Function of RTN4/Nogo during development and as transgene in adult zebrafish

(1)

Function of RTN4/Nogo during development and as transgene in adult zebrafish

Dissertation

zur Erlangung des akademischen Grades eines Doktors der Naturwissenschaften (Dr. rer. nat.)

an der Universität Konstanz,

Mathematisch-‐naturwissenschaftliche Sektion Fachbereich Biologie

vorgelegt von Manuel Alejandro Pinzón-‐Olejua

(2)

Acknowledgements

I would like to express my deep gratitude to Prof. Dr. Claudia Stuermer for her patient guidance and enthusiastic encouragement during the last years. Her willingness to give her time so generously has been very much appreciated. I would like to express my very great appreciation to Dr. Edward Malaga-‐Trillo for his valuable and constructive suggestions during the planning and development of this research work. My special thanks are extended to my colleagues from the Stuermer-‐Lab for their advice and nice discussions. I wish to thank various people for their contribution to this project; Annette Yvonne Loss for the zebrafish care; Marianne Wiechers and Ulrike Binkle for the generation of the Rtn4a and Rtn4b-‐specific antibodies.

I would also like to thank the members of the following institutes for enabling me to visit their labs and learn very valuable techniques employed in this project:

Daniele Soroldoni and Andrew Oates (Max Planck institute for Cell biology and Genetics -‐ Dresden) for their advice in the Bacterial Artificial Chromosomes recombineering;

Stefan Hans and Michael Brand (Center for Regenerative Therapies -‐ Dresden) for their support in the generation of the Cre-‐Lox transgenic lines.

Finally, I wish to thank Konstanza, Bernarda, Georg and Manuel for their support and encouragement.

(5)

Introduction

Introduction

The question why axons in the central nervous system (CNS) of mammals are unable to spontaneously regenerate is of fundamental clinical importance.

Various efforts have been undertaken to develop a cure for paraplegic patients and other CNS injuries. Even though such a cure is still not available, progress has been made in understanding the reasons underlying the poor regenerative capabilities of the mammalian CNS.

Ramon y Cajal was the first who realized that axon regeneration in the adult mammalian CNS is inhibited by factors in the environment of severed fiber tracts. Many years later, Alberto Aguayo showed that nerve grafts from the peripheral nervous system (PNS), when transplanted into the CNS, can promote axon regrowth in the injured mammalian spinal cord and optic nerve (Kandel et al., 2013). It was subsequently discovered that CNS myelin and the myelin-‐

producing cells, the oligodendrocytes, are the major inhibitory elements of the CNS (Carbonetto and Muller, 1982, Schwab, 2010). Experiments using fractionations of the myelin proteins combined with functional assays finally led to the identification of Nogo-‐A, a protein enriched in CNS myelin, whose role as a strong inhibitor of neurite development and regeneration is nowadays well-‐

known (Chen et al., 2000, GrandPre et al., 2000, Prinjha et al., 2000, Schwab, 2010).

Nogo-‐A is the longest of the three Nogo-‐gene transcripts, comprising Nogo-‐A, -‐B and –C, and belongs to the reticulon (RTN) gene family consisting of RTN-‐1, -‐2, -‐3 and -‐4 (Nogo). RTN proteins including Nogo-‐A, are predominantly structural proteins of the endoplasmic reticulum (ER) (Voeltz et al., 2006). Less than 5% of the Nogo-‐A protein reaches the plasma membrane but this small fraction seems sufficient to exert inhibition on contacting cells. There are two inhibitory domains in Nogo-‐A: The Nogo-‐66 loop within the evolutionarily conserved reticulon-‐homology domain (RHD) and the Nogo-‐A-‐specific delta20 domain (Figure 1). The respective neuronal receptors are NgR1 in complex with Lingo, Troy and p75 (for Nogo66; Figure 1) and the G-‐protein coupled sphingolipid

(6)

receptor S1PR2 (for Nogo-‐A; Figure 1), which also seems to be part of a receptor complex (Kempf et al., 2014, Schmandke et al., 2014).

Figure 1. Nogo-‐A receptors and signaling. The main signaling pathways targeted by Nogo-‐A.

Nogo-‐A-‐ Δ 20 and Nogo-‐66 both are known to activate the RhoA/ROCK pathway, resulting in depolymerization or attenuated reorganization of the actin cytoskeleton, increased actomyosin contraction, as well as reduced stabilization of microtubules. Furthermore, Nogo-‐A-‐ Δ 20 was demonstrated to inactivate CREB on pincher-‐mediated endocytosis, possibly affecting gene expression. Additionally, the Nogo-‐A-‐ Δ 20 domain is implicated in negative regulation of integrin activation. Two different receptors have been described for Nogo-‐66 (PirB and the NgR1-‐Lingo1-‐

p75/Troy-‐receptor complex); the functional receptor(s) for Nogo-‐A-‐ Δ 20 remain to be characterized. Furthermore, it is still unclear whether and if so what part of Nogo-‐A is involved in modulation of intracellular calcium transients as suggested by earlier studies using purified myelin proteins with neurite outgrowth inhibitory activity. Finally, given the fact that Nogo-‐A-‐ Δ 20 signaling was demonstrated to be highly dependent on endocytosis in vitro, it will be crucial to investigate the existence of soluble Nogo-‐A fragments in vivo and, if so, identify underlying mechanisms and/or Nogo-‐A-‐specific proteases. Taken from (Schmandke et al., 2014).

