Volume 241, Issue 10 p. 1575-1583
Research Article
Free Access

Lens regenerates by means of similar processes and timeline in adults and larvae of the newt Cynops pyrrhogaster

Takeshi Inoue

Corresponding Author

Takeshi Inoue

Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto, Japan

Department of Biophysics, Graduate School of Science, Kyoto University, Kitashirakawa-Oiwake, Sakyo-ku, Kyoto 606-8502, JapanSearch for more papers by this author
Ryo Inoue

Ryo Inoue

Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto, Japan

Search for more papers by this author
Rio Tsutsumi

Rio Tsutsumi

Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto, Japan

Search for more papers by this author
Kikuo Tada

Kikuo Tada

Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto, Japan

Search for more papers by this author
Yuko Urata

Yuko Urata

Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto, Japan

Search for more papers by this author
Chiaki Michibayashi

Chiaki Michibayashi

Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto, Japan

Search for more papers by this author
Shota Takemura

Shota Takemura

Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto, Japan

Search for more papers by this author
Kiyokazu Agata

Kiyokazu Agata

Department of Biophysics, Graduate School of Science, Kyoto University, Kyoto, Japan

Search for more papers by this author
First published: 28 August 2012
Citations: 10

Abstract

Background: It is widely accepted that juvenile animals can regenerate faster than adults. For example, in the case of lens regeneration of the newt Cynops pyrrhogaster, larvae and adults require approximately 30 and 80 days for completion of lens regeneration, respectively. However, when we carefully observed lens regeneration in C. pyrrhogaster at the cellular level using molecular markers in the present study, we found that lens regeneration during the larval stage proceeded at similar speed and by means of similar steps to those in adults. Results: We could not find any drastic difference between regeneration at these two stages, except that the size of the eyes was very different. Conclusions: Our observations suggested that larvae could regenerate a lens of the original size within a shorter time than adults because the larval lens was smaller than the adult lens, but the speed of regeneration was not faster in larvae. In addition, by repeatedly observing the regeneration in one individual transgenic newt that expressed fluorescence specifically in lens fiber cells in vivo and comparing the regeneration process at the embryonic, larval, and postmetamorphosis stages, we confirmed that the regeneration speed was the same at each of these stages in the same individual. Developmental Dynamics 241:1575–1583, 2012. © 2012 Wiley Periodicals, Inc.

INTRODUCTION

It is widely assumed that fetal and juvenile animals can regenerate faster and more effectively than adults. Indeed, there are animals, such as Xenopus laevis, that possess regenerative ability during the embryonic and larval stages, but that lose this ability as they mature (Dent,1962; Strub,1979; Muller et al.,1999; Endo et al.,2000; Colwell et al.,2003; Satoh et al.,2005; Yakushiji et al.,2007). On the other hand, some animals such as urodeles maintain high regenerative ability throughout their lives (Borgens,1982; Kawakami et al.,2004; Han et al.,2008; Muneoka et al.,2008a, b; Eguchi et al.,2011). However, even in such highly regenerative animals, juveniles regenerate lost parts faster than adults, raising the question of whether the regenerative mechanisms are different between the larval and adult stages, and/or whether juvenile cells may have higher regenerative ability than adults cells. To answer these questions, here we investigated the regeneration process of the larval lens in comparison to that of the adult lens.

Newts such as C. pyrrhogaster are urodele amphibians renowned for their remarkable regenerative ability that persists throughout their lives. They can completely regenerate lost portions of their body such as lens, limbs, jaw, heart, brain, and so on (Okamoto et al.,2007; Kurosaka et al.,2008; Maden,2008). In newts that lose a lens, the lens regenerates only from dorsal iris pigmented epithelial cells (PECs), which first dedifferentiate and subsequently differentiate into a completely different cell type to form a lens. This lens regeneration is called Wolffian lens regeneration (Wolff,1895). The lens does not regenerate from the ventral iris PECs, although both dorsal and ventral PECs show proliferative activity and transdifferentiation into lens cells (Eguchi,1988; Del Rio-Tsonis and Tsonis,2003). Clarifying the regenerative mechanisms and the differences between dorsal and ventral PECs at the cellular and molecular levels during lens regeneration provides an opportunity to investigate the boundary between regenerative and nonregenerative conditions.

