Home
Search
Collections
Journals
About
Contact us
My IOPscience
Local delivery of thyroid hormone enhances oligodendrogenesis and myelination after spinal cord injury
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2017 J. Neural Eng. 14 036014 (http://iopscience.iop.org/1741-2552/14/3/036014) View the table of contents for this issue, or go to the journal homepage for more Download details: IP Address: 193.140.216.7 This content was downloaded on 04/04/2017 at 04:42 Please note that terms and conditions apply.
You may also be interested in: Potential of human dental stem cells in repairing the complete transection of rat spinal cord Chao Yang, Xinghan Li, Liang Sun et al. Injectable hydrogels for central nervous system therapy Malgosia M Pakulska, Brian G Ballios and Molly S Shoichet Biomaterial-based drug delivery systems for the controlled release of neurotrophic factors Nima Khadem Mohtaram, Amy Montgomery and Stephanie M Willerth Metal ion-assisted self-assembly of complexes for controlled and sustained release of minocycline for biomedical applications Zhiling Zhang, Zhicheng Wang, Jia Nong et al. Hyaluronic acid limits scarring after spinal cord injury Zin Z Khaing, Brian D Milman, Jennifer E Vanscoy et al. Recent progress in stem cell differentiation directed by material and mechanical cues Xunxun Lin, Yuan Shi, Yilin Cao et al. Positive and negative cues for modulating neurite dynamics and receptor expression Melissa R Wrobel and Harini G Sundararaghavan Bioengineered fibrin-based niche to direct outgrowth of circulating progenitors into neuron-like cells for potential use in cellular therapy S Tara and Lissy K Krishnan
Journal of Neural Engineering J. Neural Eng. 14 (2017) 036014 (13pp)
https://doi.org/10.1088/1741-2552/aa6450
Local delivery of thyroid hormone enhances oligodendrogenesis and myelination after spinal cord injury Robert B Shultz, Zhicheng Wang, Jia Nong, Zhiling Zhang and Yinghui Zhong1 School of Biomedical Engineering, Science and Health Systems, Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104, United States of America E-mail: [email protected] Received 20 October 2016, revised 1 March 2017 Accepted for publication 3 March 2017 Published 30 March 2017 Abstract
Objective. Traumatic spinal cord injury (SCI) causes apoptosis of myelin-forming oligodendrocytes (OLs) and demyelination of surviving axons, resulting in conduction failure. Remyelination of surviving denuded axons provides a promising therapeutic target for spinal cord repair. While cell transplantation has demonstrated efficacy in promoting remyelination and functional recovery, the lack of ideal cell sources presents a major obstacle to clinical application. The adult spinal cord contains oligodendrocyte precursor cells and multipotent neural stem/progenitor cells that have the capacity to differentiate into mature, myelinating OLs. However, endogenous oligodendrogenesis and remyelination processes are limited by the upregulation of remyelination-inhibitory molecules in the post-injury microenvironment. Multiple growth factors/molecules have been shown to promote OL differentiation and myelination. Approach. In this study we screened these therapeutics and found that 3, 3′, 5-triiodothyronine (T3) is the most effective in promoting oligodendrogenesis and OL maturation in vitro. However, systemic administration of T3 to achieve therapeutic doses in the injured spinal cord is likely to induce hyperthyroidism, resulting in serious side effects. Main results. In this study we developed a novel hydrogel-based drug delivery system for local delivery of T3 to the injury site without eliciting systemic toxicity. Significance. Using a clinically relevant cervical contusion injury model, we demonstrate that local delivery of T3 at doses comparable to safe human doses promoted new mature OL formation and myelination after SCI. Keywords: drug delivery, thyroid hormone, hydrogel, spinal cord injury, myelination, oligodendrogenesis (Some figures may appear in colour only in the online journal)
1. Introduction
matter [2]. Studies have shown that less than 10% of axons in specific spinal tracts were necessary for locomotor, forepaw reaching, and tactile sensory functions [3]. However, prolonged and widespread oligodendrocyte (OL) loss following SCI leads to demyelination of spared axons, resulting in conduction failure [2–4]. Moreover, the demyelinated axons are susceptible to degeneration due to a lack of trophic support from myelin-forming cells [5]. Thus, replacement of lost OL
Traumatic spinal cord injury (SCI) results in significant loss of motor, sensory, and autonomic functions. Contusion injury is the most frequent form of SCI [1]. Even after severe contusive SCI, surviving axons persist in the subpial rim of white 1
Author to whom any correspondence should be addressed.
