Schwann cell myelination requires Dynein function
© Langworthy and Appel; licensee BioMed Central Ltd. 2012
Received: 20 August 2012
Accepted: 6 November 2012
Published: 20 November 2012
Skip to main content
© Langworthy and Appel; licensee BioMed Central Ltd. 2012
Received: 20 August 2012
Accepted: 6 November 2012
Published: 20 November 2012
Interaction of Schwann cells with axons triggers signal transduction that drives expression of Pou3f1 and Egr2 transcription factors, which in turn promote myelination. Signal transduction appears to be mediated, at least in part, by cyclic adenosine monophosphate (cAMP) because elevation of cAMP levels can stimulate myelination in the absence of axon contact. The mechanisms by which the myelinating signal is conveyed remain unclear.
By analyzing mutations that disrupt myelination in zebrafish, we learned that Dynein cytoplasmic 1 heavy chain 1 (Dync1h1), which functions as a motor for intracellular molecular trafficking, is required for peripheral myelination. In dync1h1 mutants, Schwann cell progenitors migrated to peripheral nerves but then failed to express Pou3f1 and Egr2 or make myelin membrane. Genetic mosaic experiments revealed that robust Myelin Basic Protein expression required Dync1h1 function within both Schwann cells and axons. Finally, treatment of dync1h1 mutants with a drug to elevate cAMP levels stimulated myelin gene expression.
Dync1h1 is required for retrograde transport in axons and mutations of Dync1h1 have been implicated in axon disease. Our data now provide evidence that Dync1h1 is also required for efficient myelination of peripheral axons by Schwann cells, perhaps by facilitating signal transduction necessary for myelination.
Motor control and sensation require that nerve impulses are rapidly and efficiently transmitted over long distances. This is achieved by axons, which relay electrical signals between the central nervous system and peripheral muscles and sensory elements, and Schwann cells, which enhance the speed and efficiency of signal propagation by ensheathing peripheral axons with insulating myelin. Peripheral nerves can be very long, reaching more than 1 m in humans, but elaboration of a far-reaching peripheral nerve network comes with an apparent cost in that it is highly susceptible to disease. More than 100 kinds of peripheral neuropathy affecting motor, sensory and autonomic systems, and numerous degenerative diseases that attack, in particular, motor neurons, have been described.
Investigation of the molecular mechanisms of peripheral nerve disease has revealed that disruption of axon transport can cause nerve dysfunction and degeneration . Molecular transport in healthy axons requires microtubules upon which cargo is carried, Kinesins, motors that generally move cargo from the neuron soma toward the axon tips, and Dynein cytoplasmic 1 heavy chain 1 (Dync1h1), which moves cargo in the opposite, retrograde direction. Disruption of any one of these components can cause disease. For example, a missense mutation that changes an amino acid within the homodimerization domain of Dync1h1 has been found in affected members of a family diagnosed with a dominant, axonal form of Charcot-Marie-Tooth (CMT) Disease  that is characterized by distal muscle weakness and atrophy, and mutations of the p150Glued subunit of Dynactin, which interacts with Dync1h1, have been identified in families with slowly progressive lower motor neuron disease and amyotrophic lateral sclerosis [3, 4]. Furthermore, dominant mutations of Dync1h1 that cause loss of proprioceptive sensory neurons have been identified in mice [5–7]. Whether disruption of Dynein-mediated molecular transport in other cellular components of peripheral nerves contributes to disease is not known.
Through a forward genetic approach using zebrafish, we found that Dync1h1 function is essential for peripheral myelination. Schwann cells were present at peripheral axons in dync1h1 mutant larvae but did not wrap them with multiple layers of membrane or express myelin genes. Genetic mosaic experiments revealed that, in addition to its role in axon transport, Dync1h1 is required in Schwann cells for efficient myelination. Deficits in peripheral myelin were improved by stimulating cAMP level, which normally mediates axon signaling necessary for myelination, raising the possibility that transduction of the myelination signal requires Dync1h1-mediated molecular transport. These insights indicate that disruption of molecular transport mechanisms might contribute to peripheral disease by affecting both axons and Schwann cells.
Dync1h1 is a subunit of the intracellular motor that transports cargoes on microtubules in a minus end-directed, or retrograde, direction. Cytoplasmic Dynein motors are crucial for a large variety of cellular functions [26, 27], including transport of molecules within axons that promote neuron survival [25, 28, 29]. Accordingly, mutations that disrupt functions of Dynein motor subunits are implicated in axon disease . Although all cells likely require cytoplasmic Dynein motor functions, in peripheral nerve disease disruption of Dynein-mediated transport has only been implicated to affect axons and not other cells that contribute to peripheral nerves, such as myelinating Schwann cells. Here we show that Dync1h1 function is essential for myelination of peripheral nerves in zebrafish.
