The generation of human cerebral organoids promises to profoundly change our understanding of human brain development by enabling its detailed study. However, current protocols have several limitations and give rise to organoids characterized by low reproducibility, small size, lack of vascularization, limited neuronal maturation, and reduced cellular diversity. These limit the use of organoids as a model system to study human brain development. Here, I propose to integrate existing 3D culturing protocols with synthetic biology and tissue engineering approaches in order to push the field beyond these limitations and study basic mechanisms of human brain development. In so doing, I expect to both generate cerebral organoids that better resemble the human brain and study basic mechanisms in an “understanding by building” framework. Specifically, I plan to (i) increase neuronal maturity by integrating the brain organoid with a smart functional vasculature system (ii) increase organoid size and cell diversity by modulating the gene expression dynamics of Hes1 (iii) increase organoid reproducibility by building an artificial notochord producing a sonic hedgehog (SHH) gradient in a time regulated fashion in order to pattern brain structures. By integrating live cell imaging, massive single-cell profiling, and electrophysiology, I will be able to study the overall effects of the three specific goals on organoid development.

Mitochondria are a fundamental component of eukaryotic cells that provide ATP through oxidative phosphorylation and play important roles in the regulation of metabolism and cell death. Accurate homeostasis of mitochondria is important for cell function, and misregulation of mitochondria in humans causes a variety of severe diseases. To achieve homeostasis, cells have to ensure that the amount of mitochondria doubles with each cell doubling period. Because mitochondria maintain their own genome encoding for essential mitochondrial proteins and RNAs, this requires a quantitative coordination of mitochondrial DNA replication, production of mitochondrial proteins, and cell growth. However, how cells couple mitochondrial biogenesis with cell growth remains unknown. Here, I propose to use budding yeast as a model organism to identify the molecular mechanism underlying the coordination of mitochondria biogenesis with cell growth and cell cycle. I will use an interdisciplinary approach, combining the microscopy based protein concentration and cell size measurements established during my postdoc, a LacO-LacI based live-cell imaging approach, and qPCRs to unravel how concentrations of mitochondrial components as well as their production rates are linked to cell cycle and cell growth. I will then use mathematical modelling to establish a quantitative model for mitochondrial homeostasis that I will test by measuring the dynamic response of mitochondria maintenance to transient perturbations. I expect that our work will then catalyze our knowledge of mitochondria maintenance in human cells and will thereby lead the way to a better understanding of mitochondrial diseases.

DNA methylation is essential for normal mammalian development. While seminal work has provided tremendous insight into the dynamic regulation of DNA methylation throughout embryogenesis, comprehensive understanding of how cell-specific methylation programs are established and maintained, and how are they involved in defining cell states in vivo through regulation of target genes, remains a formidable task. Revolutionary technologies now offer unprecedented opportunities for understanding the function of DNA methylation in specifying, memorizing and modulating embryonic programs. These powerful tools motivate further development of novel experimental systems, integrating single cell monitoring with flexible engineering of markers, reporters and perturbations. This will enable to precisely target key rare embryonic cell populations for in-depth analysis. Here, combining cutting-edge methods for single cell mapping of DNA methylation and gene expression, and by developing a novel approach for inferring spatial information from single cell genomic data, we propose to comprehensively chart the post-implantation embryo, at unprecedented resolution. To move to functional studies, we will implement our recently established reporter system that enables monitoring and isolation of cells based on endogenous locus-specific changes in DNA methylation. Specifically, we will study the developmental potential of rare epiblast cells that we identified to exhibit lower-than-expected genome-wide methylation levels. Our combined approach will open new avenues for elucidating the contribution of cell-specific DNA methylation changes to cell-state and function following implantation.

