Mammals, including human beings, can not regenerate their jaws when they are damaged by disease or accidents, whereas some amphibians can completely regenerate their jaws, although the structure of the jaw is very similar among vertebrates (Goss and Stagg,1958; Ferretti,1996; Ghosh et al.,1996). How can amphibians regenerate their jaws? What is the difference between regenerative and non-regenerative animals? Amphibians' lower jaw retains a cartilage structure called “Meckel's cartilage” even after reaching an adult stage (Fig. 1A,B). In mammals, “Meckel's cartilage” disappears during development (Richman and Diewert,1988; Chung et al.,1995). Thus, there is a possibility that the existence of Meckel's cartilage may make a difference in the regenerative ability of the lower jaw.
However, other factors must also be involved, because the regenerative ability of jaws varies even among amphibians. Urodeles such as newts (e.g., Cynops pyrrhogaster) can regenerate both their upper and lower jaws even in adulthood (Goss and Stagg,1958; Ferretti,1996; Ghosh et al.,1994,1996; Brockes,1997). On the other hand, most anuran amphibians, for example, the West African clawed frog (Xenopus tropicalis), can not regenerate their jaws at all despite the fact that they also possess Meckel's cartilage. Thus, it was of interest to compare the cellular and molecular events between newts and frogs after jaw amputation and investigate the factors causing the difference of regenerative ability between them. Previous studies on newt jaw regeneration were mainly based on histological observations. They demonstrated that, first, the cut surface of the jaw is covered by the wound epithelium until 7 days after amputation. Second, 14 days after amputation, a blastema is formed. At 28 days after amputation, cartilage starts to regenerate, and 56 days after amputation, new bone formation begins, and, finally, regenerated bone fuses at the center of the mandible (Ghosh et al.,1994). Here we compared the regenerative events at the molecular level between newts and frogs after lower jaw amputation.
Past research has suggested that, during newt lens, limb, and tail regeneration, terminally differentiated cells can dedifferentiate and regain the ability to proliferate or differentiate (Brockes,1997; Maki et al.,2007). These dedifferentiated cells contribute to the blastema, where multipotent mesenchymal cells grow and regenerate new tissues (Odelberg,2005). The high regenerative ability of newt is sometimes explained as being due to the dedifferentiation ability of newts (Okada,1991). However, there have also been some studies that showed that the events occurring at the early stage of regeneration in newts and frogs are very similar (Endo et al.,2000; Suzuki et al.,2006).
In the present study, we directly compared the regeneration processes in newts versus frogs at the cellular and molecular levels in order to investigate the factors causing the difference of jaw regenerative ability between them. First, we checked the proliferative activity of cells after amputation and then investigated the activity of genes necessary for regeneration.
Observation of Lower Jaw Bone Regeneration Using Micro-CT
The anatomical structure of the lower jaw of newts (Fig. 1) is similar to that of frogs and also that of mammals. To confirm the regenerative ability of Japanese newts and roughly determine the time-course of jaw regeneration, we observed the jaw regeneration process macroscopically, as well as by examining histological sections and performing micro-computed tomography (micro-CT) scanning. We could follow the complete process of regeneration of the lower jaw bone by micro-CT observation. Before amputation, the lower jaw bone was fused at the center of the mandible (Fig. 2A,D). Immediately after amputation, half of the mandible was absent (Fig. 2B,E), and 180 days after amputation, the regenerating lower jaw bone was almost completely fused at the center of the mandible (Fig. 2C,F; regenerated parts are indicated in sky blue). Also, there was a small region of X-ray opacity that represented tooth regeneration in the mandible (Fig. 2F, red arrowhead). However, at the same stage (180 days after amputation) in frogs (Xenopus tropicalis), there was no new bone regeneration like that in newts (data not shown).