Meanwhile, many additional inhibitory proteins were discovered in the

(7)

Introduction

and leptomeningeal cells (fibroblasts) when fiber tracts are injured (Silver and Miller, 2004). But Nogo-‐A is one of or perhaps the most potent inhibitor as was demonstrated by the fact that antibodies against the Nogo-‐A specific delta20 region allow axon regeneration in vivo when applied to the site of injury. This was shown in rodents and primates, and clinical trials are underway in human patients (Schwab, 2010).

Axon regeneration in the fish CNS and the role of Nogo

In contrast to mammals injured axons in the CNS of fish are able to regenerate (Attardi and Sperry, 1963). For instance, transection of the optic nerve in goldfish and zebrafish leads to retinal ganglion cell axon regeneration and ultimately to the recovery of vision (Gaze, 1970, Stuermer et al., 1992). This suggested that fish CNS myelin has either no inhibitors, which prevent axon regrowth. Or axons regrow despite of inhibitors which is possible when the axotomized neurons possess specific neuron-‐intrinsic properties, which empowers them to cross through inhibitory terrain (Stuermer et al., 1992).

To address this issue, experiments were performed in which the substrate properties of fish CNS myelin were compared with mammalian CNS myelin.

When offered to growing axons, the mammalian CNS myelin caused growth cone collapse but the fish CNS myelin permitted growth cone elongation (Bastmeyer et al., 1991, Wanner et al., 1995). Combined evidence from several studies indicated that fish CNS myelin / myelin proteins are growth permissive or at least by far less inhibitory than their mammalian counterparts (Diekmann et al., 2005).

These assays also showed that axons from fish and mammals responded in a similar manner: axons crossed fish CNS myelin but were inhibited by mammalian CNS myelin (Bastmeyer et al., 1991; Wanner et al., 2005; Stuermer et al., unpublished). This suggests, among others, that fish axons should have a receptor for Nogo-‐A, which, in turn, predicts that they possess the corresponding ligand, i.e., a Nogo-‐A homologue.

(8)

Whether the zebrafish S1PR2 gene serves as a Nogo receptor has not yet been addressed. However, studies searching for a zebrafish Nogo-‐A homologue have shown that the rtn4 gene was duplicated giving rise to rtn4a and rtn4b (Figure 2) (Shypitsyna et al., 2011). Zebrafish rtn4a is short and lacks the entire N-‐terminal region, which comprises Nogo-‐A delta 20 in mammals (Diekmann et al., 2005).

Zebrafish rtn4b, however, has a longer N-‐terminal domain that comprises a region homologous to the inhibitory Nogo-‐A specific domain. Functional assays comparing the effects of recombinantly expressed Nogo-‐A delta 20 (or its slightly longer version called M1-‐4) and the corresponding zebrafish rtn4b-‐derived M1-‐4 peptides are underway. In one of these assays (stripe assay), the previously discovered difference in substrate properties between the fish and mammalian Nogo peptides was confirmed: axons freely crossed the fish M1-‐4 whereas they avoided the mammalian M1-‐4 peptide lanes (Stuermer et al., unpublished).

Figure 2 Schematic representation of the human RTN4 gene and its zebrafish paralogues.

All three major isoforms encoded in the RTN4 gene in humans (RTN4A, RTN4B and RTN4C) possess the reticulon homology domain (RHD), which includes the Nogo66 domain. The N-‐

terminal region of RTN4A contains the Nogo-‐A-‐specific domain (yellow) and the neurite growth inhibitory Delta 20 (Δ20) stretch. The diagnostic M1 to M4 motifs are indicated in red. The zebrafish has two RTN4 paralogues: rtn4a and rtn4b. Rtn4a is produced in three different isoforms (Rtn4-‐l (blue), Rtn4-‐m (green) and Rtn4-‐n (orange)) with the same C-‐terminal RHD.

Rtn4b also contains the M1 to M4 N-‐terminal motifs (red) and presents a distinct RHD.

Given that there is a difference in substrate properties between zebrafish and mammalian M1-‐4, and since zebrafish axons recognize mammalian Nogo-‐A, we wondered whether axon regeneration in zebrafish might be inhibited if the mammalian Nogo-‐A would be expressed in the fish CNS. This can be addressed if one succeeds in expressing the inhibitory region of mammalian Nogo-‐A in transgenic zebrafish. This transgene expression should be directed to zebrafish

(9)

Introduction

oligodendrocytes so that Nogo-‐A should be found in the glial cells and in CNS myelin.

In this thesis, Nogo-‐A transgenic zebrafish were generated which are now ready for tests on axonal regeneration. This is discussed in Part II. Part I of this thesis explores the role of rtn4a and rtn4b in the developing zebrafish brain and predominantly its intracellular function.

(10)

Specific Introduction to Part I

Expression and function of rtn4a and rtn4b in zebrafish embryos

From the publication: Pinzon-‐Olejua et al: Essential roles of zebrafish rtn4/Nogo paralogues in embryonic development. Neural Development 2014 9:8.

