SEARCH

SEARCH BY CITATION

Keywords:

  • limb regeneration;
  • accessory limb model;
  • HGF;
  • myogenesis

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Background: Urodele amphibians have high regeneration capability that has been studied for a long time. Recently, a new experimental system called the accessory limb model was developed and becomes alternative choice for amphibian limb regeneration study. Although the accessory limb model has many advantages, an improvement was needed for some specific analysis, such as studying muscle origin. For that purpose, an accessory limb induction on nonlimb regions was attempted. Results: Accessory limb induction on a nonlimb region (flank) was possible by nerve deviation and limb skin grafting. Retinoic acid injections improved the induction rate. The induced limb possessed the same tissue context as a normal limb. Muscle cells were also abundantly observed. It is speculated that the muscle cells are derived from flank muscle tissues, because limb muscle cells are a migratory cell population and the accessory limb was induced apart from the original limb. We also found that migration of the muscle cells was regulated by Hgf/cMet signaling as in other vertebrates. Conclusions: Accessory limb induction was possible even in the nonlimb flank region. The flank-induced limb would be useful for further analysis of limb regeneration, especially for migratory cell populations such as muscle cells. Developmental Dynamics 242:932–940, 2013. © 2013 Wiley Periodicals, Inc.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Urodele amphibians regenerate various organs, including limbs. After limb amputation, a population of undifferentiated cells called a blastema emerges at the site of injury, on the amputation plane. Blastema cells arise by dedifferentiation of a part of the stump tissue. In contrast, higher vertebrates cannot form a blastema after amputation, even though their tissue components are similar to those of urodele amphibians. Cartilage, muscles, tendons, ligaments, dermis, and bones can be observed in both higher vertebrates and urodele amphibians. Therefore, higher regeneration capability in urodele amphibians cannot be associated with their specific tissue arrangement, and the factors that contribute to their higher regeneration capability remain unknown.

A recently established experimental system called the accessory limb model (ALM) has become an alternative experimental system for the study of limb regeneration in urodele amphibians (Endo et al., 2004). The ALM has some advantages in studying limb regeneration (Makanae and Satoh, 2012), and it has produced many achievements in recent years. The ALM indicates that the tissues necessary for induction of limb regeneration responses are only nerves and skin (Endo et al., 2004). Simple limb skin wounding merely results in skin healing without scar formation (Seifert et al., 2012; Kawasumi et al., 2013). When nerves are deviated to the skin wound, regeneration responses are induced instead of the skin wound healing processes (Endo et al., 2004; Satoh et al., 2007). Blastema formation can be induced by nerve deviation to the skin wound and the induced blastema shows the same features as a regular blastema (Satoh et al., 2007). However, to obtain a completely patterned limb, an additional procedure is necessary. That procedure is skin grafting from the contralateral side of a limb (Endo et al., 2004). A small piece of skin is taken from the other side of the wound. If a skin wound is created in the anterior side, a skin graft should be taken from posterior side. This piece of skin is then placed on the nerve-deviated skin wound. The skin graft is thought to supply missing positional value(s) of a limb (Makanae and Satoh, 2012). In summary, the blastema induction process can be investigated in a nerve-deviated skin wound. Investigation of limb patterning requires an additional contralateral skin graft to the nerve-deviated wound. Thus, using the ALM allows stepwise investigation of limb regeneration.

The ALM is a useful system, but there is still room for improvement to study the origin of limb muscle cells during limb regeneration. Recently, it was clearly shown that muscle cells in a regenerate were derived from muscle cells in the unamputated region (Kragl et al., 2009). Thus, migratory myogenic cells give rise to muscle tissues in a regenerate. There is still one question, should the migratory myogenic cells be derived from a limb? In case of dermis, dermis must be derived from a limb for successful limb regeneration because lack of limb dermis results in failure of limb regeneration in Ambystoma talpoideum (Thornton, 1962). It has been thought that dermis-derived blastema cells play an important role in creating the blueprint for a future limb and that lack of such a dermis-derived blastema fails to regenerate a complete limb. Unfortunately, none of the current systems used to study amphibian limb regeneration seem suitable enough to investigate the necessity for limb origin of muscle cells.