Comparison of Proliferative Ability of Cells After Amputation
We speculated that the proliferative ability of cells might be different between the two animals and that this might cause the difference of regenerative ability. In order to assess the number of proliferative cells that resulted from the stimulation by amputation, we labeled proliferating cells using BrdU pulse labeling. In order to decide the time-course for performing the BrdU experiment, we performed hematoxylin and eosin staining of the amputated newt jaw. At 4 days after amputation, some interesting phenomena could be detected. The muscle lying beneath the cut surface was weakly stained with eosin compared with intact muscle and seemed to have eroded. At the same time, the tip of Meckel's cartilage was weakly stained with hematoxylin compared with intact Meckel's cartilage. Based on these observations, we chose day 4 post-amputation to analyze the BrdU incorporation. In the intact lower jaw of newts and frogs, BrdU-positive cells could be detected in the basal cells of the epithelium (Fig. 3A,D, white arrowheads), but few mesenchymal cells were BrdU positive (Fig. 3A,D, yellow arrowheads). Four days after amputation, the cells lying under the cut surface in newts began to actively incorporate BrdU (Fig. 3B, yellow arrowheads). Fourteen days after amputation, the number of cells that incorporated BrdU in the mesenchymal space markedly increased (Fig. 3C, yellow arrowheads). These findings suggested that proliferation of cells in the mesenchymal space underlying the wound surface was activated to regenerate the lost parts. However, unexpectedly, similar phenomena were also observed in frogs (Fig. 3E,F, yellow arrowheads). In these experiments, we confirmed that there was no BrdU signal in the negative control, which was only stained with secondary antibody, or in samples that were not injected with BrdU (see Supplemental Fig. 1, which can be viewed at www.interscience.wiley.com/jpages/1058-8388/suppmat). Thus, the signals that can be seen in Figure 3 are really positive signals, and we concluded that there was no marked difference between newts and frogs regarding the number of cells that incorporated BrdU after amputation (Fig. 3).
Comparison of Proliferative Ability of Differentiated Cells
To investigate possible differences of the type(s) of cells that were stimulated to proliferate in response to amputation between newt and frog, we performed immunohistochemical staining with antibodies that recognized cell-type-specific proteins (myosin heavy chain: muscle; collagen type II: cartilage) together with BrdU pulse labeling.
In the intact mandible of newts and frogs, there were almost no BrdU-positive cells present in MHC protein-positive muscle fibers before amputation (Fig. 4A,E). Four days after amputation, some cells (mesenchymal cells) lying beneath the epithelium and muscle tissues became BrdU positive in newts (Fig. 4B, white arrowhead), but this phenomenon was not specific to newt and we could also detect some BrdU-positive cells in muscle tissues at 4 days after amputation of the frog mandible (Fig. 4F, white arrowhead).
Similar to the findings in muscle, both newt and frog Meckel's cartilage contained almost no BrdU-positive cells in the intact mandible (Fig. 4C,G). However, 4 days after amputation, some BrdU-positive cells appeared inside the Col2-containing extracellular matrix, not only in newts (Fig. 4D, white arrowhead) but also in frogs (Fig. 4H, white arrowhead). Also, in a 3D reconstruction movie, BrdU-positive cells could clearly be detected inside the collagen type II–containing extracellular matrix (see Suppl. Fig. 2) Thus, cell cycle reentry of differentiated cells was observed in both newt and frog, and we could not observe any prominent differences in BrdU incorporation experiments between the two animals.
Identification and Characterization of the MHC and Col2 Genes of Newt and Frog
Next, we investigated the mRNA levels of genes necessary for the differentiation of muscles and cartilage, and compared the levels between the two animals. In order to assess the mRNA levels after amputation, we cloned the orthologs of the MHC and type II collagen genes from newt and frog for use in detecting the mRNA levels of skeletal muscle- and cartilage-specific genes, respectively.
In the case of MHC genes, we found several MHC clones in the newt EST database generated in our laboratory. One of them (Cp-aL-002-H19) was confirmed to be expressed in skeletal muscles in the lower jaw, and we then tried to isolate its orthologous gene from Xenopus tropicalis by PCR using specific primers. We compared the clones we obtained from newt (Cp-aL-002-H19) and frog (XtMHC2) with skeletal-muscle-type MHC genes (MHC1, 2, 4, 8, 13) of other organisms. This comparison indicated that Cp-aL-002-H19 and XtMHC2 showed a very high similarity of the Smc domain at the amino acid level (see Suppl. Fig. 3A). Also, in phylogenetic analysis, these two genes were placed in the same cluster (Fig. 5A). Thus, in the experiments described below, we used these two genes as probes to detect skeletal muscle-specific gene activity.
