SEARCH

SEARCH BY CITATION

Keywords:

  • digit;
  • gene expression;
  • chick;
  • wing

Abstract

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

We report the region-specific expression of a novel gene, named mario, whose expression domain is in the distal tip of the presumptive and developing digit 2 region in the developing chick wing bud. The anterior region-specific expression of mario corresponds well with the presence of digit 2, and fate map analysis showed that mario expression at early stages represents the presumptive digit 2 region. Using mario expression as a region-specific marker for the digit 2 region, several surgical operations were performed to obtain insights into digit 2 development in the chick wing. Cell fate tracing concomitant with a zone of polarizing activity (ZPA) implantation revealed that an additional digit 2 in the ZPA implantation into the anterior or middle region of wing bud is derived from the original digit 2 region (mario-positive region). Surgical manipulations revealed that the anterior nondigit region has an inhibitory effect on digit 2 formation. Taken together, these results suggest that the most-anterior region, including the anterior necrotic zone, restricts the position of digit 2 region by limiting the anterior border of the digit 2 region and preventing its expansion. Developmental Dynamics 233:326–336, 2005. © 2005 Wiley-Liss, Inc.


INTRODUCTION

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

During limb morphogenesis in tetrapods, the anterior–posterior (AP) axis is established as a series of digits with an appropriate number and distinct morphology. Results of apical ectodermal ridge (AER) removal experiments and some fate map studies (Saunders, 1948; Stark and Seals, 1973; Summerbell, 1974; Lewis, 1975; Summerbell, 1976; Bowen et al., 1989) suggest that digit specification (development of the autopod region) in the developing chick wing bud starts at around stages 23–24. Therefore, establishment of the developmental field for digit formation would have started by these stages. Indeed, an autopod-specific homeobox gene, hoxa13, begins to be expressed at stages 22–23 (Nelson et al., 1996; Vargesson et al., 2001). In the field for digit formation, a chick wing bud produces three digits (digit 2, digit 3, and digit 4), and digit 2 develops at the anterior-most position that is the most distant from the zone of polarizing activity (ZPA), which is a group of mesenchymal cells located at the posterior border of the limb bud and is responsible for digit patterning along the AP axis (Saunders and Gasseling, 1968; Summerbell, 1979). The ZPA is thought to act through a spatial concentration gradient of morphogen that organizes the morphology of each digit (digit identity) with the concentration gradient (Wolpert, 1969; Tickle, 1981; Harfe et al., 2004), and digit 2 is produced at the lowest concentration of morphogen. Considering the importance of the concentration gradient, digit identification as well as specification of the number of digits seem to need an appropriate space of the field.

Shh (Sonic hedgehog) is thought to mediate the polarizing activity (Riddle et al., 1993; Chang et al., 1994; Lopez-Martinez et al., 1995), and high and low concentrations of shh induce additional digit 4 and digit 2, respectively (Yang et al., 1997). Taken together with the fact that its transcripts are exclusively expressed in the ZPA in the limb bud (Riddle et al., 1993) but the protein has a gradient of distribution along the AP axis (Lewis et al., 2001; Gritli-Linde et al., 2001; Zeng et al., 2001), shh seems to be a putative morphogen. A recent study in which genetic experiments were carried out has suggested that not only the spatial gradient of shh but also the length of time of exposure to shh is critical for the specification of digits (Harfe et al., 2004). In that study, digits were classified into several categories: shh-independent digits, digits made depending on the concentration gradient of shh, and digits made depending on the temporal gradient of exposure to shh. Therefore, it is likely that shh signaling determines digit identity with a complex of downstream mechanisms of the signaling. Other studies showing that shh and Gli3 double mutant mice exhibit polydactyly with a greater number of nonidentified digits (Litingtung et al., 2002; te Welscher et al., 2002) suggest that the number of digits as well as digit identity is mediated by shh signaling.

As briefly described above, the molecular basis of the positioning and identification of digits has become clear. However, since many aspects of digit identity have been examined by observing cartilage patterns after various kinds of manipulations and gene introductions, the early mechanisms to specify the field for digit formation and to establish the digit identity have remained uncertain. One of the reasons for this finding is that few early markers of the developing digit region have been identified. Hoxd cluster genes (hoxd11, 12, and 13) are a group of good markers for the digit region (Yokouchi et al., 1991; Nelson et al., 1996), but the expression domains are not specific for a digit but cross some digit primordia, and, moreover, they are expressed also in the zeugopod region. Here, we report a good candidate for the digit region-specific marker gene, mario, which is expressed specifically in the distal-peripheral region of presumptive digit 2 in the developing chick wing bud. The spatiotemporal expression domain of mario closely correlates with digit 2 formation in the wing; mario starts to be expressed in the presumptive digit 2 region at stages 22–23, when the autopod region would start to develop as mentioned above, and the expression traces the distal-peripheral thin layer of the digit 2 region throughout the period of digit development.