In addition to its activity as an inhibitor of axon growth in the adult CNS, recent studies in mice have uncovered functional roles of Nogo-‐A in neuronal development and cortical plasticity. For instance, Nogo-‐A has been demonstrated to be present in migrating neuroblasts and immature neurons in the neural tube, as well as on radially and tangentially migrating neurons of the developing cortex, affecting their motility (Mingorance-‐Le Meur et al., 2007, Mathis et al., 2010). In other studies, Nogo-‐A was found to contribute to long-‐term potentiation (LTP) in the hippocampus, ocular dominance column formation in the visual system, and size control of postsynaptic densities in cerebellar neurons (McGee et al., 2005, Petrinovic et al., 2013). Collectively, these findings suggest that Nogo-‐A negatively regulates neural plasticity in the mammalian brain (Schwab, 2010). These defects, however, do apparently not interfere with fertility and viability of the Nogo-‐A-‐knockout mouse, which shows no striking phenotype at birth (Kim et al., 2003, Simonen et al., 2003).

Much less is known concerning the role of the RTNs, especially RTN4/Nogo-‐A, in the neurodevelopment of non-‐mammalian species. In fish, such analysis is of great interest because axons regenerate successfully in the teleost CNS (Attardi and Sperry, 1963, Stuermer and Easter, 1984, Becker et al., 1997, Abdesselem et al., 2009) and because neuronal projections in the peripheral nervous system of the embryo seem to develop abnormally when rtn4a is downregulated (Brosamle and Halpern, 2009).

It has been recognized that zebrafish possess two rtn4 paralogues, rtn4a and rtn4b (Figure 2) (Diekmann et al., 2005, Shypitsyna et al., 2011). Both proteins have a conserved RHD, the hallmark of this gene family (Oertle et al., 2003), but very different N-‐terminal regions (Shypitsyna et al., 2011). A comparative study

(11)

Expression and function of rtn4a and rtn4b in zebrafish embryos Introduction

revealed that, in contrast to its mammalian counterpart, the Nogo-‐66 region in the RHD of zebrafish Rtn4, upon binding to either the zebrafish or the mouse Nogo receptor (NgR), promotes neuronal growth (Abdesselem et al., 2009). The N terminus of zebrafish Rtn4a bears no resemblance, in sequence or in length, to that of mammalian RTN4, but four short motifs—termed M1 to M4 (Figure 2)—

were found to be conserved between the N terminus of Rtn4b and the inhibitory Nogo-‐A-‐specific Delta 20 domain of mammalian RTN4 (Shypitsyna et al., 2011).

To elucidate the function of the zebrafish Rtn4b N terminus and its M1 to M4 motifs, ongoing studies in our laboratory aim to investigate the expression pattern of Rtn4b in the adult CNS and its potential ability to inhibit axon growth.

Previous work by Brösamle and Halpern (Brosamle and Halpern, 2009) addressed the role of zebrafish rtn4a using morpholino (MO)-‐based knockdown strategies and showed that downregulation of the shortest splice form, rtn4a-‐γ (Diekmann et al., 2005) (hereinafter referred to as rtn4a-‐n), led to misguidance of the posterior lateral line nerve and disorder of cranial nerves in 2-‐ and 3-‐day-‐

old embryos. Their work further suggested that Nogo–NgR interactions may contribute to axon guidance and to development of the zebrafish PNS by channeling axons through inhibitory terrain.

Our goal in the present study was to examine the expression and function of rtn4b in zebrafish embryos, particularly in light of the similarity between the N-‐

terminal region and that of mammalian Nogo-‐A/RTN4A (Figure 2). In addition we comparatively analyzed the expression of zebrafish rtn4a and its role in embryogenesis. Interestingly, and in contrast to the Nogo-‐knockout mouse, our results reveal morphological defects in the formation of the spinal cord and brain. In rtn4b-‐knockdown embryos, furthermore, the pectoral fin became absent or reduced and the lower jaw was often lost. Together, the neuronal and non-‐neuronal defects in rtn4b morphants were stronger than those in rtn4a, ultimately impairing larval motility and causing death.

(12)

Specific Introduction to Part II

Transgenesis to study the function of genes in zebrafish

A transgene is a DNA constructed piece that integrates in the genome of a recipient organism. To achieve stable transgene transmission, the DNA constructed piece should integrate into germ cells so that subsequent generations inherit the transgene. Until recently, the most common method to generate transgenic zebrafish remained the injection of linearized plasmid DNA into the cytoplasm of one cell-‐stage embryos, which could be randomly integrated into the genome. The major disadvantage is the concatameric DNA integration that in some cases can reach up to 2000 copies (Stuart et al., 1988, Iyengar et al., 1996). Nowadays, Isce-‐I meganuclease or transposon-‐mediated transgenesis have facilitated the generation of zebrafish transgenic lines by increasing the integration and transmission rates of the transgenes (Kawakami, 2007) and in the case of transposase-‐mediated transgenesis integration of foreign DNA sequences results in a single copy insertion (Figure 3).

Figure 3. Transgenesis in zebrafish. (A) The structures of the Tol2 transposable element and the minimal Tol2 vector. At the top of the illustration is the 4,682 base pair (bp), full-‐

length Tol2. RNA transcribed from Tol2 that encodes a transposase protein is shown by lines (exons) and dotted lines (introns). Black boxes and gray boxes represent coding regions and untranslated regions, respectively. Black arrowheads in boxes at both ends indicate 12 bp terminal inverted repeats (TIRs). The lower part of the figure shows the minimal Tol2 vector with the green fluorescent protein (GFP) expression cassette. The minimal Tol2 vector contains 200 and 150 bp of DNA from the left and right ends.