In the case of type II collagen genes, a full-length cDNA clone for a newt collagen type II gene (called Col2) was previously cloned from limb cartilage by Yoshizato's group (Asahina et al.,1999). We obtained it from their laboratory and then tried to clone its ortholog by PCR using specific primers. We compared the cloned Col2 genes of newt and frog with collagen type I and II genes of other organisms. In accord with the findings for MHC, the newt and frog Col2 showed a high similarity of the COLF1 domain at the amino acid level (see Suppl. Fig. 3B). In phylogenetic analysis, both genes were placed in the same cluster with the Col2 genes of other organisms (Fig. 5B). We used these two genes as probes to detect the mRNA levels of cartilage-specific gene activity.
Expression Pattern of MHC mRNA in Amputated Mandible in Newt and Frog
Uniform distribution of MHC mRNA was observed within the muscle fibers in the intact mandibles of both newt and frog by in situ mRNA hybridization with alkaline phosphatase detection, although the signal in newt was rather weak (Fig. 6A,D). We also confirmed that both sense probes gave no specific signals in intact adult lower jaws (data not shown). However, 7 days after amputation, the mRNA levels of the MHC genes in the two animals showed opposite responses. In newt, strong MHC mRNA expression appeared at the end of muscle fibers (Fig. 6B,C). According to observations of serial sections, it seemed that more than half of the muscle fibers lying under the cut surface were expressing MHC mRNA at their tips (data not shown). In frogs, we could not detect such expression of the MHC mRNA signal at the tip of the muscle fibers beneath the cut surface (Fig. 6E,F). Rather, the signal for MHC mRNA disappeared in regions close to the cut surface, although the signal of MHC protein remained. These results suggest that the two animals had opposite responses regarding the regulation of MHC mRNA expression. To quantitatively analyze the MHC mRNA expression in the newt jaw, we performed real-time RT-PCR experiments comparing the gene expression level between intact and regenerating jaws. We used the distal half of the lower jaw (not a section) of intact and regenerating newts (day 7 post- amputation) as a sample. There was no significant difference between intact and regenerating jaws, suggesting that the region where MHC mRNA is expressed is limited to just the tip of the regenerating muscle (Suppl. Fig. 4).
Dynamic Change of MHC mRNA Expression Pattern During Newt Jaw Regeneration
To more precisely define the cell types and the region where the MHC mRNA was upregulated in newts, we stained MHC mRNA using a fluorescent probe in samples from animals labeled with BrdU. In the intact newt jaw, we could not detect a specific signal of MHC mRNA in muscle fibers when we used this fluorescent probe (Fig. 7A), probably because the sensitivity of fluorescent probes is slightly lower than that of alkaline phosphatase detection. At 4 days after amputation, a strong MHC mRNA signal could be detected at the tip of muscle fibers lying beneath the cut surface (Fig. 7B). At 7 days after amputation, the region expressing MHC mRNA was expanded (Fig. 7C).
There were two types of cells present in the region that showed strong expression of MHC mRNA. First, there were some multinucleate cells that strongly expressed MHC mRNA (Fig. 7C, white arrowheads; see also Fig. 7D). Some of the cells in this population incorporated BrdU (Fig. 7E). Hereafter, these cells will be referred to as “Type 1 cells.” Second, there were some cells expressing MHC mRNA strongly at the tip of the muscle fiber (Fig. 7C, white arrows; Fig. 7F). Some of these cells incorporated BrdU (Fig. 7G). Hereafter, these cells will be referred to as “Type 2 cells.”
Comparison of Newt and Frog Regeneration With Respect to “Satellite Cell” Contribution
To elucidate whether “satellite cells” (stem/progenitor cells of muscle) contribute to these cells involved in jaw muscle regeneration, we performed immunostaining of intact and amputated newt and frog jaws using an antibody against Pax7. In the intact newt jaw, we could detect Pax7-positive cells in muscle tissues (Suppl. Fig. 5 A, red arrowheads). In the regenerating newt jaw, we could detect some cells that were double positive for Pax7 and BrdU (Fig. 8A,B, red arrowheads). However, not all cells that expressed Pax7 incorporated BrdU (Fig. 8A,B, white arrowheads); rather, there were fewer double-positive cells than Pax7-single-positive cells. On the other hand, we could not detect any Pax7-positive cells in the intact (Suppl. Fig. 5C) or amputated (Fig. 8C,D) frog jaw, although we could detect Pax7 signals in the regenerating tadpole tail (Fig. 8E, red arrows). Also we could not detect Pax7 signals in adult frog intact limbs (Suppl. Fig. 5E).