1741-2552/17/036014+13$33.00
1
© 2017 IOP Publishing Ltd Printed in the UK
R B Shultz et al
J. Neural Eng. 14 (2017) 036014
2. Materials and methods
population and remyelination of the denuded surviving axons represents a promising therapeutic strategy to promote functional recovery and improve the quality of life for patients suffering from SCI. The adult spinal cord contains oligodendrocyte precursor cells (OPCs) that have the capacity to differentiate into mature, myelin-forming OLs [6]. In addition, multipotent neural stem/ progenitor cells (NSPCs) capable of generating OLs, astrocytes, and neurons are present in the adult rat and human spinal cord [7]. However, endogenous oligodendrogenesis and remyelination processes are severely limited by the upregulation of remyelination-inhibitory molecules in the post-injury microenvironment [8–11]. These molecules include tumor necrosis factor α (TNFα), interleukin-1β (IL-1β), bone morphogenic proteins 2 and 4 (BMP-2 and 4), and chondroitin sulfate proteoglycans (CSPGs) [12–15]. While cell transplantation has demonstrated efficacy in promoting remyelination and functional recovery [16], the lack of ideal cell sources presents a major obstacle to clinical application. Potential safety concerns limit allogeneic cell transplantation, while lengthy expansion time and/or invasive isolation limit the use of autogeneic cells. Therefore, it is important to develop a therapy that can promote endogenous OPC/NSPC differentiation, maturation and remyelination. A number of growth factors/molecules have been shown to promote OL differentiation and myelination, including brainderived neurotrophic factor (BDNF), neurotrophin-3 (NT3), sonic hedgehog (Shh), and thyroid hormone 3,3′,5-triiodothyronine (T3) [17–20]. In the present study, we compared the efficacy of these molecules in promoting OL differentiation and maturation in vitro and found that T3 is the most effective compared to all the molecules. In addition, we found that T3 was effective in a wide range of 10–2000 ng ml−1. However, average T3 concentration in serum was around 1.2 ± 0.35 ng ml−1, and T3 levels in patients suffering from hyperthyroidism was found to be 5.7 ± 3.5 ng ml−1 [21]. Thus, increasing systemic T3 level to 10 ng ml−1 or even higher is likely to induce hyperthyroidism, resulting in serious side effects including anxiety, muscle weakness, heart palpitations, and thyrotoxic crisis [22, 23]. Local delivery can potentially expose the injured spinal cord tissue to therapeutic levels of T3, while avoiding deleterious side effects from systemic exposure. Hydrogel-based delivery systems have received much attention to locally deliver drugs to the injured spinal cord [24–27]. Hydrogel systems are suitable for this application because they can be fabricated from biocompatible, naturally occurring materials, and can immobilize drug or drugloaded particles at the injury site to provide sustained localized release. In addition, hydrogels exhibit mechanical properties compatible with native spinal cord tissue, minimizing the risk of additional tissue damage associated with mechanical mismatch. In this study, we developed an agarose hydrogel-based T3 delivery system capable of controlled release of physiologically relevant doses of T3, and investigated the efficacy of local delivery of T3 in promoting oligodendrogenesis using a clinically relevant unilateral cervical spinal cord contusion injury model.
2.1. Neural stem and progenitor cell (NSPC) culture
NSPCs were isolated from E17 Sprague Dawley rat embryos in accordance with protocols approved by Drexel University’s IACUC committee. The cells were isolated and cultured following a procedure modified from a protocol reported by Pedraza et al [28]. In brief, cerebrum tissue was mechanically dissociated by trituration. Then the tissue was filtered through a 70 µm cell strainer and plated in plastic tissue culture flasks. The cells were allowed to form neurospheres in basal medium (DMEM/F12, 2% B27 supplement, and 1% Penicillin/Streptomycin) supplemented with 20 ng ml−1 of basic fibroblast growth factor (bFGF, Peprotech) and plateletderived growth factor (PDGF, Peprotech). NSPCs at 3–4 passages were used in the subsequent experiments. 92.8 ± 2.7% of the isolated NSPCs were positive for Nestin, a marker for neural stem and progenitor cells. The NSPC cultures also contained a low level of glial fibrillary acidic protein+ (GFAP+) astrocytes (7.2 ± 1.3%), and negligible levels of O4+ (mature and immature OL) and myelin basic protein+ (MBP+; mature OL) cells. For differentiation experiments, NSPCs were chemically dissociated using a chemical dissociation kit (Stemcell Technologies) and plated at a density of 2.5 × 104 cells cm−2 onto polyethylenimine (PEI)-coated 48-well plates. The cells were allowed to attach overnight, and then the culture medium was replaced with basal medium containing different OL differentiation molecules. The medium was changed every 2 d. The cells were allowed to differentiate for 7 d. 2.2. Oligodendrocyte precursor cell (OPC) culture
OPCs were isolated from neurospheres via immunopanning for PGDF receptor type α (PDGFRα, early OPC marker) [29]. Briefly, non-tissue culture treated dishes were coated with 10 µg ml−1 goat anti-rabbit IgG (Jackson Immunoresearch) in 50 mM TRIS-HCl (pH 9.5) overnight at 4 C. After washing 3 times with Hank’s Balanced Salt Solution (HBSS), the dishes were incubated with 1 µg ml−1 polyclonal PDGFRα antibody (Santa Cruz) in HBSS for 2 h at room temperature. Dishes were then thoroughly rinsed, and dissociated neurospheres at passage 2 were plated onto the dishes and allowed to attach at 37 °C for 1 h. Free floating cells were removed via washing, and immunopanned cells were removed with a cell scraper and plated onto a PEI-coated T25 flask in basal medium supplemented with bFGF and PDGF for further expansion. When the cells reached 80% confluency, they were passaged at a ratio of 1:3. After 2 more passages, the cells were seeded on PEI-coated 48 well plates at a density of 3.5 × 104 cells cm−2 and allowed to attach overnight. For the differentiation experiments, medium was changed every 2 d, and cells were allowed to differentiate for 7 d. 93.8 ± 1.2% of the isolated OPCs were PDGFRα+ cells. OPC cultures also contained low levels of O4+ cells (2.6 ± 0.7%), and negligible levels of MBP+ and GFAP+ cells.
2
R B Shultz et al
J. Neural Eng. 14 (2017) 036014
of ketamine (60 mg kg−1), xylazine (6 mg kg−1) and acepromazine (0.5 mg kg−1). A partial C5 laminectomy was performed to expose half of the spinal cord. The cord was stabilized and subjected to a force of 200 kDyne, using an Infinite Horizons (IH) impactor (Precision Systems and Instrumentation) fitted with a 1.6 mm impactor tip. 10 µl of gel loaded with 2 mg ml−1 T3 was incubated in aCSF for 4 d to allow for T3 release till the stable phase of release was achieved. Immediately after injury, the gel was applied on top of the spinal cord tissue at the injury site for local delivery of T3 (n = 5). Gels were secured by applying additional agarose gel without T3 particles on top of them. Injured animals receiving blank gel treatment served as controls (n = 6). Animal weight was measured 3 d post-injury (dpi), and then weekly for 4 weeks. Both groups received intraperitoneal (i.p.) injections of 5-bromo-2′-deoxyuridine (BrdU, Sigma-Aldrich) once daily (50 mg kg−1) for 7 d starting at week 2.