The phenomena and mechanisms of regeneration in vertebrates have been studied for a long time. For example, the time course of the processes of lens regeneration in C. pyrrhogaster was reported in 1940 (Sato,1940). Previous studies have shown that the cytological changes of the dedifferentiated cells of the iris epithelium during lens regeneration in embryonic and larval newts of Notophthalmus viridescens were similar to those in the adult (Reyer,1982a, b). However, the timing of the process of lens regeneration in larval newts is still unclear, and much remains to be understood about the molecular and cellular processes that determine the onset and progression of this regeneration. Reasons for the slow progress of the research on these processes include the inability to perform genetic analysis and the long time required for sexual maturation and production of the F1 generation in newts (more than 3 years in C. pyrrhogaster). The difference of size between the adult and the larva of the newt C. pyrrhogaster (Fig. 1) complicates comparative analyses. Furthermore, the available DNA sequence data of the newt is not sufficient to analyze which genes are activated in the regenerating region, although many molecular biological approaches have been attempted in newts (Grogg et al.,2006; Maki et al.,2010). Recently, high-efficiency transgenic techniques have been established and applied in several animals using I-SceI-mediated meganuclease (Thermes et al.,2002; Ogino et al.,2006a, b; Pan et al.,2006; Soroldoni et al.,2009; Casco-Robles et al.,2010). These methods can generate entirely transgene-expressing founder animals in F0, enabling us to investigate gene function during regeneration without generating F1 lines. If lens regeneration could be analyzed during the larval stage using transgenic animals generated by this method, it should be possible to gain further insights into the cellular and molecular processes of lens regeneration at all developmental stages. Therefore, to establish that lens regeneration in larvae corresponds to lens regeneration in the adult stages, here we carefully analyzed the process of lens regeneration during the larval stage by using a transgenic technique in which the expression of a fluorescent protein was driven by the γ-crystallin promoter.

Details are in the caption following the image

Adult and larval Japanese fire-bellied newts, C. pyrrhogaster. The adult newt shown is a fully sexually matured female. The larva is stage 57 (around 3-weeks-old). It takes more than 3 years for larvae to mature to adult newts. Scale bar = 2.5 mm.

RESULTS

Histological Analysis of the Lens Regeneration Process in Larvae

To analyze the morphological features of the regenerating lens in the newt C. pyrrhogaster larva, we performed histological analysis using newt larvae that underwent lentectomy at stage 57 (around 3 weeks old, Fig. 1). As the larval eye is smaller than the adult eye (Fig. 1), a finer knife had to be used for cutting the former, so as not to damage the other parts (Supp. Fig. S1, which is available online). The methods for these procedures are described in detail in the Experimental Procedures section. Figure 2 shows bright-field sections of the larval newt eye during the process of lens regeneration. De-pigmented cells were detected in the iris at day 3–6 after lentectomy, following loosening of the intercellular connections among the iris pigmented epithelial cells (Fig. 2A). The de-pigmentation of iris epithelial cells corresponded to Sato's stage II in adult newt lens regeneration (Sato,1940; Eguchi,1988). After the completion of de-pigmentation in the dorsal iris at day 6–8 after lentectomy, a lens vesicle was formed around day 7–9 (Fig. 2B, C), and then began to elongate to form primary lens fiber cells, whose features corresponded to those seen in adult newt regeneration of Sato's stage III to VI, in the posterior side of the lens vesicle (Fig. 2D). Around 20 days later, a structure consisting of secondary lens fibers was formed (Fig. 2E), and the size of the lens was continuously enlarged thereafter by an increase of the cell number (Eguchi,1967). The timings of features such as lens vesicle formation, primary lens fiber formation, and secondary lens formation during particular stages of regeneration in larvae coincided with the timings of the corresponding features seen in the regeneration process in adult newts (Sato,1940; Eguchi,1988). Finally, growth of the lens stopped when a lens the same size as the original intact lens had been regenerated (within 1 month; Fig. 2F).