Expression of Col2 mRNA in Amputated Lower Jaw of Newt and Frog
In the intact newt lower jaw, we could detect a strong signal of Col2 mRNA in the mesenchymal cells of teeth (data not shown), but not in Meckel's cartilage or periosteum cells (Fig. 9A). At 7 days after amputation, cells that strongly expressed Col2 mRNA appeared at the cut surface of the mandible (Fig. 9B, red arrow), and tooth mesenchymal cells still maintained strong expression of Col2 mRNA (Fig. 9B, white arrows). In frogs, even in the intact lower jaw, there were detectable signals of Col2 mRNA in periosteum cells (Fig. 9C, red arrow) and in Meckel's cartilage cells (Fig. 9C, white arrows). At 7 days after amputation, the region expressing Col2 mRNA was extended both in the periosteum (Fig. 9D, red arrows) and Meckel's cartilage (Fig. 9D, white arrow). Thus, cartilage-specific mRNA was strongly up-regulated in the newt after amputation, whereas in frog it was constantly expressed before and after amputation.
Similar Cell Proliferative Activity in Newt and Frog
The incorporation of BrdU in the intact newt lower jaw mainly occurred in cells lying at the base of the epithelium, mesenchymal cells in teeth, and gland cells. After lower jaw amputation, in addition to these cells, some cells in the region of muscle tissues (Fig. 4B), and some cells embedded in Meckel's cartilage (Fig. 4D, Suppl. Fig. 2) became BrdU positive. These results indicate that the cells that newly incorporated BrdU after lower jaw amputation were reacting to the amputation stimulation, and were going to contribute to blastema formation. Past studies of limb regeneration suggested that muscle cells, mesenchymal cells, cartilage cells, and nerve cells contribute to the blastema (Muneoka et al.,1986; Lo et al.,1993; Kumar et al.,2000; Echeverri et al.,2001; Bryant et al.,2002; Echeverri and Tanaka,2002). The present results suggest that the origin of cells of the lower jaw-regenerating blastema is similar to the origin of cells in the blastema during limb- or tail-regeneration.
The simplest possible reason for differences of regenerative ability among animals would be a difference of proliferative activity of cells underlying the wound surface. Based on previous studies, we speculated that differentiated cells of newts can reenter the cell cycle, but those of frogs can not. However, we found that the number and types of cells that newly incorporated BrdU after lower jaw amputation were almost the same in frogs and newts. Unexpectedly, the combined analysis of BrdU incorporation and cell type markers clearly indicated that the differentiated cells of frog could reenter the cell cycle after amputation, like those of newt. Thus, we should consider other reason(s) to explain the difference in the regenerative ability of these animals.
MHC mRNA Is Expressed at the End of Muscle Fibers in the Newt Jaw but not in the Frog Jaw
Although we could not detect significant differences of cell proliferative activity between newt and frog, we succeeded in finding an interesting difference between the two animals regarding the level of MHC mRNA during jaw regeneration. In newt, there was remarkable expression of MHC mRNA at the tip of muscle fibers beneath the cut surface at 7 days after amputation (Fig. 6A–C). However, in the case of frogs, such expression could not be detected (Fig. 6D–F). These results suggest that the expression of MHC mRNA plays an important role in the regeneration of muscle. However, we could not detect a significant difference in the MHC gene expression level between the intact and regenerating newt jaw by real-time PCR (Suppl. Fig. 3). Taken together, these results suggest that the region expressing MHC mRNA is limited to the very tip of the muscle of the regenerating jaw.