2.3. Immunostaining
Cells were fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. After fixation, the cells were blocked and permeablized in 5% milk powder, 0.1% bovine serum albumin (BSA), and 0.1% Triton-X 100 in 0.01 M PBS. Then the cells were incubated with anti-myelin basic protein (MBP, 1:500, Covance) and anti-glial fibrillary acidic protein (GFAP, 1:2000, Dako) overnight at 4 °C to identify mature OLs and astrocytes, respectively. After washing, the cells were incubated with secondary antibodies conjugated to Alexa fluorophores (1:2000, Invitrogen) for 2 h at room temperature. For O4 staining to identify OL lineage (mature and immature), the cells were live stained with anti-O4 (1:500 R&D Systems) for 3 h, followed by Alexa fluorophore-conjugated secondary antibody for 30 min prior to fixation. All the cells were counterstained with nuclear dye 4′-6-diamidino-2-phenylindole (DAPI, Molecular Probes) for 15 min. Images were captured with a fluorescence inverted light microscope (Leica) in three random fields in each well, and four wells were used for each condition to count the cells with different staining.
2.6. Tissue processing and immunohistochemistry
Four weeks post injury, the animals were euthanized using Euthasol and perfused with 0.9% saline followed by 4% paraformaldehyde in 0.01 M PBS. Spinal cord segments were removed and post-fixed in 4% paraformaldehyde overnight, then cryoprotected in 30% sucrose for 2–3 d. The cord segments were embedded, frozen, and cut with a cryostat into 8 series of transverse cryostat sections (20 µm). Sections 160 µm apart were double immunostained for BrdU (1:20, Developmental Studies Hybridoma Bank, University of Iowa) and CC1 (1:200, Covance) to label newly formed mature OLs generated after SCI, or myelin basic protein (MBP, 1:200, Covance) and Neurofilament-H (NF-H, 1:500, Millipore) to label myelinated axons. To conduct immunostaining, sections were blocked and permeablized in a PBST solution (0.3% Triton-X 100 in PBS) containing 5% normal goat serum for 1 h, and incubated with primary antibodies overnight at 4 °C. After washing three times with PBST, tissue sections were incubated in secondary antibodies conjugated to Alexa fluorophores (1:500, Invitrogen) for 2 h at room temperature. Slides were then washed three times with PBST and mounted with DAPI Fluormount G (Southern Biotech). For BrdU/CC1 staining, tissue sections were pre-processed for antigen retrieval by incubating in 2 N HCl at 37 °C for 30 min, followed by rinsing twice with 40 mM borate buffer (pH 8.5). Tissue sections were imaged using a Zeiss AxioObserver Wide Field inverted microscope equipped with Slidebook 6 software with a stereology module. Unbiased quantification of total numbers of CC1+ and BrdU+ cells spanning 1.92 mm rostral and caudal to the injury epicenter were performed using a stereological counting technique. Total cell numbers were calculated using the equation N = ∑ Q × (1/ssf) × (1/asf) × (1/tsf), where ∑ Q is the sum of all counts, ssf is the sampling section fraction, asf is the area sampling fraction, and tsf is the thickness sampling fraction [31]. A similar counting technique was employed to quantify myelinated axons on the injured side.
2.4. Fabrication of drug delivery system and in vitro drug release
10 mg ml−1 agarose (Lonza) solution was prepared in 2 × artificial cerebrospinal fluid (aCSF) at 80 °C. 1 × aCSF contains 148.19 mM NaCl, 3.0 mM KCl, 1.41 mM CaCl2, 0.80 mM MgCl2, 0.80 mM Na2HPO4 · 7H2O, and 0.20 mM NaH2PO4 · H2O in water. T3 (Sigma-Aldrich) was dissolved in 0.1 N NaOH at a concentration of 40 mg ml−1. T3 solution was diluted in deioinized (DI) water to a final concentration of 4 or 6 mg ml−1. After agarose solution was cooled down to 37 °C, it was thoroughly mixed with an equal volume of T3 solution by pipetting, resulting in final T3 concentrations of 2 or 3 mg ml−1. Then the mixture solution was neutralized with 1 N HCl. During the neutralization step, T3 precipitated to form insoluble particles. After neutralization, the mixture solution was passed through a 27-gauge needle several times to ensure that T3 particles were uniformly distributed. Then the mixture solution was allowed to gel by cooling at 4 °C for 20 min. Agarose hydrogel loaded with T3 was incubated at 37 °C in 200 µl aCSF for quantification of T3 release. The release medium was replaced with fresh aCSF every 24 h. The amount of released T3 was determined by UV absorbance at 240 nm using an ultraviolet/visible light microplate spectrophotometer (Tecan Infinite M200 NanoQuant). Gel volumes were varied from 5–10 µl, and release studies were conducted until no detectable T3 was released. 2.5. Animal surgery
All animal procedures were approved by the IACUC committee at Drexel University and followed National Institutes of Health guidelines for the care and use of laboratory animals. A well-characterized, clinically relevant unilateral cervical contusion injury model was used in this study [30]. Briefly, adult female Sprague-Dawley rats were anesthetized with a cocktail 3
R B Shultz et al
J. Neural Eng. 14 (2017) 036014
differentiation and maturation in both OPC and NSPC cultures. This result is not surprising, as T3 has been shown to arrest OPC and NPSC division and initiate OL differentiation [34, 35], promote transcription of premyelinating genes [18], and regulate myelin basic protein-encoding gene expression [36] in precursor cell types. BDNF knockout mice have been shown to exhibit deficits in OL lineage cells as early as E17, suggesting a role for BDNF in early OL development before E17 [37]. NT3 has been shown to promote OL differentiation of E14 brain-derived neural progenitor cells and embryonic stem cells [19, 20]. In our study, the lack of response to BDNF and NT3 might in part be due to the relative maturity of our cell models, which are isolated from E17 embryos, and thus are more lineagerestricted than very early embryonic or E14 precursors. Chen et al [17] showed that BDNF treatment resulted in a significant increase in CNPase+ OLs from newborn mouse-derived neurospheres. However, this increase