mit Zinkfingernukleasen (ZFN) konnte die Transformationseffizienz um das 300-fache verbessert werden, was die Inaktivierung von Genen auch in motilen Wildtyp-Zellen ermöglichte. Damit war es möglich, die Gene für das Kanalrhodopsin-1 (ChR1), Kanalrhodopsin-2 (ChR2) und das Chlamyopsin-1/2-Gen (COP1/2) einzeln und gemeinsam auszuschalten. Eine Analyse der Phototaxis in diesen Stämmen ergab, dass die Phototaxis durch Inaktivierung von ChR1 stärker beeinträchtigt ist als durch Inaktivierung von ChR2. Um das CRISPR/Cas9-System zu verwenden, wurden die Transformationsbedingungen so angepasst und optimiert, dass der Cas9-gRNA-Komplex als in vitro hergestelltes Ribonukleoprotein in die Zellen transformiert wurde. Um die Bedingungen für präzise Genmodifikationen zu messen und zu verbessern, wurde das SNRK2.2-Gen als Reportergen für eine „Blau-Grün Test“ etabliert. Kleine Insertionen von bis zu 30 bp konnten mit kurzen Oligonukleotiden eingefügt werden, während größere Reportergene (mVenus, SNAP-Tag) mithilfe eines Donor-Plasmids generiert wurden. In dieser Arbeit konnten mehr als 20 nicht-selektierbare Gene – darunter 10 der 15 potenziellen Photorezeptorgene – mit einer durchschnittlichen Mutationsrate von 12,1 % inaktiviert werden. Insgesamt zeigt diese Arbeit in umfassender Weise, wie Gen-Inaktivierungen und Modifikationen mithilfe von ZFNs und des CRISPR/Cas9-Systems in der Grünalge C. reinhardtii durchgeführt werden können. Außerdem bietet die Sammlung der zehn Photorezeptor-Knockouts eine aussichtsreiche Grundlage, um die Vielfalt der Photorezeptoren in C. reinhardtii zu erforschen.
Gene editing is a fundamental tool in molecular biosciences in order to study the function of genes (reverse genetics). This study established zinc-finger and CRISPR/Cas9 nucleases for gene editing to target and inactivate the photoreceptor genes in C. reinhardtii. In continuation of previous work with designer zinc-finger nucleases (ZFN), the transformation efficiency could be improved 300-fold, which enabled the inactivation of genes in motile wild type cells. This made it possible to disrupt the Channelrhodopsin-1 (ChR1), Channelrhodopsin-2 (ChR2) and Chlamyopsin-1/2 (COP1/2) genes individually and in parallel. Phototaxis experiments in these strains revealed that the inactivation of ChR1 had a greater effect on phototaxis than the inactivation of ChR2. To apply the CRISPR/Cas9 system, the transformation conditions were adapted and optimized so that the Cas9-gRNA complex was successfully electroporated into the cells as an in vitro synthesized ribonucleoprotein. This approach enabled gene inactivations with CRISPR/Cas9 in C. reinhardtii. In order to measure and improve the conditions for precise gene modifications, the SNRK2.2 gene was established as a reporter gene for a ‘Blue-Green test’. Small insertions of up to 30 bp were inserted using short oligonucleotides, while larger reporter genes (mVenus, SNAP-tag) were integrated using donor plasmids. Throughout this study, more than 20 non-selectable genes were disrupted, including 10 of the photoreceptor genes, with an average mutation rate of 12,1 %. Overall, this work shows in a comprehensive way how gene inactivations and modifications can be performed in green alga C. reinhardtii using ZFNs or CRISPR/Cas9. In addition, the collection of the ten photoreceptor knockouts provides a promising source to investigate the diversity of photoreceptor genes in C. reinhardtii.

Kanalrhodopsine (ChRs) sind lichtaktivierte Ionenkanäle motiler Algen. Heterolog exprimiert erlauben sie es, Ionenflüsse durch Licht zu steuern. Bevorzugt geleitet werden von den meisten ChRs Protonen. Ausprägung und Wirkung lichtaktivierter Protonenflüsse sowie der molekulare Mechanismus protonenselektiver ChRs werden in vorliegender Arbeit untersucht und zur Entwicklung neuer optogenetischer Werkzeuge genutzt. Eine besonders hohe Protonenselektivität zeigten die grün- und rotlicht-aktivierten Kanäle CsChR und Chrimson aus den Algen Chloromonas subdivisa und Chlamydomonas noctigama. Im spektroskopisch detailliert untersuchten CrChR2 aus Chlamydomonas reinhardtii änderte sich die Protonenselektivität nach Anregung mit einem ns-Laserblitz sogar innerhalb eines Aktivierungszyklus und war insbesondere nach Öffnung des Kanals sowie in Folge der Lichtadaptation hoch. Als unentbehrlich für eine effiziente Protonenleitung erwiesen sich in allen drei Kanälen konservierte, titrierbare Reste entlang der Pore, deren individuelle Bedeutung für die Protonenleitung sich je nach Protein wesentlich unterschied. Entsprechend genügte in Chrimson der Austausch einzelner Glutaminsäuren des extrazellulären Halbkanals, dieses in einen grün- oder rotlichtaktivierten Natriumkanal zu transformieren. Aminosäuresubstitutionen der unmittelbaren Retinalumgebung verschoben hingegen das Aktionsmaximum von Chrimson röter als 600 nm und damit röter als in allen bisher beschriebenen ChRs. In Chrimson versperrt hierbei ein zusätzliches äußeres Tor den extrazellulär Halbkanal, während die Retinalbindetasche in Struktur und funktionaler Bedeutung der einzelnen Reste wesentlich jener der Protonenpumpe Bacteriorhodopsin ähnelt. Als Zwei-Komponenten-Optogenetik wurden schließlich protonen-, kationen- und anionenleitende ChRs unterschiedlicher Farbsensitivität fusioniert sowie lichtgetriebene Protonenpumpen mit protonenaktivierten Ionenkanälen kombiniert und neue optogenetische Perspektiven eröffnet.
