Progenitor-derived glia are required for spinal cord regeneration in zebrafish (original) (raw)

Development. 2023 May 15; 150(10): dev201162.

Lili Zhou,1,2 Anthony R. McAdow,1,2 Hunter Yamada,1,2 Brooke Burris,1,2 Dana Klatt Shaw,1,2 Kelsey Oonk,3 Kenneth D. Poss,3 and Mayssa H. Mokalledcorresponding author1,2,*

Lili Zhou

1Department of Developmental Biology, Washington University School of Medicine, St. Louis, MO, 63110, USA

2Center of Regenerative Medicine, Washington University School of Medicine, St. Louis, MO, 63110, USA

Anthony R. McAdow

1Department of Developmental Biology, Washington University School of Medicine, St. Louis, MO, 63110, USA

2Center of Regenerative Medicine, Washington University School of Medicine, St. Louis, MO, 63110, USA

Hunter Yamada

1Department of Developmental Biology, Washington University School of Medicine, St. Louis, MO, 63110, USA

2Center of Regenerative Medicine, Washington University School of Medicine, St. Louis, MO, 63110, USA

Brooke Burris

1Department of Developmental Biology, Washington University School of Medicine, St. Louis, MO, 63110, USA

2Center of Regenerative Medicine, Washington University School of Medicine, St. Louis, MO, 63110, USA

Dana Klatt Shaw

1Department of Developmental Biology, Washington University School of Medicine, St. Louis, MO, 63110, USA

2Center of Regenerative Medicine, Washington University School of Medicine, St. Louis, MO, 63110, USA

Kelsey Oonk

3Duke Regeneration Center, Department of Cell Biology, Duke University Medical Center, Durham, NC 27710, USA

Kenneth D. Poss

3Duke Regeneration Center, Department of Cell Biology, Duke University Medical Center, Durham, NC 27710, USA

Mayssa H. Mokalled

1Department of Developmental Biology, Washington University School of Medicine, St. Louis, MO, 63110, USA

2Center of Regenerative Medicine, Washington University School of Medicine, St. Louis, MO, 63110, USA

1Department of Developmental Biology, Washington University School of Medicine, St. Louis, MO, 63110, USA

2Center of Regenerative Medicine, Washington University School of Medicine, St. Louis, MO, 63110, USA

3Duke Regeneration Center, Department of Cell Biology, Duke University Medical Center, Durham, NC 27710, USA

corresponding authorCorresponding author.

Handling Editor: Steve Wilson

Competing interests

The authors declare no competing or financial interests.

Received 2022 Jul 27; Accepted 2023 Apr 26.

Copyright © 2023. Published by The Company of Biologists Ltd

ABSTRACT

Unlike mammals, adult zebrafish undergo spontaneous recovery after major spinal cord injury. Whereas reactive gliosis presents a roadblock for mammalian spinal cord repair, glial cells in zebrafish elicit pro-regenerative bridging functions after injury. Here, we perform genetic lineage tracing, assessment of regulatory sequences and inducible cell ablation to define mechanisms that direct the molecular and cellular responses of glial cells after spinal cord injury in adult zebrafish. Using a newly generated CreERT2 transgenic line, we show that the cells directing expression of the bridging glial marker ctgfa give rise to regenerating glia after injury, with negligible contribution to either neuronal or oligodendrocyte lineages. A 1 kb sequence upstream of the ctgfa gene was sufficient to direct expression in early bridging glia after injury. Finally, ablation of _ctgfa_-expressing cells using a transgenic nitroreductase strategy impaired glial bridging and recovery of swim behavior after injury. This study identifies key regulatory features, cellular progeny, and requirements of glial cells during innate spinal cord regeneration.

Keywords: Glia, Neural repair, Regeneration, Spinal cord injury, Zebrafish

Summary: Lineage tracing, regulatory sequence analysis and inducible cell ablation determine the contribution, molecular regulation, and requirement of bridging glia for innate neural repair during spinal cord regeneration in adult zebrafish.

INTRODUCTION

Disparate glial, neuronal and systemic injury responses underlie differential regenerative capacities of CNS tissues across vertebrates (Silver and Miller, 2004; Mokalled et al., 2016; Silver, 2016; O'Shea et al., 2017; Brennan and Popovich, 2018; Sofroniew, 2018). Whereas spinal cord injuries (SCIs) cause irreversible sensory and motor deficits in mammals (He and Jin, 2016; Silver, 2016; Sofroniew, 2018), adult zebrafish possess potent regenerative capacity and reverse paralysis within 8 weeks of complete spinal cord (SC) transection (Becker et al., 1997; Reimer et al., 2008; Goldshmit et al., 2012). Following SCI, mammalian astrocytes display multifarious injury responses that are overshadowed by anti-regenerative scar-forming cells and inhibitory extracellular molecules (Grimpe and Silver, 2004; Dias and Göritz, 2018; Sofroniew, 2018). Although the complexities of glial cell responses to injury have been extensively studied in mammals, our understanding of how glial cells respond to SCI in highly regenerative vertebrates such as zebrafish is relatively limited. Following SCI in zebrafish, specialized glial cells form a bridge that is thought to provide a physical and signaling scaffold for cellular regrowth across the lesion (Goldshmit et al., 2012; Mokalled et al., 2016). However, whether glial cells are pro-regenerative in zebrafish and how pro-regenerative glia support innate SC regeneration require further investigation.

Previous studies identified connective tissue growth factor a (ctgfa; ccn2a) as a central glial bridging factor and described the emergence of ctgfa+ gfap+ bridging glia after SCI (Mokalled et al., 2016). Ctgf is a multidomain-containing extracellular molecule that interacts with multiple signaling pathways and elicits a range of cellular responses, including cell proliferation and differentiation (Grotendorst and Duncan, 2005). After SCI, ctgfa is expressed in a number of cell types outside SC tissues around the lesion. Yet, the spatiotemporal pattern of ctgfa expression within SC tissues correlates with glial bridge formation (Mokalled et al., 2016). Following injury, ctgfa expression is first broadly induced in proliferating sox2+ progenitors around the central canal proximal to the lesion. During subsequent steps of regeneration, ctgfa transcripts localize to a subset of ventral progenitors that undergo epithelial-to-mesenchymal transition (EMT) after injury (Mokalled et al., 2016; Klatt Shaw and Mokalled, 2021; Klatt Shaw et al., 2021). Genetic evidence indicates that ctgfa functions within a multi-nodal EMT-driving gene regulatory network that is necessary and sufficient to reprogram pro-regenerative glia after injury (Klatt Shaw et al., 2021). Here, we employ a suite of _ctgf_-based genetic tools to trace the fates and functions of pro-regenerative glia during SC regeneration.

Zebrafish glia possess an astrocyte-like cell identity. However, EMT gene expression distinguishes zebrafish bridging glia from mammalian astrocytes (Klatt Shaw et al., 2021). EMT is often linked to increased plasticity and stem cell activation during tissue regeneration (Jessen and Arthur-Farraj, 2019; Wilson et al., 2020), suggesting that zebrafish glia have increased EMT-mediated plasticity and regenerative potential. In the zebrafish SC, EMT activation during bridging involves Yap/Taz, Egr1, Junbb and Spi1 transcriptional regulators (Klatt Shaw et al., 2021). Hallmarks of EMT include downregulation of epithelial markers, such as E-cadherin (Cadherin 1), and upregulation of mesenchymal genes, such as N-cadherin (Cadherin 2), driven by Twist and Zeb transcription factors (Dongre and Weinberg, 2019). Injury-induced expression of egr1 and junbb, along with yap/taz activation, correlate with ctgfa expression and are required for bridging and functional SC repair (Gervasi et al., 2012; Wu et al., 2017; Han et al., 2019; Sato et al., 2020). However, the temporal and transcriptional hierarchies of EMT driving transcriptional regulators and the mechanisms that direct glial progenitors toward specific cell fates during regeneration remain to be dissected.