The transposon vector can carry a DNA fragment, for example the GFP expression cassette in this figure, between these sequences.

(B) The synthetic transposase mRNA and a transposon donor plasmid containing a Tol2 construct with a promoter and the gene encoding green fluorescent protein (GFP) are co-‐injected into zebrafish fertilized eggs. The Tol2 construct is excised from the donor plasmid and integrated into the genome. Tol2 insertions created in germ cells are transmitted

(13)

Generation of Nogo-‐A transgenic zebrafish Introduction

fish are mosaic, and, by crossing the injected fish (founder) with wild-‐type fish, nontransgenic fish and transgenic fish heterozygous for the Tol2 insertion are obtained. In this figure, the promoter is tentatively defined as a spinal cord specific enhancer/promoter and the spinal cord of the embryo is depicted in green. Taken from Kawakami, 2007.

The Cre/lox system in zebrafish

Spatio-‐temporal transgene regulation by transgenic recombinases is a new tool for genetic research in zebrafish. Engineered transgenes permit testing of molecular mechanism such as the function of a specific RNA or protein, by expressing a desired gene (with a detectable marker) at different developmental stages and in specific cell types (Figure 4).

The most applied strategy in mouse is the use of site-‐specific recombinases such as Cre (creates recombination) and Flp (flippase). Cre promotes strand exchanges between two 34 bp loxP target sites without any additional cofactors.

LoxP sites contain two 13 bp repeats flanking an 8 bp asymmetric spacer sequence that confers directionality. Head to head orientation causes inversion of the DNA between the two sites, whereas head to tail orientation causes the irreversible excision of the DNA sequence.

Figure 4. Cre-‐mediated recombination in the red-‐to-‐green reporter line. Scheme of the recombination event. Ligand-‐dependent Cre-‐mediated recombination in cells of the red-‐to-‐ green reporter line. The chimeric CreER^T2 recombinase is retained in the cytoplasm in the absence of the ligand. After administration of tamoxifen (TAM), which is converted to the active ligand 4-‐

(14)

OHT, CreER^T2 translocates to the nucleus, where it catalyzes the recombination event. As a readout the floxed DsRed is excised and the EGFP together with the gene of interest (GOI) are expressed.

Cre-‐recombinase controlled Cre/lox site-‐specific recombination have attracted interest in the zebrafish field. Although its conventional use is a powerful technology in mouse, initial attempts to establish the Cre/lox system in other organisms showed its functionality in zebrafish (Langenau et al., 2005, Thummel et al., 2005). Nevertheless the recombination efficiency was very low (le 2007, Feng, 2007). Very recently Hans et al (2009) by using transposon-‐mediated transgenesis showed that Cre is in fact very efficient in zebrafish, indicating that the method by which transgenic zebrafish are generated might be critical (Hans et al., 2009, Hans et al., 2011). It is now believed that unambiguous site-‐specific recombination requires single copy insertions, which can only be achieved by transposon-‐mediated transgenesis or pseudotyped retrovirus. Due to the observation that localization of proteins can be controlled, chimeric Cre recombinases, fused to a ligand-‐binding domain of a steroid receptor hormone, were shown to offer temporal control of Cre-‐mediated recombination. Fusion of Cre to the mutated human ligand-‐binding domain of the estrogen receptor (CreER^T2), retains the protein in the cytoplasm, but after administration of tamoxifen (TAM), or its active metabolite 4-‐hydroxy-‐tamoxifen, the complex is internalized into the nucleus and is able to recombine genomic DNA.

Myelin-‐targeted expression of rat Nogo-‐A in transgenic zebrafish

Myelination in zebrafish is carried out by Schwann cells (PNS) and oligodendrocytes (CNS), as in all other vertebrate classes (Brosamle and Halpern, 2002). There are several myelin-‐specific proteins which function in structuring the myelin sheath around the axons. One of them is myelin basic protein (Mbp). It would thus be optimal to express Nogo-‐A under the mbp promoter so that the transgene specifically emerges in oligodendrocytes and CNS myelin.

Therefore, our goal in the present study was to target the expression of the mammalian Nogo-‐A to zebrafish oligodendrocytes. In order to achieve site-‐

(15)

Generation of Nogo-‐A transgenic zebrafish Introduction

specific recombination we use the Cre/lox system, which has been recently established in zebrafish (Hans et al., 2009). The advantage of this approach is that one generates separately a tissue-‐specific Cre driver line and an effector line, which contains the DsRed reporter silencing the expression of the gene of interest and the EGFP reporter (Figure 4). In the absence of Cre the hsp70 promoter drives the expression of DsRed2 but changes to EGFP after successful Cre-‐mediated recombination. Following this background, we aimed to generate a CreER^T2 driver line under the control of the mbp promoter as well as zebrafish rtn4a and rat Nogo-‐A effector lines under the control of the temperature–

inducible heat shock (hsp70) promoter. By crossing the CreER^T2 driver line with the rat-‐Nogo-‐A effector line the recombination takes place only in the mbp specific domain (oligodendrocytes and Schwann cells). Moreover, temporal control of recombination can be achieved by using the ligand-‐inducible CreER^T2. Site-‐specific recombination only occurs upon administration of the drug tamoxifen (TAM) or its active metabolite, 4-‐hydroxy-‐tamoxifen (4-‐OHT). In addition, because the hsp70 promoter was used to generate the effector lines, giving heat shocks to the fish can temporally control the expression of the transgene.

