The effect of retinoic acid on positional memory in the dorsoventral axis of regenerating axolotl limbs

Making a new limb out of old cells: exploring endogenous cell reprogramming and its role during limb regeneration

Cellular reprogramming is characterized by the induced dedifferentiation of mature cells into a more plastic and potent state. This process can occur through artificial reprogramming manipulations in the laboratory such as nuclear reprogramming and induced pluripotent stem cell (iPSC) generation, and endogenously in vivo during amphibian limb regeneration. In amphibians such as the Mexican axolotl, a regeneration permissive environment is formed by nerve-dependent signaling in the wounded limb tissue. When exposed to these signals, limb connective tissue cells dedifferentiate into a limb progenitor-like state. This state allows the cells to acquire new pattern information, a property called positional plasticity. Here, we review our current understanding of endogenous reprogramming and why it is important for successful regeneration. We will also explore how naturally induced dedifferentiation and plasticity were leveraged to study how the missing pattern is established in the regenerating limb tissue.

Retinoic Acid-Induced Limb Duplications

Retinoic acid (RA) and the family of molecules based on vitamin A known as retinoids have remarkable effects on limb regeneration in salamanders and newts and cause whole limb duplications in a concentration-dependent manner. They respecify all three axes of the limb-the proximodistal, the anteroposterior, and the dorsoventral axis. As a result, complete limbs can be induced to regenerate from distal amputation planes producing two limbs in tandem. Here, we describe the basic methods for undertaking these experiments as well as the use of new synthetic retinoids which have retinoic acid receptor-selective actions. These will be valuable tools in future studies on the molecular basis of limb duplications and thus our understanding of the nature of positional information in the regenerating salamander limb.

Positional Memory in Vertebrate Regeneration: A Century's Insights from the Salamander Limb

Salamanders, such as axolotls and newts, can regenerate complex tissues including entire limbs. But what mechanisms ensure that an amputated limb regenerates a limb, and not a tail or unpatterned tissue? An important concept in regeneration is positional memory-the notion that adult cells "remember" spatial identities assigned to them during embryogenesis (e.g., "head" or "hand") and use this information to restore the correct body parts after injury. Although positional memory is well documented at a phenomenological level, the underlying cellular and molecular bases are just beginning to be decoded. Herein, we review how major principles in positional memory were established in the salamander limb model, enabling the discovery of positional memory-encoding molecules, and advancing insights into their pattern-forming logic during regeneration. We explore findings in other amphibians, fish, reptiles, and mammals and speculate on conserved aspects of positional memory. We consider the possibility that manipulating positional memory in human cells could represent one route toward improved tissue repair or engineering of patterned tissues for therapeutic purposes.

ECM-mediated positional cues are able to induce pattern, but not new positional information, during axolotl limb regeneration

The Mexican Axolotl is able to regenerate missing limb structures in any position along the limb axis throughout its life and serves as an excellent model to understand the basic mechanisms of endogenous regeneration. How the new pattern of the regenerating axolotl limb is established has not been completely resolved. An accumulating body of evidence indicates that pattern formation occurs in a hierarchical fashion, which consists of two different types of positional communications. The first type (Type 1) of communication occurs between connective tissue cells, which retain memory of their original pattern information and use this memory to generate the pattern of the regenerate. The second type (Type 2) of communication occurs from connective tissue cells to other cell types in the regenerate, which don’t retain positional memory themselves and arrange themselves according to these positional cues. Previous studies suggest that molecules within the extracellular matrix (ECM) participate in pattern formation in developing and regenerating limbs. However, it is unclear whether these molecules play a role in Type 1 or Type 2 positional communications. Utilizing the Accessory Limb Model, a regenerative assay, and transcriptomic analyses in regenerates that have been reprogrammed by treatment with Retinoic Acid, our data indicates that the ECM likely facilities Type-2 positional communications during limb regeneration.

Remembering where we are: Positional information in salamander limb regeneration

Fifty years ago, Lewis Wolpert defined an important question in developmental biology: how are cell fates determined by the positions of cells within a system? He proposed that cells retain positional values as if they lie within a coordinate system and that the interpretation of these values produces patterns in development. He referred to this concept as positional information. Though initially controversial, this concept of positional information has proven to be profoundly influential in developmental biology. One area in which the influence of Wolpert's theoretical work can be clearly demonstrated is the study of limb regeneration in salamanders. Here, we review the work in limb regeneration leading up to Wolpert defining the concept of positional information and how his theory has guided regeneration research over the subsequent 50 years.

