S.-H. Jung and S. Kim contributed equally to this work.
Patterns & Phenotypes
Visualization of myelination in GFP-transgenic zebrafish
Article first published online: 13 NOV 2009
Copyright © 2009 Wiley-Liss, Inc.
Volume 239, Issue 2, pages 592–597, February 2010
How to Cite
Jung, S.-H., Kim, S., Chung, A.-Y., Kim, H.-T., So, J.-H., Ryu, J., Park, H.-C. and Kim, C.-H. (2010), Visualization of myelination in GFP-transgenic zebrafish. Dev. Dyn., 239: 592–597. doi: 10.1002/dvdy.22166
- Issue published online: 22 JAN 2010
- Article first published online: 13 NOV 2009
- Manuscript Accepted: 6 OCT 2009
- Korea Health 21 R&D Project funded by Ministry of Health and Welfare. Grant Number: A084909
- 21C Frontier Functional Human Genome Project. Grant Number: FG09-42-1
- KOSEF. Grant Number: Vascular System Research Grant
- transgenic zebrafish;
The insulation of axons in the vertebrate nervous system by myelin is essential for efficient axonal conduction. Myelination disruption and remyelination failure can cause human diseases, such as multiple sclerosis and hereditary myelin diseases. However, despite progress in understanding myelination regulation, many important questions remain unanswered. To investigate the mechanisms underlying myelination in vivo, we generated transgenic zebrafish expressing enhanced green fluorescent protein (EGFP) under the control of the mbp promoter. This transgenic fish displayed faithful EGFP expression in oligodendrocytes and Schwann cells in embryonic and adult zebrafish. Interestingly, although myelination progressed continuously in the postembryonic central nervous system, some of the spinal cord regions were filled with unmyelinated axons even in the adult spinal cord, suggesting functional differences between myelinated and unmyelinated axons. Our results suggest that this transgenic zebrafish could be a valuable animal model to study oligodendrocyte differentiation and myelination in vivo. Developmental Dynamics 239:592–597, 2010. © 2009 Wiley-Liss, Inc.
Myelin, the multilayered glial sheath around axons, is essential for the normal function of the vertebrate nervous system. Action potential propagation along myelinated axons requires the activation of voltage-gated sodium channels only in the nodal spaces (nodes of Ranvier). This allows for much faster action potential propagation in myelinated axons compared with unmyelinated axons (Giuliodori and DiCarlo,2004). Oligodendrocytes and Schwann cells are responsible for myelin synthesis in the central (CNS) and peripheral (PNS) nervous system, respectively. The importance of myelination for nervous system function is underscored by several pathological consequences of demyelinating conditions in humans, such as multiple sclerosis and hereditary myelin diseases (Franklin,2002; Scherer and Wrabetz,2008). Although considerable progress has been achieved in understanding the regulation of myelination and demyelination, many questions remain unanswered.
Zebrafish is a powerful model system for investigating vertebrate neural development, in large part due to the rapid external development and transparency of the embryos. In particular, transgenic zebrafish that carry tissue-specific promoters driving fluorescent protein expression allow visualization of cells in living zebrafish and provide a rapid, real-time in vivo system for the analysis of gene expression patterns, organ development, and neural circuit formation in the brain (Udvadia and Linney,2003; Okamoto et al.,2008). Because the mechanisms of oligodendrocyte myelination in the CNS have been especially difficult to approach using in vitro culture, several transgenic zebrafish lines have already been generated and used to investigate the development and dynamic morphogenesis of oligodendrocytes (Shin et al.,2003; Yoshida and Macklin,2005; Kirby et al.,2006; Schebesta and Serluca,2009). In particular, in vivo time-lapse imaging with Tg(nkx2.2:mgfp) transgenic zebrafish has shown dynamic oligodendrocyte progenitor behavior, suggesting the existence of a filopodium-based surveillance mechanism that regulates myelination (Kirby et al.,2006). However, none of these transgenic lines show exclusive expression of fluorescent proteins in the oligodendrocytes of the embryonic and adult zebrafish CNS.