Channelrhodopsins (ChRs) are light-gated ion channels from green algae. Expressed in host cells they are used to control ion fluxes by light and are widely applied in Neurosciences. Although generally classified as either cation or anion channels, most ChRs preferentially conduct protons. This thesis compares proton conductance of different ChRs, examines the molecular mechanism of proton selective ChRs and explores the usage of light regulated proton fluxes in two-component-optogenetics. Proton selectivity varied strongly among different ChRs and was most pronounced for the green- and red-light activated channels CsChR and Chrimson from the algae Chloromonas subdivisa and Chlamydomonas noctigama, that conducted predominantly protons even at high pH. In CrChR2 from Chlamydomonas reinhardtii proton selectivity also changed during a single activation cycle and was especially high directly after channel opening and later on following light adaptation. In all three channels efficient proton conductance depended on conserved titratable residues along the pore with different contribution of the individual side chains in each protein. The substitution of single glutamic acids in the extracellular half pore converted Chrimson into a green or red-light activated sodium channel. A single point mutation close to the retinal chromophore shifted peak absorption of Chrimson beyond 600 nm - further red than all other cation conducting ChRs. Whereas the retinal binding pocket of Chrimson resembles the proton pump Bacteriorhodpsin, the overall pore structure corresponds to other ChRs, but features an additional outer gate, that occludes the extracellular half pore and is important for both, proton selectivity and red light absorption. Finally different Two-Component-Optogenetic approaches combined proton and anion selective ChRs of distinct colour as well as light-driven proton pumps and proton-activated ion channels with major prospect for future optogenetic applications.

Mikrobielle Rhodopsine sind lichtsensitive Membranproteine und agieren als Sensoren, Biokatalysatoren oder Ionentransporter. Die Ionentransporter unterteilen sich in lichtgetriebene Ionenpumpen und in lichtaktivierte Kanalrhodopsine. Besonders die Protonenpumpe Bakteriorhodopsin steht schon lange im Fokus biophysikalischer Untersuchungen. Obwohl die Protonenpumpen seit über 40 Jahren intensiv untersucht werden, ist das Wissen über deren elektrophysiologische Eigenschaften noch immer gering. Aus diesem Grund widmete sich diese Arbeit der elektrophysiologischen Charakterisierung der mikrobiellen Rhodopsine mit dem Fokus auf Protonenpumpen. Hierfür wurden vor allem „Two-Electrode Voltage Clamp“ -Messungen (TEVC) an Oozyten des afrikanischen Krallenfrosches Xenopus leavis durchgeführt. Die Untersuchung verschiedener Protonenpumpen hat gezeigt, dass diese eine unerwartet große Diversität in ihren elektrophysiologischen Eigenschaften aufweisen. Von besonderem Interesse war die Beobachtung, dass einige Protonenpumpen neben Pumpströmen auch passive einwärts gerichtete Photoströme zeigten. Besonders deutlich war der „Pump-Kanal-Dualismus“ bei dem Gloeobacter-Rhodopsin ausgeprägt. Andere Protonenpumpen, wie das Bakteriorhodopsin oder Coccomyxa-Rhodopsin, zeigten keine einwärts gerichteten Photoströme. Das Coccomyxa-Rhodopsin wurde aufgrund seiner hohen Photostrom-Amplituden in Oozyten für eine Mutationsanalyse ausgewählt. Diese Mutationsanalyse verhalf die strukturellen Ursachen für die funktionalen Unterschiede zu identifizieren, welche sowohl zwischen den Protonenpumpen untereinander als auch gegenüber Kanalrhodopsinen beobachtet wurden. Mutationen im Gegenion-Komplex führen zu rein passiven oder inaktiven Transportern. Dagegen übernimmt der extrazelluläre Halbkanal in Protonenpumpe die Aufgabe einen passiven Protonen-Rückfluss während des Pumpzyklus zu verhindern, denn Mutationen in dieser Region verursachen passive Photoströme zusätzlich zum aktiven Pumpstrom.
Microbial rhodopsins are light-sensitive membrane proteins and operate as sensors, enzymes or ion-transporters. The ion transporters are subdivided into light-driven ion pumps and light-gated channels. Biophysical research has put focus on the proton pump bacteriorhodopsin for long time. Despite the fact that light-driven proton pumps are investigated for over 40 years, the knowledge about their electrophysiological properties is surprisingly low. For this reason, this thesis is devoted to the electrophysiological characterization of microbial rhodopsins with special focus on light-driven proton pumps. For this purpose, “Two-Electrode Voltage Clamp”-recordings (TEVC) were primarily performed using oocytes from African clawed frog Xenopus leavis. The investigation of diverse proton pumps has shown that the differences in their electrophysiological behaviors are unexpectedly high. Special interest was laid on proton pumps which show passive inward directed photocurrents when the electrochemical load exceeds a certain level. The dualism of pump and channel activity was particularly pronounced in the proton pump Gloeobacter-rhodopsin. Other proton pumps, for instance bacteriorhodopsin or Coccomyxa-rhodopsin, do not show inward directed photocurrents. Due to high photocurrent amplitudes, the Coccomyxa-rhodopsin was selected for an efficient mutagenesis study. This study allowed the identification of structural key determinants for the differences among proton pumps themselves and for the differences of proton pumps in comparison with light-gated ion channels (channelrhodopsins). Therefore, mutations of the counter-ion-complex cause inactive or purely passive transporters. The extracellular half-channel is the key element in proton pumps which prevents passive proton-backflow during the pump-cycle. Mutations in this region lead to passive leak-currents in overlap with the remaining pump-activity.

