Learning to swim: What fish can teach us about CMDs Jim Dowling, M.D., Ph.D. University of Michigan Congenital Muscular Dystrophy Family Conference August 14-15, 2010 Meet the zebrafish • • • Small, fresh water aquatic vertebrate Lifespan 1-2 years Independently swimming by day of life 3 Why zebrafish??? (part 1) It’s a numbers game… Breeding age = 3 months Average litter size = 8-13 Time between litters = 20 days Breeding age = 2-3 months Average clutch size = 100-300 Time between clutches = 1 week Breeding age = 16+ years Average litter size =1 Time between litters = 9-12 months minimum Why zebrafish??? (part 2) • Crystal clarity! – Zebrafish are optically translucent allowing for live imaging of muscle and heart Why zebrafish??? (part 3) • • Invertebrate style genetics – Large number of offspring – Can easily introduce DNA/RNA – Can do saturating mutagenesis screening Vertebrate style genome – Genome at least as complex as ours – Genome sequenced as part of the NIH genome project – All known muscular dystrophy genes are found in the zebrafish genome (Steffen et al., 2007) Zebrafish as a model for muscle disease • Muscle development begins at 20 hpf and is completed by 48 hpf – i.e. very quickly!!! • Fish muscle shares many features with human muscle – Slow and fast twitch fibers – Limb and trunk muscle – Contains the common muscle structures • What’s different? – Slow and fast fibers in separate compartments – Significantly more trunk than limb muscles – Some reagents do not work in fish Zebrafish as a model for muscle disease (cont) • Obvious phenotypic consequences from muscle dysfunction – Impaired swimming • Abnormal muscle observable in live fish – Can see it with conventional microscopy • Dowling et al. (2008) Circulation Resçearch Histopathologic changes that reflect human muscle pathologies – (dystrophic pattern in dystrophies, for example) Zebrafish as a model of muscle disease Control embryos (3 days old) Myopathy embryos (3 days old) Zebrafish models of CMDs • How do we make zf models of CMDs? • What models currently exist? • What have they been used for in the past? – In other words, what have we learned from them? • What can we use these models for in the future? – How can they help patients with CMDs? • Additional thoughts and future directions How do we make zebrafish CMD models? • Two ways to make a model 1. Transient models – – – 2. Morpholino knockdown Mutant transgene expression Effect lasts 2-5 days Stable models – Chemically induced or spontaneous mutant – Dominant transgenics – Direct gene targeting – – – Gene trap TILLING ZF method What CMD models current exist? • Transient Models (morpholino based) – LAMA2 (MDC1A) – FKRP – dystroglycan – COL6A1/A3 (UCMD) – SEPN1 (RSMD) • Stable – LAMA2 (candyfloss) – RYR1 (relatively relaxed) Zebrafish model of MDC1A • • ENU induced mutation called candyfloss Identified/characterized by Currie’s group – (Hall et al., 2007) • • • Clinical – Onset at 3 days – Progressive weakness – Death by 2 weeks Genetic – Point mutation in LAMA2 leading to premature stop codon – Absent LAMA2 staining Histopathologic – Progressive myofiber injury and degeneration – Fragility at the myotendinous junction Zebrafish model of UCMD • Experiment from my group – (Telfer et al., 2010) • • • Clinical – Severely reduced motor function at 24 and 48 hpf – Obvious morphologic abnormalities Genetics – Morpholino mediated knockdown of COL6A1 (exon 9) and COL6A3 (exon 13) – Models two common dominant mutations Histopathology – Myofiber disorganization and sarcolemmal breakdown – mitochondrial swelling, increased apoptosis – Reduced and disorganized collagen VI staining Zebrafish models of dystroglycanopathies • Experiments from two groups – Thornhill et al., 2009 – Kawahara et al., 2010 • Clinical – Reduced motor function – Morphologic changes at 24 and 48 hpf • Genetic – Morpholino knockdown of FKRP (x2 groups: Straub, Kunkel) • Histologic – Pathology in muscle, eye and brain – Reduced laminin binding – Reduced glycosylated dystroglycan Zebrafish model of core myopathies (RSMD): ryr • Identified at UM by John Kuwada and colleagues • Spontaneous mutant with abnormal swimming (named relatively relaxed) • Have impaired excitation-contraction coupling • Found to have a homozygous recessive mutation