was minimal (5%). In contrast, T3 has been shown to robustly induce OL differentiation and maturation both during development and at adult stages [38, 39], which may explain why T3 is the most effective molecule in our E17-derived progenitor models. Because we are targeting endogenous progenitors in the adult spinal cord, our E17 progenitor models are more relevant than E14 or even earlier embryonic precursors. Finally, Shh has been shown to generate more O4+ cells from neural precursors that were isolated from postnatal day 4 (P4) rat pups [40]. This is consistent with our result that Shh significantly increased percentage of O4+ cells from OPCs. However, our results show that it is not as effective as T3 in promoting oligodendrogenesis and OL maturation. BDNF, NT3, and Shh have all been shown to promote oligodendrogenesis and myelination in vivo. When fibroblasts producing BDNF or NT3 were transplanted into contused spinal cord lesions, they promoted proliferation of oligodendrocyte lineage cells, resulting in higher numbers of newly formed OLs and increased MBP+ area within fibroblast grafts as compared to control animals [41]. Similarly, when injured animals were injected with lentiviral vectors encoding for NT3 or Shh, increased myelinated axons were observed within implanted polymer bridges designed to facilitate axon growth [42]. In both cases, however, myelination was only observed within polymer or cell grafts, not in native tissue already populated with endogenous OPCs, and the observed myelination was found to be at least partly Schwann-cell mediated, which is not as ideal as OL-mediated myelination. Additionally, BDNF and NT3 could have driven myelination by influencing neuronal activity, since electrically active axons have been shown to promote OPC differentiation and myelination [43]. Shh, meanwhile, may promote oligodendrogenesis by inhibition of BMP signaling [44]. These factors make it difficult to conclusively determine whether the enhanced in vivo myelination was a result of direct effect on OPC differentiation and OL maturation. Conversely, our culture model shows T3 directly promotes both oligodendrogenesis and OL maturation from NSPCs and OPCs in a controlled environment.
2.7. Statistical analysis
All the in vitro studies were repeated 2–3 times, and the data were taken from one representative experiment of each study. One-way ANOVA followed by post-hoc analysis was used to determine any significant differences. Specifically, Tukey test was used for in vitro NSPC and OPC differentiation study analyses for multiple comparisons. Non-parametric Mann–Whitney tests were used for in vivo differentiation and myelination comparisons. Animal weight was analyzed via two-way ANOVA, followed by Sidak post-hoc analyses. Data are presented as mean ± standard deviation. 3. Results and discussion 3.1. Screening of therapeutic candidates
BDNF, NT3, Shh, and T3 have been shown to promote OL differentiation and myelination [17–20]. In this study we compared the potency of these molecules to promote OL differentiation and maturation from NSPCs and OPCs, which can be found in adult spinal cord [6, 7]. 30 ng ml−1 is commonly used for T3 in differentiation studies, whereas there was some discrepancy among reported doses for the other three molecules. Therefore, we tested the efficacy of different concentrations of BDNF, NT3 and Shh in promoting OL differentiation and maturation in NSPC cultures. The concentrations we selected have been used on published studies [17, 19, 20, 32]. After 7 d of differentiation, cells were stained for OL lineage marker O4, mature OL marker MBP, and astrocyte marker GFAP, respectively. We found that only T3 treatment significantly increased the percentage of cells expressing O4 as compared to control and all other groups (figures 1(a) and (d)). This result suggests that T3 is the most effective in promoting oligodendrogenesis from NSPCs. We further compared the potency of these molecules in promoting OL maturation. Figures 1(b) and (d) show that T3 treatment significantly increased the percentage of MBP+ (mature OL) cells compared to control, whereas none of the other molecules had any significant effect on OL maturation regardless of the concentration used. Collectively, these results suggest that T3 is the most effective in promoting OL differentiation and maturation compared to all the other molecules in NSPC cultures. We also studied the effect of these molecules on astrocyte differentiation, as astrocytes play an important role in glial scar formation after SCI [33]. Figures 1(c) and (d) show that only the highest concentration of Shh tested (500 ng ml−1) significantly inhibited astrocyte differentiation compared to control. Figures 2(a), (b) and (d) shows that immunopanned OPCs treated with 30 ng ml−1 T3 generated significantly higher percentages of O4+ and MBP+ cells compared to control and all the other treatments. In addition, 50 ng ml−1 Shh significantly increased the percentage O4+ cells compared to control (figures 2(a) and (d)). None of these treatments significantly changed percentage of cells expressing astrocyte marker GFAP (figures 2(c) and (d)). Taken together, these results suggest that T3 is the most effective in stimulating OL
4
R B Shultz et al
J. Neural Eng. 14 (2017) 036014
Figure 1. OL differentiation and maturation in NSPC culture when treated with different therapeutics. Cells treated with culture media without any therapeutics served as control. Quantification of percentage of cells for (a) O4, (b) MBP, and (c) GFAP staining. *P
Search
Collections
Journals
About
Contact us
My IOPscience
Local delivery of thyroid hormone enhances oligodendrogenesis and myelination after spinal cord injury
This content has been downloaded from IOPscience. Please scroll down to see the full text. 2017 J. Neural Eng. 14 036014 (http://iopscience.iop.org/1741-2552/14/3/036014) View the table of contents for this issue, or go to the journal homepage for more Download details: IP Address: 193.140.216.7 This content was downloaded on 04/04/2017 at 04:42 Please note that terms and conditions apply.