Regrowth of zebrafish caudal fin regeneration is determined by the amputated length

Fish have a high ability to regenerate fins, including the caudal fin. After caudal fin amputation, original bi-lobed morphology is reconstructed during its rapid regrowth. It is still controversial whether positional memory in the blastema cells regulates reconstruction of fin morphology as in amphibian limb regeneration, in which limb blastema cells located at the same proximal-distal level have the same positional identity. We investigated growth period and growth rate in zebrafish caudal fin regeneration. We found that both the growth period and growth rate differed for fin rays that were amputated at the same proximal-distal level, indicating that it takes different periods of time for fin rays to restore their original lengths after straight amputation. We also show that more proximal amputation takes longer period to reconstruct the original morphology/size than more distal amputation. Statistical analysis suggested that both the growth period/rate are determined by amputated length (depth) regardless of the fin ray identity along dorsal-ventral axis. In addition, we suggest the possibility that the structural/physical condition such as width of the fin ray at the amputation site (niche at the stump) may determine the growth period/rate.

RA Signaling in Limb Development and Regeneration in Different Species

This chapter brings together data on the role of retinoic acid (RA) in the embryonic development of fins in zebrafish , limbs in amphibians , chicks , and mice, and regeneration of the amphibian limb . The intention is to determine whether there is a common set of principles by which we can understand the mode of action of RA in both embryos and adults. What emerges from this synthesis is that there are indeed commonalities in the involvement of RA in processes that ventralize, posteriorize, and proximalize the developing and regenerating limb . Different axes of the limb have historically been studied independently; as for example, the embryonic development of the anteroposterior (AP) axis of the chick limb bud versus the regeneration of the limb bud proximodistal (PD) axis . But when we take a broader view, a unifying principle emerges that explains why RA administration to embryos and regenerating limbs results in the development of multiple limbs in both cases. As might be expected, different molecular pathways govern the development of different systems and model organisms, but despite these differences, the pathways involve similar RA signaling genes, such as tbx5, meis, shh, fgfs and hox genes. Studies of developing and regenerating systems have highlighted that RA acts by being synthesized in one embryonic location while acting in another one, exactly as embryonic morphogens do, although there is no evidence for the presence of an RA gradient within the limb . What also emerges is that there is a paucity of information on the involvement of RA in development of the dorsoventral (DV) axis . A molecular explanation as to how RA establishes and alters positional information in all three axes is the most important area of study for the future.

Hierarchical pattern formation during amphibian limb regeneration

In 1901 T.H. Morgan proposed in "Regeneration" that pattern formation in amphibian limb regeneration is a stepwise process. Since, biologist have continued to piece together the molecular components of this process to better understand the "patterning code" responsible for regenerate formation. Within this context, several different models have been proposed; however, all are based on one of two underlying hypotheses. The first is the "morphogen hypothesis" that dictates that pattern emerges from localized expression of signaling molecules, which produce differing position-specific cellular responses in receptive cells depending on the intensity of the signal. The second hypothesis is that cells in the remaining tissues retain memory of their patterning information, and use this information to generate new cells with the missing positional identities. A growing body of evidence supports the possibility that these two mechanisms are not mutually exclusive. Here, we propose our theory of hierarchical pattern formation, which consists of 4 basic steps. The first is the existence of cells with positional memory. The second is the communication of positional information through cell-cell interactions in a regeneration-permissive environment. The third step is the induction of molecular signaling centers. And the last step is the interpretation of these signals by specialized cell types to ultimately restore the limb in its entirety. Biological codes are intertwined throughout this model, and we will discuss their multiple roles and mechanisms.

Chromatin dynamics underlying the precise regeneration of a vertebrate limb - Epigenetic regulation and cellular memory

Wound healing, tissue regeneration, and organ regrowth are all regeneration phenomena observed in vertebrates after an injury. However, the ability to regenerate differs greatly among species. Mammals can undergo wound healing and tissue regeneration, but cannot regenerate an organ; for example, they cannot regrow an amputated limb. In contrast, amphibians and fish have much higher capabilities for organ-level regeneration. In addition to medical studies and those in conventional mammalian models such as mice, studies in amphibians and fish have revealed essential factors for and mechanisms of regeneration, including the regrowth of a limb, tail, or fin. However, the molecular nature of the cellular memory needed to precisely generate a new appendage from an amputation site is not fully understood. Recent reports have indicated that organ regeneration is closely related to epigenetic regulation. For example, the methylation status of genomic DNA is related to the expression of regeneration-related genes, and histone-modification enzymes are required to control the chromatin dynamics for regeneration. A proposed mechanism of cellular memory involving an inheritable system of epigenetic modification led us to hypothesize that epigenetic regulation forms the basis for cellular memory in organ regeneration. Here we summarize the current understanding of the role of epigenetic regulation in organ regeneration and discuss the relationship between organ regeneration and epigenetic memory.