Channelrhodopsin 2 (ChR2) aus dem Augenfleck von C. rheinhardtii gehört zur Gruppe der mikrobiellen Rhodopsine (Typ1-Rhodopsine). ChR2 besteht aus einem extrazellulär gelegenen N-Terminus, 7 Transmembranhelices und einem zytosolisch gelegenen C-Terminus. Der lichtreaktive Bestandteil (Chromophor) all-trans-Retinal ist via Schiff´ Base kovalent an ein Lysinrest der siebten Transmembranhelix gebunden. Bei Applikation von Blaulicht isomerisiert all-trans- zu 13-cis-Retinal, was in einer Konformationsänderung und dem Öffnen des Kanals resultiert. Abhängig vom elektrochemischen Gradienten können ein- und zweiwertige Kationen in die Zelle ein- oder aus der Zelle herausströmen. Eine retinalabhängige Stabilität konnte bereits für Bakteriorhodopsin (BR) bestätigt werden (Booth, Farooq et al. 1996, Turner, Chittiboyina et al. 2009, Curnow and Booth 2010), bezüglich ChR2 waren bisher nur wenige Daten verfügbar (Hegemann, Gartner et al. 1991, Lawson, Zacks et al. 1991). Die heterologe Expression von wildtypischem und modifiziertem ChR2 in Oozyten von X. laevis erlaubte einen detaillierteren Einblick in die retinalabhängige Stabilität und pH-abhängige Dunkelleitfähigkeit von Guanidinium. Wildtypisches Chop2 zeigte bei Zugabe von Retinal zum Inkubationsmedium, direkt nach RNA-Injektion, Stromamplituden im µA-Bereich und deutliche Fluoreszenzintensitäten. Ausschließlich endogen vorhandenes Retinal hatte verminderten Fluoreszenzen und Stromamplituden zur Folge, was auf ein geringes Vorhandensein von Chop2-Proteinen in der Plasmamembran hindeutete. Da die Inkubation über Nacht in retinalsupplementierter Lösung nur eine minimale Erhöhung des resultierenden Stromes erbrachte, deuten die in dieser Arbeit erhaltenen Ergebnisse stark auf eine verminderte Stabilität des Proteins bei fehlender Bindung des Kofaktors Retinal. Das Einfügen einer aromatischen Aminosäure (Y/F/W) an Position 159 führte zu einer, von der Retinalsupplementation unabhängigen, in beiden Ansätzen gleichwertigen Expressionsstärke. Diese äusserte sich in äquivalenten Fluoreszenzintensitäten. Die erhaltenen Stromamplituden wiesen eine starke Differenz auf: ohne Zugabe zusätzlichen Chromophors lag die Stromstärke bei nur wenigen Nanoampere, die bei Inkubation in einer retinalhaltigen Lösung über Nacht auf das Niveau von retinalsupplementierten Oozyten anstieg. Des Weiteren konnte die Zunahme der Stromamplitude innerhalb von 15 Minuten beobachtet werden, wenn die vermessenen Oozyten mit einer retinalhaltigen Lösung perfundiert wurden. Zusammengefasst weisen die Ergebnisse auf eine Stabilisierung des aromatisch substituierten Proteins hin. Bei der von Berndt et al. (2011) beschriebenen Mutante T159C konnten diese Eigenschaften nicht nachgewiesen werden. Die Modifikation der Retinalbindestelle (K257) in Verbindung mit einer aromatischen Substitution an Position 159 resultierte in deutlichen Fluoreszenzintensitäten, unabhängig von der Retinalverfügbarkeit bei, in beiden Fällen, fehlenden lichtaktivierten Strömen. Diese und die gleichwertigen Bandenstärken des Proteinimmunoblots von aromatisch substituierten ChR2-Varianten unterstützen die Hypothese der retinalunabhängigen Stabilität zusätzlich. Die Ergebnisse legen, im Falle von Chop2-WT, eine Degradation des Apoproteins nahe. Bei Einfügen einer aromatischen AS an Position 159 ist das Apoprotein davor geschützt (siehe Abb. 75). Infolge der strukturellen Similarität, dem Vorhandensein delokalisierter π-Elektronen und der räumlichen Größe der aromatischen AS ist eine strukturelle Veränderung des Apoproteins denkbar, die eine Degradation aufgrund von nunmehr unzugänglichen Ubiquitinierungsstellen verhindert. Des Weiteren besteht die Möglichkeit, dass sich bei fehlender Bindung des Kofaktors Wassermoleküle in der Nähe der Bindetasche befinden, welche von umliegenden Aminosäuren (u.a. T159, D156) unter großem Energieaufwand koordiniert werden und die strukturelle Integrität bis hin zur Degradation beeinträchtigen können. Dies könnte durch eine Erhöhung der Hydrophobizität bei Einfügen einer aromatischen Aminosäure verhindert werden. Bei Substitutionen durch eine aromatische AS (Y/W/F) an Position 159 zeigte sich ein weiteres, bisher nicht beschriebenes, Charakteristikum. Bei Perfusion der Oozyten mit einer guanidiniumhaltigen Lösung, konnten in Abhängigkeit des pH-Wertes ohne die Applikation von Licht Stöme im µA Bereich aufgezeichnet werden. Die Größe der Stromamplitude korreliert hierbei mit dem Anstieg des pH-Wertes und der Konzentration an Guanidiniumionen der perfundierten Lösung und kann durch das Hinzufügen von 1mM Lanthan reversibel geblockt werden. Des Weiteren konnten die vorgenommenen Messungen die Ergebnisse der retinalabhängigen Degradation verifizieren, da der Einstrom von Gua+ sowohl bei retinalsupplementierter Inkubation, als auch bei ausschließlich endogen vorhandenem Retinal zu beobachten war. Des Weiteren zeigte auch die Doppelmutante T159Y/K257R trotz ihres Unvermögens Retinal zu binden, die beschriebenen lichtunabhängigen Ströme. Die Ergebnisse bei Substitution durch Phenylalanin (F) stellen eine Abweichung des Musters dar. Bei Inkubation von T159F-injizierten Zellen bei ausschließlich endogen vorhandenem Retinal konnte eine stark erhöhte Guanidiniumleitfähigkeit festgestellt werden, diese kam jedoch bei retinalsupplementierter Inkubation nicht zum Tragen. Dies könnte ein Hinweis auf eine sterische Hinderung durch das gebundene Chromophor sein, die bei den Substitutionen durch Tyrosin und Tryptophan, möglicherweise durch unterschiedliche chemische Eigenschaften der AS, nicht auftreten. Die hervorgerufene pH-Abhängigkeit kann in zwei möglichen Ursachen begründet liegen: • Vorhandensein einer (de)protonierbaren Gruppe wie Histidin, Arginin oder Lysin, die als pH-Sensor dienen könnte • Deprotonierung der Schiff´ Base durch Guandininium Das Vorhandensein eines pH-Sensors konnte durch die vorgenommenen Modifikationen von H114, R115, R120 und H249 nicht bestätigt werden. Bei Substitution von K257 (in Verbindung mit T159Y) zu Arginin (R) konnte weiterhin ein pH-abhängiger Gua+-Dunkelstrom festgestellt werden. Die Modifikation zu Alanin (A) oder Glutamin (Q) hingegen resultierte im Ausbleiben der Ströme. Der Austausch einer basischen zu einer neutralen Gruppe ohne protonierbaren Rest deutet auf die Beteiligung der Schiff´ Base bzw. der Aminosäure an Position 257 am Mechanismus der Dunkelleitfähigkeit hin