You may also be interested in: Potential of human dental stem cells in repairing the complete transection of rat spinal cord Chao Yang, Xinghan Li, Liang Sun et al. Injectable hydrogels for central nervous system therapy Malgosia M Pakulska, Brian G Ballios and Molly S Shoichet Biomaterial-based drug delivery systems for the controlled release of neurotrophic factors Nima Khadem Mohtaram, Amy Montgomery and Stephanie M Willerth Metal ion-assisted self-assembly of complexes for controlled and sustained release of minocycline for biomedical applications Zhiling Zhang, Zhicheng Wang, Jia Nong et al. Hyaluronic acid limits scarring after spinal cord injury Zin Z Khaing, Brian D Milman, Jennifer E Vanscoy et al. Recent progress in stem cell differentiation directed by material and mechanical cues Xunxun Lin, Yuan Shi, Yilin Cao et al. Positive and negative cues for modulating neurite dynamics and receptor expression Melissa R Wrobel and Harini G Sundararaghavan Bioengineered fibrin-based niche to direct outgrowth of circulating progenitors into neuron-like cells for potential use in cellular therapy S Tara and Lissy K Krishnan
Journal of Neural Engineering J. Neural Eng. 14 (2017) 036014 (13pp)
https://doi.org/10.1088/1741-2552/aa6450
Local delivery of thyroid hormone enhances oligodendrogenesis and myelination after spinal cord injury Robert B Shultz, Zhicheng Wang, Jia Nong, Zhiling Zhang and Yinghui Zhong1 School of Biomedical Engineering, Science and Health Systems, Drexel University, 3141 Chestnut Street, Philadelphia, PA 19104, United States of America E-mail: [email protected] Received 20 October 2016, revised 1 March 2017 Accepted for publication 3 March 2017 Published 30 March 2017 Abstract
Objective. Traumatic spinal cord injury (SCI) causes apoptosis of myelin-forming oligodendrocytes (OLs) and demyelination of surviving axons, resulting in conduction failure. Remyelination of surviving denuded axons provides a promising therapeutic target for spinal cord repair. While cell transplantation has demonstrated efficacy in promoting remyelination and functional recovery, the lack of ideal cell sources presents a major obstacle to clinical application. The adult spinal cord contains oligodendrocyte precursor cells and multipotent neural stem/progenitor cells that have the capacity to differentiate into mature, myelinating OLs. However, endogenous oligodendrogenesis and remyelination processes are limited by the upregulation of remyelination-inhibitory molecules in the post-injury microenvironment. Multiple growth factors/molecules have been shown to promote OL differentiation and myelination. Approach. In this study we screened these therapeutics and found that 3, 3′, 5-triiodothyronine (T3) is the most effective in promoting oligodendrogenesis and OL maturation in vitro. However, systemic administration of T3 to achieve therapeutic doses in the injured spinal cord is likely to induce hyperthyroidism, resulting in serious side effects. Main results. In this study we developed a novel hydrogel-based drug delivery system for local delivery of T3 to the injury site without eliciting systemic toxicity. Significance. Using a clinically relevant cervical contusion injury model, we demonstrate that local delivery of T3 at doses comparable to safe human doses promoted new mature OL formation and myelination after SCI. Keywords: drug delivery, thyroid hormone, hydrogel, spinal cord injury, myelination, oligodendrogenesis (Some figures may appear in colour only in the online journal)
1. Introduction
matter [2]. Studies have shown that less than 10% of axons in specific spinal tracts were necessary for locomotor, forepaw reaching, and tactile sensory functions [3]. However, prolonged and widespread oligodendrocyte (OL) loss following SCI leads to demyelination of spared axons, resulting in conduction failure [2–4]. Moreover, the demyelinated axons are susceptible to degeneration due to a lack of trophic support from myelin-forming cells [5]. Thus, replacement of lost OL
Traumatic spinal cord injury (SCI) results in significant loss of motor, sensory, and autonomic functions. Contusion injury is the most frequent form of SCI [1]. Even after severe contusive SCI, surviving axons persist in the subpial rim of white 1
Author to whom any correspondence should be addressed.