In the present study, we improved the ALM and successfully induced an accessory limb on the body flank. Similar to normal ALM surgery, flank skin wounding, limb skin grafting, and nerve deviation resulted in limb formation. Retinoic acid (RA) accelerated the limb induction rate. The induced limb contained muscle tissues that were presumably derived from a flank muscle. We also investigated migratory regulation and found that it was regulated by Hgf/c-Met signaling as observed in other vertebrates. These findings and the new experimental study system are useful for further limb regeneration investigations, particularly those focusing on migratory cell populations.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Induction of Ectopic Limbs at the Flank

To determine if muscles cells in a regenerated limb are derived from limb muscle cells, we attempted to induce an accessory limb in a nonlimb region (Fig. 1). An ectopic limb could be induced on a body flank using a procedure based on the ALM established by Endo et al. (2004) and a classical report (Kiortsis, 1953). Nerves projecting to a limb have relatively longer axon fibers. These longer axons were rerouted to a flank skin wound (Fig. 1). The flank wound was always away from the limb. Two limb skin grafts (each from opposite sides of a limb) were placed beside the deviated nerve (Fig. 1A,B). RA was injected under the deviated nerve 5 days later (Fig. 1A,B). A growing structure was observed in all cases (Table 1). Ectopic limbs were inducible not only in the flank but also in other regions, such as the tail. Four weeks after surgery, blastema-like formations called “bumps” could be seen (Fig. 1C). Most bumps stopped growing and started to regress at this stage (Table 1). However, some kept growing and started limb patterning 6 weeks after surgery (Fig. 1D). The bump became flattened and was bent at the presumptive wrist/ankle region, as in normal limb regeneration and development (Fig. 1D). Ectopic flank limb formation was completed approximately 3 months after induction (Fig. 1E,F). This time span seems longer than that seen in normal accessory limb induction (Endo et al., 2004). Ectopic limbs could be induced without RA but their induction rate was greatly reduced (Table 1). In the cases of no skin grafts, limb formation was never observed, although bump formation could be seen (Table 1). As expected, simple flank skin wounding resulted in skin wound healing without any growth (Table 1).

image

Figure 1. Ectopic limbs were induced in the flank. A: A nerve from a hind limb was deviated into the skin wound on the flank and two skin grafts opposite each other were placed beside the deviated nerve. B: Schematic representation of A. Five days later, RA was injected under the deviated nerve. C,D: The surgery resulted in the formation of the bump after 4 (C) and 6 weeks (D). E: Ectopic limb was observed approximately 3 months after induction. F: Higher magnification of E. G: Skin was dissected and stained with hematoxylin–eosin. Dissected skin was not contaminated with myogenic cells. H: The expression of Pax7 and MHC mRNAs were investigated using reverse transcriptase-polymerase chain reaction. RNAs were isolated from muscle and intact skin. The expression of Pax7 and MHC mRNAs were not detected in the skin. Scale bars = 5 mm in A–E, 2 mm in F; 1 mm in G.

Download figure to PowerPoint

Table 1. Induction Rate of Ectopic Limbs in the Flank
 Wound healingOnly bump (regressed)Limb inductionTotal
  1. a

    Two limbs did not form the digits.

Lateral wound200020
Lateral wound + Nev210012
Lateral wound + Nev + Skin Graft012315
Lateral wound + Nev + Skin Graft + RA01012a22

To investigate the origin of the regenerate muscle, contamination of limb myogenic cells in the grafts was clarified (Fig. 1G,H). Histological observation of the dissected skin graft indicated no obvious muscle tissues. To confirm this, reverse transcriptase-polymerase chain reaction (RT-PCR) was performed on the dissected skin grafts (Fig. 1H). We found a complete absence of Pax7, a muscle satellite cell marker gene, and myosin heavy chain (MHC), a differentiated muscle marker gene, in the skin graft used. These results indicate that the limb skin grafts did not contain any myogenic cells.