This study defines glial cell functions during zebrafish SC regeneration. Using a battery of _ctgfa_-based genetic tools, we determined the contribution, regulation and requirement of bridging glial cells during innate SC repair. A newly generated _ctgfa:_CreERT2 transgenic line established the contribution of _ctgfa_-expressing cells to regenerative glia and glial progenitors after injury. Using a series of transgenic reporter lines, we found that regulation of ctgfa expression converges onto an EMT-driving distal enhancer element that directs _ctgfa-_dependent transcription following injury. Finally, we show that genetic ablation of _ctgfa_-expressing cells impairs glial bridging and functional SC repair. Our findings indicate the cell contributions and regulatory mechanisms that direct regenerative gliogenesis during innate SC repair.

RESULTS

Generation of _ctgfa:_CreERT2 zebrafish for genetic lineage tracing

Using _−5.5Kb-ctgfa:_EGFP and _−5.5Kb-ctgfa:_mCherry-NTR reporter lines, we previously showed that _ctgfa_-driven fluorescence recapitulates endogenous mRNA expression in ventral ependymal progenitors and in bridging glia during SC regeneration (Mokalled et al., 2016; Klatt Shaw et al., 2021). To specifically target ctgfa+ cells for permanent labeling and lineage tracing, we generated a _−5.5Kb-ctgfa:_CreERT2 line to use in combination with the previously established _ubiquitin:_loxP-GFP-STOP-loxP-mCherry transgene (Mosimann et al., 2011). The compound transgenic line _−5.5Kb-ctgfa:_CreERT2;_ubiquitin:_loxP-GFP-STOP-loxP-mCherry, referred to hereafter as _ctgfa_-Tracer, enables permanent mCherry labeling in ctgfa+ cells and their progeny following tamoxifen-induced Cre-mediated recombination (Fig. S1A). To this end, we performed SC transections on _ctgfa_-Tracer animals and control siblings, and treated with 5 mM tamoxifen at 4 days post-injury (dpi) for 24 h to induce recombination prior to the emergence of bridging glia. To assess the extent of recombination, we collected SC tissues 2 days following the end of tamoxifen treatment, which corresponds to 7 dpi. Based on EGFP (_ctgfa_-Tracer−) and mCherry (_ctgfa_-Tracer+) staining, we determined that 4% of the cells within SC tissues showed recombined _ctgfa_-Tracer+ expression in tamoxifen-treated CreERT2+ animals (Fig. S1B,D). Conversely, SC sections from vehicle-treated CreERT2+, vehicle-treated CreERT2− or tamoxifen-treated CreERT2− controls did not show recombined _ctgfa_-Tracer+ cells (Fig. S1B,D). We noted that a subset of _ctgfa_-Tracer+ cells labeled blood vessel-associated cells. However, this observation remains to be confirmed and explored. Previous studies have shown that Ctgfa is an injury-induced regeneration factor after SCI in zebrafish and that ctgfa expression is not detectable in uninjured SC tissues (Mokalled et al., 2016). To test whether _ctgfa_-Tracer activity recapitulates endogenous ctgfa expression, we treated uninjured _ctgfa_-Tracer fish with 5 mM tamoxifen for 24 h and assessed recombination 2 days following the end of tamoxifen treatment. EGFP and mCherry staining indicated no detectable recombination (_ctgfa_-Tracer+) in the absence of injury (Fig. S1C,D). Together, these studies showed that recombination in the _ctgfa_-Tracer line is injury induced, Cre dependent and tamoxifen inducible. These studies established a recombination paradigm that recapitulates the expression of endogenous ctgfa transcripts and of previously established ctgfa reporter lines after SCI (Mokalled et al., 2016; Klatt Shaw et al., 2021). We thus used the _ctgfa_-Tracer line to map the contributions of ctgfa+ cells during SC regeneration.

Contribution of ctgfa+ cells to regenerating glia during SC regeneration

To examine the fates of ctgfa+ cells during SC regeneration, we performed SCI on _ctgfa_-Tracer animals, treated with tamoxifen at 4 dpi and traced recombined cells at 7, 14 and 28 dpi. (Fig. 1A). SC sections 150 µm (proximal), 450 µm (distal) and 750 µm rostral to the lesion were analyzed using mCherry staining to label recombined _ctgfa_-Tracer+ cells, and Gfap staining for glial cell co-labeling (Fig. 1A,B). At 7 dpi, we observed infrequent Gfap+ projections that were _ctgfa_-Tracer+ proximal to the lesion, and weak Gfap+ _ctgfa_-Tracer+ signal ventral to the central canal in distal SC sections. At 14 dpi, Gfap+ _ctgfa_-Tracer+ cells were more abundant in SC tissues proximal to the lesion, and showed consistent labeling in progenitor cells ventral to the central canal in distal SC sections. By 28 dpi, the majority of _ctgfa_-Tracer+ cells within SC tissues were Gfap+ and localized to the ventral progenitors at the proximal and distal levels. We noted that immunostaining signals for filamentous Gfap intermediate filaments do not allow for exact counts of glial cell soma, and result in underestimated assessment of colocalization. Despite these limitations, quantification of Gfap+ _ctgfa_-Tracer+ colocalization indicated increased Gfap expression in recombined cells at 28 dpi relative to 7 dpi at 150 and 450 µm rostral to the lesion (Fig. 1C). For instance, at 150 µm rostral to the lesion, Gfap expression increased from 7.7% of _ctgfa_-Tracer+ cells at 7 dpi to 25.5% of _ctgfa_-Tracer+ cells at 28 dpi (Fig. 1C). Conversely, the proportions of Gfap+ _ctgfa_-Tracer+ colocalization relative to Gfap+ did not significantly change between 7 and 28 dpi (Fig. 1D), ostensibly due to increased Gfap expression in proximal tissues as SC tissues regenerate. Gfap+ _ctgfa_-Tracer+ fluorescence at 750 µm rostral to the lesion was unchanged across time points, suggesting the distal contribution of _ctgfa_-Tracer cells is minimal (Fig. 1C,D). To further address the contribution of _ctgfa_-Tracer cells relative to ctgfa expression, we performed in situ hybridization chain reaction (HCR) for ctgfa on _ctgfa_-Tracer fish at days 2 and 9 following tamoxifen treatment, which correspond to 7 and 14 dpi. SC sections at the lesion site and 150 µm from the lesion were examined (Fig. S2A). At 7 dpi, ctgfa transcript expression was elevated and the majority of _ctgfa_-Tracer cells also expressed ctgfa mRNA. In contrast, at 28 dpi ctgfa expression was lower relative to 7 dpi. At this time point, the majority of _ctgfa_-Tracer cells did not express ctgfa transcripts at the lesion site, and the domain of _ctgfa_-Tracer expression in proximal SC sections was expanded relative to ctgfa expression. These studies highlight the power of genetic lineage tracing, which allowed us to trace the fates of _ctgfa_-expressing cells at later stages of regeneration as ctgfa expression resumes toward baseline levels. Together, these experiments indicated increased contribution of _ctgfa_-derived cells to bridging glia and ventral ependymal progenitors during SC regeneration.