1741-2552/17/036014+13$33.00
1
© 2017 IOP Publishing Ltd Printed in the UK
R B Shultz et al
J. Neural Eng. 14 (2017) 036014
2. Materials and methods
population and remyelination of the denuded surviving axons represents a promising therapeutic strategy to promote functional recovery and improve the quality of life for patients suffering from SCI. The adult spinal cord contains oligodendrocyte precursor cells (OPCs) that have the capacity to differentiate into mature, myelin-forming OLs [6]. In addition, multipotent neural stem/ progenitor cells (NSPCs) capable of generating OLs, astrocytes, and neurons are present in the adult rat and human spinal cord [7]. However, endogenous oligodendrogenesis and remyelination processes are severely limited by the upregulation of remyelination-inhibitory molecules in the post-injury microenvironment [8–11]. These molecules include tumor necrosis factor α (TNFα), interleukin-1β (IL-1β), bone morphogenic proteins 2 and 4 (BMP-2 and 4), and chondroitin sulfate proteoglycans (CSPGs) [12–15]. While cell transplantation has demonstrated efficacy in promoting remyelination and functional recovery [16], the lack of ideal cell sources presents a major obstacle to clinical application. Potential safety concerns limit allogeneic cell transplantation, while lengthy expansion time and/or invasive isolation limit the use of autogeneic cells. Therefore, it is important to develop a therapy that can promote endogenous OPC/NSPC differentiation, maturation and remyelination. A number of growth factors/molecules have been shown to promote OL differentiation and myelination, including brainderived neurotrophic factor (BDNF), neurotrophin-3 (NT3), sonic hedgehog (Shh), and thyroid hormone 3,3′,5-triiodothyronine (T3) [17–20]. In the present study, we compared the efficacy of these molecules in promoting OL differentiation and maturation in vitro and found that T3 is the most effective compared to all the molecules. In addition, we found that T3 was effective in a wide range of 10–2000 ng ml−1. However, average T3 concentration in serum was around 1.2 ± 0.35 ng ml−1, and T3 levels in patients suffering from hyperthyroidism was found to be 5.7 ± 3.5 ng ml−1 [21]. Thus, increasing systemic T3 level to 10 ng ml−1 or even higher is likely to induce hyperthyroidism, resulting in serious side effects including anxiety, muscle weakness, heart palpitations, and thyrotoxic crisis [22, 23]. Local delivery can potentially expose the injured spinal cord tissue to therapeutic levels of T3, while avoiding deleterious side effects from systemic exposure. Hydrogel-based delivery systems have received much attention to locally deliver drugs to the injured spinal cord [24–27]. Hydrogel systems are suitable for this application because they can be fabricated from biocompatible, naturally occurring materials, and can immobilize drug or drugloaded particles at the injury site to provide sustained localized release. In addition, hydrogels exhibit mechanical properties compatible with native spinal cord tissue, minimizing the risk of additional tissue damage associated with mechanical mismatch. In this study, we developed an agarose hydrogel-based T3 delivery system capable of controlled release of physiologically relevant doses of T3, and investigated the efficacy of local delivery of T3 in promoting oligodendrogenesis using a clinically relevant unilateral cervical spinal cord contusion injury model.
2.1. Neural stem and progenitor cell (NSPC) culture
NSPCs were isolated from E17 Sprague Dawley rat embryos in accordance with protocols approved by Drexel University’s IACUC committee. The cells were isolated and cultured following a procedure modified from a protocol reported by Pedraza et al [28]. In brief, cerebrum tissue was mechanically dissociated by trituration. Then the tissue was filtered through a 70 µm cell strainer and plated in plastic tissue culture flasks. The cells were allowed to form neurospheres in basal medium (DMEM/F12, 2% B27 supplement, and 1% Penicillin/Streptomycin) supplemented with 20 ng ml−1 of basic fibroblast growth factor (bFGF, Peprotech) and plateletderived growth factor (PDGF, Peprotech). NSPCs at 3–4 passages were used in the subsequent experiments. 92.8 ± 2.7% of the isolated NSPCs were positive for Nestin, a marker for neural stem and progenitor cells. The NSPC cultures also contained a low level of glial fibrillary acidic protein+ (GFAP+) astrocytes (7.2 ± 1.3%), and negligible levels of O4+ (mature and immature OL) and myelin basic protein+ (MBP+; mature OL) cells. For differentiation experiments, NSPCs were chemically dissociated using a chemical dissociation kit (Stemcell Technologies) and plated at a density of 2.5 × 104 cells cm−2 onto polyethylenimine (PEI)-coated 48-well plates. The cells were allowed to attach overnight, and then the culture medium was replaced with basal medium containing different OL differentiation molecules. The medium was changed every 2 d. The cells were allowed to differentiate for 7 d. 2.2. Oligodendrocyte precursor cell (OPC) culture
OPCs were isolated from neurospheres via immunopanning for PGDF receptor type α (PDGFRα, early OPC marker) [29]. Briefly, non-tissue culture treated dishes were coated with 10 µg ml−1 goat anti-rabbit IgG (Jackson Immunoresearch) in 50 mM TRIS-HCl (pH 9.5) overnight at 4 C. After washing 3 times with Hank’s Balanced Salt Solution (HBSS), the dishes were incubated with 1 µg ml−1 polyclonal PDGFRα antibody (Santa Cruz) in HBSS for 2 h at room temperature. Dishes were then thoroughly rinsed, and dissociated neurospheres at passage 2 were plated onto the dishes and allowed to attach at 37 °C for 1 h. Free floating cells were removed via washing, and immunopanned cells were removed with a cell scraper and plated onto a PEI-coated T25 flask in basal medium supplemented with bFGF and PDGF for further expansion. When the cells reached 80% confluency, they were passaged at a ratio of 1:3. After 2 more passages, the cells were seeded on PEI-coated 48 well plates at a density of 3.5 × 104 cells cm−2 and allowed to attach overnight. For the differentiation experiments, medium was changed every 2 d, and cells were allowed to differentiate for 7 d. 93.8 ± 1.2% of the isolated OPCs were PDGFRα+ cells. OPC cultures also contained low levels of O4+ cells (2.6 ± 0.7%), and negligible levels of MBP+ and GFAP+ cells.
2
R B Shultz et al
J. Neural Eng. 14 (2017) 036014
of ketamine (60 mg kg−1), xylazine (6 mg kg−1) and acepromazine (0.5 mg kg−1). A partial C5 laminectomy was performed to expose half of the spinal cord. The cord was stabilized and subjected to a force of 200 kDyne, using an Infinite Horizons (IH) impactor (Precision Systems and Instrumentation) fitted with a 1.6 mm impactor tip. 10 µl of gel loaded with 2 mg ml−1 T3 was incubated in aCSF for 4 d to allow for T3 release till the stable phase of release was achieved. Immediately after injury, the gel was applied on top of the spinal cord tissue at the injury site for local delivery of T3 (n = 5). Gels were secured by applying additional agarose gel without T3 particles on top of them. Injured animals receiving blank gel treatment served as controls (n = 6). Animal weight was measured 3 d post-injury (dpi), and then weekly for 4 weeks. Both groups received intraperitoneal (i.p.) injections of 5-bromo-2′-deoxyuridine (BrdU, Sigma-Aldrich) once daily (50 mg kg−1) for 7 d starting at week 2.