With the increasing plethora of microbial genomes, the importance of complex microbial communities in nature has been recognized. The human microbiome has particularly attracted researchers because of its direct link to health and disease. Despite their great diversity, which is apparent from the available genomic information, little is known about the bioactive small molecules secreted by the human microbiome. Few studies have investigated the presence and ecological roles of these metabolites in the human microbiome at physiological conditions, because of the lack of appropriate methodologies.
This project attempts to address uncharacterized peptide metabolites encoded in the genomes of human-associated anaerobic bacteria, and elucidate how they contribute increasing the fitness of the host microbes. These peptides are likely to have a novel scaffold made by a unique class of oxygen-sensitive enzymes. I intend to perform multidisciplinary research composed of two approaches: A) Production of the peptide metabolites in a heterologous microbe; B) A metagenome-based assay to evaluate the population changes in the cultured microbial community upon peptide treatment. The former approach has the potential to generate natural products independent of the strain; the latter enables us to decipher the biological activities of small molecules in a complex microbial community. Innovation in both concepts will increase our understanding and the availability of microbiome-derived compounds that could improve human health. Furthermore, given the broad distribution of target peptides in anaerobes, this research will also elucidate the breadth of anaerobic chemistry inside humans.

Synaptic diversity is crucial for neuronal function. The heterogeneity of individual synaptic proteomes in a neuron underlies its input integration, compartmentalization of function, and neuronal plasticity during learning. The canonical classification of synapses based on neurotransmitter systems (e.g. the generic excitatory or inhibitory synapses) has become increasingly inadequate to explain the diverse synaptic responses in electrophysiology and plasticity. Without quantifying the variability of molecular combinations in synapses, our basic understanding of synapses is incomplete. Recently single-molecule localization microscopy has created an opportunity to interrogate individual synapses across a whole neuron. DNA Points Accumulation for imaging in Nanoscale Topography (DNA-PAINT) is ideally suited for localizing and quantifying protein copies of interest compared to conventional stochastic optical reconstruction microscopy (STORM) because of its well-defined blinking kinetics and absence of photobleaching issues. This proposal aims to quantify the proteomic heterogeneity of synapses in individual neurons using DNA-PAINT and to interrogate the relationship between plasticity and synaptic diversification using multiple approaches. It will address the following:
1. The differential localization, variability, and stoichiometry of five representative synaptic proteins in the synapses of a neuron
2. The correlation between synapse activity and their proteomes
3. A potential change in synaptic diversity after single-spine plasticity
4. The effects of global homeostatic scaling on synaptic diversity
5. The remodeling of synapse diversity by newly synthesized proteins

Our eye movements constantly generate motion patterns on our retina, yet our perception of objects in space remains stable. For the brain to recognize self-generated visual motions to construct an image that is spatially invariant of one’s own movement, retinotopic visual information has to be integrated with eye position. Primate studies have shown cortical areas containing neurons responding to different combinations of these two reference frames, and perturbations of these areas suggest a role for space perception. However, little is known about the cellular and circuit mechanisms by which reference frame transformations underlying visual stability occur. I propose to fill that void by bringing this field into mouse research where methods to monitor, perturb and dissect neurons and circuits in much more precise and targeted ways are possible. Using preparations in which visual inputs, eye positions and eye movements are experimentally tracked and controlled will allow me to isolate visual, proprioceptive and motor components. Brain-wide functional ultrasound imaging will reveal areas activated by each component alone or in combination. Calcium imaging will then provide large population responses with single cell resolution. Finally, single-cell-initiated rabies tracing and whole-cell recording will be used to delve into the circuit and synaptic mechanisms underlying the functional properties we find among different neuronal populations. This project will accelerate the field of visual stability, and more generally sensorimotor integration, by providing a new niche where cellular and circuit mechanisms can be thoroughly dissected.

Understanding how the brain works is the next great frontier in biology. One can choose among many available tools and approaches to undertake this challenge. One approach that has proven successful is to investigate the necessary role of a brain area as a first step before trying to understand the neural mechanisms by which it accomplishes its role. To identify the necessary role of a brain area, one needs to manipulate its activity with causal research methods. Optogenetics is a new causal research method that brought about a revolution in brain research conducted with rodents and small animals. Unfortunately, the same cannot be said about optogenetics when applied to animals more similar to humans, such as non-human primates. The larger primate brain poses new challenges to the success of this powerful research method. In this research project, I will develop a new optogenetic-based reversible lesion technique that addresses these challenges. Moreover, I will do so while freeing non-human primates from the physical restraints imposed on them for the practical purpose of accessing their brain. This original method will put together three new technologies: inhibitory step-function opsins, convection-enhanced delivery of opsins, and chronically implanted LED illuminators that are wirelessly activated. With this tool, it will be possible to study for the first time the causal role of brain areas of monkeys that are behaving in naturalistic environments. I will apply this tool to the study of social cognition in monkeys interacting with their peers, and borrow well-studied behavioural tasks from ethology and primatology to develop the emerging field of neuro-ethology.