By using both mario expression and cartilage pattern, we classify the duplicated digits (432 from anterior to posterior) after anterior implantation of the ZPA into two classes. Digit 3 and digit 4 originate from the nondigit region at the most-anterior part of the limb bud that are converted to have posterior identity by the implanted ZPA, and the additional digit 2 does not originate from the nondigit region but is produced by an expansion of the original digit 2 region. Furthermore, transplantation experiments and analyses of mario expression showed that the anterior nondigit region has an inhibitory effect on digit 2 formation, suggesting an important contribution of the nondigit region to the formation of the field for digit formation and to the determination of the number of digits. Our results support the idea that a certain mechanism from the anterior region as well as the posterior ZPA is pivotal for digit formation.

RESULTS

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

Novel Gene mario Produces a Noncoding RNA

A cDNA sequence for a putative noncoding RNA was named mario after the name of a cartoon character, because its unique expression in the developing limb bud (see Fig. 2) is located at the opposite side of the sonic hedgehog expression. Figure 1 demonstrates genomic structure organization for the site of mario and the surrounding genes in chicken genome and comparison of corresponding genomic sequences with some other vertebrates. These sequences obtained from databases (Ensemble and UCSC) revealed that mario sequence is located on the second chromosome and can be separated into four exons, and a presumptive transcript produced from these four exons has the polyadenylation signal at the end of the fourth exon but no significant open reading frame. We recognized two genes (BX933114 and papd1) upstream of the mario region and one gene (svil) downstream, respectively. We cloned partial cDNA sequences for these two upstream genes, BX933114 and papd, and performed in situ hybridization, but expression patterns of these genes were totally different from that of mario (data not shown), suggesting that these two upstream genes flanking mario region are in other loci different from the mario locus. Moreover, in the sequence between mario and BX933114, there were no expressed sequence tag clones that have the same direction of transcription with mario. Taken together, we estimate that mario transcribes a putative noncoding RNA.

thumbnail image

Figure 2. Expression profile of mario transcripts in the developing chick embryo. A,B: Anterior-specific expression of mario in chick wing bud at stages 23 (A) and 25 (B). CF: Expression of mario (after digit 2 rudiment has been demarcated) at stages 27 (C,D), 29 (E), and 35 (F) in the wing bud. D: A section showing mario expression at stage 27. The magnified image in D reveals that mario is only expressed in a thin layer of limb mesenchyme in the digit 2 region and that the overlying ectoderm is negative. The magnified image in F shows weak mario expression in the distal tip of the wing at stage 35. G,H: Double staining with mario/hoxd12 (G) and mario/hoxd13 (H) in stage 25 wing bud. Note that there is no overlap of expression between mario and Hoxd12/13. I,J: Expression of mario in chick leg bud at stages 27 (I) and 29 (J). All arrowheads indicate the anterior and posterior edges of the mario expressing region, and 2, 3, 4, and I, II, III, IV indicate digit number, respectively.

Download figure to PowerPoint

thumbnail image

Figure 1. Schematic representation of genomic structure organization for mario and comparison of syntenic region with other vertebrates. cDNA sequence of mario (yellow bar) can be separated into four exons that are located on chromosome 2 in the chicken genome. There are at least two loci recognizable at 5′ upstream of this mario region (papd1 in pink and an unknown gene in green, a partial cDNA sequence of which has been reported as an expressed sequence tag clone, BX933114) and one locus at 3′ downstream (svil in blue). All of them have the same direction of transcription indicated by arrows. According to comparison with genomic sequences of other vertebrates (human, mouse, and Xenopus tropicalis) obtained from databases, these regions including BX933114, papd1, and svil loci can be seen with the same order in other vertebrate genomes (each chromosome number is shown at the right side). Note that there is no sequence homologous with mario.

Download figure to PowerPoint

Genomic structure organization covering mario and its flanking genes shows us syntenic regions in other vertebrates, i.e., human, mouse, and Xenopus tropicalis. All of these genomic sequences show synteny conservation in three loci (BX933114, papd and svil), except the chicken mario locus. No sequence homologous to mario has been found even in whole genomes of other animals reported in databases.

Expression of mario in the Anterior Limb Bud Correlates With Digit 2 Formation

The expression profile of mario transcripts was examined by whole-mount in situ hybridization. Before and at stage 21, a positive signal for mario was not detected in any region in the chick embryo (not shown). At stages 22–23, mario started to be expressed at the thin layer of the anterior margin beneath the AER in both wing and leg buds (Fig. 2A; see also Fig. 5A). A chick wing has three digits, and these digits are generally defined as digit 2, digit 3, and digit 4 (see Fig. 3B). As development progresses, the anterior-specific expression of mario became stronger in the wing buds (Fig. 2B,C). The mario expression domain was in the distal-peripheral mesenchymal layer, and the ectoderm, including the AER, was negative (Fig. 2D). In late-stage wing buds, the location of mario expression along the AP axis corresponded to the developing digit 2 region, including the interdigital space between digit 2 and digit 3 (Fig. 2E). At a closing stage of digit morphogenesis (stage 35), mario expression was faint and limited to the distal tip of digit 2 (Fig. 2F) and then disappeared (not shown). To compare the anterior-specific mario expression with other region-specific marker genes, we analyzed the expression of mario, Hoxd12, and Hoxd13, using double whole-mount in situ hybridization. Hoxd12 and Hoxd13 expressions at stage 25 were detected at the posterior bud as reported previously (Yokouchi et al., 1991; Nelson et al., 1996), and both expression domains were segregated from that of mario (Fig. 2G,H). A chicken hindlimb has four digits (called digit I, II, III, and IV). Mario expression in the leg bud could be observed in a peripheral thin mesenchyme in the digit I region, including the interdigital space between digit I and digit II (Fig. 2I,J).