An external file that holds a picture, illustration, etc. Object name is develop-150-201162-g1.jpg

Contribution of ctgfa+ cells to regenerating glia during SC repair. (A) Experimental timeline to evaluate the contribution of ctgfa+ cells after SCI. _ctgfa_-Tracer animals refer to the compound transgenic line _−5.5Kb-ctgfa:_CreERT2; _ubiquitin:_loxP-GFP-STOP-loxP-mCherry. _ctgfa_-Tracer fish were subjected to complete SC transection and tamoxifen (TAM)-mediated recombination at 4 dpi to enable permanent mCherry labeling in ctgfa+ and ctgfa+-derived cells. SC tissues were collected at 7, 14 and 28 dpi. Schematic of SC tissue illustrates cross-sections at 150, 450 and 750 µm rostral to the lesion that were analyzed for histological examination. The 7 dpi time point was used to assess recombination following TAM treatment. The 14 and 28 dpi time points were used to trace the fates of ctgfa+-derived cells. (B) Immunostaining for mCherry (green), Gfap (magenta), and nuclear Hoechst (gray) at 7, 14 and 28 dpi. mCherry expression, referred to as _ctgfa_-Tracer+, is used to trace the fates of _ctgfa_-expressing cells following TAM-inducible recombination. SC sections from TAM-treated _ctgfa_-Tracer (Cre+) animals are shown. Cross-sections 150 µm (proximal) or 450 µm (distal) rostral from the lesion site are shown. High-magnification insets show select _ctgfa_-Tracer+ cells in triple-, double- or single-channel views. Arrowheads indicate _ctgfa_-Tracer+ cells. (C,D) Quantification of _ctgfa_-Tracer+ and Gfap+ colocalization at 7, 14 and 28 dpi. SC cross-sections 150, 450, and 750 µm rostral to the lesion were analyzed. _ctgfa_-Tracer+ and Gfap+ fluorescence were quantified. For each section, _ctgfa_-Tracer+ Gfap+ fluorescence was normalized to total _ctgfa_-Tracer+ in C and to total Gfap+ in D. Dots indicate individual animals and sample sizes are indicated in parentheses. Error bars represent s.e.m. ***P<0.001; **_P_<0.01 (two-way ANOVA). ns, not significant (_P_>0.05). Scale bar: 50 µm.

Contribution of ctgfa+ cells to neurons and oligodendrocytes during SC regeneration

To evaluate the contribution of ctgfa+ cells to additional regenerative cell fates after SCI, we traced and quantified _ctgfa_-derived neurons and motor neurons at 28 dpi. Using the same recombination paradigm described above, we performed mCherry staining to label recombined _ctgfa_-Tracer+ cells and co-stained with either HuC/D (Elavl3/4) to label post-mitotic neurons (Fig. 2A) or Hb9 (Mnx1) to label newly formed motor neurons (Fig. S2). By immunostaining, nuclear and cytoplasmic mCherry signals mark _ctgfa_-Tracer+ cells. Similarly, HuC/D antigens are detected in the cytoplasms and nuclei of postmitotic neurons, whereas Hb9 is exclusively nuclear in newly formed motor neurons. This allowed us to quantify the numbers of _ctgfa_-Tracer+ HuC/D+ Hoechst+ cells and _ctgfa_-Tracer+ Hb9+ Hoechst+ cells for these experiments. At 28 dpi, we occasionally observed up to three HuC/D+ _ctgfa_-Tracer+ cells per section (Fig. 2B, Fig. S2B) or a single Hb9+ _ctgfa_-Tracer+ cell per section (Fig. S2C,D), suggesting that the contribution of _ctgfa_-Tracer cells to regenerating neurons or motor neurons was negligible. At 150 µm rostral to the lesion, the HuC/D+ _ctgfa_-Tracer+ cells accounted for 4% of _ctgfa_-Tracer+ cells and 5% of HuC/D+ neurons (Fig. 2C,D). By contrast, Hb9+ neurons were not detected in any sections at either 150 or 450 µm rostral to the lesion at 28 dpi (Fig. S2E,F). We next assessed the contribution of ctgfa+ cells to myelinating oligodendrocytes by co-labeling of _ctgfa_-Tracer cells with an antibody that detects Myelin basic proteins (Mbp) (Fig. 2E). Like Gfap, immunostaining for Mbp does not allow for exact counts of oligodendrocyte cells. We thus quantified Mbp+ _ctgfa_-Tracer+ colocalization, which accounted for 10.8% of _ctgfa_-Tracer+ (Fig. 2F), and for <0.3% of Mbp+ fluorescence (Fig. 2G) at 150 µm rostral to the lesion. These studies demonstrated the contribution of ctgfa+ cells to regenerating glial cells, with minimal contribution to neuron or oligodendrocyte lineages during SC regeneration.

An external file that holds a picture, illustration, etc. Object name is develop-150-201162-g2.jpg

Contribution of ctgfa+ cells to regenerating neurons and oligodendrocytes after SCI. (A) Experimental timeline to evaluate the contribution of ctgfa+ cells after SCI. _ctgfa_-Tracer animals were subjected to complete SC transection and tamoxifen (TAM)-mediated recombination at 4 dpi to enable permanent mCherry labeling in ctgfa+ and ctgfa+-derived cells. SC tissues were collected at 28 dpi for histological examination. (B) Immunostaining for mCherry (green), HuC/D (magenta) and nuclear Hoechst (gray) at 28 dpi. mCherry expression, referred to as _ctgfa_-Tracer+, is used to trace the fates of _ctgfa_-expressing cells following TAM-inducible recombination at 4 dpi. SC sections from TAM-treated _ctgfa_-Tracer (Cre+) animals are shown. Cross-sections 150 µm (proximal) or 450 µm (distal) rostral from the lesion site are shown. High-magnification insets show select _ctgfa_-Tracer+ cells in triple-, double- or single-channel views. White arrowheads indicate _ctgfa_-Tracer+ HuC/D+ cells; yellow arrowheads indicate _ctgfa_-Tracer− HuC/D+ cells. (C,D) Quantification of _ctgfa_-Tracer+ and HuC/D+ colocalization at 28 dpi. _ctgfa_-Tracer Hoechst and HuC/D Hoechst were used to quantify the numbers of _ctgfa_-Tracer+ and HuC/D+ cells. _ctgfa_-Tracer+ HuC/D+ cells were normalized to the total number of _ctgfa_-Tracer+ cells in C, and to the total number of HuC/D+ cells in D. (E) Immunostaining for mCherry (green), Mbp (magenta) and nuclear Hoechst (gray) at 28 dpi. SC sections from TAM-treated _ctgfa_-Tracer (Cre+) animals are shown. Cross-sections 150 µm (proximal) or 450 µm (distal) rostral from the lesion site are shown. High-magnification insets show select _ctgfa_-Tracer+ cells in triple-, double- or single-channel views. White arrowheads indicate _ctgfa_-Tracer+ Mbp+ staining; yellow arrowheads indicate _ctgfa_-Tracer+ Mbp− staining. (F,G) Quantification of _ctgfa_-Tracer+ and Mbp+ colocalization at 28 dpi. _ctgfa_-Tracer+ and Mbp+ fluorescence were quantified. _ctgfa_-Tracer+ Mbp+ cells were normalized to total _ctgfa_-Tracer+ in F and to total Mbp+ fluorescence in G. For all quantifications, SC cross-sections 150, 450 and 750 µm rostral to the lesion were analyzed. Dots indicate individual animals and sample sizes are indicated in parentheses. Error bars represent s.e.m. Scale bars: 50 µm.

