Identification of Heparan-Sulfate Rich Cells in the Loose Connective Tissues of the Axolotl (Ambystoma mexicanum) with the Potential to Mediate Growth Factor Signaling during Regeneration.

Limb regeneration is the outcome of a complex sequence of events that are mediated by interactions between cells derived from the tissues of the amputated stump. Early in regeneration, these interactions are mediated by growth factor/morphogen signaling associated with nerves and the wound epithelium. One shared property of these proregenerative signaling molecules is that their activity is dependent on interactions with sulfated glycosaminoglycans (GAGs), heparan sulfate proteoglycan (HSPG) in particular, in the extracellular matrix (ECM). We hypothesized that there are cells in the axolotl that synthesize specific HSPGs that control growth factor signaling in time and space. In this study we have identified a subpopulation of cells within the ECM of axolotl skin that express high levels of sulfated GAGs on their cell surface. These cells are dispersed in a grid-like pattern throughout the dermis as well as the loose connective tissues that surround the tissues of the limb. These cells alter their morphology during regeneration, and are candidates for being a subpopulation of connective tissue cells that function as the cells required for pattern-formation during regeneration. Given their high level of HSPG expression, their stellate morphology, and their distribution throughout the loose connective tissues, we refer to these as the positional information GRID (Groups that are Regenerative, Interspersed and Dendritic) cells. In addition, we have identified cells that stain for high levels of expression of sulfated GAGs in mouse limb connective tissue that could have an equivalent function to GRID cells in the axolotl. The identification of GRID cells may have important implications for work in the area of Regenerative Engineering.

Neurotrophic regulation of epidermal dedifferentiation during wound healing and limb regeneration in the axolotl (Ambystoma mexicanum).

Adult urodeles (salamanders) are unique in their ability to regenerate complex organs perfectly. The recently developed Accessory Limb Model (ALM) in the axolotl provides an opportunity to identify and characterize the essential signaling events that control the early steps in limb regeneration. The ALM demonstrates that limb regeneration progresses in a stepwise fashion that is dependent on signals from the wound epidermis, nerves and dermal fibroblasts from opposite sides of the limb. When all the signals are present, a limb is formed de novo. The ALM thus provides an opportunity to identify and characterize the signaling pathways that control blastema morphogenesis and limb regeneration. In the present study, we have utilized the ALM to identity the buttonhead-like zinc-finger transcription factor, Sp9, as being involved in the formation of the regeneration epithelium. Sp9 expression is induced in basal keratinocytes of the apical blastema epithelium in a pattern that is comparable to its expression in developing limb buds, and it thus is an important marker for dedifferentiation of the epidermis. Induction of Sp9 expression is nerve-dependent, and we have identified KGF as an endogenous nerve factor that induces expression of Sp9 in the regeneration epithelium.

Expression of Hoxb13 and Hoxc10 in developing and regenerating Axolotl limbs and tails.

The expression of Hox complex genes in correct spatial and temporal order is critical to patterning of the body axis and limbs during embryonic development. In order to understand the role such genes play in appendage regeneration, we have compared the expression of two 5' Hox complex genes: Hoxb13 and Hoxc10 during development and regeneration of the body axis and the limbs of axolotls. In contrast to higher vertebrates, Hoxb13 is expressed not only in the tip of the developing tail, but also in the distal mesenchyme of developing hind limbs, and at low levels in developing forelimbs. Hoxc10 is expressed as two transcripts during both development and regeneration. The short transcript (Hoxc10S) is expressed in the tip of the developing tail, in developing hind limbs, and at low levels in developing forelimbs. The long transcript (Hoxc10L) is expressed in a similar pattern, with the exception that no expression in developing forelimbs could be detected. Hoxb13 and both transcripts of Hoxc10 are expressed at high levels in the regenerating spinal cord during tail regeneration, and in both regenerating hind limbs and forelimbs. The up-regulation of expression of these genes during forelimb regeneration, relative to the very low levels of expression during forelimb development, suggests that they play a critical and perhaps unique role in regeneration. This is particularly true for Hoxc10L, which is not expressed during forelimb development, but is expressed during forelimb regeneration; thus making it the first truly "regeneration-specific" gene transcript identified to date.

Vaccinia as a tool for functional analysis in regenerating limbs: ectopic expression of Shh.