thumbnail image

Figure 5. Fate maps of wing bud mesenchyme at the level of somite 17. Two cell populations on a level with somite 17 were labeled at stage 22 to assess the relationship between the mario-expressing region and digit 2 primordium. A: Mario expression in a stage 22 wing bud. Red dotted lines indicate the extension line from somite 17 on the wing bud region. Two arrowheads indicate the anterior and posterior edges of the mario-expressing domain, respectively. Mario starts to be expressed on a level with somite 17. B: Diagram of DiO injection in a stage 22 wing bud. Green capillaries indicate injection sites, which are located on the extension line from the anterior or posterior edge of somite 17. C: Distribution of labeled cells at 48 hr after DiO injection. The oval-shaped dotted line is an outline of digit 2 rudiment estimated in a brightfield. The digit 2 primordium was located between the two cell populations labeled. D: mario expression in the same bud as that shown in C. E: Diagram illustrating DiO labeling on a level with somite 17 simultaneous with zone of polarizing activity (ZPA) implantation. A ZPA graft, shown by a red square, was implanted into the anterior edge of the wing bud at the somite 15–16 boundary. F: Distribution of labeled cells at 48 hr after DiO labeling on a level with the somite 16–17 boundary, simultaneous with ZPA implantation. G: mario expression in the same bud as that shown in F. Note that the anterior edge of the expanded mario-expressing region agrees with DiO-labeled cells in F and that an additional digit 2-expressing mario is located on posterior side of the labeled cells. H: Diagram illustrating DiO labeling with ZPA implantation into the apex of wing bud at somite 18. I: DiO-labeled cells in H are visualized after 48 hr. J: mario expression in the same bud as that shown in I, indicating that the posterior edge of mario expression corresponds with the distal end of DiO-labeled cells in I.

Download figure to PowerPoint

thumbnail image

Figure 3. Relationship between mario expression and digit 2/digit 2 formation. A,B: Anterior-specific expression pattern of mario at stage 25 (A) and skeletal pattern at stage 35 (B) in normal wing buds. C,D: Mario expression at stage 25 is expanded by an anteriorly grafted zone of polarizing activity (C), and a typical skeletal pattern in the resulting wing bud includes an additional digit 2 (D). E–H: Modification of mario expression and skeletal pattern by application of a bead soaked in retinoic acid. E,F: A low concentration (10 μg/ml) of retinoic acid expands the mario-expressing region (E, compare with Fig. 4A) and induces an additional digit 2 (F). G,H: A higher concentration (1.0 mg/ml) results in diminished expression of mario (G) and duplicated pattern 4334 (H). Arrowheads indicate the anterior and posterior edges of the mario-expressing region.

Download figure to PowerPoint

thumbnail image

Figure 4. Apical ectodermal ridge (AER) and fibroblast growth factor (FGF) signaling effects on mario expression. A: At 24 hr after removal of the anterior one-third portion of the AER, mario expression is undetectable in the operated wing bud. B: AER removal results in complete deletion of digit 2. C: SU5402 (4 mg/ml) application down-regulates mario expression when grafted into the anterior side of a wing bud. D: Anterior elongation of mario expression after the implantation of an FGF2 bead. Compare the mario-expressing domain with that shown in Figure 1C. The asterisk indicates the position of the bead, and the arrowhead shows the posterior border of the mario-expressing region.

Download figure to PowerPoint

Grafting of the ZPA into the anterior margin resulted in a duplicated pattern, including two digit 2 elements (Fig. 3D), as described previously. At 24 hr after the ZPA implantation, mario expression was expanded anteriorly (Fig. 3C) compared with normal expression (Fig. 3A), suggesting that this anterior-specific expression is related to the region for digit 2. To further investigate the relationship between mario expression and digit 2, we applied retinoic acid (RA) -soaked beads into the anterior margin of the wing bud, because RA is known to induce an additional digit pattern in a concentration-dependent manner. A low concentration of RA (10 μg/ml) always induces only digit 2 (Fig. 3F, 12 of 12 cases), and at this dose of RA, mario expression was expanded anteriorly after 24 hr (Fig. 3E, 5 of 5 cases). A higher dose of RA (1 mg/ml) greatly diminished mario expression (Fig. 3G, 4 of 4 cases), always resulting in a duplicated pattern of 4334 or 434, which has no digit 2 (Fig. 3H, 9 of 9 cases).