To describe the induced accessory limb on the flank in detail, the flank limbs were sectioned and subjected to histological analysis (Fig. 2A–D). Cartilage was visualized by Alcian blue staining (Fig. 2A), and calcified bones were visualized by Alizarin Red (Fig. 2B). Most of the flank limbs were deformed, and the pattern of digits was abnormal (Fig. 2A,B). We speculated that this was caused by the RA supplement, which is known to cause digit duplication (Thoms and Stocum, 1984). Although the induced limbs were deformed, basic limb patterning was observed. Joints, stylopods, zeugopods, and autopods were identifiable in the flank limbs (Fig. 2A,B), and their histology was similar to that of a normal limb (Fig. 2C,D). The induced limb possessed the necessary limb tissues. Epidermis, dermis, cartilage (that would eventually form the radius and ulna), and hematocytes were apparent (Fig. 2C,D). Axon fibers, which were recognized using an anti-acetylated tubulin antibody, were well penetrated into the flank limb (Fig. 2E,F). The histological observations indicated that a limb induced on a flank possesses the same structures as a normal limb. In the ALM, the accessory limbs induced at the stylopod level formed a normally patterned zeugopod and autopod. Accessory limbs induced at the zeugopod level did not form the zeugopod but formed only autopod elements (Satoh et al., 2010). Conversely, regenerated structures associated with the stylopod, zeugopod, and autopod appeared in the flank limb.

image

Figure 2. Ectopic limbs on the flank have similar structures with a normal limb. A: Cartilage of the ectopic limb harvested 3 months after the induction was stained with Alcian blue. B: Hardened bone of induced limb was stained Alizarin red. Arrows indicate induced limbs (A,B). C–F: Transverse sections at the level of the zeugopod were stained with Alcian blue and hematoxylin–eosin (C,D) or by an anti-acetylated tubulin antibody (red) (E,F). Higher magnification of the areas enclosed in squares in A, C, and E are shown in (B), (D), and (F), respectively. Arrow indicates hematocytes (D). Scale bar = 10 mm in A, 5 mm in C,E, 500 μm in D,F.

Download figure to PowerPoint

Muscles Are Present in Ectopic Limbs on a Flank

Next, we focused on the muscle in the induced limb. Histological observation suggested that muscles were induced in the ectopic limb (Fig. 2C,D). To confirm the presence of muscle with molecular marker genes and investigate myogenesis in the ectopic blastema, the expression pattern of myogenic marker genes in the ectopic limb was tested by in situ hybridization and immunohistochemistry. We compared transverse sections of a growing flank bump (Fig. 1C) with that of the fully patterned flank limb (Fig. 1E). In the growing flank bump, there were histologically undifferentiated cells (Fig. 3A). Cells expressing cardiac actin mRNA, which serves as a marker for early myogenesis (Nicolas et al., 1998; Satoh et al., 2005), could be detected throughout the growing bump (Fig. 3C). The expression of the mRNA-encoding MHC, which serves as marker for late myogenesis, was barely detectable (Fig. 3E,G). In the fully patterned flank limb, histologically differentiated cells were observed (Fig. 3B). Cardiac actin, MHC, and MHC proteins were detected (Fig. 3D,F,H). We also investigated the expression pattern of these two genes in regular regenerating blastemas (Fig. 4). Cardiac actin was broadly expressed between the early bud (EB) and late bud (LB) stages and was detectable before MHC expression (Fig. 4A,D,D′,G). In the blastema at the LB and the pallet (PS) stage, cardiac actin expression seemed up-regulated in the dorsal and ventral regions (Fig. 4G,J). MHC mRNA and protein were still not detected at the LB stage of blastemas; however, these were detected in the proximal (nonamputated) region (Fig. 4B,C,E,F,H,I). MHC expression was slow to appear, but we confirmed that MHC was detectable in the muscle tissue of digit stage (DS) blastemas (Fig. 4N,N′,O,O′) and cardiac actin gene expression was more restricted at this stage (Fig. 4M,M′). The myogenic marker gene expression pattern in the ectopic limb on a flank was similar to that of the regenerating blastema. These data indicated that myogenesis occurs normally during flank limb formation and that muscle tissues exist in the flank limb.