2.3. Immunostaining
Cells were fixed with 4% paraformaldehyde in PBS for 15 min at room temperature. After fixation, the cells were blocked and permeablized in 5% milk powder, 0.1% bovine serum albumin (BSA), and 0.1% Triton-X 100 in 0.01 M PBS. Then the cells were incubated with anti-myelin basic protein (MBP, 1:500, Covance) and anti-glial fibrillary acidic protein (GFAP, 1:2000, Dako) overnight at 4 °C to identify mature OLs and astrocytes, respectively. After washing, the cells were incubated with secondary antibodies conjugated to Alexa fluorophores (1:2000, Invitrogen) for 2 h at room temperature. For O4 staining to identify OL lineage (mature and immature), the cells were live stained with anti-O4 (1:500 R&D Systems) for 3 h, followed by Alexa fluorophore-conjugated secondary antibody for 30 min prior to fixation. All the cells were counterstained with nuclear dye 4′-6-diamidino-2-phenylindole (DAPI, Molecular Probes) for 15 min. Images were captured with a fluorescence inverted light microscope (Leica) in three random fields in each well, and four wells were used for each condition to count the cells with different staining.
2.6. Tissue processing and immunohistochemistry
Four weeks post injury, the animals were euthanized using Euthasol and perfused with 0.9% saline followed by 4% paraformaldehyde in 0.01 M PBS. Spinal cord segments were removed and post-fixed in 4% paraformaldehyde overnight, then cryoprotected in 30% sucrose for 2–3 d. The cord segments were embedded, frozen, and cut with a cryostat into 8 series of transverse cryostat sections (20 µm). Sections 160 µm apart were double immunostained for BrdU (1:20, Developmental Studies Hybridoma Bank, University of Iowa) and CC1 (1:200, Covance) to label newly formed mature OLs generated after SCI, or myelin basic protein (MBP, 1:200, Covance) and Neurofilament-H (NF-H, 1:500, Millipore) to label myelinated axons. To conduct immunostaining, sections were blocked and permeablized in a PBST solution (0.3% Triton-X 100 in PBS) containing 5% normal goat serum for 1 h, and incubated with primary antibodies overnight at 4 °C. After washing three times with PBST, tissue sections were incubated in secondary antibodies conjugated to Alexa fluorophores (1:500, Invitrogen) for 2 h at room temperature. Slides were then washed three times with PBST and mounted with DAPI Fluormount G (Southern Biotech). For BrdU/CC1 staining, tissue sections were pre-processed for antigen retrieval by incubating in 2 N HCl at 37 °C for 30 min, followed by rinsing twice with 40 mM borate buffer (pH 8.5). Tissue sections were imaged using a Zeiss AxioObserver Wide Field inverted microscope equipped with Slidebook 6 software with a stereology module. Unbiased quantification of total numbers of CC1+ and BrdU+ cells spanning 1.92 mm rostral and caudal to the injury epicenter were performed using a stereological counting technique. Total cell numbers were calculated using the equation N = ∑ Q × (1/ssf) × (1/asf) × (1/tsf), where ∑ Q is the sum of all counts, ssf is the sampling section fraction, asf is the area sampling fraction, and tsf is the thickness sampling fraction [31]. A similar counting technique was employed to quantify myelinated axons on the injured side.
2.4. Fabrication of drug delivery system and in vitro drug release
10 mg ml−1 agarose (Lonza) solution was prepared in 2 × artificial cerebrospinal fluid (aCSF) at 80 °C. 1 × aCSF contains 148.19 mM NaCl, 3.0 mM KCl, 1.41 mM CaCl2, 0.80 mM MgCl2, 0.80 mM Na2HPO4 · 7H2O, and 0.20 mM NaH2PO4 · H2O in water. T3 (Sigma-Aldrich) was dissolved in 0.1 N NaOH at a concentration of 40 mg ml−1. T3 solution was diluted in deioinized (DI) water to a final concentration of 4 or 6 mg ml−1. After agarose solution was cooled down to 37 °C, it was thoroughly mixed with an equal volume of T3 solution by pipetting, resulting in final T3 concentrations of 2 or 3 mg ml−1. Then the mixture solution was neutralized with 1 N HCl. During the neutralization step, T3 precipitated to form insoluble particles. After neutralization, the mixture solution was passed through a 27-gauge needle several times to ensure that T3 particles were uniformly distributed. Then the mixture solution was allowed to gel by cooling at 4 °C for 20 min. Agarose hydrogel loaded with T3 was incubated at 37 °C in 200 µl aCSF for quantification of T3 release. The release medium was replaced with fresh aCSF every 24 h. The amount of released T3 was determined by UV absorbance at 240 nm using an ultraviolet/visible light microplate spectrophotometer (Tecan Infinite M200 NanoQuant). Gel volumes were varied from 5–10 µl, and release studies were conducted until no detectable T3 was released. 2.5. Animal surgery
All animal procedures were approved by the IACUC committee at Drexel University and followed National Institutes of Health guidelines for the care and use of laboratory animals. A well-characterized, clinically relevant unilateral cervical contusion injury model was used in this study [30]. Briefly, adult female Sprague-Dawley rats were anesthetized with a cocktail 3
R B Shultz et al
J. Neural Eng. 14 (2017) 036014