Channelrhodopsin 2 (ChR2) from the eyespot of C. rheinhardtii belongs to the group of microbial-type rhodopsins (Type1-rhodopsins). It consists of an extracellular N-terminus, seven transmembranehelices and a cytosolic C-terminus. The light-reactive element (Chromophor) all-trans-retinal is covalently bound to a lysine of the seventh transmembrane helix via Schiff´ Base. The isomerisation from all-trans- to 13-cis-Retinal after illumination leads to a conformational change within the protein, resulting in the opening of the channel and thereby in an in- or efflux of mono- and divalent cations, depending on the electrochemical gradient. For bacteriorhodopsin (BR) a retinal dependent stability could be confirmed (Booth, Farooq et al. 1996, Turner, Chittiboyina et al. 2009, Curnow and Booth 2010), concerning ChR2 only limited data is available (Hegemann, Gartner et al. 1991, Lawson, Zacks et al. 1991). Heterologous expression of wildtype (WT) and mutant ChR2 in the oocytes of X. laevis revealed a more detailed insight into retinal dependent stability and pH-dependent conductance of guanidinium without the application of light.

Funktionelle Expression von ChR2 in Pflanzen In der vorliegenden Arbeit konnte erstmalig die funktionelle Expression des licht-aktivierten Channelrhodopsin-2 aus Chlamydomonas reinhardtii in höheren Pflanzen gezeigt werden. Obwohl die erfolgreiche Transformation auf der Basis der Integration einer Expressionskassette für WT-ChR2 in Pflanzen genetisch nachgewiesen werden konnte, war ein funktioneller Nachweis nicht möglich. Demgegenüber war die funktio-nelle Expression aller getesteten ChR2-Mutanten im transienten Expressionsansatz er-folgreich und konnte schließlich auf der Basis der im Rahmen dieser Arbeit generierten Konstrukte auch für stabil transformierte Arabidopsis-Pflanzen bestätigt werden. ChR2 wurde in Arabidopsis-Protoplasten sowie Tabak-Epidermis- und Mesophyllzellen an der Plasmamembran lokalisiert, zeigte jedoch aufgrund der Überexpression eine starke Überladung des Endomembransystems. Elektrophysiologische Messungen mit Hilfe der Einstichtechnik belegten, dass ChR2 sowohl in Arabidopsis-Keimlingen als auch im Tabakmesophyll funktionell ist, wobei sich die erzeugten Blaulicht-vermittelten Depolarisationen weitaus erfolgreicher im Ta-baksystem darstellten. Alle eingesetzten ChR2-Mutanten waren funktionell und zeigten in Einstichmessungen mit Oozytendaten korrelierende Kinetiken. Die Mutante C128A wurde hinsichtlich der erzielten lichtinduzierten Membranpotentialdepolarisationen als effektivste ChR2-Variante identifiziert. Calcium-Messungen mit dem Reporterprotein Aequorin lieferten keinen Beweis für einen direkt durch ChR2-C128A vermittelten Calcium-Einstrom in Arabidopsis-Protoplasten. Jedoch konnte ein cytosolischer Calcium-Anstieg ca. 3min nach Blau-lichtapplikation beobachtet werden. Dies deutet darauf hin, dass die durch ChR2 vermittelten Membranpotentialänderungen zu einer Aktivierung endogener, Calcium-permeabler Ionenkanäle führen könnte. Für die ChR2-L132C Mutante konnte allerdings in ersten Messungen ein direkter Calcium-Anstieg nach Lichtgabe beobachtet werden. Transkriptionelle Änderungen aufgrund ChR2-basierter, elektrischer Signalmuster In RNA-Seq-Analysen mit transient transformierten Tabakblättern konnte die Bedeu-tung der Signalsignatur elektrischer bzw. Calcium-basierter Signale verifiziert werden: Die Applikation zweier in ihrer Form gänzlich unterschiedlicher elektrischer Signal-muster lieferte ein signifikant unterschiedlich reguliertes Set an Genen, wobei einige wenige durch beide Behandlungen induziert werden konnten. Langanhaltende Depolari-sationen regulierten deutlich mehr Gene und waren daher in ihrer Wirkung weitaus ef-fektiver als kurze, repetitive Depolarisationen. Die bioinformatische Analyse dieser Daten zeigte, dass die Nachahmung eines im Zuge der Pathogenantwort bekannten, langen Depolarisationspulses Gene der Flagellin-induzierten Signaltransduktion adressierte, während kurze, wiederkehrende Pulse mit gleichem Informationsgehalt diese nicht regulierten
Functional expression of ChR2 in plants The current work for the first time proves the functional expression of the light gated ChR2 derived from the unicellular green algae Chlamydomonas reinhardtii in higher plants. In line with its function ChR2 was localized primarily at the plasma membrane in Ara-bidopsis protoplasts and Tobacco epidermal and mesophyll cells. However, overexpres-sion as well as codon-usage often resulted in overloading of the endomembrane system such as Golgi and ER. Electrophysiological recordings by means of the impalement technique proved ChR2 function in Arabidopsis seedlings as well as Tobacco mesophyll cells. Blue light induced depolarization, however, was more efficient in the Tobacco system. In contrast to the WT protein all ChR2-mutants tested in this study were shown to be functional. Channel kinetics gained from Xenopus oocyte TEVC measurements corre-lated well with observed depolarization and repolarization kinetics of individual mu-tants. Highest depolarization levels were generated by the mutant C128A. Using the Calcium reporter Aequorin, measurements provided no evidence for ChR2-C128A mediated Calcium-influx in Arabidopsis