Mapping the gene regulatory sequences that direct ctgfa expression after SCI

EMT activation localizes to ventral glial progenitors after SCI. The EMT-driving gene regulatory network, which includes ctgfa, is necessary and sufficient to promote glial bridging and functional SC repair (Mokalled et al., 2016; Klatt Shaw et al., 2021). To improve our understanding of the regulatory mechanisms that induce localized ctgfa expression and EMT activation after SCI, we analyzed the cis_-regulatory elements that direct injury-induced ctgfa transcription in regenerating SC tissues. Using −5.5Kb-ctgfa:EGFP, we previously showed that EGFP fluorescence recapitulates endogenous mRNA expression in ventral glial progenitors and in bridging glia during SC regeneration (Mokalled et al., 2016). To identify specific regulatory elements within the −5.5 Kb genomic region of −5.5Kb-ctgfa:EGFP, stable transgenic lines expressing an EGFP cassette downstream of the −4 Kb- or −3 Kb genomic regions were generated (−_4Kb-ctgfa:EGFP and −3Kb-ctgfa:EGFP, respectively) (Fig. 3A). Multiple independent lines were generated for each transgene to control for possible positional effects that may impact transgene expression. To establish stable reporter lines, transgenesis was first confirmed by performing EGFP PCR, and then by assessing EGFP fluorescence in developing zebrafish embryos (Fig. S3A). At 3 days post-fertilization, we observed comparable EGFP expression in −5.5Kb-, −4Kb- and −3Kb-ctgfa:EGFP. To determine the expression of ctgfa reporter lines after SCI, adult reporter animals were subjected to complete SC transections and examined for EGFP expression at 10 dpi (Fig. 3B). ctgfa_-driven EGFP and Gfap expression were assessed 150 µm (proximal) and 450 µm (distal) rostral to the lesion. Similar to −_5.5Kb_-ctgfa:EGFP, −_4Kb-ctgfa:EGFP transgenic lines showed EGFP expression at 10 dpi, whereas EGFP was not detectable in −3Kb-ctgfa:EGFP lines (Fig. 3C). These results suggested that sequences contained within a 971 bp _cis_-regulatory element (referred to hereafter as 1 Kb enhancer element) located between 3 and 4 Kb upstream of the ctgfa translational start site are necessary to direct ctgfa expression after SCI.

An external file that holds a picture, illustration, etc. Object name is develop-150-201162-g3.jpg

Regulation of ctgfa expression during SC regeneration. (A) A series of transgene constructs were used to identify the _cis_-regulatory region that drives _ctgfa_-dependent EGFP expression after SCI. A minimum of three stable lines were generated for each transgene. (B) EGFP and Gfap immunostaining was used to assess reporter expression at 10 dpi and in uninjured control tissues. SC tissues from −5.5Kb-ctgfa:EGFP, −4Kb-ctgfa:EGFP and −3Kb-ctgfa:EGFP transgenic animals are shown. Cross-sections are shown at 150 µm (proximal) and 450 µm (distal) rostral to the lesion. Dashed lines delineate central canal edges. Arrows point to EGFP expression in ventral ependymal progenitors. (C) Quantification of EGFP fluorescence in −5.5Kb-ctgfa:EGFP, −4Kb-ctgfa:EGFP and −3Kb-ctgfa:EGFP transgenic animals. SC cross-sections 150 µm (proximal) and 450 µm (distal) rostral to the lesion were quantified. Dots indicate individual animals and sample sizes are indicated in parentheses. Error bars represent s.e.m. **P<0.01; *_P_<0.05 (unpaired _t_-test with Welch's correction). ns, not significant (_P_>0.05). Scale bar: 50 µm.

A distal enhancer element directs glial ctgfa expression after SCI

To test whether the putative 1 Kb enhancer element is sufficient to promote _ctgfa_-driven EGFP expression after SCI, we generated 1Kb-ctgfa:EGFP transgenic lines (Fig. 4A). We first tested whether larval EGFP expression in this line recapitulated _−5.5Kb_-, _−4Kb_- and _−3Kb_-ctgfa:EGFP expression at 3 days post-fertilization, finding that it did (Fig. S3A). To minimize the impact of positional effects on transgene expression, three independent 1Kb-ctgfa:EGFP lines were generated and are referred to as L1, L2 and L3. SC tissues from adult 1Kb-ctgfa:EGFP transgenic fish were transected to examine EGFP expression following injury. At 10 dpi, EGFP was not detectable in uninjured SC tissues, but was comparable to −5.5Kb- and −4Kb-ctgfa:EGFP expression in SC sections 150 µm (proximal) and 450 µm (distal) rostral to the lesion (Fig. 4B,C). This expression pattern was recapitulated in lines L1, L2 and L3. Bioinformatics analysis of the 1 Kb genomic region that directs ctgfa expression after injury revealed predicted binding sites for Tead (three binding sites), JunB (two binding sites) and Spi1 (six binding sites) transcription factors (Fig. S3B). Tead factors, which associate with the Yap/Taz co-activators to control gene transcription, are known upstream regulators of ctgfa expression in multiple systems. Consistent with their role in promoting cell proliferation and stem cell maintenance (Vassilev et al., 2001; Zhao et al., 2008), in situ hybridization showed that ctgfa expression was downregulated upon transient Yap/Taz knockdown in injured SC tissues (Fig. S3C). ctgfa expression was also downregulated in junbb CRISPR mutants (Fig. S3D), but not in spi1a mutants (Fig. S3E). These results indicated that a 1 Kb enhancer element directs injury-induced ctgfa expression during SC regeneration as well as larval expression, and suggested that ctgfa expression after SCI is induced by cooperative regulation of Tead and JunB transcription factors.

An external file that holds a picture, illustration, etc. Object name is develop-150-201162-g4.jpg

A distal enhancer element directs glial ctgfa expression after SCI. (A) Sequences within a 1 Kb enhancer element located 3-4 Kb upstream of the ctgfa transcriptional start site were combined with a 96 bp mouse Fos minimal promoter and subcloned upstream of an EGFP-expressing cassette. The clone was co-injected into one-cell-stage wild-type embryos, and three founders were isolated for propagation. (B) EGFP and Gfap immunostaining was used to assess reporter expression at 10 dpi and in uninjured tissue. SC tissues from three independent lines of 1Kb-ctgfa:EGFP transgenic animals (L1, L2 and L3) are shown. Cross-sections are shown at 150 (proximal) and 450 (distal) µm rostral to the lesion. Dashed lines delineate central canal edges. Arrows point to EGFP expression in ventral ependymal progenitors. (C) Quantification of EGFP fluorescence in −5.5Kb-ctgfa:EGFP and 1Kb-ctgfa:EGFP-L1 transgenic animals. SC cross-sections 150 µm (proximal) and 450 µm (distal) rostral to the lesion were quantified. Dots indicate individual animals and sample sizes are indicated in parentheses. Error bars represent s.e.m. ns, not significant (_P_>0.05; unpaired _t_-test with Welch's correction). (D) Schematic of wild-type ctgfa locus (ctgfaWT allele) and the targeting strategy adopted to delete sequences within the 1 Kb enhancer element upstream of ctgfa (ctgfastl683 allele). (E) Quantitative RT-PCR for ctgfa in ctgfastl683/ stl683 relative to ctgfaWT. eef1a1l1 was used as a loading control. Relative expression was normalized to eef1a1l1 expression and to expression levels in ctgfaWT siblings. SC tissues from four animals were pooled into a single biological replicate. Four biological replicates were used. Dots represent biological replicate pools. Error bars represent s.e.m. *P<0.05 (unpaired _t_-test with Welch's correction). Scale bar: 50 µm.

To test whether sequences within the 1 Kb enhancer element are required to drive endogenous ctgfa expression, we generated a CRISPR deletion line that lacks the putative enhancer region, referred to hereafter as ctgfastl683 (Fig. 4D, Fig. S4A). By quantitative RT-PCR, ctgfa expression was 50% lower in ctgfastl683/stl683 relative to wild-type siblings (Fig. 4E). However, qRT-PCR indicated that levels of the EMT marker genes cdh1, cdh2, twist1a and twist1b were not changed in ctgfastl683/stl683 animals (Fig. S4B). Moreover, swim endurance assays showed ctgfastl683/stl683 had comparable swim capacity relative to their wild-type siblings (Fig. S4C). These findings indicated that the 1 Kb enhancer element located 3-4 Kb from the ctgfa translational start site is necessary to achieve endogenous ctgfa expression levels after SCI, but is dispensable for the activation of an EMT-driving gene regulatory network and for functional SC repair.