For the generation of CreER^T2 driver lines we inserted the mbp promoter fragment into a Tol2 plasmid carrying a mCherry-‐T2A-‐CreER^T2. The effector lines were established by inserting the rat Nogo-‐A into an existing Tol2 vector that carries all required elements, such as the Kozak sequences important for the initiation of the translation process, as well as the polyadenilation signal, which is necessary for nuclear export, translation and stability of mRNA.

Stable transgenic lines were successfully generated for this study using Tol2-‐

mediated transgenesis. Coinjection of Tol2 vectors in combination with Tol2 transposase mRNA, resulted in random insertion of the Tol2-‐flanked transgene into the zebrafish genome. Founders were screened for F1 transmission. Three positive founders, for each construct, were chosen to generate stable lines.

The use of appropriate controls in these experiments will be crucial. For instance, studying axon growth and regeneration in the zf-‐mbp:rat-‐Nogo-‐A line will tell us whether rat Nogo-‐A is inhibitory under similar conditions as in

(16)

mammals (i.e. when Nogo-‐A is expressed in myelinating cells of the CNS). In addition, studying axon regeneration in the zf-‐mbp:zf-‐rtn4a line will help us confirm the non-‐inhibitory properties of zebrafish rtn4 and exclude the possibility that the inhibition eventually observed in the zf-‐mbp:rat-‐Nogo-‐A line is somehow due to the choice of promoter used, rather than to rat Nogo-‐A itself.

Finally, a transgenic line expressing only the EGFP protein was considered in order to control for unspecific effects by the fluorescent protein, and also as a future experimental tool to visualize myelination, or even remyelination, in real time.

(17)

Aims of the work

Aims of the work

In sum, the work described in this thesis aims to examine the expression and function of rtn4a and rtn4b in zebrafish embryogenesis, particularly in light of the similarity between the zebrafish Rtn4b N-‐terminal region and that of mammalian Nogo-‐A/RTN4A, and to generate transgenic fish expressing mammalian Nogo-‐A in order to analyze the influence of the expression of the protein and to functionally characterize the effect of Nogo-‐A on axon regeneration.

Thus, we aim to:

Part I: Explore the expected and yet unidentified functions of Nogo-‐A paralogues in fish development, not related to axonal growth inhibition.

Part II: Generate stable transgenic fish lines expressing rat-‐Nogo-‐A in oligodendrocytes and Schwann cells by Cre/lox recombination and transposon mediated transgenesis. This fish will be suitable to analyze their axon regeneration potential after an optic nerve lesion.

(18)

Material and Methods

Part I: Expression and function of rtn4a and rtn4b in zebrafish

From the publication: Pinzon-‐Olejua et al: Essential roles of zebrafish rtn4/Nogo paralogues in embryonic development. Neural Development 2014 9:8.

Zebrafish husbandry

Zebrafish (Danio rerio) were maintained at 28°C under a 14-‐hour light, 10-‐hour dark cycle (Westerfield, 1995). Developmental stages are indicated based on those described by Kimmel et al. (Kimmel et al., 1995) and in hours and days postfertilization (hpf and dpf, respectively). Some embryos were raised in fish water containing 0.003% 1-‐phenyl 2-‐thiourea to prevent pigmentation (Karlsson et al., 2001). A zebrafish reporter line expressing GFP under the control of the sonic hedgehog gene promoter tg(shh:gfp) was obtained from Max-‐Planck-‐

Institute Developmental Biology (Tübingen, Germany). tg(hb9:gfp)-‐transgenic zebrafish expressing GFP in motor axons were provided by D Meyer (University of Innsbruck, Austria). tg(Isl1:gfp) zebrafish expressing GFP in cranial motor neurons were provided by S Higashijima (Okazaki Institute for Integrative Bioscience, Higashiyama, Japan) and tg(brn3c:mgfp) zebrafish expressing membrane-‐targeted GFP in retinal axons were provided by H Baier (University of California, San Francisco, USA).

Whole-‐mount in situ hybridization

Whole-‐mount in situ hybridization was performed as described previously (Westerfield, 1995). We cloned 1.3 kb of rtn4a-‐l, 1 kb of rtn4a-‐m and 0.9 kb of rtn4a-‐n (including the full open reading frames (ORFs) and 393 bp from the 3′UTR and 1.5 kb from the rtn4b N terminus, including the M1 to M4 motifs) in pCRII TOPO (Invitrogen, Carlsbad, CA, USA) and used them as templates for the synthesis of two independent RNA in situ hybridization probes with the DIG RNA Labeling Kit (Roche Applied Science, Penzberg, Germany). Transcription

(19)

Material and Methods

patterns were visualized using an Axioplan 2 compound microscope (Carl Zeiss Microscopy, Oberkochen, Germany) using Nomarski (differential interference contrast) optics, photographed with a Zeiss Color Axiocam and further processed using Adobe Photoshop 12.0 software (Adobe Systems, San Jose, CA, USA).