Because cytoplasmic Dynein had not been previously implicated in peripheral myelination we were surprised to discover that dync1h1 mutant zebrafish larvae did not express peripheral mbp, a myelin gene. Schwann cells were present at peripheral axons in mutant larvae, so the lack of mbp expression did not result from absence of Schwann cells. However, the Schwann cells did not wrap axons normally, sometimes ensheathing large diameter axons in only about one turn of Schwann cell membrane, and portions of some large diameter axons were not wrapped by any Schwann cell membrane. Schwann cells of mutant larvae also failed to express pou3fI and egr2b, which encode transcription factors essential for myelin gene expression. One interpretation of these data is that, in the absence of dync1h1 function, Schwann cells fail to progress from a premyelinating, progenitor state to a fully differentiated, myelinating state.
Consistent with the known role of Dync1h1 in axon transport, peripheral axons in dync1h1 mutant larvae had hallmarks of degeneration. Additionally, mutant larvae had fewer pLLn axons than wild type and mutant axons were, on average, smaller than wild type. Therefore, the failure of myelination could be an indirect consequence of an axonal defect. We attempted to test this possibility by creating different combinations of wild-type and mutant axons and Schwann cells. When we combined wild-type Schwann cells with mutant axons, the Schwann cells expressed Mbp indicating that Dync1h1 function is not required in axons for myelination. We also combined wild-type axons with mutant Schwann cells. In the majority of cases Mbp was either not evident or only present at very low levels. However, in 25% of the cases in which we analyzed wild-type pLLn axons, the associated mutant Schwann cells appeared to express Mbp at approximately normal levels. We interpret these data to mean that, although Schwann cells lacking Dync1h1 function can express Mbp when in the presence of wild-type axons, Dync1h1 in Schwann cells increases the efficiency and strength of Mbp expression.
If Dync1h1 function within Schwann cells enhances the efficiency of myelination, then stimulation of signaling pathways necessary for myelination might rescue the defects associated with the loss of Dync1h1 function. We tested this by treating mutant larvae with forskolin, which elevates cAMP levels. Whereas rat Schwann cells cultured with axons readily myelinate them, in the absence of axons Schwann cells do not express myelin genes. This requirement for axons can be circumvented by treating Schwann cells with forskolin or cAMP analogs . Elevation of cAMP activates protein kinase A (PKA), and PKA inhibition interferes with myelination  indicating that PKA is a necessary component of the signal transduction pathway. A major target of the cAMP response is Pou3f1/Oct6, which may be activated following PKA phosphorylation of cAMP response element binding protein (CREB) and NF-κB transcription factors [32, 33]. Pou3f1/Oct6 promotes transcription of Egr2/Krox20, which encodes a transcription factor that, with Sox10, drives expression of myelin genes [35–38]. Remarkably, forskolin fully restored egr2b and mbp expression in dync1h1 mutant larvae, consistent with the possibility that Dync1h1 promotes myelination via a cAMP-dependent pathway.
How might Dync1h1 function in Schwann cells promote myelination? One possibility is that Dync1h1 is required for retrograde transport of signaling molecules from the Schwann cell-axon interface to the nucleus to regulate gene expression. Dynein-mediated retrograde transport of signaling endosomes along axons has been proposed to play key roles in neuron specification, axon outgrowth and neuron survival [25, 28]. The best known example is internalization and trafficking of Trk receptors following neurotrophin stimulation [39–42] but signaling endosomes appear to contain entire signaling complexes with components of the Ras-MAP kinase, PLCγ and PI3 kinase pathways [39–41]. Transcription factors also might be included in signaling endosomes. CREB, translated from axon-localized mRNA in response to NGF, is transported to the cell nucleus to promote neuron survival  and nuclear translocation of NF-κB depends on Dynein function in various cell types . Both CREB and NF-κB promote Schwann cell myelination [32, 33], making them candidates for factors that that are actively transported within Schwann cells following activation of a cAMP-dependent signaling pathway.
Another possibility is that Dync1h1 is required for signaling by the G protein-coupled receptor (GPCR) Gpr126, which is essential for peripheral myelination in both zebrafish and mice [45, 46]. Upon ligand binding, GPCRs stimulate activity of membrane-bound adenylyl cyclase via G proteins, to produce cAMP. Ligand binding also induces GPCR endocytosis and, although this is generally considered to be a mechanism for signal attenuation, in some circumstances cAMP production is enhanced by GPCR endocytosis [47, 48]. Dynein is important for receptor sorting in early endosomes . Together with the fact that forskolin treatment rescues the myelination defects of both gp126 and dync1h1 mutant larvae, these observations raise the possibility that Dync1h1-mediated internalization and endosomal sorting of Gpr126 is essential for its signaling activity.