Ablation of ctgfa+ cells impairs SC regeneration

Previous studies showed that ctgfa mutations impair glial bridging, axon regrowth and functional SC repair in zebrafish (Mokalled et al., 2016). However, because Ctgfa is a secreted matricellular molecule, it remained unclear from these reverse genetic studies whether _ctgfa_-expressing glial cells are required for SC regeneration in zebrafish. To address this outstanding question, we employed previously generated ctgfa:mCherry-Nitroreductase (ctgfa:mCherry-NTR) transgenic animals for cell-ablation studies (Klatt Shaw et al., 2021). In this system, _ctgfa_-driven expression of the bacterial NTR enzyme catalyzes the reduction of the prodrug metrodinazole (MTZ) into a cytotoxic product that induces cell death (Curado et al., 2008). To ablate ctgfa+ progenitors and early bridging glia, we subjected ctgfa:mCherry-NTR (Tg+) fish and their wild-type (Tg−) siblings to complete SC transections, followed by two consecutive treatments with 0.1 mM MTZ for 24 h at 4 and 7 dpi (Fig. 5A). Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays, immunohistochemistry for PCNA, and ctgfa in situ hybridization were performed at 9 dpi to assess the extents of cell death, cell proliferation and ctgfa expression using this ablation regimen (Fig. 5A). TUNEL staining revealed that the profiles of mCherry+ TUNEL+ cells were significantly increased in MTZ-treated NTR animals compared with vehicle-treated siblings (Fig. 5B,C). Concomitant with increased cell death, the proportions of mCherry+ PCNA+ cells were elevated threefold in MTZ-treated NTR fish relative to controls (Fig. 5D,E). To assess ctgfa+ cell ablation further, we found that ctgfa expression was markedly reduced in MTZ-treated NTR fish (Fig. 5F,G). These studies established a genetic ablation model for ctgfa+ cells and were subsequently used to examine the requirement for ctgfa+ cells during SC regeneration.

An external file that holds a picture, illustration, etc. Object name is develop-150-201162-g5.jpg

Ablation of ctgfa+ cells during SC regeneration. (A) Experimental timeline to establish a genetic ablation model for ctgfa+ cells after SCI. Transgenic _ctgfa:_mCherry-2A-NTR animals and control siblings were subjected to complete SC transection and metronidazole (MTZ) treatment at 4 and 7 dpi to ablate ctgfa+ cells. TUNEL staining, PCNA staining and ctgfa in situ hybridization were performed to evaluate the extent of cell death and cell proliferation at 9 dpi. (B,C) TUNEL staining in _ctgfa:_mCherry-2A-NTR animals (Tg+) 9 dpi. Tissue sections from vehicle- and MTZ-treated animals are shown. Representative micrographs shown in B are 150 µm rostral to the lesion. The proportion of mCherry+ TUNEL+ cells was normalized to total nuclei number and shown in C. (D,E) PCNA staining in _ctgfa:_mCherry-2A-NTR animals (Tg+) at 9 dpi. Tissue sections from vehicle- and MTZ-treated animals are shown. Representative micrographs shown in D are 450 µm rostral to the lesion. The proportion of mCherry+ PCNA+ cells was normalized to total nuclei number and shown in E. In B,D, arrowheads in high-magnification insets indicate mCherry+ TUNEL+ and mCherry+ PCNA+ cells in triple-, double- or single-channel views. (F,G) ctgfa in situ hybridization in _ctgfa:_mCherry-2A-NTR animals (Tg+) at 9 dpi. Tissue sections from vehicle- and MTZ-treated animals are shown. Representative micrographs shown in F are 450 µm rostral to the lesion. Dashed red lines delineate central canal edges. Quantification of ctgfa+ expression domains is shown in G. For all quantifications, dots represent individual animals and animal numbers are indicated in parentheses. Error bars represent s.e.m. **P<0.01; *P<0.05 (unpaired _t_-test with Welch's correction). Scale bars: 100 µm.

To determine the impact of ctgfa+ cell ablation during SC regeneration, we subjected ctgfa:mCherry-NTR (Tg+) fish and their wild-type (Tg−) siblings to complete SC transections, followed by two consecutive treatments with 0.1 mM MTZ for 24 h at 4 and 7 dpi. Glial bridging, axon regrowth, swim endurance and swim behavior assays were then performed at 14 dpi (Fig. 6A). For glial bridging, Gfap immunostaining was used to determine the areas of glial bridges at the lesion site relative to the areas of intact SC tissues (Fig. 6B,C). Percent bridging at the lesion site averaged 20% in MTZ-treated Tg− controls, and decreased to 6% in MTZ-treated Tg+ animals (Fig. 6D). For anterograde axon tracing, biocytin was applied at the hindbrain level and biocytin-labeled axons were traced at 600 µm (proximal) and 1500 µm (distal) caudal to the lesion (Fig. 6E,F). In this assay, axon regrowth was attenuated by 68% in proximal SC tissues from MTZ-treated Tg+ animals compared with MTZ-treated Tg− controls (Fig. 6G). These results indicated ctgfa+ cells are required for the cellular regeneration of glial and axonal bridges after SCI.

An external file that holds a picture, illustration, etc. Object name is develop-150-201162-g6.jpg

ctgfa+ cells are required for SC regeneration. (A) Experimental timeline to evaluate the requirement of ctgfa+ cells after SCI. Transgenic _ctgfa:_mCherry-2A-NTR animals and control siblings were subjected to complete SC transection and metronidazole (MTZ) treatment at 4 and 7 dpi to ablate ctgfa+ cells. Swim assays were performed and SC tissues were collected at 14 dpi to assess regeneration. (B-D) Glial bridging in _ctgfa:_mCherry-2A-NTR animals (Tg+) and wild-type siblings (Tg−) at 14 dpi. Tg+ and Tg− animals were treated with MTZ. Representative immunohistochemistry shows the Gfap+ bridge at the lesion site in C. Percentage bridging represents the cross-sectional area of the glial bridge at the lesion site relative to the intact SC 750 µm rostral to the lesion in D. (E-G) Anterograde axon tracing in _ctgfa:_mCherry-2A-NTR animals (Tg+) and wild-type siblings (Tg−) at 14 dpi. Biocytin axon tracer was applied rostrally and analyzed at 600 µm (proximal) and 1500 µm (distal) caudal to the lesion (E). Representative images show the extent of biocytin labeling 600 µm (proximal) caudal to the lesion in F. Axon growth was first normalized to the extent of biocytin at 450 and 750 µm rostral to the lesion, and then to labeling in Tg− controls at the proximal level in G. (H) Swim endurance assay for _ctgfa:_mCherry-2A-NTR animals and wild-type siblings at 14 dpi. This behavioral test was used to assess the capacity of _ctgfa_-ablated animals to swim in an enclosed swim tunnel under increasing water current velocities. The time at exhaustion for each fish is shown. (I,J) Swim behavior assays were used to assess the performance of _ctgfa:_mCherry-2A-NTR animals and wild-type siblings at minimal water current velocity at 14 dpi. Average Y-position in the tunnel (I) and burst frequency (J) were quantified at 10 cm/s water current velocity. For all quantifications, dots represent individual animals and animal numbers are indicated in parentheses. Error bars represent s.e.m. ***P<0.001; **P<0.01; *P<0.05 (unpaired _t_-test with Welch's correction for D and G; one-way ANOVA with multiple comparisons for H-J). Scale bars: 100 µm.