Understanding and exploiting the sequence - structure - function relationship is one of the fundamental goals of bioinformatics. Systematic predictions of protein structure, function and interactions could provide a global understanding of the protein interaction networks that underlie cell’s life. Conversely, being able to design molecules, peptides or proteins with a prescribed fold or function is a promising lead for drug design. Unfortunately, current state-of-the-art remains far away from this goal. Indeed, direct approaches based on physical interactions, such as Ab Initio Molecular Dynamics (MD) or Fragment-based folding are often either too computationally expensive or inaccurate. Recently, important progress were achieved in structure, function and protein-protein interactions prediction by the introduction of coevolution-based approaches such as Direct Coupling Analysis (DCA). Though powerful, such models are strongly limited in practice because they require many sequences from the same family in order to accurately predict structure. One natural explanation is that they lack general knowledge of proteins biochemistry, such as amino-acid similarity and stereotypes of interactions. Here, we propose to leverage recent advances in the fields of transfer learning and deep generative models to design new coevolution models that share knowledge between protein families, and thus would be more efficient than naive DCA. Beyond structure prediction, we also propose new general-purpose architectures that aim at learning directly the sequence - structure - function relationship, and discuss applications to protein function prediction and sequence/fold designs.

This project aims at clarifying how brain connectivity fluctuations over time relate to mood changes and its interplay to cognitive control. Three specific aims are defined. 1) I will clarify how aberrant mood responses contribute to cognitive control abilities. 2) I will identify how mood fluctuations over short timescales of days to months impact cognitive control abilities. 3) I will investigate how dynamic changes in brain connectivity relates to mood changes and its interplay with cognitive control.
I will use reinforcement learning algorithms and computational models of mood to describe cognitive control-mood interplay and determine how aberrant mood reactions relate to impaired behavioural adjustment. To this end, I will investigate cognitive control-mood interplay in a large sample of healthy adolescents and adolescents with clinical and subclinical depression as impaired mood reactivity is mostly pronounced in this population. Behavioural data of cognitive control–mood interplay will be used conjointly with functional resting state brain imaging data. Repeated, longitudinal assessment at the behavioural and brain level will elucidate dynamics of network configuration over time and how they impact on mood changes and its interaction with cognitive control. With this project I aim to reach a mechanistic understanding of cognitive control-mood interplay and associated brain networks. The project has also implications for the understanding of symptoms of depression, which is a major, leading cause of disability.

Classical ensemble experiments have been used to characterize DNA damage responses that protect bacterial cells against the toxic and mutagenic effects of DNA damage and stress conditions. More recently, examining the underlying gene regulatory mechanisms at the level of single cells and single molecules revealed unexpected stochastic effects that cause cellular heterogeneity in the damage response. These observations provoke the question whether variations in gene expression modulate DNA repair activities and the rates of mutagenesis of individual cells.

I will address this fundamental question using the E. coli adaptive response to DNA alkylation damage as a model. It has been shown that the regulator protein Ada exhibits extreme variation in gene expression between cells in an isogenic population. Using single-cell expression reporters and single-molecule tracking, I will investigate how this variation affects the expression and DNA repair activities of the genes that are regulated by Ada, namely aidB, alkA and alkB. I will then link phenotypic and genetic variation by whole-genome sequencing of single-cell isolates. This will uncover the impact of DNA repair heterogeneity on mutagenesis at the genome level. Finally, I will investigate the role of the adaptive response in antibiotic-induced mutagenesis.

Because the DNA repair pathways are highly conserved, my findings will also impact our understanding of DNA repair and mutagenesis in eukaryotes.