Axolotls, with their extensive abilities to regenerate as adults, provide a useful model in which to study the mechanisms of regeneration in a vertebrate, in hopes of understanding why other vertebrates cannot regenerate. Although the expression of many genes has been described in regeneration, techniques for functional analysis have so far been limited. In this paper we demonstrate a new method for efficient overexpression of foreign genes in axolotls. Using vaccinia virus expressing beta-galactosidase microinjected into regenerating limbs, we show that vaccinia can infect both dividing and nondividing limb cells. The site of infection remains discrete and there is no secondary spread of infection to nearby cells. beta-Gal is expressed at high levels in blastema cells for about a week and in differentiated cells for longer. Blastemas that have been injected with vaccinia at different stages regenerate normally. As a test of the utility of vaccinia for functional analysis in regeneration, we constructed a virus expressing Shh and injected it into the anterior of regenerating limbs. Ectopic Shh expression caused extra digits, carpals, and tarsals in the hands and feet of regenerating limbs, suggesting that despite differences in the timing of expression and the eventual pattern, the function of Shh appears to be similar to that in the developing limbs of other vertebrates. Our results demonstrate that vaccinia virus is an excellent vector for ectopically expressing genes for secreted proteins and is a useful tool to study the function of signaling molecules during the process of regeneration in urodeles.

Expression of Mmp-9 and related matrix metalloproteinase genes during axolotl limb regeneration.

One of the earliest events in limb regeneration is the extensive remodeling of the extracellular matrix (ECM). Matrix metalloproteinases (MMPs) are a family of matrix degrading enzymes that have been identified in both normal and disease states. Using RT-PCR and cDNA library screening, we have isolated sequences homologous to four different Mmp genes. The spatial and temporal expression of one of these, Mmp-9, has been analyzed during axolotl limb regeneration. Northern blot analysis identifies a 3.8 kb transcript that is abundantly expressed during regeneration, and whole-mount in situ hybridization has uncovered an unusual bi-phasic expression pattern. The first phase begins at 2 hours after amputation, and expression is confined to the healed wound epithelium. This phase continues for 2 days, showing peak expression at 14 hours after amputation. This early phase may be needed to retard reformation of the basal lamina of the epidermis, and thereby facilitate the epidermal-mesenchymal interactions required for successful regeneration. The second phase begins a few days later when a small blastema has formed. During this phase, expression is in the mesenchyme, localized to cells around the tips of the cut skeletal elements. This expression is maintained through several stages until redifferentiation begins. The timing and position of the second phase of expression is consistent with a role for Mmp-9 in the removal of damaged cartilage matrix. We have also discovered that the time of onset of Mmp-9 expression is sensitive to denervation, which causes a delay of several hours. Finally, retinoids, known for their dramatic effects on the pattern of regenerating limbs, can cause a down regulation of Mmp-9 expression. Dev Dyn 1999;216:2-9.

Sonic hedgehog (shh) expression in developing and regenerating axolotl limbs.

Sonic hedgehog (shh) expression is detectable in the posterior mesenchyme of many developing vertebrate limbs. We have isolated an RT-PCR fragment from the axolotl, Ambystoma mexicanum, that has high identity to other vertebrate shh genes. We describe the localization of this transcript during development and regeneration and in response to tissue grafts and retinoic acid (RA) exposure in the axolotl. Even though axolotl digits show a reversed polarity of differentiation (anterior [A] to posterior [P]) when compared to other tetrapods (P to A), shh is nevertheless expressed on the posterior margin of developing and regenerating limb buds. When A cells are grafted adjacent to P cells, an ectopic domain of shh is induced. Exposure to retinoic acid (RA), a molecule known to alter pattern in all three limb axes in urodeles, results in ectopic expression of shh in anterior cells of the regeneration blastema. Prior to this induced expression in response to RA, there is an earlier response by the endogenous domain of shh, which is downregulated within the first few hours of exposure.

Analysis of the pattern of QM expression during mouse development.