image

Figure 3. Muscle marker genes were expressed in the growing flank limb. A,B: Transverse sections of a growing flank bump (A) and the ectopic limb (B) were stained with Alcian blue and hematoxylin–eosin stain. B was sectioned at the level of the autopod of an induced limb (Fig. 1E). C,E,G: The boxed areas of the adjacent sections to A. D,F,H: The boxed areas of the sections adjacent to B. C,D: Sections were analyzed for a maker of early myogenesis (cardiac actin) by in situ hybridization and shows that the pattern of expression of cardiac actin (blue). E,F: The pattern of expression of MHC, a maker of terminal myogenesis expression, is shown (blue). G,H: Immunohistochemical analysis of MHC expression (green). Nuclei were counterstained with Hoechst dye (blue). The arrow in G indicates cells expressing MHC. Scale bars = 1 mm in A, 5 mm in B, 500 μm in C–H.

Download figure to PowerPoint

image

Figure 4. A–O′: Expression of muscle differentiation markers in the blastema. Gene expression patterns of myogenic marker genes in longitudinal (A–F,G–O) and transverse sections (D′–F′,M′–O′) of regenerating blastemas are shown. The expression of cardiac actin (A,D,D′,G,J,M,M′) and MHC (B,E,E′,H,K,N,N′) mRNAs was detected by in situ hybridization (blue). C,F,F′,I,L,O,O′: Immunohistochemistry for MHC (green), and counter-stained with Hoechst (blue). Cardiac actin was first expressed in entire blastema, and then split into the dorsal and ventral region. MHC mRNA and protein were expressed in the DS blastema. Scale bars = 5 mm.

Download figure to PowerPoint

HGF/c-Met Signaling Regulates Muscle Cell Migration in the Axolotl Limb

The data described above suggested that myogenesis similar to normal regeneration takes place in the accessory limb grown on a flank. However, detailed information about myogenic cell migration into a limb field (blastema) is still unknown. Hgf/cMet signaling regulates the migration of limb muscle cells in other vertebrates (Christ and Brand-Saberi, 2002). Therefore, we tested whether HGF/c-Met signaling regulated muscle cell migration during limb regeneration in the axolotl as in other vertebrates (Fig. 5). Limbs were amputated at the stylopod level and regenerated in the presence of a c-Met kinase inhibitor. Longitudinal sections of middle bud stage blastemas were prepared (Fig. 5). The inhibitor treatment appeared not to affect blastema formation itself (Fig. 5A,B). To visualize migratory myogenic cells, Pax7 expression was analyzed using immunohistochemistry. Pax7 is expressed in muscle satellite cells and migratory muscle precursor cells (Christ and Brand-Saberi, 2002). Few Pax7-positive cells were detected in the inhibitor-treated blastema (Fig. 5C). In the proximal region (stump) of the inhibitor-treated limb, Pax7-positive cells were still observed (Fig. 5C′). Pax7-positive cells were also detected in the control (dimethyl sulfoxide [DMSO] treated) blastema (Fig. 5D). The c-Met kinase inhibitor inhibited migration of the myogenic precursor cells into the blastema, suggesting that Hgf/cMet signaling regulates the migration process of limb muscle cells. Finally, we confirmed temporal expression of HGF and c-Met mRNAs in the regenerating blastema (Fig. 2E). HGF mRNA expression in blastema could be detected at all times, and c-Met mRNA expression was high at the EB stage and reduced at other stages.

image

Figure 5. The c-Met kinase inhibitor II inhibits the migration of the myogenic precursor cells into the blastema. A,B: Longitudinal sections of blastemas were stained with Alcian blue and reverse transcriptase-polymerase chain reaction (RT-PCR). A,C: An LB stage blastema grew in the presence of c-Met kinase inhibitor. B,D: An MB stage blastema treated with dimethyl sulfoxide as a control. (C) and (D) depict the boxed areas of the adjacent sections in A and B, respectively. C: Few Pax7-positive cells were detected in the inhibitor-treated blastema. (C′) In contrast, Pax7-positive cells were detected at the limb stump. D: Pax7-positive cells are present in the control blastema. E: The expression of hgf and c-met mRNAs were investigated using reverse transcriptase-polymerase chain reaction. RNAs were isolated from early bud (EB), middle bud (MB), late (LB), pallet stage (PS), and intact skin (skin). The expression of hgf mRNA was not detected in the normal skin. The level of c-met expression was high at the EB stage and decreased at subsequent stages. Ribosomal protein L19 served as the internal control. Scale bars = 5 mm in A,B, 1 mm in C,E.