Myelin basic protein (MBP), the second most abundant protein in CNS myelin, is responsible for adhesion of the cytosolic surfaces of multilayered compact myelin (Brosamle and Halpern,2002; Boggs,2006; Harauz and Musse,2007). As in mammals, mbp in zebrafish is reportedly expressed specifically in the myelinating oligodendrocytes and Schwann cells in the CNS and PNS (Brosamle and Halpern,2002). To investigate the mechanisms of myelination in vivo, here we generated Tg(mbp:egfp) transgenic zebrafish expressing enhanced green fluorescent protein (EGFP) under the control of the mbp promoter. We observed faithful EGFP expression in the oligodendrocytes and Schwann cells of embryonic and adult transgenic zebrafish. Interestingly, we found that myelination progressed continuously in the postembryonic CNS. However, despite this continued myelination, some spinal cord regions were filled with unmyelinated axons, suggesting functional differences between the myelinated and unmyelinated axons in the adult spinal cord.
The mbp Promoter Drives EGFP Expression in Zebrafish Oligodendrocyte Myelination
A zebrafish orthologue of mbp is expressed in myelinating, mature oligodendrocytes and Schwann cells of zebrafish, suggesting that mbp is a useful marker for myelination (Brosamle and Halpern,2002; Kazakova et al.,2006; Li et al.,2007; Lyons et al.,2009). To visualize oligodendrocytes and their myelin processes in living zebrafish, we first isolated ∼ 2 kb of the 5′-upstream region of the zebrafish mbp by polymerase chain reaction (PCR) from the zebrafish genomic DNA template, and cloned it into the pGEM T-easy vector (Promega). The putative regulatory region of mbp was confirmed by sequence analysis and subcloned into the Tol2-EGFP vector to drive EGFP expression (Fig. 1A). The resulting plasmid, pTol2mbp:EGFP DNA, was linearized and injected into one-cell stage zebrafish embryos together with synthetic transposase mRNA. Injected embryos were raised to 4 dpf and viewed with a stereomicroscope equipped with an epifluorescence optic.
Most embryos injected with pTol2mbp::EGFP DNA had EGFP+ cells which have an oligodendrocyte morphology in the CNS. This result indicated that EGFP was accurately expressed under the control of the mbp regulatory sequences (data not shown). We raised injected fishes to adulthood, mated them with wild-type fish, and screened the progeny for EGFP expression. The obtained transgenic founder male fish were crossed with wild-type zebrafish, and F1 EGFP+ fish [designated Tg(mbp:egfp)] were raised to adulthood. EGFP was expressed in the Tg(mbp:egfp) transgenic fish in a pattern similar to that of the endogenous mbp gene in the ventral hindbrain (arrowheads) and Schwann cells of the PNS (lateral line, arrows) at 4 days postfertilization (dpf; Fig. 1B,C).
F1 progeny were further analyzed for EGFP expression by confocal microscopy. Consistent with known mbp expression patterns (Brosamle and Halpern,2002), large numbers of EGFP+ oligodendrocytes were mainly detected in the ventral hindbrain and spinal cord (Fig. 1D–F) at 5 dpf. In the spinal cord, most EGFP+ cells were found ventrally, and fewer cells were found scattered in the dorsal spinal cord (Fig. 1D,F). EGFP expression in myelinating Schwann cells was also detected but weaker than that in CNS oligodendrocytes, and EGFP expressing cells were not continuously located in the peripheral nervous system. Because we used only approximately 2 kb of the cloned 5-upstream region, we think it is possible that this region lacked some regulatory elements for the strict regulation of EGFP expression in myelinating Schwann cells (Fig. 1F, arrows).
MBP mRNA is reportedly transported from the cell body into the myelinating oligodendrocyte processes, allowing visualization of the course of the myelinated fiber tracts by in situ hybridization of the MBP transcript (Brosamle and Halpern,2002). Consistently, we detected EGFP fluorescence in the myelinated fiber tracts of the ventromedial hindbrain bundle, hindbrain commissural axons (Fig. 1G), and posterior commissure, which was symmetrically arranged in the hindbrain (Fig. 1H).