protoplasts. However, a delayed Cal-cium increase approximately 3min following blue light application suggests that ChR2 mediated membrane potential changes activated endogenous Calcium permeable ion channels. Preliminary experiments using the L132C mutant showed cytosolic Calcium elevation directly after light stimulation. All ChR2 mutants tested in this study are suited for generating defined depolarization patterns in plants, whereas simulation of specific Calcium signatures seems possible with ChR2 variants based on the L123C mutation. Transcriptional changes based on ChR2-mediated, electrical patterns Transcriptional analyses of transiently transformed Tobacco leaves verified the im-portance of signature shape of electrical or Calcium-based signals in plants: Application of two blue light triggered depolarization patterns differing in shape revealed two sets of significantly differentially regulated genes, although a small number of genes was regulated by both stimuli. Sustained depolarizations regulated more genes and thus turned out to be more effective than short, repetitive depolarizations. Bioinformatic analyses of the data generated by RNA-Seq revealed the power of elec-trical signal application. Mimicking known electrical pathogen responses by applying a sustained depolarization pattern indeed addressed genes involved in flagellin signaling. In contrast, repetitive depolarization pulses regulated a completely different set of genes

Since Channelrhodopsins has been described first and introduced successfully in freely moving animals (Nagel et al., 2003 and 2005), tremendous impact has been made in this interesting field of neuroscience. Subsequently, many different optogenetic tools have been described and used to address long-lasting scientific issues. Furthermore, beside the ‘classical’ Channelrhodopsin-2 (ChR2), basically a cation-selective ion channel, also altered ChR2 descendants, anion selective channels and light-sensitive metabotropic proteins have expanded the optogenetic toolbox. However, in spite of this variety of different tools most researches still pick Channelrhodopsin-2 for their optogenetic approaches due to its well-known kinetics. In this thesis, an improved Channelrhodopsin, Channelrhodopsin2-XXM (ChR2XXM), is described, which might become an useful tool to provide ambitious neuroscientific approaches by dint of its characteristics. Here, ChR2XXM was chosen to investigate the functional consequences of Drosophila larvae lacking latrophilin in their chordotonal organs. Finally, the functionality of GtACR, was checked at the Drosophila NMJ. For a in-depth characterisation, electrophysiology along with behavioural setups was employed. In detail, ChR2XXM was found to have a better cellular expression pattern, high spatiotemporal precision, substantial increased light sensitivity and improved affinity to its chromophore retinal, as compared to ChR2. Employing ChR2XXM, effects of latrophilin (dCIRL) on signal transmission in the chordotonal organ could be clarified with a minimum of side effects, e.g. possible heat response of the chordotonal organ, due to high light sensitivity. Moreover, optogenetic activation of the chordotonal organ, in vivo, led to behavioural changes. Additionally, GtACR1 was found to be effective to inhibit motoneuronal excitation but is accompanied by unexpected side effects. These results demonstrate that further improvement and research of optogenetic tools is highly valuable and required to enable researchers to choose the best fitting optogenetic tool to address their scientific questions
Seit dem Channelrhodopsine das erste Mal beschrieben und erfolgreich in lebende Tiere eingebracht wurden (Nagel et al., 2003 und 2005), kam es zu einem beträchtlichen Fortschritt in diesem interessanten Gebiet der Neurowissenschaften. In der nachfolgenden Zeit wurden viele verschiedene optogenetische Werkzeuge beschrieben und zur Bearbeitung neurowissenschaftlicher Fragestellungen angewandt. Des Weiteren haben neben dem „klassischen“ Channelrhodopsin-2 (ChR2), ein im Wesentlichen Kation selektiver Kanal, auch modifizierte ChR2 Abkömmlinge, Anion selektive Kanäle und Licht sensitive metabotrope Proteine, die opotogenetische Werkzeugkiste erweitert. Dennoch greifen die meisten Wissenschaftler trotz der Vielfalt an optogenetischen Werkzeugen meist noch zu Channelrhodopsin-2, da seine Wirkungseigenschaften sehr gut erforscht sind. In der nachfolgenden Arbeit wird ein weiterentwickeltes Channelrhodopsin, Channelrhodopsin2-XXM (ChR2XXM), beschrieben. Aufgrund seiner vielfältigen Eigenschaften stellt es ein vielversprechendes Werkzeug dar, vor allem für zukünftige neurowissenschaftliche Forschungsarbeiten. Hierbei wurde ChR2XXM eingesetzt, um zu untersuchen welche Auswirkungen das Fehlen von Latrophilin im Chordotonal Organ von Drosophilalarven hat. Schließlich wurde noch die Funktionalität von GtACR an der neuromuskulären Endplatte der Drosophila überprüft. Für die umfassende Charakterisierung wurden elektrophysiologische und verhaltensbasierte Experimente an Larven durchgeführt. Es konnte gezeigt werden, dass ChR2XXM aufgrund einer erhöhten Affinität zu dem Chromophore Retinal, im Vergleich zu ChR2 ein besseres zelluläres Expressionsmuster, eine bessere zeitliche Auflösung und eine erheblich höhere Lichtsensitiviät aufweist. Durch den Einsatz von ChR2XXM konnte, aufgrund der hohen Lichtsensitiviät, mit nur minimalen Nebeneffekten, wie z.B. mögliche Wärmeaktivierung des Chordotonalorgans, der Einfluss von Latrophilin (dCIRL) auf die Signaltransmission im Chordotonalorgan, aufgeklärt werden. Ferner führte eine optogenetische, in vivo, Aktivierung des Chordotonalorgans zu Verhaltensänderungen. Zusätzlich konnte gezeigt werden, dass GtACR1 zwar effektiv motoneuronale Erregung inhibieren kann, dies aber von unerwarteten Nebeneffekten begleitet wird. Diese Ergebnisse zeigen auf, dass weitere Forschung und Verbesserungen im Bereich der optogenetischen Werkzeuge sehr wertvoll und notwendig ist, um Wissenschaftlern zu erlauben das am besten geeignetste optogenetische Werkzeug für ihre wissenschaftlichen Fragestellungen auswählen zu können