Aggression, manifested in violence and brutality, poses major risks to our society. While our understanding of the neural correlates of aggression progressed substantially, how aggression is developed and why, remains an open question. Part of this gap is due to the view of the neural circuitry of aggression as fixed, making aggressive behaviors inevitable. Yet, recent advances in the anatomical and genetic basis reveal that aggression has an experience dependent aspect, holding great promise for unraveling the mechanisms of aggression. Aimed at revealing the mechanisms that control aggressive behaviors, this study leverages experience dependent plasticity in tandem with a fusion of circuit and cellular level approaches.
Using a combination of advanced technologies including 2-photon imaging in head fixed mice, single cell RNA sequencing, a combination of 2-photon imaging with in-situ hybridization and gene editing with CRISPR/CAS9, three questions are addressed: 1. What are the changes in neuronal activity following experience dependent plasticity? 2. What are the genetic changes that are associated with the alterations found in neuronal activity? 3. What links these genetic changes to neuronal activity and aggression?
Joining cellular level and circuit level approaches to examine how aggression is developed by experience dependent plasticity should yield a new understanding on the relationship between genes, neuronal function and aggressive behavior. By discovering these basic features in the mechanism underlying aggression, the knowledge obtained from this study could open the path to a new era in which prevention of pathological aggression will become feasible.

Functional imaging and optogenetic activation of neural circuits during behavior of untethered animals would allow detailed investigation into the closed-loop interaction of sensory inputs, brain, and motor outputs of behaviors in naturalistic conditions. Previous work investigating the role of Drosophila central complex in navigation and the optic tectum of larval zebrafish in prey capture have been informative, but tethering the animals in such experiments has limited the extent to which circuit mechanisms for multi-sensory, closed-loop control could be investigated. I propose to achieve cellular resolution neural activation and recording in freely moving Drosophila and larval zebrafish, by developing a system for aiming a fast volumetric two-photon microscope to precisely follow, with minimal latency, a region in the animal’s brain. The “feedback based lock-on module” provides low latency feedback to aim the imaging volume of the microscope. Integrating the lock-on module to a fast two-photon microscope based on Bessel beam optics along with an optogenetic stimulator will enable closed-loop analysis of neural mechanism of animal behaviors. The system will then be used to study neural activities of the Drosophila central complex during idiothetic path integration and optic tectum of larval zebrafish during prey capture after expressing GECIs and optogenetic channels into the various sets of neurons using different binary expression systems like GAL4 and LexA.

Transposable elements (TEs) comprise a substantial proportion of all vertebrate genomes, including more than half of our own. They are particularly active in the germ line and during development, and impose an significant evolutionary burden on organisms. As such, numerous mechanisms have evolved to control their spread, including the KRAB zinc finger (KZF) family - the largest and most diverse group of transcription factors in mammals. Strong evidence points to an arms race between TEs and KZFs, and their genome frequencies are highly correlated across diverse vertebrate species. However, KZFs are involved in numerous biological processes beyond TE repression, and we do not yet understand how coevolution with TEs has produced this functional diversity in vertebrates. In addition, whilst the KZFs are mostly limited to tetrapods, other large ZF subfamilies exist outside of this lineage, and the roles of many of these are completely unknown.

Using a comparative and functional genomics approach, I will investigate the dynamics of TE-ZF coevolution, in order to better understand how this phenomenon has shaped the architecture and regulation of vertebrate genomes. Specifically, I will test the hypothesis that TE insertion near ZFs leads to co-regulation of their activity, and furthermore, that this process drives the genomic clustering of ZFs. In parallel, I will use functional techniques and long read re-sequencing to undertake a comprehensive study of of a large, un-described and mysterious family of ZFs in zebrafish. In doing so, I hope to open new lines of research in this essential model organism of embryogenesis and development.

The genome editing field is currently limited in its control over how double-strand breaks can be selectively repaired by non-homologous end joining (NHEJ) or homology-directed repair (HDR) pathways. By providing a donor DNA template, HDR offers potential for corrective therapies targeting a much larger number of genetic diseases which cannot be cured by either single nucleotide substitutions or gene inactivation alone. Yet the efficiency of HDR remains considerably lower than that of NHEJ, hampering its usefulness for therapeutic applications. The use of linear plasmids for enhanced homologous recombination in this proposal directly aims to tackle this challenge and should drive efficient genome editing towards clinical translation.