Mario Expression Depends on the AER

To analyze the dependence of mario expression on AER signaling, we removed the anterior AER from the wing bud at stage 21, when expression of mario has not yet started, and analyzed mario expression at 24 hr after the removal. Mario expression was not detected (Fig. 4A), and AER removal led to a reduction of digit 2 and the distal part of the radius in the resulting limb (Fig. 4B). SU5402, an antagonist of fibroblast growth factor (FGF) signaling, caused reduction of mario expression (Fig. 4C) as did AER removal, and, in contrast, FGF2 application in the presence of the AER elongated mario expression anteriorly (Fig. 4D), suggesting that the AER-dependency of mario expression involves FGF signaling.

Fate of Early mario-Expressing Region

The correlation between mario expression and digit 2 suggests that mario expression can be used as a digit 2-specific marker. Indeed, mario expression at later stages was restricted to the distal tip of digit 2, but we have not confirmed whether the mario-expressing region at early stages (Fig. 5A, see also Fig. 2A) is the presumptive digit 2 region, because we cannot see any digit primordia at these early stages. To examine how the anterior–peripheral region of the chick wing bud, which expresses mario, contributes to digit 2 formation, we followed the fate of the mario-expressing region. Some previous studies using excision experiments and fate map analyses (Saunders, 1948; Stark and Seals, 1973; Summerbell, 1979; Wilson and Hinchliffe, 1987; Bowen et al., 1989; Vargesson et al., 1997) suggested that the digit 2 region probably originates in the space opposite somite 17 (the chick wing bud is located in the space opposite somites 15 to 20) in early-stage embryos (see Fig. 5A). Lipophilic dye (DiO) was injected into two distal-marginal points on a level with the boundary of somites 16–17 and somites 17–18 (Fig. 5B) at stage 21–22 just before mario starts to be expressed, and DiO-labeled cells were traced. After 48 hr, the distal top of digit 2 primordium was located between the two populations of labeled cells (Fig. 5C). Furthermore, these points approximately corresponded to the anterior and posterior ends of the mario-expressing region (Fig. 5D). From these experiments, we assume that the presumptive digit 2 region is located at somite 17 level, where mario expression starts, and we decided to use mario expression as a marker for the digit 2 region. In experiments described below, the region opposite somites 15–16 is called the anterior nondigit region, the region opposite somite 17 is called digit 2 region, and the region at the more-posterior part (at the level of somites 18–19) is called the digit 3–4 region.

Expansion of the Digit 2 Region After Implantation of the ZPA

As shown in Figure 3C, mario expression is extended anteriorly after ZPA implantation. There are possible explanations for this extension. Cells inside the original mario-expressing region may proliferate additionally, resulting in anterior elongation of the mario-expression region. Alternatively, cells in the more-anterior region than the original mario-expressing region may be converted to express mario, resulting in anterior addition of the mario-expressing region. To elucidate this point, the anterior edge of the putative digit 2 region (the same point as in Fig. 5B) was labeled with DiO after ZPA implantation into the anterior base of the nondigit region (opposite somite 15; Fig. 5E). In the wing bud in which the ZPA was grafted, the labeled cells were located on the anterior side of the additional digit 2 primordium (Fig. 5F) that corresponds to the anterior boundary of the mario-expressing region (Fig. 5G), suggesting that the original mario-expressing region is expanded after ZPA implantation and that the additional digit 2 arises from the elongation of the original digit 2 region. In other words, our findings suggest that the ectopic digits after ZPA implantation into the anterior base (somite 15 level) can be classified into two types: one is additional digits for which the anterior nondigit region (somites 15–16 level) are converted to the posterior digit lineage (digit 3 and digit 4), and the other is an expanded digit (digit 2) that originates in the digit 2 region (somite 17 level).

On the other hand, when the posterior edge of the putative digit 2 region (the same point as in Fig. 5B) was labeled with DiO after ZPA implantation into the apex of the wing bud (opposite somite 18; Fig. 5H), the labeled cells were located on the posterior boundary of the mario-expressing region (Fig. 5I,J), indicating that the original mario-expressing region is shifted anteriorly after this type of ZPA implantation. These results suggest that implanted ZPA into the middle induced posterior digits between the digit 2 region and ZPA itself and that the digit 2 in the duplication arises from the original digit 2 region. Taken together, it is possible that the mario-expressing region, the so-called digit 2 region, is not induced ectopically by the implanted ZPA and that digit 2 in the ZPA-implanted limb is derived from the original digit 2 region even if the limb bud gets the new positional information from the additional ZPA implanted anteriorly or posteriorly.

Anterior Nondigit Region Cannot Generate Digit 2

Data shown in Figure 5E–G suggest that the anterior nondigit region generates digit 3 and digit 4 but does not generate digit 2. To obtain further data supporting this idea, we performed three types of surgical experiments.