Download figure to PowerPoint

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

In the present study, we succeeded in inducing accessory limbs on the body flank. The limb possessed muscle cells similar to those in a normal regenerating limb. This experimental system appears useful for further studies designed to reveal what kinds of tissues should be derived from limbs and what kinds of tissues do not need to be derived from limbs.

Ectopic Limb Induction in the Flank

We succeeded in inducing an ectopic limb on the flank of an axolotl. The most efficient method requires the injection of RA in addition to the deviation of a nerve and the explant of limb skin (Table 1; Fig. 1). In urodele amphibian limb regeneration, RA affects the proximodistal (Maden, 1982; Crawford and Stocum, 1988), dorsoventral, and anteroposterior (Kim and Stocum, 1986; Ludolph et al., 1990) axes. Thus, RA injection would generate variation of the positional values, which is assumed as an essential factor for limb regeneration (Makanae and Satoh, 2012). In the present study, the induced flank limb showed multiple digits (Figs. 1 2), suggesting that the RA injection disturbed the positional values. The effects of RA on axolotl limb pattering in limb regeneration have been well studied (Maden, 2002; Maden and Hind, 2003). Induction of the flank limb was achieved using zeugopodial skin (see the Experimental Procedures section) and the induced limb, for the most part, possessed the stylopodium (Fig. 2B). This is consistent with the proximalization effect of RA. Duplication of the anterior–posterior axis can usually be seen in the autopodium of the induced limb (Fig. 1A,B). However, it was very hard to determine digit identity without molecular marker genes, as the number of phalanges was easily changed by housing conditions. Comparison of the digit duplication pattern with previous achievements (Stocum and Thoms, 1984; Thoms and Stocum, 1984; Kim and Stocum, 1986) will be the subject of future studies.

An ectopic limb can be induced by skin wounding, nerve deviation, and skin grafting (Endo et al., 2004). Although the ALM blastema induced by a skin wound and nerve deviation without a skin graft regresses and finally disappears, completely patterned ectopic limbs can be induced in the presence of dorsoventral and anteroposterior positional values created by skin grafts (Makanae and Satoh, 2012). In the case of a flank limb, when only a piece of skin (the anterior or the posterior side) is grafted to the site of a skin wound to which a nerve has been deviated, ellipsoid-shaped incomplete structures were induced and eventually disappeared (data not shown). This may have occurred because of loss of positional values. Lack of either anterior or posterior skin tissue causes loss of the positional values, leading to the failure of limb regeneration. It is very likely that nonlimb skin does not have such limb positional information. Nerve deviation alone to the flank wound could not result in limb formation (Table 1). A classic study (Thornton, 1962) demonstrated that head skin grafting to a limb prevents limb regeneration in Ambystoma talpoideum. This indicates that dermal (skin) fibroblasts must be derived from a limb to carry positional information.

Muscle Cells in the Induced Accessory Limb on the Flank

The present study demonstrated that a limb ectopically induced on a flank possessed muscle tissues. It also demonstrated that myogenic cells in regenerating limbs were supplied from the muscle tissue of the limb stump and that dermal fibroblasts do not contribute to skeletal muscle (Kragl et al., 2009). Because dermal fibroblasts are likely to carry positional information, blastema cells derived from the dermis must be derived from the limb. In this regard, it was unclear whether the skeletal muscle of regenerating limbs must also be derived from the limb. In the present study, it was shown that the induced accessory limb on the flank possessed abundant muscle tissues (Fig. 3). The flank skin wound was created 3–10 mm from the limb, and the induction point of the accessory limbs was always more than 5 mm from a limb. Moreover, skin grafts did not contain muscle cells (Fig. 1G,H). Therefore, it is very likely that the muscle cells in the induced flank accessory limbs were derived from flank muscle tissues. This is not inconsistent with a previous report wherein muscle cells in an induced flank limb in a chick embryo were supplied from lateral nonlimb region somites (Heymann et al., 1996). In the chicken flank limb, Hgf/c-Met signaling plays a role in muscle migration (Heymann et al., 1996). As was the case with chicken, muscle migration in axolotl limb regeneration also appears to be controlled by Hgf/c-Met signaling (Fig. 5). Given the chick flank limb experiment and conserved mechanism of muscle migration, it is reasonable to consider that axolotl flank muscle tissues provide muscle precursor cells to the induced limb field.