Dissertação de mestrado em Biotecnologia Farmacêutica, apresentada à Faculdade de Farmácia da Universidade de Coimbra
Nos últimos anos têm sido desenvolvidas várias ferramentas para permitir o controlo de neurónios específicos, possibilitando o estudo da sua função. Estas novas ferramentas superam a falta de selectividade e o fraco controlo temporal proveniente do uso de estimulação eléctrica no controlo de actividade neuronal. A optogenética refere-se á integração de óptica e genética para obter um ganho ou perda de função em eventos bem definidos dentro de células específicas em tecido vivo. A capacidade de “ligar” e “desligar” neurónios utilizando luz é de facto uma tecnologia inovadora que oferece uma solução para limitações passadas. A optogenética, considerada por vários especialistas como ‘’método do ano’’ e ‘’inovação da década’’, em 2010, é utilizada para hiperpolarizar ou despolarizar neurónios alvo, de uma forma menos invasiva, utilizando luz e usufruindo de uma alta resolução espacial e escala temporal na ordem dos milissegundos. Esta técnica tem permitido o mapeamento e estudo de redes neuronais com uma grande eficácia. A ‘’Channelrhodopsin-2’’ (ChR2) é um canal catiónico sensível à luz, derivado da microalga Chlamydomonas reinhardtii. Na última década, a ChR2 tornou-se o arquétipo central e a principal ferramenta da optogenética. Actualmente, a caixa de ferramentas optogenética está em contínua actualização, com contribuições de estratégias de engenharia protéica, tais como mutagénese dirigida e a construção de quimeras com troca de domínios de diferentes espécies de channelrhodopsin. No entanto, alguns aspectos da forma ‘’wild-type’’ da ChR2 ainda requerem atenção e melhoramento. Estes incluem o seu espectro de acção, cinética, níveis de expressão, inactivação, condutância e exactidão de pico de absorção. Em termos de propriedades espectrais, poucas variantes desta proteína têm sido geradas e completamente caracterizadas com sucesso. No entanto, o aprimorar do espectro de activação da ChR2 e do formato do respectivo pico de absorção são algumas das propriedades mais desejadas. A ChR2 é excitada preferencialmente com comprimentos de onda de luz azul (470nm), o que limita o seu uso em material biológico de alta taxa de difusão, tal como o cérebro. Luz de excitação com maiores comprimentos de onda diminui a difusão de luz produzida por tecidos biológicos, e não é absorvida pela hemoglobina, assim, formas da ChR2 ‘’red-shifted’’, a absorver luz vermelha ou mesmo perto de infravermelha, são ferramentas desejáveis para a excitação de tecidos profundos. Alem disto, variantes ‘’blue-shifted’’ são também ferramentas atrativas para desenvolver, XXI dado que a combinação de várias ChR2 que apresentem sensibilidades a diversos comprimentos de onda permitiriam a estimulação de diferentes populações neuronais sem interferência entre si. Neste projecto, realizámos um desenho ab-initio para produzir quatro novas variantes de ChR2, usando uma abordagem de mutagénese dirigida no ambiente do cromófero da ChR2 alterando de forma radical os resíduos alvo. As mutações foram selecionadas com a aplicação de Time Dependent – Density Functional Theory (TDDFT) para prever o espectro de absorção dos mutantes selecionados da ChR2. O ‘’colour tuning’’ da ChR2 foi alcançado em quatro novas variantes criadas. Em particular, fomos capazes de gerar três variantes ‘’red-shifted’’ e uma ‘’blue-shifted’’. Após caracterização espectral, as variantes F217D e F269D apresentaram um ‘’red-shift’’ significativo de 90nm, a variante L221D apresentou um ‘’red-shift’’ de 180nm, a variante F269H apresentou um ‘’blue-shift’’ de 20nm. Apesar dos nossos resultados, é necessária uma caracterização protéica adicional, tal como a avaliação do tráfego membranar em neurónios e as características electrofisiológicas destes novos mutantes para determinar as proriedades cinéticas do canal. Neste trabalho, também conseguimos definir e descrever com sucesso a expressão e purificação da ChR2 ‘’wild-type’’ e de todas as quatro novas variantes no sistema eucariótico de expressão heteróloga - Pichia pastoris. Por fim, o nosso estudo valida as previsões de Time- Dependent Density Functional Theory e revela que abordagens de simulação biofísica podem ser utilizadas com vista à criação de variantes de ChR2 inteligentemente desenhadas. O desenho de novas variantes ChR2, seguindo a lógica racional aplicada, é uma abordagem poderosa e fiável para obter proteínas optimizadas para estratégias biotecnológicas. Os resultados originais obtidos com este trabalho demonstram potential para aplicações futuras, já que novas e melhoradas variantes de ChR2 continuarão a desempenhar um papel central no desenvolvimento e implementação da optogenética