QM, a novel gene that was originally identified as a putative tumor suppressor gene, has since been cloned from species encompassing members of the plant, animal, and fungal kingdoms. Sequence comparison indicates that QM has been highly conserved throughout eukaryotic evolution. QM is a member of a multigene family in both mouse and man, is expressed in a broad range of tissues, and is downregulated during adipocyte differentiation. Jif-1, a chicken homolog of QM, has been reported to interact with the protooncogene c-Jun, and to inhibit transactivation of AP-1 regulated promoters in vitro. Furthermore, disruption of the yeast QM homolog is lethal. Although these studies suggest that the QM gene product plays an important role within the normal cell, the precise role of QM has remained elusive. In this study, a thorough analysis of the pattern of QM expression during mouse development was undertaken, using the techniques of whole mount in situ hybridization and whole mount immunohistochemistry, in combination with conventional immunohistochemical analysis of tissue sections. QM is expressed in numerous embryonic tissues, and is differentially expressed throughout the embryo. The cytoplasmic localization of QM is consistent with its reported association with ribosomes, and inconsistent with its previously hypothesized function as a direct modulator of the nuclear protooncogene c-Jun. QM is expressed in the developing epidermis, and is particularly strong within developing limbs. Analysis of embryos of various stages of gestation indicate that QM is downregulated in the surface ectoderm of the embryo as development proceeds. QM protein is not detectable within either nucleated or enucleated red blood cell precursors. QM is strongly expressed within chondrocytes within the transition zone of developing limb cartilage, as well as within differentiated keratinocytes of the suprabasal regions of the epidermis. Furthermore, within both cartilage and skin, there is an inverse relationship between QM expression and proliferative capacity. This pattern of QM expression suggests that this novel gene product may be involved in processes such as posttranslational protein processing which are essential for differentiation of specific tissues during embryogenesis.

Expression of Msx-2 during development, regeneration, and wound healing in axolotl limbs.

Msx genes are transcription factors that are expressed during embryogenesis of developing appendages in regions of epithelial-mesenchymal interactions. Various lines of evidence indicate that these genes function to maintain embryonic tissues in an undifferentiated, proliferative state. We have identified the axolotl homolog of Msx-2, and investigated its expression during limb development, limb regeneration, and wound healing. As in limb buds of higher vertebrates, axolotl Msx-2 is expressed in the apical epidermis and mesenchyme; however, its expression domain is more extensive, reflecting the broader region of the apical epidermal cap in amphibians. Msx-2 expression is downregulated at late stages of limb development, but is reexpressed within one hour after limb amputation. Msx-2 is also reexpressed during wound healing, and may be essential in the early stages of initiation of the limb regeneration cascade.

Expression of HoxD genes in developing and regenerating axolotl limbs.

Hox genes play a critical role in the development of the vertebrate axis and limbs, and previous studies have implicated them in the specification of positional identity, the control of growth, and the timing of differentiation. Axolotl limbs offer an opportunity to distinguish these alternatives because the sequence of skeletal differentiation is reversed along the anterior-posterior axis relative to that of other tetrapods. We report that during early limb development, expression patterns of HoxD genes in axolotls resemble those in amniotes and anuran amphibians. At later stages, the anterior boundary of Hoxd-11 expression is conserved with respect to morphological landmarks, but there is no anterior-distal expansion of the posterior domain of Hoxd-11 expression similar to that observed in mice and chicks. Since axolotls do not form an expanded paddle-like handplate prior to digit differentiation, we suggest that anterior expansion of expression in higher vertebrates is linked to the formation of the handplate, but is clearly not necessary for digit differentiation. We also show that the 5' HoxD genes are reexpressed during limb regeneration. The change in the expression pattern of Hoxd-11 during the course of regeneration is consistent with the hypothesis that the distal tip of the regenerate is specified first, followed by intercalation of intermediate levels of the pattern. Both Hoxd-8 and Hoxd-10 are expressed in non-regenerating wounds, but Hoxd-11 is specific for regeneration. It is also expressed in the posterior half of nerve-induced supernumerary outgrowths.

Cell cycle length affects gene expression and pattern formation in limbs.

The relationship between growth and pattern specification during development remains elusive. Some molecules known to function as growth factors are also potent agents of pattern formation. This raises the possibility that growth factors could act in pattern formation via an effect on the cell cycle. We have tested the significance of the length of the cell cycle for gene expression and pattern formation in developing chick limb buds by locally slowing the cell cycle. When anterior cell cycles are lengthened by reversible inhibition of DNA replication or by other means, some genes characteristic of the posterior polarizing region are expressed, and digit duplication is observed. Conversely, when posterior cell cycles are slowed, expression of some posterior-specific genes is inhibited, but the pattern is normal. These results indicate that control of the length of the cell cycle could play a primary role in pattern formation by influencing the complement of genes expressed in a particular region of the embryo.

Nerve dependency of regeneration: the role of Distal-less and FGF signaling in amphibian limb regeneration.