differentiation and maturation in both OPC and NSPC cultures. This result is not surprising, as T3 has been shown to arrest OPC and NPSC division and initiate OL differentiation [34, 35], promote transcription of premyelinating genes [18], and regulate myelin basic protein-encoding gene expression [36] in precursor cell types. BDNF knockout mice have been shown to exhibit deficits in OL lineage cells as early as E17, suggesting a role for BDNF in early OL development before E17 [37]. NT3 has been shown to promote OL differentiation of E14 brain-derived neural progenitor cells and embryonic stem cells [19, 20]. In our study, the lack of response to BDNF and NT3 might in part be due to the relative maturity of our cell models, which are isolated from E17 embryos, and thus are more lineagerestricted than very early embryonic or E14 precursors. Chen et al [17] showed that BDNF treatment resulted in a significant increase in CNPase+ OLs from newborn mouse-derived neurospheres. However, this increase was minimal (5%). In contrast, T3 has been shown to robustly induce OL differentiation and maturation both during development and at adult stages [38, 39], which may explain why T3 is the most effective molecule in our E17-derived progenitor models. Because we are targeting endogenous progenitors in the adult spinal cord, our E17 progenitor models are more relevant than E14 or even earlier embryonic precursors. Finally, Shh has been shown to generate more O4+ cells from neural precursors that were isolated from postnatal day 4 (P4) rat pups [40]. This is consistent with our result that Shh significantly increased percentage of O4+ cells from OPCs. However, our results show that it is not as effective as T3 in promoting oligodendrogenesis and OL maturation. BDNF, NT3, and Shh have all been shown to promote oligodendrogenesis and myelination in vivo. When fibroblasts producing BDNF or NT3 were transplanted into contused spinal cord lesions, they promoted proliferation of oligodendrocyte lineage cells, resulting in higher numbers of newly formed OLs and increased MBP+ area within fibroblast grafts as compared to control animals [41]. Similarly, when injured animals were injected with lentiviral vectors encoding for NT3 or Shh, increased myelinated axons were observed within implanted polymer bridges designed to facilitate axon growth [42]. In both cases, however, myelination was only observed within polymer or cell grafts, not in native tissue already populated with endogenous OPCs, and the observed myelination was found to be at least partly Schwann-cell mediated, which is not as ideal as OL-mediated myelination. Additionally, BDNF and NT3 could have driven myelination by influencing neuronal activity, since electrically active axons have been shown to promote OPC differentiation and myelination [43]. Shh, meanwhile, may promote oligodendrogenesis by inhibition of BMP signaling [44]. These factors make it difficult to conclusively determine whether the enhanced in vivo myelination was a result of direct effect on OPC differentiation and OL maturation. Conversely, our culture model shows T3 directly promotes both oligodendrogenesis and OL maturation from NSPCs and OPCs in a controlled environment.
2.7. Statistical analysis
All the in vitro studies were repeated 2–3 times, and the data were taken from one representative experiment of each study. One-way ANOVA followed by post-hoc analysis was used to determine any significant differences. Specifically, Tukey test was used for in vitro NSPC and OPC differentiation study analyses for multiple comparisons. Non-parametric Mann–Whitney tests were used for in vivo differentiation and myelination comparisons. Animal weight was analyzed via two-way ANOVA, followed by Sidak post-hoc analyses. Data are presented as mean ± standard deviation. 3. Results and discussion 3.1. Screening of therapeutic candidates
BDNF, NT3, Shh, and T3 have been shown to promote OL differentiation and myelination [17–20]. In this study we compared the potency of these molecules to promote OL differentiation and maturation from NSPCs and OPCs, which can be found in adult spinal cord [6, 7]. 30 ng ml−1 is commonly used for T3 in differentiation studies, whereas there was some discrepancy among reported doses for the other three molecules. Therefore, we tested the efficacy of different concentrations of BDNF, NT3 and Shh in promoting OL differentiation and maturation in NSPC cultures. The concentrations we selected have been used on published studies [17, 19, 20, 32]. After 7 d of differentiation, cells were stained for OL lineage marker O4, mature OL marker MBP, and astrocyte marker GFAP, respectively. We found that only T3 treatment significantly increased the percentage of cells expressing O4 as compared to control and all other groups (figures 1(a) and (d)). This result suggests that T3 is the most effective in promoting oligodendrogenesis from NSPCs. We further compared the potency of these molecules in promoting OL maturation. Figures 1(b) and (d) show that T3 treatment significantly increased the percentage of MBP+ (mature OL) cells compared to control, whereas none of the other molecules had any significant effect on OL maturation regardless of the concentration used. Collectively, these results suggest that T3 is the most effective in promoting OL differentiation and maturation compared to all the other molecules in NSPC cultures. We also studied the effect of these molecules on astrocyte differentiation, as astrocytes play an important role in glial scar formation after SCI [33]. Figures 1(c) and (d) show that only the highest concentration of Shh tested (500 ng ml−1) significantly inhibited astrocyte differentiation compared to control. Figures 2(a), (b) and (d) shows that immunopanned OPCs treated with 30 ng ml−1 T3 generated significantly higher percentages of O4+ and MBP+ cells compared to control and all the other treatments. In addition, 50 ng ml−1 Shh significantly increased the percentage O4+ cells compared to control (figures 2(a) and (d)). None of these treatments significantly changed percentage of cells expressing astrocyte marker GFAP (figures 2(c) and (d)). Taken together, these results suggest that T3 is the most effective in stimulating OL
4
R B Shultz et al
J. Neural Eng. 14 (2017) 036014
Figure 1. OL differentiation and maturation in NSPC culture when treated with different therapeutics. Cells treated with culture media without any therapeutics served as control. Quantification of percentage of cells for (a) O4, (b) MBP, and (c) GFAP staining. *P