The requirement of “limbness” during limb regeneration in urodele amphibians would be an informative insight toward achieving limb regeneration in higher vertebrates. A description of which types of cells can be prepared from any tissue and which type of tissues should be derived from a limb is important for future tissue engineering. The experimental system reported here is a very powerful system for obtaining this kind of information.

EXPERIMENTAL PROCEDURES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

Animals and Surgical Procedures

Animals were obtained from the Ambystoma Genetic Stock Center (University of Kentucky) and private breeders. They (4.5–8.0 cm, snout to tail tip) were maintained at 20–22°C in dechlorinated water. Animals were anesthetized using 0.1% ethyl 3-aminobenzoate methanesulfonate salt (Sigma, MS222), pH 7.0, for surgical procedures.

Limb amputation was performed on both upper and lower limbs. ALM surgery was performed as reported (Endo et al., 2004; Makanae and Satoh, 2012). Briefly, skin was dissected from the anterior side of an upper arm. Nerves were isolated and rerouted from the ventral side to the wound site with forceps and scissors, and nerve ends were placed at the center of the wound. After surgery, the animals were placed on ice to recover from the surgery.

To induce a superfluous limb on a flank, nerves from a limb were deviated to a skin wound created in a flank region, and two limb skin grafts (each on the opposite side of a limb; we usually used anterior and posterior skin) were placed beside the deviated nerve. To avoid confusing on orientation of skin grafts, skin grafts were always prepared from a zeugopdial part. Orientation of a stylopod is a little confusing (Makanae and Satoh, 2012). After surgery, the animals were placed on ice to recover from surgery. Five days later, approximately 5 μl vitamin A palmitate (Oil, Wako) was injected under the deviated nerve.

Kinase Inhibition

To test the effect of inhibiting the protein kinase activity of c-Met during limb regeneration, we amputated limbs and kept the animals in the presence of 0.9 μM c-Met Kinase Inhibitor (Calbiochem) or II or DMSO (Nacalai Tesque, Inc.) in dechlorinated water. Breeding water was changed every 2 weeks. At least five independent experiments were performed and gained the same results.

RT-PCR

Total RNA was prepared from regenerating blastemas and intact skin from forelimbs and hindlimbs using TriPure reagent (Roche). Total RNA was used as a template for first-strand cDNA synthesis using oligo (dT) primers. Superscript III reverse transcriptase (Invitrogen) was used for the extension according to the manufacturer's instructions. Each PCR cycle was performed as follows: 96°C for 15 sec; 58°C for 30 sec; 72° for 30 sec, and a final extension for 5 min at 72°C. We used 35 cycles (24 cycles for ribosomal protein L19) for each assay. The following primers were used: Ribosomal protein L19 forward, CATGGGCACTGGTAAGAGAAAAG; Ribosomal protein L19 reverse, GCGGCGCAAGATTCTCAT; hgf forward, GAGTTCTGTGCGGTTAAACC; hgf reverse, GCAAGTCCTTGCATTTGTAG; c-met forward, TGAAAAGCTCTTAACCTGCA; c-met reverse, AACAGATCCTGTCTTAGAATGG; MHC forward, CGCCACACATCTTCTCCAT, MHC reverse; GTCAGCAGAGGCCAGTTTTC; pax7 forward; GCGAGAAGAAAACCAAGCAC, pax7 reverse; GTCCGAATAGCTGGTGAAGC.