Details are in the caption following the image

Histological features during larval lens regeneration. A: The depigmentation of iris epithelial cells corresponds to Sato's stage II in adult newt lens regeneration. B: Completion of depigmentation at day 6–8 (Sato's stage III in adults). C: Lens vesicle formation seen at day 7–9 (Sato's stage IV in adults). D: Primary lens fiber cells (plf) in the posterior side of the lens vesicle were formed at day 11–13 (Sato's stage VI in adults). E: At around day 20, the structure of the secondary lens fiber was formed (Sato's stage VII in adults). F: The size of the lens was continuously enlarged by an increase of the cell number, and lens regeneration was completed within 1 month (F). The approximate times to reach each stage after lentectomy are indicated in the respective panels. Upper direction is dorsal, and lower direction is ventral. co, cornea; di, dorsal iris; le, lens epithelium; lv, lens vesicle; nr, neural retina; plf, primary lens fiber; slf, secondary lens fiber; vi, ventral iris. Scale bar = 100 μm.

Cell Division and Cell Differentiation During Lens Regeneration in Larvae

To further investigate the regeneration process of the larval lens at the cellular level compared with the regeneration process in the adult, we conducted histological analysis using molecular markers. In adults, it was reported that cell cycle re-entry occurs at day 4–6 (stage II), and massive cell growth in the regenerating lens is detected at day 6–12 (stage III and IV; Sato,1940; McDevitt,1982; McDevitt and Brahma,1982, 1990; Borst and McDevitt,1987; Mitashov et al.,1992; Mizuno et al.,2002) Therefore, we first investigated the rate of cell division during larval lens regeneration. When we performed immunohistochemistry using cell cycle markers, such as proliferating cell nuclear antigen (PCNA) and phosphohistone H3 (pH3), for S-phase and M-phase respectively, a small number of PCNA-positive cells and pH3-positive cells were detected in the iris at day 3–6, which was consistent with the findings in adults (Eguchi and Shingai,1971). A high level of cell division was detected in the anterior side of the lens vesicle at around at day 7–9 (Fig. 3A, bracket). When nuclear DNA of the lens vesicle cells was stained with Hoechst 33342, the fluorescence intensity of the Hoechst 33342 was higher in the anterior region of the lens vesicle cells than in the posterior region, and thus overlapped with the region where the PCNA-positive and pH3-positive cells were detected (Hoechst in Fig. 3A, bracket), suggesting that this stage in larvae might correspond to Sato's stage IV. Next we analyzed the cell differentiation to determine the timing of the primary and secondary lens fiber formation in larvae. Using antibody against αA-crystallin protein to examine the expression pattern during lens regeneration in larval newts, αA-crystallin protein was first detected at day 11–13 after lentectomy (Fig. 3B). The timing of expression of the αA-crystallin protein and the morphological features at that time corresponded to Sato's stage VI, which occurs at day 12–15 after lentectomy in adult newts (Sato,1940; McDevitt,1982; McDevitt and Brahma, 1982; Borst and McDevitt,1987; McDevitt and Brahma,1990; Mitashov et al.,1992; Mizuno et al.,2002). At day 12–15, strong signals were detected in the secondary lens fiber cells that began to surround the primary lens fiber cells in the posterior region of the regenerating lens, and thereafter the signal of αA-crystallin in the primary lens fiber cells gradually became weaker as the lens grew (Fig. 3B). Finally, we analyzed the denucleation of lens fiber cells at a later stage of lens regeneration in larvae, because it is known that the nuclei of lens fiber cells are shed and the denucleated lens fiber cells are a hallmark of lens maturation during development and regeneration (Vrensen et al.,1991; Wride,1996; Rong et al.,2002; Wang et al.,2010). Nuclei were detected by Hoechst 33342 staining in regenerating lens at 20 days after lentectomy (corresponding to Sato's stage IX), although the regenerating lens was segregated from the dorsal iris (left panel in Fig. 3C). In contrast, at day 24–26 after lentectomy (corresponding to Sato's stage X) no nuclei were found in the lens fiber cells (right panel in Fig. 3C). In addition, no proliferative cells were detected at this stage (data not shown). These results suggest that regeneration of the lens was completed within 26 days after lentectomy in larval newts, which was around 2 months earlier than lens regeneration was completed in adult newts.