To further investigate the identity of the EGFP+ cells, we labeled transverse spinal cord sections of 5 dpf Tg(mbp:egfp) transgenic larvae with anti-MBP and anti-Sox10 antibodies specific to mbp (Lyons et al.,2005) and oligodendrocyte lineage cells including oligodendrocyte precursor cells (OPCs) and oligodendrocytes (Park et al.,2005), respectively. Most EGFP+ cells were found in the spinal cord white matter, where mature oligodendrocytes were located (Fig. 2A,B, arrows). Because the anti-MBP antibody labeled the oligodendrocyte processes but not the cell bodies, we did not observe any MBP+, EGFP+ cell bodies. However, we did detect abundant MBP+, EGFP+ oligodendrocyte processes in the ventral and, to a lesser extent, dorsolateral spinal cord (Fig. 2A, arrowheads). All EGFP+ cells located in the spinal cord white matter were Sox10+ (Fig. 2B, arrows), but the Sox10+ OPCs in the spinal cord gray matter were MBP− (Fig. 2B, arrowheads). This result indicated that EGFP was expressed in the mature oligodendrocytes but not in the OPCs.
To investigate whether the EGFP+ cells developed as neurons or radial glia, we labeled transgenic animal sections with antibody markers for neurons (anti-Hu) and radial glia (anti-BLBP). We never observed colocalization of EGFP with these markers (Fig. 2C,D), indicating that EGFP+ cells exclusively mark oligodendrocytes. We also labeled transverse hindbrain sections of 5 dpf Tg(mbp:egfp) transgenic larvae with anti-Sox10 and anti-Hu antibodies. As was the case in the spinal cord, most of the EGFP+ cells in the hindbrain were Sox10+ oligodendrocytes (Fig. 2E) and we never observed colocalization of EGFP and Hu (Fig. 2F). Therefore, consistent with endogenous mbp expression, we conclude that Tg(mbp:egfp) transgenic fish express EGFP specifically in the oligodendrocyte population, indicating that Tg(mbp:egfp) transgenic fish can be used as a valuable animal model to study oligodendrocyte differentiation and myelination in vivo.
Continuous Myelination Occurs in the Postembryonic Zebrafish Spinal Cord
OPCs reportedly generate myelin-forming oligodendrocytes in the corpus callosum and cortical gray matter of the normal adult mouse brain, indicating that axon myelination continues to occur (Rivers et al.,2008). We previously revealed that the number of oligodendrocyte lineage cells increases continuously into adulthood in the zebrafish spinal cord (Park et al.,2007), suggesting that axon myelination continues in the postembryonic zebrafish CNS. To investigate whether oligodendrocytes myelinate axons continuously in the postembryonic spinal cord, we examined whether adult Tg(mbp:egfp) transgenic fish expressed EGFP fluorescence. By 3 months postfertilization (mpf), the spinal cord of the adult Tg(mbp:egfp) transgenic fish displayed EGFP fluorescence, indicating that mbp was still expressed in the adult CNS (Fig. 3).
We next labeled transverse spinal cord sections of Tg(mbp:egfp) transgenic fishes at postembryonic stages with an anti-acetylated tubulin antibody, which labels CNS axon fibers (Fig. 4). By 10 dpf, the majority of EGFP+ cells were located in the ventral spinal cord and fewer cells were located in the dorsolateral white matter (Fig. 4A). Only a small portion of the acetylated-tubulin+ axon bundles were myelinated by EGFP+ processes in the spinal cord ventral and dorsolateral white matter, and most of the other axons were not yet myelinated (Fig. 4A). However, consistent with our previous observation that the number of olig2+ oligodendrocyte lineage cells increases continuously throughout the postembryonic stages (Park et al.,2007), the number of EGFP+ oligodendrocytes and myelinated axon bundles increased continuously into adulthood (Fig. 4A–D). In the spinal cord of 3-month-old adult zebrafish, most of the axon bundles located in the ventral and dorsolateral white matter of the spinal cord were surrounded by thick EGFP+ processes, indicating that they were highly myelinated (Fig. 4D,E). However, the lateral and dorsal top regions of the spinal cord were filled with unmyelinated axons, and only a few oligodendrocytes were observed in these regions (Fig. 4D,F). Interestingly, these nonmyelinated regions of the spinal cord were occupied by highly branched radial glial processes that were labeled by anti-BLBP antibody (Fig. 4D,G). Myelinated axons displayed well-organized bundles that were separated from each other by myelin processes (Fig. 4E). Nonmyelinated axons displayed a different morphology, with an irregular shape without bundle formation (Fig. 4F). The EGFP+ oligodendrocyte number was slightly increased in the spinal cord of 6-month-old zebrafish compared with that of 3-month-old zebrafish, but the myelination pattern was consistently maintained (Fig. 4D,H). Taken together, our data indicate that MBP-expressing oligodendrocytes are generated in the postembryonic spinal cord until at least 6 months of age, and myelination occurs continuously into adulthood except in the nonmyelinated regions occupied by radial glial processes in zebrafish.