Cloning full-‐length rtn4a and rtn4b cDNAs

The rtn4a full coding sequence was amplified by RT-‐PCR from 1-‐dpf zebrafish embryo total RNA with the following primers: forward rtn4a-‐fw 5′-‐

atgcagccgcaggagtacat-‐3′ and reverse rtn4a-‐rv 5′-‐ggctgccgggtcacgact-‐3′. The rtn4b cDNA was amplified with forward primer rtn4b-‐fw 5′-‐gtcctgagctgcgctatttc-‐3′ and reverse primer rtn4b-‐rv 5′-‐gttatttagtaggcagcggtgtg-‐3′ by RT-‐PCR from total RNA extracted from adult zebrafish optic nerve. First-‐strand cDNA was synthesized under standard conditions with the SuperScript First-‐Strand Synthesis System (Invitrogen) using an oligo(dT) primer. All of the above-‐mentioned PCR experiments were done with Phusion High-‐Fidelity DNA Polymerase (Finnzymes/Thermo Fisher Scientific, Espoo, Finland). Full-‐length cDNAs were cloned into a PCR2.1 TOPO vector (Invitrogen) and sequenced.

Morpholino knockdowns and mRNA rescue

The following MOs were purchased from Gene Tools (Philomath, OR, USA) and designed to target independent sequences at the 5′ UTRs and the start codon of the zebrafish rtn4a and rtn4b, including known splice variants based on the following sequence data obtained from the GenBank database.

rtn4a-‐l, 5′-‐taaagtaacttcaagatgcgccgga-‐3′ (position on mRNA −55/−30) and 5′-‐tcgtggagcttatttgatcatccat-‐3′ (position on mRNA 1/25) [GenBank:AY555039.1];

rtn4a-‐m, 5′-‐cgtgcatcggtcatatatccagtca-‐3′ (position on mRNA −18/+7) and 5′-‐

ttatctgaattggcgtgcatcggtc-‐3′ (position on mRNA −5/+20) [GenBank:AY555042.1]; rtn4a-‐n, 5′-‐ctcgctcattctgcgatcagacagcc-‐3′ (position on mRNA −25/0) and 5′-‐gctccaccacttgtttggaatccat-‐3′ (position on mRNA 1/25) [GenBank:AY555043.1]; rtn4b, 5′-‐ccactgcgggagaactcagaacagc-‐3′ (position on mRNA −81/−57, for better distinction, rtn4b-‐MO-‐1) and 5′-‐

(20)

gctcgttctgtgtcctccatcggga-‐3′ (position on mRNA −5/+20, rtn4b-‐MO-‐2) [RefSeq:NM_001040335.1]; control, 5′-‐aacgaacgaacgaacgaacgaacgc-‐3′

In addition to ATG-‐targeting MOs, as described by Brösamle and Halpern (Brosamle and Halpern, 2009), we used MOs directed against 5′UTR sequences of the rtn4a splice variants.

All microinjections were performed at early cleavage stages (one-‐ to four-‐cell stage) using a manual micromanipulator (Narishige, Tokyo, Japan) coupled to a Transjector 5246 (Eppendorf, Hamburg, Germany) under a Stemi 2000 stereomicroscope (Carl Zeiss Microscope). After running specificity and dose-‐

dependency controls, MOs were injected at a concentration of 0.5 or 1.0 ng/nl in 13 Danieau buffer (58 mM NaCl, 0.7 mM KCl, 0.4 mM MgSO4, 0.6 mM Ca(NO3)2, 5.0 mM 2-‐[4-‐(2-‐hydroxyethyl)piperazin-‐1-‐yl] ethanesulfonic acid [pH 7.6]) and 0.125% Phenol Red (Sigma-‐Aldrich, St Louis, MO, USA).

For MO rescue experiments, rtn4a-‐l was cloned in frame with GFP into the EcoRI/ApaI restriction sites of pGFP-‐N1. Rtna-‐l-‐gfp, rtn4a-‐l and rtn4b ORF cDNAs were subcloned into the EcoRI/XbaI (rtn4al-‐gfp), EcoRI/XbaI (rtn4a-‐l) or EcoRI/StuI (rtn4b) restriction sites of pCS2+ (provided by Z Varga, University of Oregon, Eugene, OR, USA) and transcribed in vitro using the mMESSAGE mMACHINE SP6 kit (Ambion, Austin, TX, USA).

For mRNA synthesis, DNA templates were linearized with BssHII. After synthesis, template DNA was removed by DNaseI digestion of the rtn4a-‐l and rtn4b mRNAs. rtn4a-‐l or rtn4b MO at 1.0 ng/nl in 13 Danieau buffer were coinjected with capped mRNAs at 20 or 100 pg/nl at a 1:1 ratio in 0.05 M KCl and 0.125% Phenol Red. For overexpression experiments, mRNAs were microinjected at 100 pg/nl. At least 200 embryos per experiment were microinjected (5-‐nl injection volume) and kept in E3 medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl2 and 0.33 mM MgSO4) at 28°C. Quantification of phenotypes was carried out on 200 embryos per experiment, from among which a smaller number were selected for detailed analysis. Images were acquired using a SteREO Lumar.V12, Axioplan 2 or confocal laser scanning microscope

(21)

LSM 710 (Carl Zeiss Microscopy). Images were further processed using Adobe Photoshop 12.0 software.