To evaluate the functional impact of ablating ctgfa+ cells after SCI, we first assessed the swim capacities of ablated animals in an enclosed swim tunnel under increasing water current velocities (Fig. 6H) (Burris et al., 2021). In this swim endurance assay, control animals showed comparable swim functions, averaging between 22 min for vehicle-treated Tg+ fish and 27 min for MTZ-treated Tg− fish. However, swim endurance was significantly diminished in MTZ-treated Tg+ animals, with a swim time average of 17 min. We then tracked the swim behavior of ablated fish under a constant, low current velocity of 10 cm/s (Burris et al., 2021). Fish position in the swim tunnel (Y position) and burst frequency were quantified to assess overall swim competence. In these swim behavior assays, MTZ-treated Tg+ animals stalled in the back quadrant of the swim tunnel (Fig. 6I), and displayed 32% less-frequent bursts under low current velocity (Fig. 6J). In control experiments, treating Tg+ animals with two doses of MTZ in the absence of injury did not impact the swim capacity of MTZ-treated Tg+ animals (Fig. S5A,B). Similarly, local application of recombinant C-terminal Ctgf peptides onto the SC lesions of MTZ-treated Tg+ animals was not sufficient to measurably rescue the swim capacity of animals with ctgfa+ cell ablation (Fig. S5C,D). Together, these experiments indicated that ablation of ctgfa+ cells impairs cellular and functional recovery after SCI, and that ctgfa+ glial cells play an essential pro-regenerative role during SC regeneration in zebrafish.

DISCUSSION

This study explores the contribution, regulation and requirements of glial cells during innate spinal cord repair in zebrafish. Our genetic lineage-tracing and cell-ablation results are consistent with a model in which _ctgfa_-expressing progenitor and glial cells contribute to and are required for glial bridging and functional recovery after SCI.

Following SC transection injuries, the severed SC stumps retract away from the lesion core. Rapid activation and proliferation of ependymal progenitors seal the severed central canal at both ends of the lesion. As regeneration proceeds, the SC stumps are attached by newly formed regenerate tissues, the central canal reconnects, and ependymal progenitors regain their epithelial-like, central canal-lining morphology. Our genetic lineage-tracing studies showed that ctgfa+ cells give rise to regenerating glia, including bridging glia, with minimal contribution to neuronal or oligodendrocyte lineages. These results are consistent with a model in which ctgfa+ progenitors differentiate into glia after SCI. Our lineage-tracing studies employed tamoxifen labeling at 4 dpi, a time point at which ctgfa is highly expressed but not restricted to ventral ependymal progenitors. Specifically, ctgfa expression at 4 dpi marks early bridging glia at the lesion site in addition to additional cell types around injured SC tissues. We thus could not rule out the possibilities that ctgfa+ glia labeled at 4 dpi proliferate and give rise to more bridging glia, or that ctgfa+ cells around the lesion transdifferentiate into glia during SC regeneration. We also showed that _ctgfa_-driven lineage tracing labels ventral ependymal progenitors at 28 dpi, suggesting that ctgfa+ progenitors undergo self-renewal and contribute to the newly formed ventral ependyma within regenerate tissues. An alternative possibility is that ctgfa+ glia may undergo dedifferentiation into ependymal progenitors at later stages of regeneration. We propose that devising future strategies that enable differential targeting of ctgfa+ progenitors and ctgfa+ glia are needed to improve our understanding of the potency and proliferative capacities of progenitor and glial cells during SC regeneration.

Previous studies have implicated Fgf and Ctgf signaling in glial bridge formation in adult zebrafish. The glial bridging functions of Fgf signaling were established using transgenic dominant-negative Fgf receptor manipulations, sprouty mutants and Fgf8 injection (Goldshmit et al., 2012). ctgfa genetic loss of function was shown to impair glial bridging, axon regrowth and functional SC repair (Mokalled et al., 2016). However, because Fgf and Ctgf are secreted extracellular proteins, the glial requirements of these molecules during bridging and of glial bridging during SC regeneration remained unclear. Previous studies suggested axon regrowth proceeds independently of the projection of glial processes across the lesion (Dervan and Roberts, 2003; Briona et al., 2015; Wehner et al., 2017). Specifically, using a _gfap_-driven Nitroreductase-expressing transgenic line, Wehner et al. showed that axon regrowth was unaffected following Gfap+ cell ablation in larval zebrafish (Wehner et al., 2017). Here, we show that ablation of ctgfa+ cells is sufficient to impair glial bridging, axon regrowth and functional recovery after SCI, indicating ctgfa+ cells are required during SC repair. Noting the technical limitations of the NTR-MTZ system in simultaneously achieving complete and specific cell ablation, it is possible that a small proportion of Gfap+ cells escaped cell ablation in the larval studies by Wehner et al., or that our _ctgfa_-driven studies ablated non-neural ctgfa+ cells in addition to ctgfa+ bridging glia. However, regenerating axons often associate with elongated bipolar astrocytes across vertebrates, including zebrafish and mammals (White et al., 2008; Filous et al., 2010; Goldshmit et al., 2012; Zukor et al., 2013; Mokalled et al., 2016). Bridging glia from adult zebrafish retain an elongated morphology (Goldshmit et al., 2012; Mokalled et al., 2016), and possess a mesenchymal signature that correlates with increased plasticity and immature astrocytic cell identity (Klatt Shaw et al., 2021). Similarly, in adult mice a small proportion of elongated astrocytes, termed ‘astroglial bridges’, correlate with increased axon regrowth under genetic manipulations, such as PTEN deletion (Zukor et al., 2013). We thus propose that the requirements for axon regrowth may differ between larval and adult zebrafish, and that future comparative studies will shed light on the molecular similarities and differences between murine elongated astrocytes and zebrafish bridging glia.

Our findings demonstrate that glial bridging is an effective, natural mechanism of SC regeneration and establish glial cell responses are pro-regenerative and indispensable to achieve SC repair. We suggest that further investigation into glial cell fates will springboard translational applications to improve bridging and regeneration in the mammalian CNS.

MATERIALS AND METHODS

Zebrafish

Adult zebrafish of the Ekkwill, Tubingen and AB strains were maintained at the Washington University Zebrafish Core Facility. All animal experiments were performed in compliance with institutional animal protocols. Male and female animals between 4 and 6 months of age and of ∼2 cm in length were used. Experimental fish and control siblings of similar size and equal sex distribution were used for all experiments. SC transection surgeries and regeneration analyses were performed in a blinded manner, and two to four independent experiments were repeated using different clutches of animals for all experiments. Control and experimental animals in which SC tissues were transected were housed in equal numbers (four to seven fish) in 1.1 l tanks. The following previously published zebrafish strains were used: Tg(ctgfa:mCherry-NTR)stl650 (Klatt Shaw et al., 2021), Tg(−5.5Kb-ctgfa:EGFP)pd96 (Mokalled et al., 2016), junbbstl672 (Klatt Shaw et al., 2021) and spi1astl669 (Klatt Shaw and Mokalled, 2021). yap1/taz crispants were generated as previously described (Klatt Shaw and Mokalled, 2021). Newly constructed strains are described below.

Generation of transgenic Tg(ctgfa:CreERT2) zebrafish

The following primers were used to amplify a 5.5 kb genomic DNA region upstream of the ctgfa translational start site: ClaI forward primer 5′-atcgattttggctactttcagctaagactgg-3′ and ClaI reverse primer 5′-atcgattctttaaagtttgtagcaaaaagaaa-3′. The genomic fragment was cloned into a PCR2.1-TOPO vector, then subcloned into a ClaI-digested PCS2-CreERT2 plasmid to generate a ctgfa:CreERT2 clone. The clone was co-injected into one-cell-stage wild-type embryos with I-SceI. Multiple founders were isolated and propagated. The full name of this line is Tg(ctgfa:CreERT2)stl652. ctgfa:CreERT2 animals were crossed into the _ubiquitin:_loxP-GFP-STOP-loxP-mCherry transgene (Mosimann et al., 2011) to generate ctgfa:Tracer animals.