First, we tested whether the nondigit region can compensate removed digit 2 region. When the anterior half of a wing bud was surgically removed (Fig. 6A), mario expression disappeared (Fig. 6B) and the resulting wing showed a complete defect of digit 2 (Fig. 6C; 8 of 8 cases). To determine whether the anterior nondigit region compensates digit 2 in this type of ablation, the anterior nondigit region was joined to the anterior face of the posterior bud after removal of the anterior half of a wing bud (Fig. 6D). This surgical operation failed to reproduce digit 2 (Fig. 6F; 2 of 14 cases), and mario expression was not recovered (Fig. 6E). The grafted nondigit region is thought to receive the ZPA signal, because the distance from the ZPA to the grafted nondigit region seems to be normal. Indeed, when the digit 3–4 region was joined (Fig. 6G), mario expression was rescued (Fig. 6H) and digit 2 was compensated (Fig. 6I; 15 of 18 cases). Nevertheless, the anterior nondigit region could not compensate digit 2, suggesting that this region is not able to reproduce digit 2.

thumbnail image

Figure 6. Manipulations to examine whether the anterior nondigit region can compensate the ablated digit 2 region. A,D,G: Diagrams in the first column illustrate three types of manipulations, respectively. B,E,H: Pictures in the second column show mario expression at 24 hr after each operation. C,F,I: Examples of resulting skeletal patterns in the operated wings are shown in the third column. A–C: The excised anterior half of the wing bud opposite somites 15–17 (A) did not reproduce the expression of mario (B) and exhibited no digit 2 (C). D–F: When the anterior region opposite somites 15–16 was joined to the anterior face of the posterior bud (D), mario expression was not detected after 24 hr (E) and digit 2 was not reproduced (F). G–I: When the posterior digit region opposite somite 18 was joined with the posterior bud (G), mario expression was detected after 24 hr (H) and the resulting bud newly differentiated digit 2 (I).

Download figure to PowerPoint

For the second experiment to investigate the ability of the anterior nondigit region to form digit 2, we implanted the ZPA directly into each region. A wing bud was dissected and all of the parts were ablated, except for the anterior nondigit region (Fig. 7A), digit 2 region (Fig. 7D), or digit 3–4 region (Fig. 7G), and then the ZPA was grafted on the distal-posterior face of each region. The digit 2 region with the ZPA always formed digit 2 with other digits (Fig. 7F; 21 of 30 cases). The digit 3–4 region with the ZPA formed two digits in a sample and sometimes reproduced digit 2 (Fig. 7I; 5 of 15 cases). In contrast, the anterior nondigit region with the ZPA never formed digit 2 (Fig. 7C; 0 of 44 cases), although this combination frequently produced two digits (29 of 44 cases). We also examined mario expression at 24 hr after these experiments. No expression of mario was detectable (Fig. 7B) after the experiment shown in Figure 7A, but the experiments in Figure 6B,C induced mario expression at the anterior margin (Fig. 7E,H), consistent with the data of cartilage pattern.

thumbnail image

Figure 7. Manipulations to examine whether the zone of polarizing activity (ZPA) can reproduce digit 2 in the anterior nondigit 2 region. A,D,G: Diagrams in the first column demonstrate combinations of each region opposite somites 15–16 (A), 17 (D), 18 (G), and the ZPA. B,E,H: The second column shows mario expression at 24 hr after each operation. C,F,I: The third column shows examples for the resulting skeletal pattern in the operated wings.

Download figure to PowerPoint

To determine whether the anterior nondigit region also prevents digit 2 formation itself, the digit 2 region was directly grafted into the anterior nondigit region. An additional cartilage element that sometimes fused into host radius lost digit 2 identity (Fig. 8A; 5 of 6 cases). This loss of digit identity is probably due to the inhibitory effect of the anterior nondigit region, because a digit 2-like structure with joints was formed when the digit 2 region was grafted into the flank region (Fig. 8D; 5 of 6 cases). Sometimes cartilage pattern is not confirmative, but the above idea was supported by gene expression analysis. In the same experiment as Figure 8A, both mario (Fig. 8B) and hoxa13 (Fig. 8C) expressions were negative in the grafted tissue. On the other hand, in the implantation into the flank region, both mario and hoxa13 were expressed in the grafted tissue (Fig. 8E,F).