Details are in the caption following the image

Cell division and cell differentiation during lens regeneration in larvae. A: Proliferating cell nuclear antigen (PCNA), phosphohistone H3 (pH3), and nuclear staining with Hoechst 33342 at day 7–9. Dividing cells were detected in the anterior side of the lens vesicle (bracket). B: Immunostaining with anti-αA-crystallin antibody during lens regeneration. The αA-crystallin protein was first detected from day 11–13 after lentectomy in the primary lens vesicle. C: Nuclear staining with Hoechst 33342 visualizing the denucleation of lens fiber cells at a later stage of lens regeneration in larvae. Nuclear DNA was still detected at day 21–24 (asterisk). At day 24–26 after lentectomy, no nuclei were found in the lens fiber cells. The approximate times to reach each stage after lentectomy are as indicated in the respective panels. Dorsal is at the top, and ventral is at the bottom. co, cornea; di, dorsal iris; le, lens epithelium; lv, lens vesicle; plf, primary lens fiber; slf, secondary lens fiber. Scale bars = 50 μm in A; 100 μm in B, C.

Size of Regenerating Lens

Next, we calculated the size of the lens during regeneration to evaluate the timing of the completion of the lens regeneration in larvae. Because the newt lens is spherical, we estimated the lens size by measuring the diameter and the cross-sectional area at the midplane of the lens using rostral-to-caudal serial sections. We found that the diameter of the lens in intact animals was 286.4 ± 7.6 μm (n = 5), and the size of the regenerating lens gradually increased from day 10–12 (average diameter ± SEM; 103.5 ± 28.1 μm, n = 5), when the primary lens fiber started to be formed, until day 24–26 (292.4 ± 6.4 μm, n = 6), when the denucleation was complete (Fig. 4A; Supp. Table S1, which is available online). The diameter of the lens at day 24–26 was almost the same as that of the intact lens (292.4 ± 6.4 μm and 286.4 ± 7.6 μm, respectively). Along with the diameter of the regenerating lens, the area at the midplane of the regenerating lens was increased from day 10–12 (9,600 ± 5,600 μm2, n = 5) until day 24–26 (64,000 ± 2,900 μm2, n = 6) after lentectomy (Fig. 4B; Supp. Table S2, which is available online). Furthermore, the number of lens epithelial cells at the midplane of the regenerating lens (Reyer et al.,1994) gradually increased from day 11–13 (28.2 ± 1.6 cells, n = 5) until day 24–26 (48.8 ± 1.4 cells, n = 6; Fig. 4C, Supp. Table S3, which is available online). There was no significant difference in size or epithelial cell number between intact lenses and lenses after 26 days of regeneration, suggesting that the regeneration of the lens in larvae was completed within 26 days after lentectomy.

Details are in the caption following the image

The size of the regenerating larval lens. A: Diameter at the midplane of the regenerating lens (μm). B: Area at the midplane of the regenerating lens (μm2). C: The number of lens epithelial cells at the midplane of the regenerating lens visualized by nuclear staining (cells). *P < 0.05; **P < 0.005; NS: not significant.