Generation of transgenic Tg(−4Kb-ctgfa:EGFP) and Tg(−3Kb-ctgfa:EGFP) zebrafish

The following forward primers were used to amplify 4 and 3 Kb genomic region upstream of the ctgfa translational start site: ClaI 4 Kb forward primer 5′-ccatcgataggcagcaatagcgtcagat-3′ and ClaI 3 Kb forward primer 5′-ccatcgattttgacccctctcagtgaa-3′. A common reverse primer was used to amplify 4 and 3 Kb genomic regions: ClaI reverse primer 5′-ccatcgatttctttaaagtttgtagcaaaaaaga-3′. The genomic fragments were cloned into a PCR2.1-TOPO vector, then subcloned into a ClaI-digested PCS2-EGFP plasmid. Clones were co-injected into one-cell-stage wild-type embryos with I-SceI. A minimum of three founders were isolated and propagated for each transgene. The full names of these lines are Tg(_−4Kb_-ctgfa:EGFP)stl656 and Tg(_−3Kb_-ctgfa:EGFP)stl657. Animals were analyzed as hemizygotes.

Generation of transgenic Tg(1Kb-ctgfa:EGFP) zebrafish

The following primers were used to amplify a 1Kb genomic region upstream of the ctgfa translational start site: ClaI forward primer 5′-ccatcgatccacaaggctattgcaacg-3′ and ClaI reverse primer 5′-ccatcgatagtatgcacctattcactgag-3′. The genomic fragments were cloned into a PCR2.1-TOPO vector, then subcloned into ClaI-digested PCS2-fos-EGFP plasmid. The clone was co-injected into one-cell-stage wild-type embryos with I-SceI. Three founders were isolated and propagated. The full names of these lines are Tg_(1Kb_-ctgfa:EGFP-L1)stl658, Tg_(1Kb_-ctgfa:EGFP-L2)stl659 and Tg_(1Kb_-ctgfa:EGFP-L3)stl660. Animals were analyzed as hemizygotes.

Generation of ctgfastl683/stl683 zebrafish

CRISPR/Cas9 design, mutagenesis and screening was performed as previously described (Klatt Shaw and Mokalled, 2021). Briefly, crRNA guide RNA sequences were selected using CHOPCHOP (https://chopchop.cbu.uib.no/). For dgRNA #1 (5′-GTAACTATTACAGGTTCAGCAGG-3′), the PAM site is 80 bp upstream of 1 Kb enhancer element). For dgRNA #2 (5′-ACTGGTAGGAGTGACGAGGACGG-3′), the PAM site 37 bp downstream of 1 Kb enhancer end. Alt-R tracrRNA and crRNA gRNAs (IDT, 1072534) were annealed at a final concentration of 50 µM into dgRNA duplexes. Alt-R S.p. Cas9 nuclease V3 (IDT, 1081059) was diluted in 1 M HEPES at pH 7.5, 2 M KCl to a working concentration of 25 μM. Annealed dgRNA duplexes were diluted 1:1 in duplex buffer (IDT, 11-05-01-03) to a working concentration of 25 µM. Equal volumes of dgRNA were added to Cas9 protein and incubated at 37°C for 5 min. One nanoliter of CRISPR/Cas9 solutions was injected at the one-cell stage. The full name of this allele is ctgfastl683.

Spinal cord transection and treatments

Zebrafish were anaesthetized in 0.2 g/l of MS-222 buffered to pH 7.0. Fine scissors were used to make a small incision that transects the spinal cord 4 mm caudal to the brainstem region. Complete transection was visually confirmed at the time of surgery. Injured animals were also assessed at 2 or 3 dpi to confirm loss of swim capacity post-surgery.

Tamoxifen treatment

Tamoxifen (Sigma-Aldrich, T5648) was reconstituted into a 1 mM stock solution in ethanol. ctgfa:Tracer fish were treated with a 5 µM working solution at 4 dpi for 24 h.

MTZ treatment

To ablate ctgfa+ cells, ctgfa:mCherry-NTR fish and wild-type siblings were subjected to complete SC transections, followed by two consecutive treatments with 0.1 mM MTZ in fish water (Sigma-Aldrich, M1547) for 24 h at 4 and 7 dpi. TUNEL assay, PCNA staining and ctgfa in situ hybridization were performed at 9 dpi to assess the extents of cell death, cell proliferation and ctgfa expression using this ablation regimen. Glial bridging, axon tracing and swim function were assessed at 14 dpi.

CTGF-CT treatment

Treatment with recombinant CTGF-CT was performed as previously described (Mokalled et al., 2016; Saraswathy et al., 2022). Lyophilized human CTGF-CT peptide (eBioscience, 14-8503-80) was reconstituted in ddH2O to a concentration 50 ng/µl. Zebrafish were anaesthetized using MS-222. Two microliters (100 ng) of reconstituted peptides were injected at 9 dpi adjacent and lateral to the SC lesion site. Two microliters of ddH2O were injected for vehicle controls.

Quantitative RT-PCR

For RNA collection, 2 mm SC sections including the lesion site plus additional rostral and caudal tissue proximal to the lesion were collected at 10 dpi. Tissue samples were homogenized in Trizol, then incubated for 3 min at room temperature in chloroform. RNA was pelleted following isopropanol incubation (>2 h at −80°C), then resuspended in DEPC-treated water. For cDNA synthesis, 1 µg of RNA was converted into cDNA with the Maxima First Strand cDNA Synthesis Kit (Thermo Fisher Scientific, K1672) according to the manufacturer's specifications. Quantitative PCR was completed using the Luna polymerase master mix (NEB, M3003) using gene-specific primers. Previously published primers were designed to flank introns and were confirmed to not amplify product from genomic DNA (Klatt Shaw et al., 2021). To determine primer efficiency, a standard curve was generated for each primer set using cDNA pooled from 1 week post-injury SCs. Only primer sets with a calculated amplification efficiency per cycle of 1.8-2.2 were used. qRT-PCR was performed on a Bio-Rad CFX Connect Real-Time System. Expression fold change for each gene was calculated using the ΔCq method and normalized to fold change eef1a1l1 expression compared with controls. For all qRT-PCR experiments, four biological replicates were used per genotype and four SC tissues were pooled for each replicate. Each biological replicate was run in technical duplicate, and the average log2 fold change of these technical replicates is shown.

Histology

Sixteen-micron-thick cross cryosections or 25-µm-thick longitudinal cryosections of paraformaldehyde-fixed SC tissues were used. Tissue sections were imaged using a Zeiss AxioVision compound microscope or a Zeiss Axioscan.Z1 slide scanner for in situ hybridization, and a Zeiss LSM 800 confocal microscope or a Zeiss Axioscan.Z1 slide scanner for immunofluorescence.

For in situ hybridization, a previously cloned ctgfa probe was used (Mokalled et al., 2016). Linearized vectors were used to generate digoxygenin-labeled cRNA probes. In situ hybridization assays were performed as previously described (Mokalled et al., 2016).

In situ HCR was performed as previously described (Choi et al., 2018). Briefly, ctgfa HCR probes were designed using a python script (IDT). Tissue sections were hybridized for 48 h at 37°C, then incubated in amplification buffer for 1 h. Tissue sections were then incubated in amplification buffer for 24 h at room temperature before proceeding to immunohistochemistry. The hybridization, wash and amplification buffers were purchased from Molecular Instruments.