Visualization of Oligodendrocyte Processes in Living Zebrafish
Transgenic zebrafish expressing EGFP under a tissue-specific promoter have been used to directly visualize and analyze dynamic developmental processes, including cell migration and morphogenesis. Several lines of GFP transgenic zebrafish have been generated to investigate the mechanisms of myelination and demyelination. Tg(olig2:egfp) transgenic fish express EGFP in motor neurons, oligodendrocyte lineage cells, and some of the interneuron population during development (Shin et al.,2003; Park et al.,2004), but only in oligodendrocyte lineage cells and a discrete population of radial glia in the postembryonic spinal cord (Park et al.,2007). Although Tg(olig2:egfp) transgenic fish show faithful EGFP expression in the oligodendrocyte lineage cells in the postembryonic spinal cord, this model is somewhat difficult to use to study myelination because the reporter is expressed in several different cell types during development. Tg(olig1:egfp) and Tg(plp:egfp) transgenic zebrafish express EGFP in oligodendrocyte lineage cells, but its expression in adult oligodendrocytes and their processes has not been determined (Yoshida and Macklin,2005; Schebesta and Serluca,2009).
Compared with these transgenic lines, our Tg(mbp:egfp) transgenic fish have several advantages for investigating the myelination and demyelination processes. Tg(mbp:egfp) transgenic fish expressed EGFP exclusively in oligodendrocytes and Schwann cells in the CNS and PNS, and oligodendrocyte-specific EGF expression was maintained in adult zebrafish. Oligodendrocyte processes surrounding the axon bundles were clearly visualized throughout the postembryonic stages. Our Tg(mbp:egfp) transgenic fish is, therefore, likely to serve as a valuable tool for the direct identification of myelination mutants in mutagenesis screens and for the in vivo recording of myelination and demyelination processes using time-lapse confocal microscopy.
Continuous Axon Myelination in the Postembryonic Spinal Cord
A study of myelination in the corpus callosum of the mouse brain previously revealed that myelination continues from birth until at least P240, at which time ∼70% of axons are still not fully myelinated (Sturrock,1980). A recent study using Pdgfra-creERt2/Rosa26-YFP double transgenic mice further revealed that adult OPCs generate mature, myelinating oligodendrocytes until at least 8 months of age. Furthermore, more than 20% of the oligodendrocytes were generated after 7 weeks of age in the adult mouse corpus callosum (Rivers et al.,2008). Volumetric imaging studies indicated that it takes at least 4 decades before myelination is complete in human brain (de Graaf-Peters and Hadders-Algra,2006). Although the function of these late-myelinating axons has not been determined, these previous data indicate that axon myelination continues in the postembryonic CNS. Our recent study in zebrafish revealed that the number of olig2+ oligodendrocyte lineage cells increased continuously into adulthood, suggesting that newly generated oligodendrocytes myelinate axons continuously in the postembryonic spinal cord (Park et al.,2007).