Immunohistochemistry

Anesthetized embryos (6 to 24 hpf) were fixed in 4% paraformaldehyde (PFA) in phosphate-‐buffered saline (PBS) for 2 hours at room temperature (RT) or overnight at 4°C. Embryos/larvae older than 48 hpf were fixed in PFA for 30 minutes at RT, washed in PBS-‐Tween 20 (PBST) and permeabilized in acetone for 7 minutes at −20°C. The following antibodies and concentrations were used for whole-‐mount immunohistochemistry: polyclonal anti-‐neurolin, 1:500 (Diekmann and Stuermer, 2009); monoclonal antiacetylated tubulin, 1:1,000 (Sigma-‐Aldrich) and the monoclonal anti-‐HuC/HuD neuronal protein (16A11) 1:1,000 (Molecular Probes, Sunnyvale, CA, USA). For staining with the polyclonal Rtn4a antibody (IK964, which was generated in our laboratory) diluted 1:250 (Abdesselem et al., 2009), PFA fixation was not used. Instead, embryos were incubated on ice in 50% methanol in PBS, pH 7.4 (2 minutes), 100% MeOH (5 minutes) and 50% MeOH in PBS (2 minutes). To generate a polyclonal antibody against zebrafish Rtn4b, the rtn4b-‐M1-‐M4 region (Shypitsyna et al., 2011) was amplified by PCR from a pCR2.1 TOPO vector containing the rtn4b ORF. Forward rtn4b-‐M1-‐fw 5′-‐GGGAATTCTAGCCCGTCTCCAGACCTGCTCCAGGA-‐3′ and reverse rtn4b-‐M4-‐rv 5′-‐GGGTCGACCTA-‐CTGCAGACCCTGGAGCAGCTCTGCC-‐3′ primers containing EcoRI and SalI restriction enzyme sites were designed to amplify 490 bp, including the M1 to M4 motifs. The PCR product was digested with EcoRI and SalI and cloned in frame into the pGEX-‐4T-‐3 glutathione S-‐transferase expression vector (GE Healthcare Life Sciences, Freiburg, Germany) after the thrombin cleavage site. The recombinant protein was used to immunize a rabbit to produce the polyclonal antibody K1121. The immunopurified Rtn4b antibody was used at a dilution of 1:500.

Nuclei were counterstained with 100 ng/ml DAPI, together with the secondary antibody, for 30 minutes at RT. The secondary antibodies were cross-‐purified with fluorophore-‐conjugated goat anti-‐rabbit and cyanine 3 or Alexa Fluor 488–

coupled anti-‐mouse antibodies in which specimens were incubated overnight at

(22)

4°C. For analysis of Rtn4 expression levels, embryos were dechorionated, deyolked, lysed and analyzed by Western blotting. Blots were exposed to polyclonal Rtn4a antibody (IK964; diluted 1:10,000) and polyclonal Rtn4b antibody (K1121; diluted 1:1,000) and to a monoclonal antibody against GFP (diluted 1:2,000 to detect Rtn4al-‐GFP; Roche Applied Science).

Bromodeoxyuridine labeling

To label cells in the S-‐phase, embryos were immersed in 10 mM BrdU (Sigma-‐

Aldrich) in 1% dimethyl sulfoxide in E3 medium. Embryos were incubated for 1 hour at 28°C and washed in E3 medium (three timer for 5 minutes), fixed in 4%

PFA overnight at 4°C and dehydrated in methanol at −20°C. After gradual rehydration, embryos were permeabilized with proteinase K (10 μg/ml) followed by postfixation with 4% PFA, washed with PBST, blocked with 10%

normal goat serum in PBST for at least 2 hours at room temperature and incubated with mouse anti-‐BrdU-‐fluorescein isothiocyanate antibody (1:200;

Sigma-‐Aldrich) in 4% blocking solution overnight at 4°C.

Acridine orange staining

To get an impression of the extent of apoptosis, 1-‐dpf live embryos were incubated in 2 μg/ml acridine orange (Sigma-‐Aldrich) for 30 minutes, followed by three rinses in E3 medium. Embryos were anesthetized in 0.016% Tricaine methanesulfonate (MS-‐222; Sigma-‐Aldrich) and photographed (Zeiss Lumar.V12 stereomicroscope).

Motility tests

To evaluate the escape response, 3-‐dpf embryos were touched with the tip of a fine needle twice at the dorsal tip of the tail. Embryos that did not react were classified as nonmotile. Three groups of at least 50 embryos were tested in each experiment.

(23)

Quantifications

To quantify total cell numbers and axon branching of motor neurons in tg(hb9:gfp), control and rtn4b-‐MO1-‐injected embryos, six representative specimens from each group were fixed at 1 and 2 dpf, respectively, and their trunk regions were scanned by confocal microcopy. All fluorescent cells (trunk segments 15 to 18) and axonal projections (trunk segments 5 to 8 and 15 to 18) were counted in z-‐stack confocal reconstructions. Embryos exhibiting aberrant branching and mistakes in pathfinding of their motor axons were classified as mild, and those, which in addition showed defasciculation were categorized as strong. The size of the eye and the area covered by RGCs, as well as the areas of the optic tectum, forebrain and neuropil, were determined in tg(Brn3c:mgfp) control, rtn4a-‐l and rtn4b MO1-‐injected embryos, with 10 representative specimens at 3 and 5 dpf. Areas were measured in ImageJ software (National Institutes of Health, Bethesda, MD, USA) by using ventral and dorsal z-‐plane projections of the head. Data are represented as mean values, and error bars indicate the standard error of the mean. Data were analyzed using analysis of variance (ANOVA) and paired t-‐test was used after determining whether the sample datasets conform to a normal distribution. P-‐values are indicated as follows: *P≤0.05. **P≤0.01.