Dlx-3, a homolog of Drosophila Dll, has been isolated from an axolotl blastema cDNA library, and its expression in developing and regenerating limbs characterized. The normal expression pattern, and the changes that occur during experimental treatments, indicate a correlation between Dlx-3 expression and the establishment of the outgrowth-permitting epidermis. Dlx-3 is expressed at high levels in a distal-to-proximal gradient in the epidermis of developing limb buds, and is upregulated in the apical ectodermal cap (AEC) during limb regeneration. Expression is maximal at the late bud stage of regeneration, coincident with the transition from the early phase of nerve dependency to the later phase of nerve independence. Dlx-3 expression in the epidermis is rapidly downregulated by denervation during the nerve-dependent phase and is unaffected by denervation during the nerve-independent phase. We investigated this relationship between nerves and Dlx-3 expression by implanting FGF-2 beads into regenerates that had been denervated at a nerve-dependent stage. Dlx-3 expression was maintained by FGF-2 after denervation, and regeneration progressed to completion. In addition, we detected FGF-2 protein in the AEC and in nerves, and observed that the level of expression in both tissues decreases dramatically in response to denervation. We conclude that both limb development and regeneration require a permissive epidermis, characterized by Dlx-3 and FGF expression, both of which are maintained by FGF through an autocrine loop. The transformation of the limb epidermis into a functional AEC that produces and responds to FGF autocatalytically, is presumed to be induced by FGF. Since nerves appear to be a source of this priming FGF, it is possible that a member of the FGF family of growth factors is the elusive neurotrophic factor of limb regeneration.

Molecular mechanisms in the control of limb regeneration: the role of homeobox genes.

Axolotls are unique among vertebrates in their ability to regenerate lost appendages as adults. They provide the opportunity to study the mechanism of regeneration in vertebrates and are an inspiration to pursue the goal of appendage regeneration in humans. In this article, we review data on the role of homeobox-containing genes in the regulation of limb regeneration. As a group, these genes are important in pattern formation in the primary body axis, developing limbs and regenerating limbs. To date, a total of 22 homeobox genes have been identified as being expressed in regenerating limbs. Nearly all of these are also expressed during limb regeneration, further supporting the view that limb development and regeneration involve similar regulatory mechanisms. Our recent results on the expression of HoxA genes demonstrate that once a blastema has formed, subsequent outgrowth and pattern formation are similar to those of limb development. In contrast to developing limbs, reexpression of the HoxA genes in regeneration occurs by a non-colinear mechanism that likely is related to the necessity of mature limb cells to undergo dedifferentiation in order to give rise to the blastema. These studies also indicate that the pattern is respecified by a distal-first mechanism during regeneration in contrast to the apparent proximal-to-distal sequence observed in developing limbs. Expression of the HoxA genes is altered coordinately in response to retinoic acid in a manner consistent with the transformation of a distal blastema to a proximal blastema. Given the recent increase in studies of the molecules involved in regeneration, it is likely that many of the functionally important regeneration genes will be identified and characterized in the near future.

Digit induction by Hensen's node and notochord involves the expression of shh but not RAR-beta 2.

It is well established that Hensen's nodes can induce the formation of supernumerary digits after grafting into the anterior margin of the developing limb bud. The recent finding that distinct mesodermal cell populations are segregated within the node has made it possible to isolate different prospective cell types in an attempt to correlate digit-inducing ability with cell fate. We find that the prospective notochord cells contained within Hensen's node are able to induce supernumerary digits, whereas presumptive somite cells cannot. This early difference in inducing ability persists into later stages of development: epithelial somites are unable to induce while notochord from all lengths of the neuraxis continues to induce. Using probes to retinoic acid receptor-beta 2 and sonic hedgehog (shh) we find no evidence to support the idea that inducing tissues generate extra digits by releasing retinoic acid into adjacent limb tissue but find that the inducing ability of a tissue correlates with its expression of shh.

Regulation of HoxA expression in developing and regenerating axolotl limbs.