Unlike circular plasmids which are a ubiquitous tool used throughout modern molecular biology, linear plasmids are an underappreciated biological phenomenon despite being widespread in nature. What constitutes their biological function and what genes they encode is largely unknown. The ability to manipulate linear plasmids in vivo will provide a powerful tool to study and understand their biological significance for the first time. Linear plasmids are double-stranded DNA genetic elements characterised by terminal proteins covalently attached to the 5'-end of each linear DNA strand. In particular, the feature of a terminal protein covalently linked to DNA is a strong prospect for biotechnology applications: I will exploit such linear plasmid systems in mammalian cells for enhanced homologous recombination.

The overall aim of the proposed work is to develop new physical chemistry/bioelectrochemistry techniques in order to better understand the electron-transfer mechanisms of conductive biofilms at the electrode surface. This will have immediate benefits in the development of the promising renewable biophotovoltaic (BPV) technology as well as long-term significance for improvements in understanding the role of key biological electron-transfer species across a broad range of systems.

BPVs use photosynthetic biofilms to harvest solar energy and then convert this into electrical current. They are, however, limited in power output due to unexpectedly slow electron-transfer at the biofilm/electrode interface, and it remains unclear why this is. There is an urgent need to improve fundamental understanding of these electron-transfer mechanisms in order to direct future development of BPVs and remove this bottleneck for their maximum efficiency.

Here, we propose the combination of a number of sophisticated and highly-specialised surface-study techniques in combination with the bioelectrochemistry, synthetic biology and electrode design already in place in the host group, to study the key species (such as the individual redox proteins and complexes) both individually and in situ within model lipid bilayers and finally within the biofilms themselves. Importantly, this will allow us to build up a mechanistic picture of each aspect of these complex buried biointerfaces and hence distinguish between different electron-transfer mechanisms (e.g. via pili aromatic amino acids, outer-membrane redox proteins or diffusional shuttles) that are currently proposed.

Most odontocete cetaceans (toothed whales, dolphins, and porpoises) produce echolocation and communication sounds using bilateral pairs of sound-generating phonic lips. The production of high-frequency echolocation clicks and analysis of the returning echoes allow foraging odontocetes to detect, localize, and capture prey. Communication between odontocete conspecifics is mediated by family-specific sounds, with dolphins producing low-frequency whistles and porpoises emitting high-frequency social clicks. Previous research on the lateralization of sound production in dolphins suggested a left hemisphere bias for echolocation and a right hemisphere bias for conspecific communication; however, similar to other behavioral studies of cetacean lateralization, hemispheric specialization was inferred, rather than measured. Here, I aim to perform the first assessment of hemispheric specialization for the production and perception of echolocation and communication sounds in the bottlenose dolphin (Tursiops truncatus) and harbor porpoise (Phocoena phocoena). Using bioacoustics approaches, novel species-customized electroencephalography-functional near-infrared spectroscopy (EEG-fNIRS) arrays, and functional magnetic resonance imaging (fMRI), I will directly and indirectly measure neural activity underlying motor control and auditory sensation in these odontocetes. In addition to providing unprecedented insights into cetacean brain function, this project has the potential to redefine the limits of neuroscience and animal behavior research through the introduction of portable, non-invasive, and customizable EEG-fNIRS equipment that can be adapted for taxonomically diverse species.

Cytokines act as essential regulators of immunity by eliciting signaling outputs from cell surface receptors. There have been extensive efforts to use cytokines for a wide range of immune disease, including cancer, but their functional pleiotropy has limited their clinical utility. Precedent studies in the Garcia lab, demonstrated that surrogate cytokine ligands can alter signaling output by modifying receptor dimer geometry. For more precise topological control of receptor-ligand interactions and signaling, we hope to create a new class of modular binding scaffold proteins which are applicable to diverse receptor targets. The new scaffolds will be computationally designed and engineered to have three major characteristics: (i) high binding specificity to a target receptor, (ii) self-assembly to form dimers or trimers, (iii) drug-like properties including minimal immunogenicity. These scaffolds will be used for manipulating the geometry of cell surface receptors to fine-tune of signaling outputs. These novel scaffolds can also be utilized to drive the formation of non-natural receptor pairings to elicit completely new functional outputs. Proof-of-concept experiments from Garcia lab have shown that non-natural receptor dimers of receptor tyrosine kinases (RTK) can induce different signaling outputs from each of their parental receptors. We believe that the new scaffolds can efficiently combine any cytokine receptor and RTK to form diverse combinations of dimeric or trimeric receptor pairs. In this way, it will help us to broaden our knowledge on kinase-linked receptor signaling and also have a chance of finding new agonistic and therapeutically meaningful protein ligands.

