Hugo Bellen's research interests include (1) neuronal communication and maintenance, (2) development of tools to control transcript and protein levels in adult neurons to assess which proteins are required for neuronal function and survival, and (3) the creation of genome-wide libraries to manipulate genes in vivo. His lab uses Drosophila melanogaster as a model system because most biological processes are evolutionarily conserved and studies in fruit flies provide many important clues about the aging process in animals and humans.
Our interest in synaptic transmission prompted us to study the function of the VapB protein. We discovered that the protein does not play a prominent role in neurotransmission but is required for the development and maintenance of the neuromuscular junction (NMJ). It was then discovered that the loss of its human homolog causes amyotrophic lateral sclerosis. We therefore decided to investigate the biological processes leading to the demise of aging neurons in these and other mutants. Our research strives to (1) determine the key biological processes initiating the demise of neurons in aging flies, (2) elucidate the molecular processes that impair the function of the affected organelles, (3) find ways to improve neuronal function in degenerating neurons, and (4) develop methods in which neurodegenerative processes may be suppressed. We tackle these problems by combining forward genetics, electrophysiological paradigms, transmission electron microscopy, cell biology, biochemistry, pharmacology, and genomic approaches.
Our model neuron, the fly photoreceptor (PR), offers many advantages. Its electrical activity can be gauged upon light exposure via electroretinograms (ERGs). Unlike vertebrate PRs, the fly PRs sense light and directly project their axons into two different layers of the fly brain. The stereotyped arrangements of cell bodies of the PRs in fly ommatidia makes performing light and electron microscopy in the retina simple, while stereotyped projections of the PR terminals in the fly brain allow ultrastructural analyses of defined axons and synapses. The ERGs provide information about the functional properties of the neurons; the morphological studies provide cell biology information, including the types of organelles that have been affected.
To define the key processes that initiate the demise of neurons we performed large-scale, forward unbiased mosaic genetic screens in adult flies. Our strategy was to create random lethal mutations, induced on chromosomes carrying a transgene that promoted recombination between specific chromosome arms (FRT) upon expression of a recombinase (Flipase). This allowed us to produce homozygous mutant clones in the eyes of heterozygous adult flies. We then monitored the demise of PR neurons over a period of up to six weeks.
A high-throughput screening procedure led to the rapid identification of numerous mutations causing a functional loss of PRs. The development of a new mapping strategy, using a BAC (bacterial artificial chromosome) transgenic technique, P[acman], permits, in combination with whole-genome sequencing, rapid mapping of these mutations.
Once we identify the gene, we define the cells or organelles in which the proteins are expressed, assess when the loss of the gene leads to the first symptoms, and attempt to unravel the molecular processes that lead to the demise of the neurons by studying the presymptomatic stages of neurodegeneration. Finally, we use genetic and pharmacologic approaches to attempt to suppress the neurodegenerative features associated with the deleterious mutations.
Results from these screens have caused us to focus on nuclear genes whose products are trafficked to the mitochondria and on genes that are involved in autophagy and endolysosomal trafficking. However, we also study two other key proteins, NMNAT and VapB, as we isolated the first fly mutations in these genes and have been trying to unravel how they cause the demise of neurons.
Forty percent of the genes we have identified so far from our unbiased genetic screens encode mitochondrial proteins, indicating that mitochondria play a critical role in numerous forms of neurodegeneration. We have isolated mutations in 34 different genes (Figure 1), 18 of which have human homologs associated with diseases in infants and adults. These diseases, which exhibit both shared and distinct clinical presentations, led us to investigate the phenotypes associated with mutations in these genes in Drosophila. Based on characterization and comparison of these mutants, we have developed common phenotypic profiles and identified common underlying mechanisms. We found that mutations in very different genes and affecting very different processes can cause strikingly similar phenotypes, while genes affecting the same pathway or process can lead to very different phenotypes. These observations permit us to classify the genes based on various criteria and develop targeted strategies to suppress the neurodegenerative phenotypes for seemingly unrelated genes and processes.
So far, we have identified five genes that affect lysosome function. Mutations in subunits of the retromer complex (Vps26 and Vps35) reveal that rhodopsin 1 (Rh1), a light-sensing protein in PRs, is recycled from endosomes upon activation and endocytosis. The loss of retromer-dependent recycling leads to constitutive Rh1 lysosomal degradation, burdening the endolysosomal pathway, with subsequent PR dysfunction. Loss of Crag, the first guanylate exchange factor (GEF) for Rab11 identified, leads to increased Rh1 lysosomal degradation and PR degeneration.