thumbnail image

Figure 8. Inhibitory effect of the anterior nondigit region on digit 2 formation. A–F: Presumptive digit 2 region was grafted into the nondigit region (A–C) or the flank region (D–F). A,D: Resulting skeletal patterns. B,C,E,F: mario expression (B,E) and hoxa13 expression (C,F) were analyzed at 24 hr after implantation. The arrowhead indicates the grafted tissue. G–L: Relationship between cell death in the ANZ and mario expression; application of a phosphate buffered saline-soaked (G,H), a higher dose (0.1 mg/ml) of bone morphogenetic protein-4 (BMP4) -soaked (I,J), or a low dose (0.01 mg/ml) of BMP4-soaked (K,L) bead into the anterior bud. G,H: In the control experiment, the ANZ visualized by Nile blue staining (G) is located next to the mario-expressing region (H). I,J: A remarkable increase in cell death was detected by 0.1 mg/ml of BMP4 application (I), and mario was significantly down-regulated with BMP4 (J). K,L: The 0.01 mg/ml of BMP4 application barely increased cell death (K), but mario was significantly down-regulated (L). M: A model for regulation of mario expression and digit 2 formation in the early wing bud. The apical ectodermal ridge (AER) is necessary for mario expression. Because fibroblast growth factor (FGF) can induce ectopic mario expression and inhibition of FGF signaling down-regulates mario expression, mario needs FGF as an AER factor for induction and maintenance of its expression. On the other hand, the nondigit region has an inhibitory effect on digit 2 formation. This inhibitory effect on digit 2 formation, which can be mediated by BMP4, prevents anterior expansion of mario and restricts the digit 2 region.

Download figure to PowerPoint

Although the molecular mechanism underlying the inhibitory effect of the anterior nondigit region on digit 2 formation remains obscure, there is an interesting spatial correspondence between the mario-expressing region and the anterior necrotic zone (ANZ). In normal development, the ANZ visualized by Nile blue staining (Fig. 8G) is located at the anterior margin nearby the mario-expressing region (Fig. 8H). As reported previously (Francis et al., 1994; Francis-West et al., 1995; Yokouchi et al., 1996), bmp4 is expressed in the anterior mesenchyme that coincides with the anterior nondigit region, and bone morphogenetic proteins (BMPs) are known to play a key role in the process of cell death in the developing limb bud (Ganan et al., 1996; Kawakami et al., 1996; Yokouchi et al., 1996; Zou and Niswander, 1996; Macias et al., 1997; Merino et al., 1998; Guha et al., 2002). Application of high concentration of BMP in the anterior wing bud increased cell death (Fig. 8I), and Mario expression was down-regulated at 24 hr after the BMP-bead implantation (Fig. 8J). Although lower concentration of BMP hardly increased cell death (Fig. 8K), this concentration of BMP also down-regulated mario expression (Fig. 8L).

DISCUSSION

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

Mario as a Marker Gene for Digit 2 Region

The results presented here suggest that mario expression is associated with digit 2/digit 2 formation in the chick wing bud. In all manipulations, including AER removal, ZPA implantation, RA application, and tissue ablation and transplantation, the presence and absence of digit 2 coincided with that of mario expression. These results indicate that digit 2 formation can be predicted by observation of mario expression. The expression of mario showed a temporal correlation with digit 2/digit 2 formation. Stages 22–23, when mario starts to be expressed in the wing bud, is the time when the proximal part of the autopod is determined according to results of AER-removal experiments and fate map studies (Saunders, 1948; Summerbell, 1974, 1976; Lewis, 1975; Bowen et al., 1989). Because mario continues to be expressed until stage 35, when the most-distal part of digit primordia is formed, the temporal period of mario expression seems to agree with the entire stage for autopod formation. The spatial pattern of mario expression also shows correspondence with digit 2/digit 2 formation. Mario is expressed in a narrow layer of distal mesenchyme just beneath the AER, and the expression is always anterior-specific. At stages 22–23, mario starts to be expressed at the level of somite 17, although the expression is faint. Tissue ablation experiments (Saunders, 1948; Summerbell, 1979; Wilson and Hinchliffe, 1987; and not shown) revealed that presumptive digit 2 region in wing buds of stages 21–22 is located at the distal region opposite somite 17. When DiO is injected into two distal points opposite the anterior and posterior edges of somite 17, the two populations of DiO-labeled cells sandwich the digit 2 primordium, and the mario expression appears inside the two populations after 48 hr (Fig. 5C). It is thought that the localization of mario transcripts represents the location of presumptive digit 2 even in the early wing bud before the formation of digit primordia. The digit primordia become observable at around stage 27, and distinct morphology in each individual digit becomes conspicuous at around stage 29. At these later stages, mario is expressed in anterior–distal–peripheral cells, overlying digit 2 cartilage primordium. Considering that mario is not expressed in the cartilage primordium itself of digit 2, mario expression does not mark digit 2 cartilage specifically, but it exclusively represents the digit 2 region, including digit 2 primordia and the surrounding interdigital tissue. Taken together with the finding that mario expression is not correlated with (formation of) the other digits, i.e., digit 3 and 4, we conclude that mario expression can be used as a specific marker for the digit 2 region.