Consistent with these previous reports, our data here directly reveal that axon myelination by MBP+ oligodendrocytes occurs continuously in the postembryonic spinal cord (Fig. 4), raising the question of the function of late-myelinating axons. However, as in the corpus callosum of the mouse brain (Sturrock,1980), approximately 40% of the axons were unmyelinated in the spinal cord of 3-month-old zebrafish (Fig. 4). We also observed that there were two populations of axons in the spinal cord of adult zebrafish. Axons in the ventral and dorsolateral spinal cord white matter were highly myelinated, while other regions of the spinal cord white matter, occupied by highly branched radial glial processes, were filled with nonmyelinating axons. The occupation of the nonmyelinated regions by radial glial fibers suggests that there are functional differences between myelinated and nonmyelinated axons in the adult spinal cord. Previously, it has been shown that the density of the radial glial processes is highest in the dorsolateral fasciculus (DLF), which has unmyelinated axon fascicules. This is in contrast to the dorsal funiculus in the adult frog spinal cord (DF), which normally contains very large myelinated dorsal root axons (Miller and Liuzzi,1986). Interestingly, from the study of dorsal root axonal regeneration in the adult frog spinal cord, it has been shown that both sensory and motor axons are regenerated effectively within the DLF. By contrast, fewer sensory and motor axons are regenerated in the DF, suggesting that unmyelinated axons and radial glial processes provide the cellular environment for the efficient regeneration of damaged axons in the adult spinal cord (Liuzzi and Lasek,1986). These reports also suggest the possibility that unmyelianted axon bundles and radial glial processes observed in our study may provide a more permissive cellular environment for axonal regeneration after injury in the adult zebrafish spinal cord.
Zebrafish mbp regulatory region were PCR amplified from the zebrafish genomic DNA template with the following primer set: forward primer, 5′-GTC GAC CAG ATG CTG AGA TGT GAC TAC TGC AAA TGA-3′ and reverse primer, 5′-GGA TCC GTT GAT CTG TTC AGT GGT CTA CAG TCT GGA-3′. Both primers were designed with the sequences obtained from ensemble search with mbp cDNA sequence (AY860977). Amplified DNA fragment contains approximately 2 kb of mbp regulatory region and 80 bp of 1st exon. Amplified DNA was subcloned into the pGEM T-easy vector (Promega) for the sequence analysis, and then digested with SalI/BamHI and inserted into the XhoI/BamHI site of the Tol2-GFP vector, from which the CMV promoter was removed.
Generation of Tg(mbp:egfp) Zebrafish
Fertilized eggs were injected with 1 nl of a mixture containing 25 ng/μl of the Tol2 mbpP-GFP DNA and 25 ng/μl of the transposase mRNA (Kotani et al.,2006). Injected embryos were raised to adult fish and crossed with wild-type. F1 embryos were screened for GFP fluorescence under the fluorescent dissecting microscope MZ16FA (Leica) at 4 dpf.
In Situ RNA Hybridization
The 5- to 180-dpf fishes were anesthetized until movement had ceased. The brains, internal organs, and some of the trunk muscles were dissected in embryo medium (EM) (15 mM NaCl, 0.5 mM KCl, 1 mM CaCl2, 1 mM MgSO4, 0.15 mM KH2PO4, 0.05 mM NH2PO4, 0.7 mM NaHCO3) and fixed in 4% paraformaldehyde overnight.
For immunohistochemistry, we used the following primary antibodies: rabbit anti-MBP (1:100; Lyons et al.,2005), rabbit anti-Sox10 (1:1,000; Park et al.,2005), mouse anti-HuC/D (16A11, 1:20, Molecular Probes), mouse anti-BLBP (1:100; Kim et al.,2008), and anti-acetylated Tubulin (1:100, Sigma). For fluorescent detection of antibody labeling, we used Alexa Fluor 488 and Alexa Fluor 568 goat anti-mouse or goat anti-rabbit conjugate (1:500, Molecular Probes). Fluorescence images were collected using a Zeiss LSM 510 laser scanning confocal microscope.
