10.1517/17460440902835483 2009 Informa UK Ltd ISSN 1746-0441 507All rights reserved: reproduction in whole or in part not permitted
ZebrafishmusculardiseasemodelstowardsdrugdiscoveryHiromi HirataNagoya University, Graduate School of Science, Furo-cho, Chikusa-ku, Nagoya 464-8602, Japan
Background: Zebrafish is an amenable vertebrate model useful for the study of development and genetics. Small molecule screenings in zebrafish have successfully identified several drugs that affect developmental process. Objective: This review covers the basics of zebrafish muscle system such as muscle development and muscle defects. It also reviews the potential use of zebrafish for chemical screening with regards to muscle disorders. Conclusion: During embryogenesis, zebrafish start to coil their body by contracting trunk muscles 17 h postfertilization, indicating that a motor circuit and skeletal muscle are functionally developed at early stages. Mutagenesis screens in zebrafish have identified many motility mutants that display morphological or functional defects in the CNS, clustering defects of acetylcholine receptors at the neuromuscular junctions or pathological defects of muscles. Most of the muscular mutants are useful as animal models of human muscle disease such as muscle dystrophy. As zebrafish live in water, pharmacological drugs are easily assayable during development, and thus zebrafish may be used to determine novel drugs that mitigate muscle disease.
Keywords: development, locomotion, muscle, mutant, zebrafish
Expert Opin. Drug Discov. (2009) 4(5):507-513
Zebrafish have three distinct cell types of axial musculature; fast-twitch muscle cells (fast muscle), slow-twitch muscle cells (slow muscle) and muscle pioneer cells. The fast and slow muscles are identical to white and red muscles, respec-tively. Muscle pioneers, which are necessary for correct outgrowth of primary motor axons, are a specialized subset of slow muscle cells in fish. Each muscle cell originates from somites but takes a distinct lineage and is finally located at a distinct part of the axial musculature.
After gastrulation (5 10 h postfertilization [hpf ]), somites are formed from paraxial mesodermal tissue, in other words, from presomitic mesoderm [1,2]. This segmentation process occurs every 30 min in zebrafish from the anterior to the posterior direction on both sides of the neural tube and notochord. Each somite gives rise to myotomes, dermatomes and sclerotomes, which respectively differentiate into muscles, dermis and bones. In parallel with the somitogenesis, hedgehog signals from midline induce medial myotome to differentiate into adaxial cells, which are muscle precursors that give rise to slow muscles and muscle pioneer cells [3,4]. Most adaxial cells migrate radially to the lateral surface to form a single-layer muscle underneath the skin, where they differentiate into slow muscle cells, whereas a subset of the adaxial cells remains at the medial loca-tion, elongates its shape to span from medial to lateral myotome and differentiates into muscle pioneer cells [5-7]. The rest of the myotomal cells differentiate into fast muscle cells.
In mammalian musculature, slow and fast muscle cells exist as a mosaic pattern. Each muscle cell type can be histologically distinguished by ATPase
1. Development of skeletal muscle
2. Development of motility
3. Zebrafish motility mutants
caused by muscle or
neuromuscular junction defects
4. Chemical biology in zebrafish
6. Expert opinion
508 ExpertOpin.DrugDiscov.(2009) 4(5)
activity of myosin or reductase activity of mitochondria. In fish, in contrast, these two populations are easily distinguishable, because they are not mixed in location. The slow muscle is only a superficial single cell layer of fibers, whereas the rest of the muscle constitutes fast muscle. In slow muscle, a network of electrical coupling to share synaptic currents is important to drive rhythmic swimming, whereas fast muscle generates action potentials to mediate rapid escape behavior [8-10]. It may be beneficial to examine cell type-specific injury and recovery of muscle in zebrafish.