Some functional analyses, including overexpression experiments of a putative full-length mario gene that we cloned resulted in no effects on digit formation (not shown), so that we have little information on the function of mario and its contribution on digit 2 formation. As far as we investigated, mario has no significant open reading frame, and no reported sequence in any databases has been found as a sequence homologous to that of mario. It is possible that mario functions as a noncoding RNA, although it also remains possible that we have only cloned a part of mario transcript. Mario expression is not wing bud-specific; mario is also expressed in the leg bud. If mario has a role in digit formation, these most anterior digits (digit 2 in the wing bud and digit I in the leg bud) may be formed by the same mechanism. The issue of whether the most-anterior digit of the wing in modern birds is digit 1 or digit 2 is a fascinating and important issue that has been debated over a long period (Wagner and Gatuthier, 1999). The expression of mario may support recent evidence that chick wing digits are 1, 2, 3 (Vargas and Fallon, 2004), and further investigation of the function of mario may lead to the establishment of a molecular approach to this issue.

Grafted ZPA Organizes Additional Digits by Different Mechanisms

Because the ZPA gives rise to a complete additional digit pattern when grafted anteriorly, the ZPA is able to change the normal prospective fate of the anterior mesenchyme to the posterior one and re-set up an additional AP axis. It was reported previously that this alteration of prospective fate after implantation of the ZPA occurs in the cell population beside an exogenous ZPA (Saunders and Gasseling, 1968; Tickle et al., 1975) and that the influence of the grafted ZPA does not reach throughout the anterior half of the bud (Yang et al., 1997). Our cell fate tracing analysis after ZPA implantation in the anterior nondigit region (Fig. 5) revealed that the additional digit 2 arises from the expanded original digit 2 region, which normally gives rise to digit 2, but that the additional digit 3 and digit 4 originate from the anterior nondigit region, which normally forms no digits. Our data suggest that the ectopic digits 3 and 4 after ZPA implantation are additional digits in which the nondigit regions are converted to the posterior digit lineage and that the ectopic digit 2 is an expanded digit that originates in the normal digit 2 region. It is thought that ZPA can repattern digit 3 and digit 4 in the anterior nondigit region. That the combination of the ZPA and the digit 2 region alters digit 2 into digits 3 and 4 (Fig. 7) indicates that the ZPA can repattern digits 3 and 4 also in the digit 2 region. This finding coincides with the report that shh can reproduce additional digits 2, 3, and 4 in the presumptive digit 2 region, in a concentration- and time-dependent manner (Tickle et al., 1975; Yang et al., 1997; Harfe et al., 2004). These results do not necessarily mean independency of digit 2 from the function of the ZPA, because the ZPA could reproduce digit 2 in the presumptive digit 3–4 region; rather, our data suggest multiple functions of the ZPA in digit identification and determination of the number of digits. Indeed, the ZPA has been shown to be involved in cell proliferation as well as in construction of the AP axis (Aono and Ide, 1988; Niswander et al., 1994; Laufer et al., 1994; Sun et al., 2000). This growth promotion activity may be an indirect function of the ZPA, mediated under the control of a positive feedback loop between the ZPA and the AER. One good candidate for this mediator is FGF; FGF application mimics the cell proliferation activity of the ZPA in vitro (Aono and Ide, 1988), and local application of FGFs gives rise to additional digit 2s, without shh induction (Riley et al., 1993; Wada and Nohno, 2001). The importance of FGF signaling on digit 2 formation was also supported by our data that simple FGF application is enough to induce the elongation of mario expression (Fig. 4D) and that inhibition of FGF signaling by SU5402 is sufficient to reduce mario expression (Fig. 4C).