Further study of this event in newt lower jaw regeneration revealed that there are two types of cell populations in which MHC mRNA is up-regulated. The first type are cells that express MHC mRNA strongly and MHC protein weakly and are multinucleate (Fig. 7C, white arrowheads; D). We called these cells “Type 1 cells.” The second type (Type 2 cells) are cells that express MHC mRNA strongly, and are present only at the tip of muscle fibers (Fig. 7C, white arrows; F). There were some cells that incorporated BrdU among the Type 1 cells and Type 2 cells (Fig. 7E,G; white arrows). These results suggest that Type 1 cells might be derived from some kind of stem/progenitor cells and are going to differentiate to muscle fibers. On the other hand, Type 2 cells appeared to be cells present in the cut muscles remaining in the amputated lower jaw. They expressed MHC mRNA strongly in a restricted region at the tip of the muscle fibers, and the nuclei located there were entering S phase. Type 2 cells might correspond to newt myotubes in vitro that reenter the cell cycle when stimulated by serum (Tanaka et al.,1997). Altogether, these results suggest that there are two types of cell lineages that contribute to muscle regeneration. The first is a lineage from stem cells (Type 1 cells; Fig. 10A), and the second is a lineage from muscle fibers (Type 2 cells; Fig. 10B).
In the case of frogs, the cells defined as “Type 1 cells” and “Type 2 cells” in newts could not be detected. The fact that there were BrdU-incorporating cells but no Type 1 cells in the amputated frog jaw suggests that in frogs, stem/progenitor cell-like cells can enter S phase, but can not fuse or produce MHC mRNA (Fig. 10A). There were also no Type 2 cells in the amputated frog jaw, which suggests that frog muscle fibers are unable to express MHC mRNA at the tip of the amputated muscle fibers and, therefore, can not repair the muscle or produce cells to regenerate muscle (Fig. 10B).
The Role of “Satellite Cells” in Newt and Frog After Lower Jaw Amputation
To investigate the role of “satellite cells” in newt and frog lower jaws after amputation and whether “Type 1 cells” are derived from stem/progenitor cells of muscle, we performed immunohistochemistry using antibody that recognizes Pax7. Recently, there was a report demonstrating the existence of Pax 7-expressing satellite cells in newt, and also investigating their proliferative ability and multipotency (Morrison et al.,2006). In our study, Pax7-positive cells could be detected in the regenerating as well as intact newt jaw. In the intact jaw, Pax7-positive cells were present in muscle tissues, and almost none of these cells incorporated BrdU (Suppl. Fig. 5A, red arrowheads). On the other hand, at 7 days after amputation, there were some cells lying beneath the wound epithelium that were Pax7-positive and incorporated BrdU. Also, these kinds of cells were clustered, like “Type 1 cells” (Fig. 8A,B; red arrowheads). These results strongly suggest that these “Type 1 cells” in newts were derived from the satellite cell population. However, in the case of frog, we could not detect Pax7-positive signals in either the intact or amputated jaw, although there were many cells incorporating BrdU in the amputated jaw (Fig. 8C,D, white arrows; Suppl. Fig. 5C). In the case of the regenerating tail in tadpoles, we could detect Pax7-positive signals (Fig. 8E, red arrows). Also, it was reported that frogs can regenerate their tails at the tadpole stage, and that satellite cells play a crucial role in muscle regeneration in this case (Chen et al.,2006).
These observations lead us to speculate that in frog muscle, the expression of Pax7 decreases as the developmental stage advances and the developing frogs lose their ability to regenerate muscles. Moreover, they suggest that the cellular activities of both satellite cells and differentiated muscles may have important roles in the regenerative ability of animals.
Difference of Col2 mRNA Expression Pattern in Lower Jaw Regeneration of Newt and Frog
In the intact newt lower jaw, we could detect Col2 mRNA in mesenchymal cells in teeth but not in Meckel's cartilage or periosteum cells (Fig. 9A). It is known that type I collagen is localized in newt tooth dentine (Kogaya,1999). The present results suggest that the Col2 gene plays an important role in newt tooth development or maintenance. In addition, 7 days after amputation, we could detect cells strongly expressing Col2 mRNA at the surface of the amputated bone (Fig. 9B). This finding suggests that these cells with up-regulated Col2 mRNA expression are the origin of regenerated cartilage.
In contrast to newts, in the case of the intact frog jaw, we could already detect a signal of Col2 mRNA expression in Meckel's cartilage and periosteum cells (Fig. 9C). At 7 days after amputation, the region of periosteum cells expressing Col2 mRNA was extended, whereas the pattern and strength of Col2 mRNA expression in Meckel's cartilage were not markedly changed (Fig. 9D). This finding suggests that the periosteum cells expressing Col2 mRNA were the origin of the partially regenerated cartilage in the frog.