In Vivo Analysis of Regeneration Process in Transgenic Newts

To determine whether the regeneration speed in embryonic, larval and adult C. pyrrhogaster is the same or not, we analyzed the regeneration process sequentially in one individual newt by comparing the regeneration process in the same individual at the embryonic (stage 36), larval (stage 57), and postmetamorphosis stages using a transgenic newt expressing a Xenopus laevis γ-crystallin promoter-driven fluorescent marker gene expressed specifically in lens (Smolich et al.,1994; Offield et al.,2000). The method we used for obtaining transgenic animals using I-SceI meganuclease is widely applicable for performing cellular as well as genetic analysis in F0 animals (Casco-Robles et al.,2010). As an initial control, we analyzed the auto-fluorescence in newts and detected strong green fluorescence in intact newts but no red auto-fluorescence (data not shown). Therefore, we decided to use the red fluorescent protein tdTomato as a transgenic marker (Shaner et al.,2004). Using it, we analyzed the regeneration process sequentially in one individual newt by comparing the regeneration process at the embryonic stage, larval stage, and postmetamorphosis stages using an I-SceI meganuclease-mediated transgenic newt with a tdTomato fluorescent marker in the lens. Figure 5A shows the experimental design for assessing the imaging of the regenerating lens after lentectomy and the intervals between successive lentectomies. Figure 5B shows the results for one transgenic individual. After lentectomy, γ-crystallin promoter re-activation was detected from day 11 (Sato's stage VI) in the embryonic stage, larval stage, and postmetamorphosis stages in this individual (Fig. 5B). When we analyzed other transgenic individuals in the same way, the fluorescence was detected at the around same time (day 10–12) during each stage in a given individual, although there were slight differences among individuals (Supp. Table S4, which is available online). These results suggested that the lens regeneration speed was essentially unchanged throughout the life of the newt C. pyrrhogaster.

Details are in the caption following the image

In vivo tracing of the regenerating lens in larvae using transgenic newts. A: Experimental design for performing imaging of regenerating lens after lentectomy. The regeneration process was assessed sequentially in an individual newt by comparing the regeneration process in the same individual repeatedly at the embryonic stage, larval stage, and postmetamorphosis stages using γ-crystallin promoter-driven tdTomato transgenic newts. B: Regenerating lens of embryonic, larval, and postmetamorphosis stages in one individual. The γ-crystallin promoter-driven fluorescence was not detected 10 days after lentectomy. The onset of the expression of the fluorescence occurred at 11 days after lentectomy at the embryonic, larva, and postmetamorphosis stages. Arrows indicate tdTomato γ-crystallin promoter-driven fluorescence of the newly formed lens. Scale bar = 200 μm.

DISCUSSION

Close Similarity Between the Early Process of Lens Regeneration in Larval and Adult C. pyrrhogaster

Although lens regeneration in adult newts has been studied by many scientists since its discovery more than 100 years ago (Collucci,1891; Wolff,1895), there have been few studies on the process of lens regeneration in larval newts (Reyer,1982a, b). To clarify the cellular processes of larval lens regeneration, we carefully investigated the process of lens regeneration histologically and molecular biologically in larvae to compare this process with that of lens regeneration in adult newts described previously (Sato,1940; McDevitt,1982; McDevitt and Brahma,1982; Borst and McDevitt,1987; McDevitt and Brahma,1990; Mitashov et al.,1992; Mizuno et al.,2002; Del Rio-Tsonis and Tsonis,2003; Henry and Tsonis,2010). Figure 6 shows a summary of the process of lens regeneration in C. pyrrhogaster larvae. Our results revealed that the lens of larvae regenerates in the early phase (from stage I to stage X) through similar stages with similar timing to those of adults, namely delamination, cell cycle re-entry, depigmentation, cell division, and primary and secondary lens fiber formation in the dorsal iris. The onset of expression of αA-crystallin protein occurred after the onset of expression of αA-crystallin mRNA, in accord with the reported gap between αA-crystallin mRNA expression onset and protein expression onset in adult C. pyrrhogaster (Mizuno et al.,2002). Furthermore, we found that the timing of γ-crystallin promoter re-activation in the regenerating lens, which was a critical landmark during lens regeneration for judging the regeneration speed, was similar among embryo, larva, and juvenile (Figs. 5, 6). In contrast, the timing of the completion of the lens during its regeneration in larvae was earlier than that in the adult. The reason for the difference of the lens maturation timing between larvae and adults appeared to be due to the difference of the size of the lens: the diameter of the larval lens was around 300 μm (Fig. 4), while that of the adult lens was over 1,000 μm (Eguchi et al.,2011), and therefore the volume of the adult lens was more than 30 times larger than that of the larval lens. Thus, the time required after completion of the denucleation of lens fiber cells (Sato's Stage X) to complete of the growth of the small lens by increasing the number and size of lens cells might be shorter than that required for the large lens (Fig. 4; Hornsby and Zalik,1977). Importantly, by using nuclear staining with Hoechst 33342 to detect denucleation of interior lens fibers as a maturation marker, we could monitor the transition of the cell nuclei of lens fibers in the larval lens regenerants (Fig. 3C; Vrensen et al.,1991; Wride,1996; Rong et al.,2002; Wang et al.,2010). In intact and regenerated lens after stage X, we could not find any proliferative cells among lens fiber or lens epithelial cells, suggesting that the newt can control the proper lens size by controlling the number of cells produced during regeneration.