Anterior Nondigit Region Has an Inhibitory Effect on Digit 2 Formation

The finding that the implanted ZPA converts cells in the anterior nondigit region into only digit 3 and digit 4 suggests that the anterior nondigit region cannot organize digit 2. This idea is supported by the following results. First, the anterior nondigit region never gave rise to mario expression and digit 2 formation when it was joined to the digit 3–4 region (Fig. 6). Second, the anterior nondigit region with grafted ZPA sometimes formed digit 3 and digit 4, but this combination never formed digit 2 (Fig. 7). Not only does the ZPA have the ability to reorganize digit 2 in the anterior nondigit region, but this anterior region is also likely to have a property intolerant of digit 2 formation. When the digit 2 region was transplanted into the anterior nondigit region, the grafted tissue failed to express either hoxa13 (an autopod marker) or mario (a digit 2 marker), and the cartilage element derived from the graft lost morphological features of a digit (Fig. 8A). It is thought that this defect is not due to a long distance of the graft position from the ZPA, because this graft autonomously maintained marker gene expression in the flank region. Because the anterior nondigit region is located next to the digit 2 region, this inhibitory effect may be involved in limitation on the space of the field for digit formation, by restricting the position of digit 2 (Fig. 8M). The ANZ, one of the programmed cell death regions in the developing limb bud, is located in the anterior nondigit region and in the anterior base of wrist at later stages. The mario-expressing region appears to be next to the ANZ (Fig. 8G,H), although it remains unclear whether these two domains are overlapped or not. Interestingly, Xt mice, which show an anteriorly extended digit region, exhibit a significant decrease in cell death in the ANZ (Aoto et al., 2002), and talpid2 chick mutants, whose digit region are wider than that of wild-type and they lack the normal pattern of cell death in the ANZ and PNZ (Dvorak and Fallon, 1991). These findings suggest a relationship between the ANZ and the space for digit formation. Down-regulation of bmp4, whose expression domain is located in the nondigit region (Francis et al., 1994; Francis-West et al., 1995; Yokouchi et al., 1996), was also shown in the Xt and talpid2 limbs (Aoto et al., 2002; Bastida et al., 2004). Taken together with the fact that the bmp4 expressed there is necessary for apoptotic cell death (Yokouchi et al., 1996; Guha et al., 2002), these results suggest that bmp4 is a good candidate for this inhibitory effect. Mario expression was remarkably down-regulated after application of high concentration of BMP4 (Fig. 8J), and cell death increased in this case (Fig. 8I). On the other hand, FGF2, which has an effect opposite to that of BMPs (Niswander and Martin, 1993), extended the expression domain of mario (Fig. 4D). FGF2 application has been reported to diminish the cell death in the ANZ, resulting in additional digit 2s (Riley et al., 1993; Montero et al., 2001). These findings are compatible with the idea that bmp4 and the ANZ in the anterior nondigit region specify the anterior border of the digit region (Fig. 8M). However, the digit 2 region transplanted into the anterior edge of a bud lost molecular and morphological features of digit 2 but was not excluded by cell death. It was reported that an application of a low concentration of BMP4 (0.01 mg/ml) cannot cause extension of the necrotic zone over the distal mesenchyme in the limb (Bastida et al., 2004). Our anterior application of 0.01 mg/ml of BMP4 also caused only a faint increase of dead cells around the bead (Fig. 8K), although mario expression was down-regulated also in this case (Fig. 8L), suggesting that a certain downstream rather than cell death also mediates BMP signal for limiting mario expression into the digit 2 region. It is therefore possible that a signaling cascade other than programmed cell death in the anterior nondigit region also contributes to specification of the anterior border of the digit region (Fig. 8M).

EXPERIMENTAL PROCEDURES

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

Cloning a Putative Full-Length mario

A 3′ end clone (approximately 450 bp) for mario transcript was cloned by a differential-display reverse transcriptase-polymerase chain reaction technique as described previously (Tamura et al., 2003). We then performed a 5′ rapid amplification of cDNA ends experiment, screenings using several kinds of cDNA libraries made from chick limb buds, and a genome with information obtained from genomic databases (Ensemble and UCSC). According to these experiments, the putative full length of mario transcript (NCBI accession no. AY885261) was estimated to be approximately 4 kb with no significant open reading frame, suggesting that mario transcript is a noncoding RNA product whose sequence has no significant motif or domain and no homology with any other genes, as far as we searched.

Whole-Mount In Situ Hybridization

The original clone for mario was used for a probe, and a hoxa13 clone for a probe was kindly provided by A. Kuroiwa. To synthesize antisense RNA probes, the mario and hoxa13 clones were digested with BamHI and XhoI and then transcribed with T7 and T3 RNA polymerase, respectively. Whole-mount in situ hybridization was carried out using standard methods.

Surgical Manipulations, DiO Labeling, and Nile Blue Staining

Fertilized chicken eggs (White Leghorn from Iwaya farm, Sendai, Japan) were routinely incubated at 38°C until a certain stage (Hamburger and Hamilton, 1951) for each manipulation. Wing buds were observably stained with Nile blue diluted in phosphate buffered saline. Preparation of grafts and the subsequent manipulations were performed with fine-sharpened tungsten needles. We performed a variety of surgical manipulations, including ZPA grafting, AER removal, and application of several molecules. For application of RA (Sigma), beads (AG1-X2 formate form, Bio-Rad Laboratories, Ltd.) were soaked in 0.01 mg/ml or 1 mg/ml RA in dimethyl sulfoxide and implanted beneath the anterior AER (Tamura et al., 1990). AG1-X2 beads were also used as carriers of SU5402 (4 mg/ml, from Calbiochem). Affigel blue beads (Bio-Rad Laboratories, Ltd.) were used as carriers of FGFs (1 mg/ml for FGF2 (human recombinant), FGF4 and FGF8 from Genzyme/TECHNE), and BMP4 (0.1 or 0.01 mg/ml human recombinant BMP4, from R&D System). For DiO labeling, DiO (Sigma) diluted in EtOH was administered by injection using a pulled micropipette.

For each experiment, some embryos were allowed to develop for more 24 or 48 hr to observe gene expressions or cell death, and others were allowed to develop for up to 10–11 days to observe the skeletal pattern by Alcian blue staining. Nile blue staining for visualization of cell death was performed as previously described (Saunders and Gasseling, 1962; Omi and Ide, 1996).

Acknowledgements

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

We thank Dr. Atsushi Kuroiwa for the hoxa13 plasmid. We also thank Drs. Hiroyuki Ide, Hiroaki Yamamoto, and Sayuri Yonei-Tamura for critical comments on the manuscript. K.T. was funded by the Ministry of Education, Science, Sports, and Culture, Japan.

REFERENCES

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