(24)

Part II: Rat-‐Nogo transgenic Zebrafish

Germ line transformation

For germ line transformation, plasmid DNA and Tol2-‐transposase mRNA were injected into fertilized eggs (F0), raised to adulthood and crossed to wild-‐type zebrafish from the AB line, which is the primary background of all transgenic and mutant fish that come from the Zebrafish International Resource Stock Center (ZIRC), in Oregon, USA. The red-‐to green reporter and the neuronal Cre driver lines were kindly provided by Dr. Stefan Hans from the Brand’s Lab in the CRTD-‐

Dresden (Hans et al., 2009, Hans et al., 2011).

Bacterial artificial chromosomes (BAC) recombineering

All necessary regulatory regions were obtained by screening the CHORI BAC library. The zebrafish BAC clones CH73-‐343K20 and CH211-‐22D6 containing rtn4a and mbp respectively, were identified. Mapping the BAC clones onto the zebrafish genome confirmed that the BACs contained 95,1 Kb upstream of mbp, and 18,5 kb upstream of rtn4a. To create the appropriate DNA vectors for fish transgenesis, we modified the relevant BAC clones via Escherichia coli-‐based homologous recombination and Flp-‐FRT-‐mediated site-‐specific recombination approaches (Sarov et al., 2006). During this process, the genes of interest were inserted downstream of the regulatory region.

Generation of driver line

To create the pTol mbp:mCherry-‐T2A-‐CreER^T2 plasmid, 5 kb of the mbp regulatory region were amplified by PCR with restriction sites EcoNI at the 5’

end and FseI at the 3’ end. PCR products were sequentially subcloned into the EcoNi-‐FseI site of the pTol:mCherry-‐T2a-‐CreER^T2 (Hans et al., 2009).

For the mbp:CreER^T2 line, F1 embryos were screened by PCR using mbp(ttgccaacgttgtaggctactacc) and Cre(tagagcctgttttgcacgttcacc)-‐specific primers that result in an 867 base pair fragment. Positive embryos were examined under a fluorescent microscope and positive embryos were raised. Out

(25)

of 16 PCR-‐positive F0 fish 8 lines showed a distinctive CreER^T2 expression pattern. Three lines were established and carriers were identified by either PCR or cross to the red-‐to-‐green reporter line.

Generation of effector lines

Rat Nogo-‐A was amplified by PCR with Fw: Rv: primers including SalI and SpeI restriction sites. The PCR product was digested with the appropriate restriction enzymes and further ligated into pCR4BLUNT-‐TOPO-‐EGFP-‐T2A-‐ntr previously digested with XhoI and XbaI. Positive clones were sequenced. The new EGFP-‐

T2A-‐Nogo-‐A was digested with restriction enzymes StuI and AscI and ligated with the pTol:hsp70-‐DsRed-‐2-‐EGFP (Hans et al., 2009), previously digested with SmaI and AscI restriction enzymes. The pTol-‐hsp70-‐DsRed-‐rtn4-‐T2A-‐EGFP plasmid was generated in the same manner.

F1 embryos were examined under a fluorescent microscope and positive embryos were raised. This way, ten independent F0 were identified and tested for recombination in the presence of Cre. As they all showed efficient recombination, the strongest allele was chosen to establish the rat-‐Nogo-‐A-‐red-‐

to-‐green transgenic line, which now shows efficient recombination in the third generation.

PCR analysis of transposase mediated transgenesis -‐ Micro-‐injection and excision assay

Transposase mRNA was synthesized as described previously (Kawakami 2004;

Kawakami et al. 2004b). Approximately 1 nanoliter of a DNA–RNA solution containing 25 ng/ml circular DNA of a transposon-‐donor plasmid and 25 ng/ml transposase mRNA were injected into fertilized eggs. Approximately 24 h after the injection, DNA samples were prepared from the injected embryos and the transient excision assay was performed as described (Kawakami and Shima 1999; Kawakami 2004). The excision products were amplified by using primers Tp-‐Fw (gctactacatggtgccattcct) and Tp-‐Rv (ggcacgacaggtttcccgac).

(26)

Pharmacological treatments and heat induction

For pharmacological treatments the following stock solutions were made and stored at -‐20°C: 50 mM tamoxifen (TAM; Sigma, T5648) in DMSO; 25 mM 4-‐

hydroxy-‐tamoxifen (4-‐OHT; Sigma, H7904) in ethanol. For embryo treatments, dilutions of these chemicals were made in embryo medium as follows: TAM: 5 and 0.5; 4-‐OHT: 0.5 mM. At mid-‐gastrulation (8 hpf) or at 30 hpf embryos were transferred into petri dishes containing the treatment solution. Larvae (5 dpf onwards) were incubated overnight in 4-‐OHT. For control treatments, sibling embryos were incubated in corresponding dilutions of DMSO and ethanol. All incubations were conducted in the dark. For heat induction, embryos were transferred into fresh petri dishes. After removal of excess embryo medium, 42°C embryo medium was added and the petri dishes were kept for two hours in a 39°C incubator. Afterwards, embryos were returned to the 28.5°C incubator.

GFP expression in embryos

GFP expression in embryos was analyzed by using a fluorescence stereo-‐

microscope Lumar (Zeiss), and photos were taken.

Function of RTN4/Nogo during development and as transgene in adult zebrafish