Details are in the caption following the image

Comparison of the staging of lens regeneration between adults and larvae. From Sato's stage I to stage X, the regeneration process was similar between adults and larvae, and thereafter the completion of regeneration in larvae occurred earlier than that in adults. Adult staging refers to Sato's staging (Sato,1940; Eguchi,1988)

We confirmed that newt larvae could regenerate a lens faster (approximately 30 days) than adults (approximately 80 days). However, our detailed observations of lens regeneration at the cellular level using molecular markers revealed that lens regeneration in larvae proceeds at a similar speed and through similar steps to those of adults. We could not find any drastic difference between larvae and adults, except that the size of the eyes was very different. Thus, larvae can regenerate a lens of the original size in a shorter time than adults, because the lens of larvae is smaller than that of adults and therefore requires less time for its maturation, although the speed of the steps is not faster in larvae. In our study of transgenic newts, we used one marker gene to investigate the timeline during lens regeneration through a newt's life. Although it would be better to analyze the lens regeneration process at several time points using several markers specific for different stages during the lens regeneration, at present it is impossible to produce transgenic newts expressing multiple transgenes. In addition, we should carefully re-investigate whether similar phenomena are observed in other regeneration systems.

Combined Use of Transgenesis and Larval Regenerants May Yield Further Insights

Next, we traced and compared the regeneration process of the lens noninvasively to determine the regeneration speed at different life stages by using the transgenic technique. Transgenic techniques in newts have been established by several groups (Makita et al.,1995; Ueda et al.,2005; Casco-Robles et al.,2010). Importantly, the advantage of the I-SceI meganuclease-mediated method over the methods used in those studies is its high efficiency of transgene expression (Thermes et al.,2002; Ogino et al.,2006a, b; Pan et al.,2006; Casco-Robles et al.,2010,2011) in nonmosaic founder animals, which means that F0 founders can be used to monitor and manipulate gene regulation during regeneration as well as development. In this study, we demonstrated that regeneration speed is unchanged throughout the life of C. pyrrhogaster by repeated analysis of the regeneration process in one individual transgenic newt. Our results also suggest that similar cellular and molecular mechanisms occurred in the repair process of lens regeneration in larvae and adults. Therefore, clarifying the regeneration process in larvae might shed light on principles of regeneration that also hold true in adults. Finally, one of the most difficult issues impeding a genetic approach for the investigation of regeneration mechanisms in newts has been thought to be the long time needed for sexual maturation of the F1 founders. As we verified here that larval lens regeneration occurs by means of cellular processes similar to those of the adult lens regeneration, the combination of the transgenic technique and regeneration analysis using larvae presented here might be amenable to modifications enabling the functional characterization of the genes involved in regeneration in newts in various organs such as limbs, jaw, heart, brain, and so on, without the need to wait for a couple of years for such maturation of F1 founders. Although we could not exclude the possibility of integration-site sequence-dependence of the efficiency of expression of exogenous genes in the newt, the in vivo imaging presented here using one individual repeatedly made it possible to compare the promoter activity during different stages of an individual's life independent of any variations of the transgene integration among individuals.