Over the last few years, several tools have been developed to allow the control over specific types of neuron to enable the study of their function. These novel tools aim to overcome the lack of selectivity and the poor temporal control that derives from trying to control neuronal activity with electrical stimulation. Optogenetics refers to the integration of optics and genetics to obtain gain or loss of function in well-defined events and within specific cells in living tissue. The capacity to turn neurons “on and off” using light is indeed a groundbreaking technology that has become a solution for past limitations. Considered by many, “method of the year” and “breakthrough of the decade”, in 2010, optogenetics is used to hyperpolarize or depolarize specific targeted neurons using light in a less invasive manner, with high spatial resolution and a temporal resolution on the scale of milliseconds. This technique has allowed the mapping and study of neuronal networks with demonstrated efficacy. Channelrhodopsin-2 (ChR2) is a light-gated cation channel, derived from the microalga Chlamydomonas reinhardtii. In the last decade, ChR2 has become the central archetype and the main tool of optogenetics. Presently, the optogenetic toolbox is under continuous update, with contributions from protein engineering strategies, such as site-directed mutagenesis and construction of chimeras with domain swaps between channelrhodopsins of different species. However, some aspects of the wild-type form of ChR2 still require attention and enhancement. These include its action spectra, kinetics, expression levels, inactivation, conductance and absorption peak sharpness. In terms of spectral properties, few variants of this protein have been successfully generated and fully characterized. Nevertheless, tuning of ChR2 activation spectra and absorption peak sharpness are one of the most sought after properties. ChR2 is optimally excitable at a wavelength of blue light (470nm), which limits its use in high light-scattering biologic material, such as the brain. However, long-wavelength excitation light decreases the scattering of light produced by biological tissues and is not absorbed by haemoglobin. Thus, a red-shifted form of ChR2, absorbing red or even near infrared light would be a desirable tool for the excitation of relatively deep tissues. Furthermore, blue-shifted variants would also be attractive tools to develop, since the combination of ChR2 proteins with well separate wavelength sensitivities, combined with multicoloured optics, would permit the stimulation of different neuronal populations with no XXIII interference between them. In this project, we performed ab-initio design to produce four new ChR2 variants, using a radical site-directed mutagenesis approach on target residues in the environment of the ChR2 chromophore. The mutations were selected with the application of Time Dependent – Density Functional Theory (TDDFT) to predict the absorption spectra of ChR2 selected mutants. We achieved successful colour tuning of ChR2 with our four newly created variants. In particular, we were able to generate three red-shifted and one blue–shifted variant. After spectral characterization, the F217D and F269D variants presented a significant 90nm red shift, the L221D variant had a 180nm red shift and the F269H variant presented a 20nm blue shift. Despite our results, additional protein characterization is needed, such as the assessment of membrane trafficking in neurons and an electrophysiological characterization to determine channel kinetic proprieties for each of the variants. In this work, we were also able to define and describe the successful expression and purification of wild type ChR2 and of all the new four variants using the eukaryotic Pichia pastoris heterologous expression system. Finally, our study validates Time-Dependent Density Functional Theory predictions and reveals that biophysical simulation approaches may be used towards the creation of intelligently designed ChR2 variants. The design of new ChR2 variants, following our applied rationale, is a powerful and reliable approach to obtain enhanced proteins for biotechnological strategies. The original output obtained here shows potential for future optogenetic application, as new and improved ChR2 variants will continue to play a central role in the development and implementation of optogenetics

Research Doctorate - Doctor of Philosophy (PhD)
Understanding how the brain is connected is an important first step if we are to successfully treat its conditions. Currently, much of what we know about the brain connectome has been discovered through neuroanatomical tracing studies. This approach however, lacks functionality in that we cannot determine the physiological consequences of manipulating a given pathway or its overall necessity to a specific behavioural outcome. Dysregulated social and emotional functioning, including hypothalamic-pituitary-adrenal (HPA) axis responses, are cardinal features of depressive disorders. Not surprisingly, HPA axis activity is tightly regulated by limbic structures including the medial (MeA) and central amygdala (CeA) which act to increase hypothalamic neuroendocrine output. The lack of direct connectivity between these amygdaloid structures and the neuroendocrine hypothalamus has led to suggestions the MeA and CeA mediate HPA axis responses to stress via indirect relays. While historically, the CeA has received considerable attention in regards to HPA axis regulation, both traditional and modern tracing techniques provide anatomical evidence of direct MeA to hypothalamic neuroendocrine connectivity. At this time however, functional evidence supporting this connection is lacking. Fortunately, recent advances in technology now allows for a functional assessment of individual pathways within the brain connectome. Through the application of optogenetics, the primary goal of this thesis is to investigate the functional characteristics of a direct MeA to neuroendocrine hypothalamus projection. In this regard, I demonstrate the capacity of the MeA to drive stress neuroendocrine responses through direct, functional projections to the hypothalamus. Moreover, that a group of MeA neurons, which may be derived from a neuroendocrine lineage, provide direct glutamatergic input to corticotropin-releasing factor (CRF) neurons, which sit at the apex of the HPA axis. Together, these results suggest therapeutically targeting the MeA, as opposed to the CeA, may prove to be more successful in the treatment of mood related disorders.