The Critical Difference Between Newt and Frog After Jaw Amputation
The results of the present study indicate that the reason why newts can regenerate their lower jaw but frogs can not does not depend on the ability of cells to proliferate. Two cell lineages appeared to contribute to newt muscle regeneration. The first lineage consisted of cells derived from some kind(s) of stem cells (mesenchyme cells, satellite cells, or side population cells) and the second lineage was derived from muscle tissues. Also, MHC mRNA was expressed in both cell lineages. However, in the case of frog, although the cells that incorporated BrdU existed, expression of MHC mRNA did not occur. These results suggest that newts can produce cell-specific mRNA during the regeneration process from very early stages, and this ability plays an important role in making newts able to regenerate various tissues and appendages. In the case of lens regeneration, previous studies demonstrated that the expression of lens-specific mRNA is required for lens cell differentiation or transdifferentiation (Agata et al.,1983,1993).
The findings of the present study suggest that Pax7-positive cells (satellite cells) exist in the adult newt jaw but not in the adult frog jaw, and these cells play an important role in the regeneration of muscle cells (Fig. 10A,C). A previous report demonstrated that transplantation of Pax7-positive cells into the limb blastema of frog rescued muscle regeneration, but the rescued muscle was not able to achieve a completely normal muscle equivalent to an intact limb muscle (Satoh et al.,2005). All these observations taken together suggest that Pax7-positive satellite cells (Type 1 cells in newt) are necessary but not sufficient for muscle regeneration. In the case of newt, two types of cells (Type 1 cells, Type 2 cells) participate in muscle regeneration, as mentioned above. These facts lead us to speculate that, to regenerate complete muscle tissues, “Type 2 cells” are required for the regeneration of core muscle tissues, and that “Type 1 cells” recruited by these core regenerating tissues are also essential (Fig. 10C).
The present findings provide a novel view of vertebrate regeneration, and also provide an important clue for investigating the basis of the ability of newts to regenerate various tissues and appendages. To confirm the differences of gene activation during muscle regeneration, we are now making transgenic newts and frogs that carry the GFP gene with muscle-specific promoters and enhancers.
Usefulness of Micro-CT for Amphibian Regeneration Research
It has been reported that dentary bone regenerates and fuses at the center of the mandible in newts 6 months after amputation, as demonstrated using 2-dimensional X-ray analysis (Graver,1974). We detected the regenerated dentary bone fused at the center of the mandible and also regenerating teeth in more detail using 3-dimensional analysis (Fig. 2F, red arrowhead). It is also known that regeneration of the lower jaw bone is restricted to the dentary bone, and the innermost prearticular bone is replaced by cartilage (Ghosh et al.,1996). We could also clearly observe that only the dentary bone regenerated (Fig. 2F). A study using micro-CT to investigate limb bone regeneration in newt has been reported (Stock et al.,2003). Altogether, the results indicate that micro-CT is more useful for analyzing hard tissue regeneration than 2-dimensional X-ray analysis. However, fully regenerated jaws at 180 days were not observed in newts that had undergone more than a few micro-CT scans (data not shown). Moreover, the regeneration stage of individuals whose CT images had been taken several times after amputation was markedly delayed compared with the stages at comparable times in a past report (Ghosh et al.,1996). This suggests that the rate of regeneration is adversely affected by X-ray irradiation. Therefore, it is very important to consider the conditions and dose of X-rays when using the micro-CT system for regeneration research.
Adult Japanese fire-bellied newts, Cynops pyrrhogaster, were collected from rice field or ponds in Okayama and Hyogo prefectures, Japan. Adult West African frogs, Xenopus tropicalis, were a kind gift from the National BioResource Project (Institute for Amphibian Biology, Graduate School of Science, Hiroshima University). Both were kept in plastic containers in tap water at room temperature and fed twice a week.
Surgical Procedures and BrdU Injection
Animals were anesthetized in 0.1% tricaine (3-aminobenzoic acid ethylester methanesulphonate salt; Sigma, Poole, UK) in tap water. The distal half of the lower jaw was amputated using fine scissors, leaving the tongue and hyoid apparatus intact. After the amputation, the cut surface was trimmed to make it flat. After surgery, animals were allowed to recover from anesthesia in a shallow solution of 0.5% sulfamerazine (Sigma, Poole, UK) for 12–24 hr and thereafter kept individually in tap water in containers. The animals were sacrificed and the amputated jaws were collected at various times. In some experiments, 24 hr before sacrifice, animals were injected intraperitoneally with BrdU (250 μg/g body weight).