Even though we used the Xenopus laevis γ-crystallin promoter sequence in this study, the expression of the γ-crystallin promoter-driven tdTomato fluorescence was specifically detected in the lens (Fig. 5B). In addition, we confirmed the same specific expression pattern in several transgenic newts driven by the same Xenopus tropicalis promoter sequence (data not shown). These results suggest that the genomic sequences of Xenopus, whose nucleic acid sequence has been well analyzed (Brown,2004; Hellsten et al.,2010), might be useful for monitoring the gene expression in C. pyrrhogaster, whose huge genomic sequence remains unknown (Baldari and Amaldi,1976; Bozzoni and Beccari,1978; Garner,2002).

There are two aspects of regeneration that are widely generally accepted. The first is that simpler animals may have stronger regenerative ability than more complex animals. However, when we survey regenerative ability across various animal phyla, we cannot find such clear boundaries between organisms ranging from simple multicellular organisms to complex mammals (Agata and Inoue,2012). Instead, there are both regenerative and nonregenerative animals in each phylum. The second is that fetal and juvenile animals can regenerate faster and more robustly than adults. In this study, we showed that the speed of regeneration was not faster in larvae even though larvae could regenerate a lens of the original size within a shorter time than adults. Rather, because the adult lens is larger than the larval lens, it may take more time to acquire the number of lens fiber cells after stage X by cell proliferation in adults. Here we have clearly demonstrated by the transgenic technique that larval and adult cells showed similar regeneration speed during the process of lens regeneration in the newt C. pyrrhogaster. In the case of frogs, it is known that Xenopus loses regenerative ability after metamorphosis (Dent,1962; Strub,1979). This raises the question of whether the regenerative ability is gradually lost during development or whether metamorphosis may cause the loss of this ability. Re-investigation of regenerative ability and speed in a precise manner may provide much useful information for future applications of regenerative medicine.

EXPERIMENTAL PROCEDURES

Animals

Adult Japanese fire-bellied newts, C. pyrrhogaster, were collected from rice field or ponds in Kyoto and Hyogo prefectures, Japan. Newts were kept in plastic containers in tap water at room temperature and fed twice a week. All newts were maintained and manipulated according to a protocol approved by the Animal Care and Use Committee of the Kyoto University.

Preparation of Fertilized Eggs and Larval Newts

One week before spawning, 100 U of gonadotropin (Asuka Pharmaceutical) in 0.6% NaCl solution was injected intra-peritoneally in females every 2 days. After two injections, the females were put in water tanks with several pieces of soft plastic sheet as an egg-laying scaffold and allowed to spawn eggs. Some embryos at stage 36 were used for regeneration experiments. After the larval newts hatched naturally from the eggs, they were kept individually in small plastic containers in tap water at room temperature and fed twice a week, and then stage 57 (3-week-old) larval or juvenile newts were used for all experiments.

Lentectomy of Embryonic and Larval Newts

The embryonic and larval eyes were so small that it was necessary to use a finely sharpened knife for lentectomy in order not to injure the iris. To produce such a knife, a razor's edge was cut, and then the back of the cut razor's edge was sharpened using a grinder as shown in Supp. Fig. S1A. A larva anesthetized with 0.1 mg/ml of ethyl 3-aminobenzoate methanesulfonic acid was fixed with appropriately bent staples on a 2% agarose gel, and then the lens was removed after cutting the cornea (Supp. Fig. S1B). When we evaluated the completion of the lens regeneration of the transgenic newts, lens size was measured until it stopped changing. To compare the regenerating lens between larvae and adults, the criteria of the Sato's stages were used (Sato,1940).