Homeobox genes are important in the regulation of outgrowth and pattern formation during limb development. It is likely that homeobox genes play an equally important role during limb regeneration. We have isolated and identified 17 different homeobox-containing genes expressed by cells of regenerating axolotl limbs. Of these, nearly half of the clones represent genes belonging to the HoxA complex, which are thought to be involved in pattern formation along the proximal-distal limb axis. In this paper we report on the expression patterns of two 5' members of this complex, HoxA13 and HoxA9. These genes are expressed in cells of developing limb buds and regenerating blastemas. The pattern of expression in developing axolotl limb buds is comparable to that in mouse and chick limb buds; the expression domain of HoxA13 is more distally restricted than that of HoxA9. As in developing mouse and chick limbs, HoxA13 likely functions in the specification of distal limb structures, and HoxA9 in the specification of more proximal structures. In contrast, during regeneration, HoxA13 and HoxA9 do not follow the rule of spatial colinearity observed in developing limbs. Instead, both genes are initially expressed in the same population of stump cells, giving them a distal Hox code regardless of the level of amputation. In addition, both are reexpressed within 24 hours after amputation, suggesting that reexpression may be synchronous rather than temporally colinear. Treatment with retinoic acid alters this Hox code to that of a more proximal region by the rapid and differential downregulation of HoxA13, at the same time that expression of HoxA9 is unaffected. HoxA reexpression occurs prior to blastema formation, 24-48 hours after amputation, and is an early molecular marker for dedifferentiation.

Developmental regulation of a matrix metalloproteinase during regeneration of axolotl appendages.

Removal of specific extracellular matrix (ECM) components has been implicated in the initiation of salamander limb regeneration. Remodeling of the ECM at the distal stump is necessary for the release of cells that eventually contribute to the blastema. Several matrix metalloproteinases (MMPs) have been well characterized as important to various physiological and pathological processes, such as bone remodeling and tumor invasion. The goal of this study is to identify and characterize MMPs that modulate the ECM during appendage regeneration in the axolotl Ambystoma mexicanum. By analyzing axolotl tissue extracts using gelatin-substrate gels, we have identified a 90-kDa gelatinase/collagenase that is upregulated within hours after limb amputation. This gelatinase shows a dramatic elevation in activity during the dedifferentiation and blastema stages. Its activity declines by the palette stage and returns to its basal level by the digit stages. The increase in activity of the 90-kDa gelatinase in response to amputation is independent of the nerve supply and the wound epithelium but these factors affect its subsequent downregulation. In addition to the blastema, the 90-kDa gelatinase can be detected in the stump at least 4 mm proximal to the regenerate. Similar regulation of the 90-kDa gelatinolytic activity is observed during tail regeneration and flank would healing. We suggest that this 90-kDa gelatinase/collagenase may play a role in the initiation and rapid growth phase of axolotl regeneration and wound healing.

Reciprocal changes in Hox D13 and RAR-beta 2 expression in response to retinoic acid in chick limb buds.

The finding that retinoic acid (RA) has major effects on pattern formation in developing chick limbs has led to the conclusion that RA plays a central role in the normal development of the limb. In addition, it has been demonstrated that treatment with exogenous RA leads to changes in expression of homeobox-containing Hox complex genes within the developing limb. This has been used to argue that, since RA activates 5' Hox D complex genes in the same temporal and spatial sequence as in normal development, Hox genes might be regulated by RA during normal limb development. In this study, we further examine the temporal and spatial changes in expression of Hox D13 and RAR-beta 2 transcripts in wing buds in response to local application of RA in vivo. We confirm reports that RAR-beta 2 expression is induced early and locally by RA. However, we find that the effect of RA on Hox D13 gene expression at the distal end of wing buds at stage 25/26 is downregulation of transcript expression. Furthermore, we find that activation of ectopic Hox D13 expression in response to implantation of RA beads along the anterior margin at stage 20/21 is indirect. Finally, the effects of RA exposure on Hox D13 expression appear to directly correlate with effects on the pattern of distal skeletal elements.

Regeneration of HoxD expression domains during pattern regulation in chick wing buds.

The expression domains of genes located at the 5' end of the HoxD (formerly Hox-4) complex appear to correlate with pattern along both the proximal-distal (PrDi) and the anterior-posterior (AP) axes of the developing limb bud, and it has been suggested that the HoxD gene products are involved in the specification of positional information during limb development. The apical ectodermal ridge is required for limb outgrowth and is thought to influence mesodermal cells at the distal end of the limb bud in a region within which patterning events occur. In this paper, we examine the expression of 5' HoxD genes during PrDi pattern regulation in chick wing buds. In limbs undergoing pattern regulation, we demonstrate that the domains of HoxD11 and HoxD13 gene expression are "regenerated" within 24 hr of removal of the distal mesenchyme. In contrast, in limbs which will not form distal structures, HoxD13 expression becomes reduced.