The following primary antibodies were used: mouse monoclonal anti-myosin heavy chain IgG (Upstate, Lake Placid, NY), mouse monoclonal anti-collagen type II IgG (Chemicon, Temecula, CA), rat monoclonal anti-bromodeoxyuridine IgG (Oxford Biotechnology, Kidlington, Oxfordshire), and mouse monoclonal anti-Pax7 IgG (Developmental Studies Hybridoma Bank, Iowa City, IA). For immunofluorescence studies, primary antibodies were detected with appropriate species-specific Alexa Fluor-or FITC-conjugated secondary antibodies (Invitrogen, Aurora OH).
Specimens were fixed overnight at 4°C in 4% paraformaldehyde in 70% PBS, and decalcified with 0.5 M EDTA (pH 7.4) for 1 week. For detection of Pax7, tissues were fixed at 4°C in 2% paraformaldehyde and equilibrated in 30% sucrose and embedded in Tissue-Tek (O.C.T compound, Sakura). Frozen sections (14 μm thick) were mounted on adhesive slides. Other samples were dehydrated with a graded series of ethanol, embedded in paraffin wax, and sectioned at 5–10 μm. After dewaxing, sections were incubated in 2 × SSC buffer with 0.05% Triton X-100 and 0.05% saponin (MP Biomedicals, Solon, OH). The sections were rinsed with 2 × SSC and blocked in TNB blocking buffer (0.1 M Tris-HCl, pH 7.5, 0.15 M NaCl, 0.5% Blocking reagent; Perkin Elmer, Waltham, MA) for 1 hr. Sections were incubated with appropriate primary antibodies (anti-myosin heavy chain, anti-collagen type II, or anti-Pax7) overnight at 4°C and washed several times in TNT Buffer (0.1 M Tris-HCl, pH 7.5, 0.15 M NaCl, 0.05% Tween-20), and incubated with the appropriate secondary antibody for 1 hr at room temperature. After color detection was performed, in the case of double staining of BrdU, the samples were treated with 2 M HCL at room temperature for 5 min and reacted overnight in 4°C with rat anti-BrdU (Oxford Biotechnology Ltd., Kidlington, Oxfordshire). Fluorescein-conjugated goat affinity-purified antibody to rat IgG (Cappel, Aurora, OH) was used as the secondary antibody. Finally, the sections were counterstained with Hoechst 33258 (Sigma, St. Louis, MO) and mounted with fluorescent mounting medium (DakoCytomation, Carpinteria, CA).
EST clone Cp_aL_002_H19 was used as a clone for newt myosin heavy chain (CpMHC2) cDNA. A plasmid containing newt α1(II) collagen cDNA (Asahina et al.,1999) was a kind gift from K. Yoshizato (Hiroshima Univ). Xenopus tropicalis myosin heavy chain 2 (XtMHC2) cDNA was cloned by polymerase chain reaction (PCR) with the following gene-specific oligonucleotide primers: 5′ CTGAATGACACTGTCATTGGC 3′ and 5′ CAGCTGATCTTC- AAGGGTG 3′, corresponding to the homologous region of the CpMHC2 insert. Xenopus tropicalis collagen type II (XtCol2) cDNA was cloned by PCR using primers 5′ GAAGCCACTTTGAAATCCC 3′ and 5′ GG- AAAGTACTCAGGTCCTTAG 3′ corresponding to the homologous region of newt collagen type II (CpCol2) cDNA.