Finally, Dync1h1 might facilitate ErbB2/ErbB3 receptor signaling. ErbB receptors and the related EGF receptor are endocytosed upon ligand binding and trafficked through endosomes . In fact, endocytosis and endosomal trafficking can transport ErbB2 to the nucleus of cultured cells  where, in principle, it could influence gene expression. Signaling mediated by ErbB2 and ErbB3 receptors promotes Schwann cell proliferation and migration . We found that pLLn Schwann cells of dync1h1 did not increase in number between 3 and 4 dpf as in wild type, and mutant Schwann cells also migrated more slowly and in less direct fashion. These apparently common roles open the possibility that Dync1h1-mediated endosomal trafficking influences ErbB receptor signaling.
The work reported here shows that Dync1h1, a protein that is required for axon transport and implicated in diseases that affect peripheral axons, is necessary for myelination of peripheral axons by Schwann cells in zebrafish. Our genetic mosaic data provide evidence that efficient myelination requires Dync1h1 function in both Schwann cells and axons. Our observations raise the possibility that mutations of Dync1h1 cause nerve disease, not only by causing damage to axons but also by disrupting formation or maintenance of myelin.
The University of Colorado Denver Institutional Animal Care and Use Committee approved all zebrafish studies. Zebrafish stains used include dync1h1 hi3684Tg  (Amsterdam et al., ), Tg(mnx1:GFP) ml2 , Tg(sox10:mRFP) vu234 , Tg(elavl3:Kaede) rw0130a  and AB. Embryos produced by paired matings were raised at 28.5°C, maintained in egg water or embryo medium, and staged according to hpf or dpf. Homozygous mutants for the dync1h1 hi3684Tg allele were created by pair-wise crossings of dync1h1 hi3684Tg+/− adults. Embryos younger than 48 hpf were genotyped using Hi3684_5E01: 51-AAACCTACAGGTGGGGTCTTTC-31 and Hi3684_5E02: 51-GCTACAACTACGAGCAAGTCAACC-31 as primers for PCR to amplify the mutant dync1h1 hi3684Tg allele (protocol available at Zebrafish International Resource Center). A second primer set 51-TCTTTAGCGTCGTCCTCCAG-31 and Hi3684_5E02 was used to amplify the wild-type allele.
Experiments were performed as described previously . Probes used include sox10, erbb3, mbp, egr2b and pou3f1. Following in situ hybridization, tissues were fixed with 4% paraformaldehyde, equilibrated in 70% glycerol and mounted on glass coverslips for whole-mount imaging. Images were collected using a Zeiss Axio Observer equipped with DIC optics, Retiga Exi digital color camera and Volocity software (Improvision/PerkinElmer). All images were imported into Adobe Photoshop software and image processing was limited to changes in resolution, levels, contrast, brightness and cropping.
Embryos and larvae were fixed in 4% paraformaldehyde, 8% sucrose, 1X PBS overnight at 4°C. For whole-mount immunocytochemistry, embryos and larvae were incubated in ddH2O for 4 hours, blocked in PBS-TX (1X PBS, 1% Triton X) containing 10% sheep serum and 10% BSA for 1 hour at room temperature (RT) and incubated in primary antibody diluted in antibody solution (2% sheep serum, 2% BSA, PBS-TX) for 24 hours at 4°C. After washing several hours in PBS-TX, secondary antibodies diluted in antibody solution were applied for 24 hours at 4°C. Embryos and larvae were washed in PBS-TX for at least 4 hours. For immunocytochemistry on sections, embryos and larvae were embedded in 1.5% agar/5% sucrose, frozen with 2-methyl-butane chilled by immersion in liquid nitrogen, and sectioned using a cryostat microtome (20 μm). Sections were re-hydrated with 1 × PBS and pre-blocked for 30 minutes in 2% sheep serum/BSA-1 × PBS. The sections were incubated with primary antibody overnight at 4°C, washed extensively with 1 × PBS and incubated with the appropriate fluorescent secondary antibody for 2 hours at RT. Once the secondary antibody was washed off, sections were covered with Vectashield (Vector Laboratories, Burlingame, CA, USA). Primary antibodies used were rabbit anti-Sox10 (1:200) , mouse anti-acetylated tubulin (1:5000, catalog T7451, Sigma-Aldrich, St. Louis, MO, USA) and rabbit anti-Mbp (1:200) . Secondary antibodies used were Alexa Fluor 405-, 488- and 568-conjugated goat anti-rabbit; Alexa Fluor 405- and 568-conjugated goat anti-mouse (1:200, Life Technologies, Carlsbad, CA, USA).