Immunohistochemistry

Immunohistochemistry of tissue sections was performed as described previously for tissue sections (Satoh et al., 2007) using an anti-Pax7 primary antibody (DSHB, 1/500), anti-MHC (DSHB, 1/50), anti-mouse IgG-Alexa 488 (Invitrogen), anti-mouse IgG-Alexa 594 (Invitrogen), and anti-rabbit IgG-Alexa 488 (Invitrogen). Nuclei were stained with Hoechst 33258 solution (Doindo). Images were captured using an Olympus BX51 microscope system.

In Situ Hybridization

To synthesize an antisense RNA probe, templates were synthesized using PCR with Ex Taq DNA polymerase (TaKaRa) and transcribed with T7 RNA polymerase (for cardiac actin, TaKaRa) or SP6 RNA polymerase (for MHC, TaKaRa). Specimens were fixed overnight at room temperature (RT) in 4% paraformaldehyde/phosphate buffered saline (PBS) and were then decalcified in 10% ethylenediaminetetraacetic acid at RT. Samples were treated with 5 μg/ml Proteinase K and refixation was performed by treating samples with 4% paraformaldehyde/PBS. The samples were hybridized in a solution containing RNA probes at 63°C overnight. After hybridization, the sections were washed with 50% formamide, 5× saline-sodium citrate (SSC) for 30 min at 63°C twice and with 50% formamide, 1.5× SSC for 30 min at 63°C three times. Sections were then washed with TBST (10 mM Tris-HCl, 150 mM NaCl, 0.1% Tween 20) at RT. Blocking was performed using 0.5% Blocking Reagent (Roche) for 30 min. The samples were then incubated with alkaline phosphatase (AP) -conjugated anti-digoxigenin-AP antibody (Roche, 1/1,000) for 2 hr at RT and then washed three times with TBST for 10 min. The staining reaction was performed using nitro-blue terazolium chloride (Wako) and 5-bromo-4-chloro-3′-indolylphosphatase p-toluidine salt (Wako) as a substrate for AP in a buffer containing 100 mM NaCl, 100 mM Tris (pH9.5), 20 mM MgCl2, and 0.001% Tween 20.

Histology

Dried tissue sections were immersed in tap water to remove Optimum Cutting Temperature (OCT) Compound (Sakura Finetek), stained with Alcian blue solution (Wako) for 3 min, washed with water, and then stained with hematoxylin (Wako) for 5 min and the washed with tap water for several minutes and stained with eosin (Wako) solution for 5 min, followed by washing with 70% ethanol. Sections were then dehydrated with ethanol and mounted using Softmount (Wako, Richmond, VA).

To visualize the cartilage of the regenerated skeletal elements in whole-mount preparations, we fixed samples in 10% Formalin Neutral Buffer Solution (Wako) overnight. We then incubated the samples in 70% ethanol, 1% HCl for 3–4 hr, after which they were stained with 0.1% Alcian blue in 70% ethanol, 1% HCl for 2–3 days. Samples were washed with 4% KOH without agitation for 2 hr at room temperature, followed by a solution of 2% KOH/50% glycerol for 1–3 days. Samples were cleared in 100% glycerol before photography.

To visualize cartilage and bone matrix of the regenerated skeletal elements in whole-mount preparations, we fixed samples in 10% Formalin Neutral Buffer Solution (Wako) overnight. We then incubated the samples in 90% ethanol, 10% acetic acid for 3–4 hr, after which they were stained with 0.1% Alcian blue in 90% ethanol, 10% acetic acid for 2–3 days. Samples were washed with 4% KOH for 2 hr at room temperature. They were stained with 0.4% Alizarin red in 4% KOH overnight, and the incubated in a solution of 2% KOH/50% glycerol for 1–3 days. Samples were cleared in 100% glycerol before to photography.

ACKNOWLEDGMENTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES

A.H. was funded by the Japan Society for the Promotion of Science and A.S. received a JSPS Grant-in-Aid for Young Scientists (B).

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. RESULTS
  5. DISCUSSION
  6. EXPERIMENTAL PROCEDURES
  7. ACKNOWLEDGMENTS
  8. REFERENCES