Primary antibodies used in this study were: rabbit anti-dsRed (Clontech, 632496; 1:200), mouse anti-GFAP (ZIRC, ZRF1; 1:500), mouse anti-HuC/D (Invitrogen, A21271; 1:500), mouse anti-Hb9 (Developmental Studies Hybridoma Bank, AB2145209; 1:50), rabbit anti-Mbp (a generous gift from B. Appel's lab, University of Colorado, Aurora, USA, 1:500), chicken anti-GFP (Aves Labs, GFP-1020; 1:1000) and rabbit anti-PCNA (GeneTex, GTX124496; 1:500). Secondary antibodies used in this study were: Alexa Fluor 488, Alexa Fluor 594, and Alexa 647 goat anti-rabbit or anti-mouse antibodies (Jackson ImmunoResearch, 1:200, 103-545-155, 111-605-003, 111-585-003, 115-605-003 and 115-585-003). Nuclear Hoechst staining was performed according to the manufacturer's recommendation (Thermo Fisher Scientific, H3570). TUNEL staining was performed following the manufacturer's protocol (Thermo Fisher Scientific, C10617).

Quantification

Colocalization analysis

Colocalization of _ctgfa_-Tracer with Gfap and of _ctgfa_-Tracer with Mbp were performed using the JACoP plugin in Fiji. Orthogonal projections of individual image stacks were generated using Zen software. The polygon selection tool was used to outline the spinal cord perimeters to define regions of interests. Thresholds were user-defined for the _ctgfa_-Tracer channel and Gfap/Mbp channel. M1 and M2 coefficients were calculated. M1 is defined as the ratio of the summed intensities of pixels from the _ctgfa_-Tracer channel for which the intensity in the Gfap/Mbp channel is above zero to the total intensity in the Gfap/Mbp channel. M2 is defined as the ratio of the summed intensities of pixels from the _ctgfa_-Tracer channel for which the intensity in the Gfap/Mbp channel is above zero to the total intensity in the _ctgfa_-Tracer channel. M1 and M2 coefficients were multiplied by 100 to obtain percentage expression. Two-way ANOVA and multiple comparisons were performed using Prism software to determine statistical significance of swim times between groups.

Cell counting

Cell counting of _ctgfa_-Tracer with HuC/D and of _ctgfa_-Tracer with Hb9 were performed using a customized Fiji script (adapting ITCN: Image based Tool for counting nuclei; https://imagej.nih.gov/ij/plugins/itcn.html). Orthogonal projections of individual image stacks were generated using Zen software. A customized Fiji script incorporated user-defined inputs to define channels (including Hoechst) and to outline SC perimeters. To quantify nuclei, the following parameters were set in ITCN counter: width, 15; minimal distance, 7.5; threshold, 0.4. Once nuclei were identified, user-defined thresholds of individual cell markers were used to mask the image and identify nuclei located inside the masked regions. xy coordinates were extracted for each nucleus for cell counting. Raw counts and xy coordinates from Fiji were processed using a customized R script. Two markers were considered overlapping if they shared nuclei with same xy coordinates.

Quantification of ctgfa:EGFP expression

Images were thresholded by an operator unaware of the treatment groupings, and the areas of the thresholded signal were measured. To calculate percentage EGFP+ area, the thresholded areas were normalized to total SC area for each section. Unpaired _t_-tests with Welch's correction were performed using Prism software to determine statistical significance between groups.

Quantification of ctgfa in situ hybridization

Images were converted into a 16-bit tiff and a horizontal line was drawn at the center of the central canal to separate the dorsal and ventral sections of the SC. Images were thresholded by an operator unaware of the treatment groupings, and the areas of the thresholded signal were measured in ventral SC tissues. To calculate percentage ctgfa+ area, the thresholded areas were normalized to total SC area for each section. Unpaired _t_-tests with Welch's correction were performed using Prism software to determine statistical significance between groups.

Glial bridging

GFAP immunohistochemistry was performed on serial transverse sections. The cross-sectional area of the glial bridge (at the lesion site) and the area of the intact SC (750 µm rostral to the lesion) were measured using ImageJ software. Bridging was calculated as a ratio of these measurements. Unpaired _t_-tests with Welch's correction were performed using Prism software to determine statistical significance between groups.

Axon tracing

Anterograde axon tracing was performed on adult fish at 28 dpi. Fish were anaesthetized using MS-222 and fine scissors were used to transect the cord 4 mm rostral to the lesion site. Biocytin-soaked Gelfoam Gelatin Sponge was applied at the new injury site (Gelfoam, Pfizer, 09-0315-08; biocytin, saturated solution, Sigma-Aldrich, B4261). Fish were euthanized 6 h post-treatment and biocytin was detected histologically using Alexa Fluor 594-conjugated streptavidin (Molecular Probes, S-11227). Biocytin-labeled axons were quantified using the ‘threshold’ and ‘particle analysis’ tools in the Fiji software. Four sections per fish at 600 µm (proximal) and 1500 µm (distal) caudal to the lesion core, and two sections 450 and 750 µm rostral to the lesion, were analyzed. Axon growth was normalized to the efficiency of biocytin labeling rostral to the lesion for each fish. Percentage axon growth was then normalized to the rostral level of the control group. Unpaired _t_-tests with Welch's correction were performed using Prism software to determine statistical significance between groups.

Swim endurance assays

Zebrafish were exercised in groups of 8-12 in a 5 l swim tunnel device (Loligo, SW100605L, 120 V/60 Hz). After 10 min of acclimation inside the enclosed tunnel, water current velocity was increased every 2 min and fish swam against the current until they reached exhaustion. Exhausted animals were removed from the chamber without disturbing the remaining fish. Swim times at exhaustion were recorded for each fish. Results are expressed as mean±s.e.m. One-way ANOVA and multiple comparisons were performed using the Prism software to determine statistical significance of swim times between groups.

Swim behavior assays

Zebrafish were divided into groups of five in a 5 l swim tunnel device (Loligo, SW100605L, 120 V/60 Hz). Each group was allowed to swim for a total of 15 min under zero to low current velocities (5 min at 0 cm/s, 5 min at 10 cm/s, and 5 min at 20 cm/s). The entire swim behavior was recorded using a high-speed camera (iDS, USB 3.0 color video camera) with following settings: aspect ratio, 1:4; pixel clock, 344; frame rate, 70 frames/s; exposure time: 0.29; aperture, 1.4-2; maximum frames; 63,000. Movies were converted to 20 frames/s and analyzed using a customized Fiji macro. For each frame, animals/objects >1500 px2 were identified, and the xy coordinates were derived for each animal/object. Frame were independently, and animal/object tracking was completed using a customized R Studio script. The script aligned coordinates and calculated swim metrics considering three separate frame windows (frames 0-6000 at 0 cm/s; frames 6001-12,000 at 10 cm/s, and frames 12,001-18,001 at 20 cm/s). One-way ANOVA and multiple comparisons were performed using the Prism software to determine statistical significance of swim times between groups.

Supplementary Material

Supplementary information

Acknowledgements

We thank V. Cavalli, A. Johnson and L. Solnica-Krezel for discussions; B. Appel for sharing the Mbp antibody; and the Washington University Zebrafish Shared Resource for animal care.

Footnotes

Author contributions

Conceptualization: M.H.M.; Methodology: L.Z.; Formal analysis: L.Z.; Investigation: L.Z., A.R.M., H.Y., B.B., D.K.S., K.O.; Resources: K.D.P.; Data curation: L.Z.; Writing - original draft: M.H.M.; Writing - review & editing: K.D.P., M.H.M.; Visualization: L.Z.; Supervision: K.D.P., M.H.M.; Funding acquisition: K.D.P., M.H.M.

Funding

This research was supported by grants from the National Institutes of Health (R21 NS124635 and R21 NS096617 to K.D.P.; R01 NS113915 and R01 NS123708 to M.H.M.). Deposited in PMC for release after 12 months.

Data availability

All relevant data can be found within the article and its supplementary information.

References

Articles from Development (Cambridge, England) are provided here courtesy of Company of Biologists