We also found that loss of subunits of the voltage-gated Ca2+ channel (VGCC), a complex that has only been studied in the context of synaptic transmission, causes an age-dependent accumulation of autophagic vacuoles in PR terminals. Loss of the mouse VGCC homologs leads to similar autophagic defects and degeneration of neurons, indicating that VGCC plays a previously uncharacterized evolutionarily conserved role in autophagy and lysosomal degradation.
Mutations in NMNAT (NAD synthase nicotinamide mononucleotide adenylyltransferase) cause the quickest and most severe PR loss of all genes identified so far. Overexpression of NMNAT provides strong protective effects against light-induced PR degeneration in flies and protects against spinocerebellar ataxia 1 (SCA1)-induced neurodegeneration, suggesting a broad neuroprotective function. NMNAT protects neuronal function and survival partly through a proteasome-mediated pathway, independent of its enzymatic activity. NMNAT displays strong chaperone activity both in biochemical assays and cultured cells, and shares significant structural similarity with a known yeast chaperone.
We are unraveling the precise mechanism by which this protein acts as a potent neuroprotective agent by studying genetic and protein interactors of NMNAT. To exploit NMNAT's neuroprotective functions in human diseases, especially Alzheimer's disease, we are collaborating with Hui-Chen Lu (Baylor College of Medicine), Huda Zoghbi (HHMI, Baylor College of Medicine), and scientists at MD Anderson Cancer Center to develop potential drugs.
VapB and Amyotrophic Lateral Sclerosis
Mutations in the human gene VapB were shown to cause amyotrophic lateral sclerosis 8 (ALS8), a severe neuromuscular disease in which motor neurons stop functioning. In Drosophila, we unraveled a new signaling pathway whereby the human VapB protein, which is an endoplasmic reticulum (ER)-associated protein, is cleaved, secreted, and circulated in the blood (Figure 2). In collaboration with Michael Miller (University of Alabama) and Hiroshi Tsuda (McGill University), we showed that its amino-terminal MSP (major sperm protein) domain binds to growth cone guidance receptors on muscles and regulates glutamate receptor clustering as well as mitochondrial dynamics in a cell nonautonomous manner. A point mutation associated with ALS8 blocks the secretion of VapB. Other familial forms of ALS, as well as sporadic cases, also exhibit diminished levels of VapB, indicating that defects in this novel signaling pathway may underlie multiple forms of this disease. We are now pursuing experiments to determine if VapB can be used as a diagnostic tool and a therapeutic molecule for ALS.
Genome-wide Tools to Manipulate Flies
We have been developing technology and reagents for the fly community for the past 25 years. These resources are popular to manipulate the fly genome. We helped develop enhancer detector transposable elements that permitted the cloning of hundreds of genes and also led to the development of the UAS-GAL4 system by Andrea Brand (University of Cambridge, UK) and Norbert Perrimon (HHMI, Harvard Medical School), the most widely used Drosophila technology for expressing a gene/protein/RNAi of interest in a tissue-specific manner. In collaboration with Allan Spradling (HHMI, Carnegie Institution of Washington) and Roger Hoskins (Lawrence Berkeley National Laboratory), we are using transposable elements to generate a library of transgenic flies with tagged genes, permitting the mutational analysis of nearly 10,000 genes in vivo. We also developed the transformation vector P[acman], which allows the insertion of transgenes up to 250 kb and the integration of the transgene at defined sites to avoid positional effects. This vector also allows elegant manipulations of the DNA in bacteria using recombineering. The vector also permitted the creation of a library of more than 100,000 transformation-ready P[acman] clones covering the whole fly genome. Recently, we developed a new transposable element collection of MiMIC insertions that permits genomic tagging of thousands of proteins as well as the temporal and spatial knockdown of genes and proteins in a reversible manner in specific cells at all stages of development and in adult flies.
Grants from the National Institute of Health (NIGMS), the Robert A. and Renee E. Belfer and Huffington Foundations, and Target ALS provided partial support for these projects.
As of April 6, 2016