Intestinal stem cells, which reside in crypts, fuel self-renewal, maintenance, and repair of the epithelial lining of the gut. The regulatory mechanisms that govern maintenance and differentiation of intestinal stem cells, as well as structural and functional differences along regions of the intestine are poorly understood. DNA methylation (DNAme) holds potential in deciphering these mechanisms by revealing gene regulation and clonal structure of cell populations.

The goal is to investigate the dynamics, diversity, and function of DNAme patterns regulating intestinal epithelial homeostasis at single-cell resolution. This proposal will survey single-cell DNAme of intestinal crypts at three anatomical scales: dynamics within individual crypts, diversity between neighboring crypts, and region-specific diversity from proximal to distal sections of the small intestine. Our hypothesis, which builds on preliminary data, is that DNAme patterns of intestinal cells are largely determined by the clonal history of these cells and therefore hold promise as an endogenous lineage tracing tool. To test this hypothesis, I will use a range of single-cell sequencing techniques and develop computational methods to pursue three specific aims:

Aim 1: Determine the single-cell DNAme dynamics within intestinal crypts.

Aim 2: Analyze the diversity of DNAme patterns between crypts along different regions of the intestine.

Aim 3: Disentangle the contribution of cell-type and clonal history in DNAme patterns.

This proposal combines computational with experimental approaches, and leverages existing collaborations between the van Oudenaarden and Clevers labs at the Hubrecht Institute.

The brain has a remarkable capacity for combining long-term memory with exquisite plasticity allowing for adaptation to the environment. A growing aspect of neural plasticity is modulation of myelination. How can myelination be altered? Myelination can be altered by the generation of new myelinating oligodendrocytes or alternatively by mature myelinating oligodendrocytes rendering their existing myelin. The current dogma is that continuous generation of oligodendrocytes underlies myelin remodeling—even though the contribution of mature oligodendrocytes remains poorly understood. Here I propose a multidisciplinary approach, combining my expertise in human cell biology with Dr. Chan’s extensive proficiency in oligodendrocyte biology to study the cellular and molecular mechanisms underlying myelin alterations upon induced neuronal activity and after injury. I plan to explore the myelinogenic potential of mature oligodendrocytes and their contribution to myelin remodeling. This strategy combines genetic fate-mapping and manipulation of oligodendrocyte lineage cells, in vivo pharmacogenetic neuronal stimulation (DREADDs) with single cell, temporal-resolved gene expression analysis and behavioral assessment. More specifically, genetically altered progenitors unable to generate new oligodendrocytes and mature oligodendrocytes will be labeled and profiled using single cell RNA-seq. This will allow for the longitudinal assessment of genes involved in activity-dependent myelination and the consequence of this manipulation. Knowledge of myelin plasticity is essential for the development of new treatments for demyelinating disease such as multiple sclerosis.

Neurons are the most diverse class of nervous system cell types, based on thorough characterization of single cells into molecular taxonomies. Individually neurons are fascinating and complicated, but at the core of computation are their interactions. The unique arrangement of circuits between specialized neuronal types results in a wide range of sophisticated functions and complex behaviors. While female and male brains largely perform the same sophisticated functions, the neurobiological basis for gender differences in some behaviors has fascinated researchers over decades. Sexual dimorphism in the amygdala (specifically the posterodorsal medial amygdala, MePD) shapes gender differences in reproductive behavior, aggression or social recognition, with well-established contribution of gonadal steroid hormones. At the core of this dimorphism are neuroanatomical differences of the MePD, such as regional and cell volume, cell number and cell circuitry. While the phenomenon is well established using relatively simple methods, a more global, systematic characterization of differences in circuits in the context of cell types and gene expression has not been technically possible. To this end, I propose to develop and apply an integrated approach for simultaneous connectomics and transcriptomics measurements of the medial amygdala at single cell resolution. This study will reveal the contribution of circuits, cell types and gene expression to gender-based amygdala dimorphism. Further, the proposed framework to understanding circuit implementations in the context of cell types will advance other studies of nervous system function and behavioral neuroscience.