A t 3, 4, or 6 dpf larvae were anesthetized with Tricaine, placed on ice, and fixed in a solution of 2% glutaraldehyde, 4% paraformaldehyde and 0.1 M sodium cacodylate, pH 7.4. Fixation was accelerated using a Biowave Pro Laboratory Microwave with ColdSpot (Ted Pella, Inc, Redding, CA, USA) maintained at 15°C. Membranes were enhanced using either secondary fixation with OsO4, uranyl acetate, and imidazole , or secondary fixation using OsO4-TCH-OsO4. Electron micrographs were collected using a FEI Techai G2 BioTwin microscope, transferred to Adobe Photoshop and image processing was limited to resolution, contrast, and cropping. The axon area and the length of the longest axis of each pLLn axon was measured using Volocity software.
Embryos were dechorionated and incubated in embryo medium containing 50 μM forskolin (F6886, Sigma) dissolved in dimethyl sulfoxide (DMSO). Control embryos were treated with an equal concentration of DMSO (0.1% in embryo medium). Treatments were from 45 to 52 hpf. Following treatment, embryos were placed in embryo medium without drug, fixed at the indicated times, and processed for in situ RNA hybridization or immunohistochemistry.
Donor embryos were prepared by pair-wise crossing of either AB or Tg(sox10:mRFP) adults. Host embryos were prepared by pair-wise crossing of either dync1h1+/− or dync1h1+/−;Tg(mnx1:GFP) adults. Donor embryos and host embryos at blastula stage were positioned into the molded wells of an agarose plate. Transplantation was performed as previously described  with approximately 20 donor cells transplanted into the host embryo. Wild-type Tg(sox10:mrfp) donors were injected with Alexa488-conjugated 10,000 MW dextran (Invitrogen) at cleavage stage to enable discrimination between donor and host cells. Alternatively, wild-type Tg(elavl3:Kaede) donors were transplanted to Tg(sox10:mRFP);dync1h1 −/− hosts. At 5 dpf, genetic mosaic larvae were fixed, processed for whole-mount immunocytochemistry, and imaged as described above. Larvae with Alexa488 dextran staining in the cell bodies of pLLn neurons or in spinal cord motor neurons, which could contribute wild-type axons to peripheral nerves, were removed from the study and not analyzed for expression of Mbp.
We purchased a previously described antisense Morpholino oligonucleotide (MO) designed to block translation of dync1h1 and consisting of the sequence CGCCGCTGTCAGACATTTCCTACAC  from Gene Tools, LLC (Corvallis, OR, USA. The MO was dissolved in water and diluted prior to injections. We injected 1 to 2 nL into the yolk just below the single cell of fertilized embryos. All MO-injected embryos were raised in embryo medium at 28.5°C.
At 16 hpf, embryos were embedded in low melting point agarose and mounted in a heated chamber (28.5°C) of a motorized stage. Z-stack images were obtained every 10 minutes from 16 to 30 hpf using a PerkinElmer UltraVIEW VoX Confocal System coupled with a Zeiss Axio Observer inverted compound microscope fitted with a 40X oil immersion objective (NA = 1.3). Using Volocity software (PerkinElmer, Waltham, MA, USA), images were processed using deconvolution and contrast enhancement. Four-dimensional volumes were assembled at a speed of 6 to 8 time points/second and exported as QuickTime movie files.
cyclic adenosine monophosphate
cAMP response element binding protein
Days post fertilization
Dynein cytoplasmic 1 heavy chain 1
Central nervous system
G protein-coupled receptor
Hours post fertilization
Myelin basic protein
Motor axon exit point
Polymerase chain reaction
Protein kinase A
Posterior lateral line nerve
Peripheral nervous system
Standard error of the mean
Transmission electron microscopy
Peripheral nervous system.
We thank Jimann Shin, Danette Nicolay and members of the Appel lab for advice and discussions; Sarah Casper and Christina Kearns for assistance with experiments; Sarah Kucenas and Will Talbot for providing cDNA clones; JoAnn Buchanan for technical advice and providing the OsO4-TCH EM protocol;, Dot Dill for electron microscopy technical support, and Wendy Macklin, Laura Opincariu and Christina Kearns for comments on the manuscript. Materials were also provided by the Zebrafish International Resource Center, supported by NIH-NCRR grant P40 RR012546. The University of Colorado Denver Zebrafish Core Facility is supported by NIH grant P30 NS048154. The electron micrographs were generated in the EM core facility of the Department of Cell and Developmental Biology, supported by NIH grant P30 NS048154. This work was supported by NIH Grant RO1 NS062717 and the Gates Frontiers Fund.
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.