We thank Dr. K. Kawakami for providing of Tol2 transposon vector, and Korea Zebrafish Organogenesis Mutant Bank (ZOMB) for providing zebrafish lines. This work was supported by a grant (code A084909) from the Korea Health 21 R&D Project funded by Ministry of Health & Welfare, the Republic of Korea, and by a grant (FG09-42-1) from 21C Frontier Functional Human Genome Project and by Vascular System Research Grant from KOSEF.
- 2006. Myelin basic protein: a multifunctional protein. Cell Mol Life Sci 63: 1945–1961. .
- 2002. Characterization of myelination in the developing zebrafish. Glia 39: 47–57. , .
- 2006. Ontogeny of the human central nervous system: what is happening when? Early Hum Dev 82: 257–266. , .
- 2002. Why does remyelination fail in multiple sclerosis? Nat Rev Neurosci 3: 705–714. .
- 2004. Myelinated vs. unmyelinated nerve conduction: a novel way of understanding the mechanisms. Adv Physiol Educ 28: 80–81. , .
- 2007. A tale of two citrullines--structural and functional aspects of myelin basic protein delamination in health and disease. Neurochem Res 32: 137–158. , .
- 2000. Multicolor whole-mount in situ hybridization. Methods Mol Biol 137: 139–148. , .
- 2006. A screen for mutations in zebrafish that affect myelin gene expression in Schwann cells and oligodendrocytes. Dev Biol 297: 1–13. , , , , , , .
- 2008. Notch-regulated perineurium development from zebrafish spinal cord. Neurosci Lett 448: 240–244. , , , , , , .
- 2006. In vivo time-lapse imaging shows dynamic oligodendrocyte progenitor behavior during zebrafish development. Nat Neurosci 9: 1506–1511. , , , , , , .
- 2006. Transposon-mediated gene trapping in zebrafish. Methods 39: 199–206. , , , .
- 2007. Olig1 and Sox10 interact synergistically to drive myelin basic protein transcription in oligodendrocytes. J Neurosci 27: 14375–14382. , , , .
- 1986. Dorsal root axonal regeneration in the adult frog spinal cord. A model of vertebrate CNS regeneration. Neurochem Pathol 5: 237–253. , .
- 2005. erbb3 and erbb2 are essential for Schwann cell migration and myelination in zebrafish. Curr Biol 15: 513–524. , , , , , , , , .
- 2009. Kif1b is essential for mRNA localization in oligodendrocytes and development of myelinated axons. Nat Genet 41: 854–858. , , , .
- 1986. Regional specialization of the radial glial cells of the adult frog spinal cord. J Neurocytol 15: 187–196. , .
- 2008. Transgenic technology for visualization and manipulation of the neural circuits controlling behavior in zebrafish. Dev Growth Differ 50( suppl 1): S167–S175. , , .
- 2004. Spatial and temporal regulation of ventral spinal cord precursor specification by Hedgehog signaling. Development 131: 5959–5969. , , .
- 2005. Oligodendrocyte specification in zebrafish requires notch-regulated cyclin-dependent kinase inhibitor function. J Neurosci 25: 6836–6844. , , , .
- 2007. An olig2 reporter gene marks oligodendrocyte precursors in the postembryonic spinal cord of zebrafish. Dev Dyn 236: 3402–3407. , , , .
- 2008. PDGFRA/NG2 glia generate myelinating oligodendrocytes and piriform projection neurons in adult mice. Nat Neurosci 11: 1392–1401. , , , , , , , .
- 2009. olig1 Expression identifies developing oligodendrocytes in zebrafish and requires hedgehog and notch signaling. Dev Dyn 238: 887–898. , .
- 2008. Molecular mechanisms of inherited demyelinating neuropathies. Glia 56: 1578–1589. , .
- 2003. Neural cell fate analysis in zebrafish using olig2 BAC transgenics. Methods Cell Sci 25: 7–14. , , , , .
- 1980. Myelination of the mouse corpus callosum. Neuropathol Appl Neurobiol 6: 415–420. .
- 2003. Windows into development: historic, current, and future perspectives on transgenic zebrafish. Dev Biol 256: 1–17. , .
- 2005. Oligodendrocyte development and myelination in GFP-transgenic zebrafish. J Neurosci Res 81: 1–8. , .