Development of neural circuits and muscles in zebrafish is very fast and embryos show three distinct stereotyped behaviors (spontaneous coiling, touch-evoked coiling and swimming) by 36 hpf [11,12]. The earliest locomotion consists of repetitive, spontaneous alternating coiling of the trunk. This simple slow coiling is independent of mechanosensory stimulation and abruptly starts at 17 hpf, reaches a peak frequency of 0.3 1 Hz at 19 hpf and declines to < 0.1 Hz by 26 hpf. After 21 hpf, embryos respond to mechanosen-sory stimulation with fast trunk coils. The initiation of this touch-evoked coiling indicates that neural networks from somatic sensory neurons to motoneurons through interneurons as well as skeletal muscles are functionally connected even before 24 hpf to execute escape behaviors. By 26 hpf, mechanosensory stimulation initiates swimming, which is defined as a forward movement with rhythmic tail flips by at least one body length. The frequency of muscle contractions during swimming increases from 7 Hz at 26 hpf to 30 at 36 hpf, the latter being similar to the frequency of swimming of adult zebrafish .
The escape behavior followed by a mechanosensory stimulation can be divided into several steps from sensory perception to muscle contraction. In zebrafish embryos, two types of mechanosensory neurons perceive touch stimuli. Head and yolk stimulation are transduced by trigeminal sensory neurons, whereas tail stimulation activates RohonBeard mechanosensory neurons in the trunk [13,14]. RohonBeard cells die within 4 days in development and, in larval stage, the function is taken over by the dorsal root ganglia . These sensory neurons project to interneuronal networks located in the spinal cord and hindbrain to produce motor rhythm . This motor pattern alternatively activates motor neurons in each side of the spinal cord. Motor terminals release acetylcholine at the neuromuscular junctions (NMJs) and depolarize the muscle membrane [9,17,18]. Depolarizations of the plasma membrane spread down the transverse-tubules (t-tubules), which are invaginations of the plasma membrane, and cause conformational changes of the dihydropyridine receptor (DHPR), a voltage sensor located in the t-tubule membrane . The DHPRs then trigger opening of ryano-dine receptors (RyR) in the adjacent sarcoplasmic reticulum
(SR) membrane to allow Ca2+ release from the SR to the cytosol . Increased cytosolic Ca2+ activates troponin C that initiates actin/myosin sliding, thus causing muscle contraction . The cytosolic Ca2+ levels are rapidly decreased by sarco(endo)plasmic reticulum Ca2+-ATPase (SERCA), a calcium pump expressed in the SR of skeletal muscle that enables fast relaxation . As activation of muscle is caused by rhythmic motor outputs, slow and fast bilateral alternation of muscle contractions generates coiling and swimming behavior, respectively.
Zebrafish is the vertebrate most amenable to forward genetic screens. Generation of zebrafish homozygous mutants was first described in 1981 by Streisinger and his colleague, who developed zebrafish as a genetic model of vertebrates [23-27]. In the early 1990s two large-scale mutagenesis screens were successfully performed in Tbingen, Germany, and Boston, US, and > 4,000 mutants were identified [28,29]. In these screening, 166 behavioral mutants that displayed abnormal touch-evoked swimming were reported . Among them, in 63 mutations, striation of somitic muscle fibers was reduced by birefringence observation, indicating that these mutants have defects in structural arrangement of actin and myosin. In fact, the responsible genes of this class of mutants are structural components of muscle. On the other hand, some other locomotion mutants displaying normal birefringence under polarized light are defective in Ca regulation in muscle or formation of CNS or NMJ.
Mutations in dystrophin cause embryonic-onset, progressive degeneration of skeletal muscle and impaired locomotion [31,32]. Many muscle fibers in the mutants were detached from the myoseptum, which is an attachment site of myofibrils located at somite boundaries. Similarly, laminin alpha2 mutants display disconnection of muscle fibers . Although these two mutants show similar fiber detachment and reduced muscle birefringence, the morphological defects in dystrophin mutants are caused by sarcolemmal rupture, whereas the muscle atrophies in the laminin mutants are mechanically induced by spontaneous muscle contraction . Another mutant, which is linked to titin locus, also shows sarcomere defects [34,35]. As mutations of the dystrophin, laminin or titin gene in humans are responsible for muscular dystrophy, these zebrafish mutants could serve as animal models for human muscular dystrophy [36-38].
Some other mutants with reduced birefringence of skeletal muscle also show morphological defects of muscle. In hsp901 mutants, thick filaments composed of myosin pro-teins are absent and the sarcomere structures are defi-cient . The mutant embryos lack both spontaneous and stimulus-evoked muscle contractility. Similarly, unc45b mutations impair skeletal myofibril formation and locomotion . Hsp901 and its cochaperone Unc45b