Dissertação de Mestrado em Biologia Celular e Molecular, apresentado ao Departamento de Ciências da Vida da Faculdade de Ciências e Tecnologia da Universidade de Coimbra
To better understand the neuronal system, beyond simply “listening” to cells one should be able to “communicate” with them. Until recently, approaches used to establish this communication while relying on the direct manipulation of neurons, have remained technically unsophisticated. Chemical and drug applications provide some degree of cell specificity at the cost of poor spatial and temporal resolution, while direct electrical stimulation of cells, even if allowing for more precise temporal control, lacks in cell type selectivity and may result in tissue damage. In 2005, a new technical approach - Optogenetics - emerged as an answer to these limitations. Based on previously described proteins from the opsin family, such as Channelrhodopsin-2 (ChR2), several groups demonstrated that it was possible to directly manipulate neurons with light. These channels are capable of changing their permeability to certain ionic species upon illumination of specific wavelength and intensity. Using optogenetics, it became possible to activate discrete populations of neurons (-genetic) in a less invasive fashion using light (opto-), under precise temporal control (millisecond time scale) and high spatial resolution (fiber optical light spot). For the past half-decade, great effort has been invested into improving the proprieties of these optogenetic channels and tools. Several enhancements have been achieved in terms of kinetics, for example: Chronos, ChETA and SFO variants; ion permeability: in the eNpHR 3.0 and iC1C2 variants; and in tuning the absorption spectra towards red-shifted variants as in: VChR1 and Crimson. This latter feature has taken up 10 great interest in the optogenetics field due to the scarceness of variants with nonoverlapping absorption spectra. With the recent determination of the crystal structure of ChR it is now possible to apply Time Dependent – Density Functional Theory (TDDFT) to predict the absorption spectra of the Channelrhodopsin using computational algorithms. Therefore, we propose to intelligently manipulate candidate residues to produce color-tuned variants of channelrhodopsin-2, taking advantage of TD-DFT calculations in order to predict the absorption spectra of these variants and, finally, using directed site mutagenesis to generate these mutants and to optimize their expression in a heterologous system. This work describes the successful expression and purification of ChR2 in the Pichia pastoris expression system, and the generation and partial characterization of two novel ChR2 mutants. As a final goal, the determination of ChR2 absorption spectra was set, however, we could not achieve sufficient amount of protein for this analysis, and further optimization is required for the expression and concentration of purified protein samples. Further analysis using electrophysiology is also required to functionally determine the kinetic proprieties of the channel.
Para melhor compreender o sistema nervoso, para além de simplesmente “ouvir” as células, é também necessário “comunicar” com estas. Até recentemente, técnicas usadas para estabelecer esta comunicação eram caracterizadas pela sua falta de sofisticação técnica. Aplicação de drogas e outros químicos conferem alguma especificidade no que toca ao tipo de célula activada, no entanto, apresenta uma baixa resolução temporal e espacial. Por outro lado, electrofisiologia, apesar de permitir obter uma enorme resolução espacial e temporal, peca quando consideramos a especificidade celular obtida e a grande probabilidade de causar dano nos tecidos a serem estudados. Em 2005, a técnica “Optogenética” emergiu como uma solução para os problemas técnicos apresentados anteriormente. Esta técnica baseia-se no uso de uma proteína da família das “opsinas”, “Channelrhodopsin-2” (ChR2), que vários grupos mostraram que é capaz de controlar actividade neuronal quando luz incide na mesma. Num mecanismo que envolve a captação de energia luminosa e convertendo-a em mudanças conformacionais da proteína, esta alterna entre estados permeáveis e não permeáveis, sendo possível controlar população geneticamente definidas (-genética) usando luz (opto-), com uma enorme precisão espacio-temporal. Durante a última década, grande investimento foi desenvolvido no que toca a melhorar as propriedades deste canais “optogenéticos”. Grandes melhoramentos foram feitos relativamente a características cinéticas, por exemplo: Chronos, ChETA e variantes SFO; permeabilidade iónica: eNpHR 3.0 e iC1C2; e modificação do espectro 12 de absorbção para maiores comprimentos de onda: VChR1 e Chrimson. Esta última caracteristica têm gerado grande interesse na área já que, até à data, não existem variantes que não apresentem espectros de absorção totalmente distintos. Com a recente determinação da estrutura da ChR é possível aplicar teorias de mecânica quantica como “Time Dependent – Density Function Theory” (TD-DFT) para prever o espectro de absorpção da proteína usando métodos computacionais. Assim, propomos mutar, intelegetemente, resíduos na proteína de forma a obter variantes que respondam a maiores comprimento de onda, tirar partido de TD-DFT para prever os seus espectros e finalmente validados em laboratório, expressando estas variantes num sistema heterologo. Esta tese descrever a expressão e purificação de ChR2 no sistema heterologo de Pichia pastoris. Como objectivo final, a determinação do espectro de absorção da proteína foi delineado, no entanto devido à baixa quantidade de proteína isolada tal não foi possível e assim futuras optimização ao protocolo serão necessárias, assim como várias outras análises, nomeadamente no que toca as propriedades cinéticas da proteína usando técnicas de electrofisiologia.