Histological Sectioning and Immunohistochemistry

Specimens were fixed overnight at 4°C in 4% paraformaldehyde, 10% methanol in 70% phosphate buffered saline (PBS), and equilibrated in 20% sucrose and embedded in Tissue-Tek (O.C.T. compound, Sakura) for immunohistochemistry. To prepare histological sections, samples were dehydrated with a graded series of ethanol, embedded in paraffin wax, and sectioned at 10 μm. For immunohistochemistry, frozen sections (10 μm thick) were mounted on adhesive slides. The sections were rinsed with 70% PBS containing 0.1% Triton X-100 and blocked in 70% PBS containing 0.1% Triton X-100 and blocking reagent (Roche) for 1 hr at room temperature. After that, the sections were incubated with appropriate primary antibody: anti-αA-crystallin (Developmental Studies Hybridoma Bank), anti-PCNA (Sigma), or anti-pH3 (Upstate Biotechnology), overnight at 4°C, washed several times in 70% PBS containing 0.1% Triton X-100, and incubated with the appropriate secondary antibody (Alexa488-labeled anti-mouse IgG(H+L) antibody, or Alexa594-labeled anti-rabbit IgG(H+L) antibody; Invitrogen) and 1 μg/ml of Hoechst 33342 (Invitrogen) for 1 hr at room temperature. Fluorescence was detected with an FV10 confocal microscope (Olympus).

Image Analysis for Calculation of the Regenerating Lens Size

To calculate the lens size during lens regeneration, at least five images of the coronal sections from a head with a regenerating lens were prepared. The regenerating lenses of the same Sato's stage as determined by morphological features were pooled as groups. The image of the section of the center of the spherical lens was acquired with Metamorph software (Molecular Devices) and the diameter and area were analyzed with ImageJ software (National Institutes of Health).

Vector Construction

The I-SceI-pBSII vector plasmid and CMV-EGFP/I-SceI-pBSII vector plasmid were provided by Dr. H. Ogino of Nara Institute of Science and Technology (Smolich et al.,1994; Thermes et al.,2002; Ogino et al.,2006a, b). The 2.3-kb fragment of the Xenopus laevis γ-crystallin promoter sequence was provided by Dr. H. Yokoyama of Tohoku University. The tdTomato sequence was obtained from Dr. R. Y. Tsien (Shaner et al.,2004). The γ-crystallin-tdTomato/I-SceI-pBSII vector was generated by introducing the tdTomato cassette into the γ-crystallin/I-SceI-pBSII after it was digested with BamHI and EcoRI restriction enzymes. A schematic drawing of the vector construct is shown in Supp. Fig. 2.

Generation of Transgenic Newts and Imaging

The production of γ-crystallin promoter-driven tdTomato-expressing transgenic newts and CMV promoter-driven EGFP-expressing transgenic newts as a positive control was carried out as previously described (Ogino et al.,2006a, b; Casco-Robles et al.,2010,2011). Images of newts were acquired using a fully automated fluorescence stereoscopic microscope Leica M205 FA with a super-high pressure mercury lamp (EL6000), and a Plan Apo objective (WD 61.5 mm) controlled using Leica LAS-AF6000 3D visualization software. Fluorescent and bright-field images were acquired using a Leica DFC 360 FX monochrome camera. A TXR filter set (560/40) was used for tdTomato fluorescence imaging. A motorized stage was used to collect a tile scan of higher magnification images.

Statistical Evaluation

Quantitative data were analyzed by one-way analysis of variance (ANOVA). For the calculation of lens size, data from at least five images of five regenerating eyes of at least three individuals were averaged.

Acknowledgements

We thank Dr. Elizabeth Nakajima for careful reading of the manuscript. We thank Dr. Hajime Ogino (Nara Institute of Science and Technology) and Dr. Hitoshi Yokoyama (Tohoku University) for their generous gift of γ-crystallin promoter sequence-containing vector and CMV-EGFP/I-SceI-pBSII plasmid vector. We also thank Dr. Roger Y. Tsien (University of California at San Diego) for a generous gift of tdTomato-containing vector. This work was supported by a research grant from the Brain Science Foundation to T.I., Takeda Science Foundation to T.I., a Grant-in-Aid for Global COE Program A06, a Grant-in-Aid for Creative Scientific Research, and a Grant-in-Aid for Scientific Research on Innovative Areas from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.