In Situ Hybridization
The digoxigenin-labeled RNA probes used in this study, CpMHC2, CpCol2, XtMHC2, and XtCol2, were prepared using a DIG RNA labeling kit according to the manufacturer's protocol (Roche, Basel Switzerland) by using each cDNA clone (see above) as the template. Sections were made by the same procedure as described for immunohistochemistry. After dewaxing, sections were digested with 1 μg/ml proteinase K in 70% PBS at 37°C for 10 min, fixed again in 4% paraformaldehyde in 70% PBS at 4°C for 20 min, and acetylated with acetic anhydride in triethanolamine buffer (Hayashi et al.,1978). The slides were covered with a coverslip, and hybridized overnight at 55°C with heat-denatured DIG-labeled probes in 50% de-ionized formamide, 5 × SSC, 0.1% Tween-20, 0.1 mg/ml heparin sulfate, 1 mg/ml yeast tRNA, and 10% dextran sulfate. After hybridization, the slides were washed in 50% formamide, 5 × SSC and 0.1% Tween-20 3 times for 1 hr each at 55°C, and in 2 × SSC twice for 15 min each at room temperature. The slides were rinsed in Buffer-1 (0.1 M maleic acid, 0.15 M NaCl, 0.1% Triton X-100), incubated in Buffer-2 (1% Blocking reagent, Roche, in Buffer-1) for 30 min and then in Buffer-2 containing alkaline phosphatase (AP)-conjugated anti-digoxigenin antibody (Roche) diluted 1:300 at 4°C overnight. They were rinsed in TPBS (0.1% TritonX-100 in PBS) 3 times for 15 min each, and in TMN buffer (0.1 M NaCl, 0.1 M Tris-HCl, pH 9.5, 0.05 M MgCl2) for 10 min. A mixture of BCIP/NBT was used for color development of the alkaline phosphatase-conjugated anti-DIG-antibody (Roche). In the case of fluorescence in situ hybridization, we used a TSA kit (Invitrogen) for color detection. In the case of double or triple immunohistochemical staining, the sections used for detection of mRNA signals were incubated with TNB blocking buffer, and reacted with the appropriate first antibody overnight at 4°C. The slides were washed with TNT buffer 3 times for 10 min each, and reacted with the appropriate secondary antibody. Finally, the sections were counterstained with Hoechst 33258 (Sigma) and mounted with fluorescent mounting medium (DakoCytomation).
The expression of MHC mRNA was analyzed by quantitative real-time PCR. Total RNA was isolated using an RNeasy mini kit according to the manufacturer's protocol (Qiagen, Chatsworth, CA). Total RNA samples were prepared from approximately 20 mg each of tissue from intact newt distal and proximal jaw and regenerating (day 7 postamputation) distal and proximal jaw. Isolated total RNA was reverse transcribed to cDNA with M-MLV reverse transcriptase (Invitrogen) in the presence of oligo (DT) 12-18 primer (Invitrogen) and dNTPs for 50 min at 42°C. For quantitative real-time PCR, each synthesized cDNA was analyzed using the LightCycler system (Roche). The FastStart DNA Master SYBR Green Kit (Roche) and the LightCycler System were used as described in the manufacturer's manuals. Oligonucleotide primers 5′ CGACCTCTCCAGCATAAAGATCGA- GTT 3′ and 5′ GCTCTCGCCAGTCCTGTCGAAG 3′ were used for MHC, and 5′ GCATGCTGTGACTGCTACACAAAAG 3′ and 5′ GCTGGAATGATATTCTGGTTTGCAC 3′ were used for GAPDH. Each reaction was carried out in a total volume of 20 μl in a glass capillary tube containing 2 μl of cDNA sample, 3 mM MgCl2, 10% LightCycler-DNA Master SYBR Green buffer (Taq DNA polymerase, reaction buffer, deoxynucleotide triphosphate mix, and SYBR Green dye), and each primer at 0.5 μM. The data represent the mean ± SD calculated from three independent experiments.
For micro-CT analysis, animals were anesthetized and kept moist with wet paper towels. A LaTheta LTC-100 was used for obtaining X-ray images, and 3D reconstruction was done using VGStudio MAX software.
We thank Professor Katsutoshi Yoshizato for providing the newt Col2 cDNA clone. We thank Dr. Nobuyasu Maki for teaching us techniques for using newt samples. We thank Dr. Shigeru Kuratani for letting us use the micro-CT instrument. We thank Professor Yoshirou Yaoita for providing us with Xenopus tropicalis. We thank Hiroshi Kubota for teaching us techniques for handling Xenopus. We thank Dr. Elizabeth Nakajima for careful reading of the manuscript. This study was supported by the Naito Foundation, Global COE for Evolution and Biodiversity, and Project for Realization of Regenerative Medicine as well as a Grant-in-Aid for Creative Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan, to K.A.