in ryr1b • Mutations cause significantly reduced levels of RYR1 in fast muscle • Genetically and histologically a model for recessive core myopathies – RYR1 and SEPN1 related myopathies Hirata et al. 2007 What did we learn from these models? • Candyfloss (MDC1A) – LAMA2 important for the myotendinous junction – Like mice, LAMA2 zebrafish have an early onset, severe phenotype • UCMD model – Confirmed mitochondrial proton pore hypothesis (and confirmed potential therapeutic role for molecules that act at the MPP) – Unlike mice, UCMD zebrafish have an early onset, severe phenotype • FKRP morphants – Demonstration of ability to model dystroglycanopathies in the zebrafish – Defined key techniques for future experimentation • Core myopathy models (SEPN1 and RYR1) – Confirmed key role of excitation-contraction coupling in muscle function – Established/strengthened linked between SEPN1 and RYR1 related myopathies How can we use these models in the future? • Identification of new/novel aspects of disease pathogenesis • Identification of disease biomarkers • Platform for rapid identification/development of novel therapeutics Therapy development in zebrafish models of CMDs • Two stable genetic models of CMDs in zebrafish – Candyfloss (MDC1A) – Relatively Relaxed (ryr) • Morpholino models recapitulate clinical, genetic, and pathologic features of CMDs • No new therapies yet developed in these models – Verification of efficacy of CsA in UCMD model • Zebrafish have excellent potential for MTS and HTS Telf er et al., 2010 Drug screening in the zebrafish • ZF and drug screening – – – – • Large number of offspring Frequent mating Easily absorb drugs in media Translucent body plan plus many GFP markers Muscle specific phenotypes for drug screens – Birefringence – Motor function – Other targets? (for example, cardiac phenotypes) HTS and birefringence in zebrafish • Birefringence – Property of muscle when placed under polarized light – Abnormal birefringence is a measure of impaired myofiber integrity and/or organization – Abnormal birefringence reported in all muscular dystrophy ZF yet described • Birefringence and CMDs – Abnormal in candyfloss (LAMA2) – Abnormal in MO knockdown of COL6A1 (UCMD model) – Abnormal in MO knockdown of FKRP (dystroglycan model) FKRP morphants (Kunkel) Drug screen of birefringence in zebrafish Embryos are fertilized and collected De-chorionate embryos Place in 24 well dish with one drug in each well (16 embryos per drug) Allow exposure to drug for 3 day s (change drug each day) Examine birefringence at day of life 4 Secondary analysis of positive hits (dose response; post-sympt Rx; survival; etc) Re-test positive hits with a large sample size Positive hit = no fish with abn birefringence at day of life 4 Evaluate birefringence at days of life 5-7 Wash away durg Birefringence in UCMD model +/- cyclosporin A An example of a successful screen using birefringence with a DMD model 3 day washout Drug 44: 0/16 with sap Drug 44: 2/16 with sap Drug 50: 4/16 with sap Drug 50: 4/16 with sap Drug screening and motor function in zebrafish • CMD models in zebrafish all have abnormal motor function – Impaired touch evoked escape response (starting at 3 dpf in candyfloss, earlier in MO models) – Impaired swimming in all models 3 day old control 3 day old ryr Antioxidant treatment strategy Identify ryr embryos (DOL3) Place in 12 well dish with antioxidant or placebo (DMSO). Replace solution daily Record daily swim speeds Determine survival (lethality for ryr at 12-14 days) Noldus Zebrafish System Proof of principle: Antioxidant treatment in ryr zebrafish Untreated CTL CTL ryr ryr Antioxidant treated Future Directions- Zebrafish Models • What do we have – Excellent model of MDC1A and RYR1 – Transient models of UCMD and ZKRP dystroglycanopathy • Not really amenable to large scale drug screening – Sound, quantitative assays for drug screening • What do we need – Standardization of phenotypes and assays – “stable” models for UCMD, dystroglycanopathies, SEPN1, and laminopathies A few parting ideas… • What validation is (or should be) required for novel concepts regarding pathogenesis that are developed in the zebrafish? • What are the step(s) (or what should they be) between drug discovery in non-murine models and clinical trial?