FGF, BMP, and RA signaling are sufficient for the induction of complete limb regeneration from non-regenerating wounds on Ambystoma mexicanum limbs

Some organisms, such as the Mexican axolotl, have the capacity to regenerate complicated biological structures throughout their lives. Which molecular pathways are sufficient to induce a complete endogenous regenerative response in injured tissue is an important question that remains unanswered. Using a gain-of-function regeneration assay, known as the Accessory Limb Model (ALM), we and others have begun to identify the molecular underpinnings of the three essential requirements for limb regeneration; wounding, neurotrophic signaling, and the induction of pattern from cells that retain positional memory. We have previously shown that treatment of Mexican axolotls with exogenous retinoic acid (RA) is sufficient to induce the formation of complete limb structures from blastemas that were generated by deviating a nerve bundle into an anterior-located wound site on the limb. Here we show that these ectopic structures are capable of regenerating and inducing new pattern to form when grafted into new anterior-located wounds. We additionally found that the expression of Alx4 decreases, and Shh expression increases in these anterior located blastemas, but not in the mature anterior tissues, supporting the hypothesis that RA treatment posteriorizes blastema tissue. Based on these and previous observations, we used the ALM assay to test the hypothesis that a complete regenerative response can be generated by treating anterior-located superficial limb wounds with a specific combination of growth factors at defined developmental stages. Our data shows that limb wounds that are first treated with a combination of FGF-2, FGF-8, and BMP-2, followed by RA treatment of the resultant mid-bud stage blastema, will result in the generation of limbs with complete proximal/distal and anterior/posterior limb axes. Thus, the minimal signaling requirements from the nerve and a positional disparity are achieved with the application of this specific combination of signaling molecules.

Mechanisms of urodele limb regeneration

This review explores the historical and current state of our knowledge about urodele limb regeneration. Topics discussed are (1) blastema formation by the proteolytic histolysis of limb tissues to release resident stem cells and mononucleate cells that undergo dedifferentiation, cell cycle entry and accumulation under the apical epidermal cap. (2) The origin, phenotypic memory, and positional memory of blastema cells. (3) The role played by macrophages in the early events of regeneration. (4) The role of neural and AEC factors and interaction between blastema cells in mitosis and distalization. (5) Models of pattern formation based on the results of axial reversal experiments, experiments on the regeneration of half and double half limbs, and experiments using retinoic acid to alter positional identity of blastema cells. (6) Possible mechanisms of distalization during normal and intercalary regeneration. (7) Is pattern formation is a self-organizing property of the blastema or dictated by chemical signals from adjacent tissues? (8) What is the future for regenerating a human limb?

Retinoic acid receptor regulation of epimorphic and homeostatic regeneration in the axolotl

Salamanders are capable of regenerating amputated limbs by generating a mass of lineage-restricted cells called a blastema. Blastemas only generate structures distal to their origin unless treated with retinoic acid (RA), which results in proximodistal (PD) limb duplications. Little is known about the transcriptional network that regulates PD duplication. In this study, we target specific retinoic acid receptors (RARs) to either PD duplicate (RA treatment or RARγ agonist) or truncate (RARβ antagonist) regenerating limbs. RARE-EGFP reporter axolotls showed divergent reporter activity in limbs undergoing PD duplication versus truncation, suggesting differences in patterning and skeletal regeneration. Transcriptomics identified expression patterns that explain PD duplication, including upregulation of proximal homeobox gene expression and silencing of distal-associated genes, whereas limb truncation was associated with disrupted skeletal differentiation. RARβ antagonism in uninjured limbs induced a loss of skeletal integrity leading to long bone regression and loss of skeletal turnover. Overall, mechanisms were identified that regulate the multifaceted roles of RARs in the salamander limb including regulation of skeletal patterning during epimorphic regeneration, skeletal tissue differentiation during regeneration, and homeostatic regeneration of intact limbs.

Beryllium nitrate inhibits fibroblast migration to disrupt epimorphic regeneration

Epimorphic regeneration proceeds with or without formation of a blastema, as observed for the limb and skin, respectively. Inhibition of epimorphic regeneration provides a means to interrogate the cellular and molecular mechanisms that regulate it. In this study, we show that exposing amputated limbs to beryllium nitrate disrupts blastema formation and causes severe patterning defects in limb regeneration. In contrast, exposing full-thickness skin wounds to beryllium only causes a delay in skin regeneration. By transplanting full-thickness skin from ubiquitous GFP-expressing axolotls to wild-type hosts, we demonstrate that beryllium inhibits fibroblast migration during limb and skin regeneration in vivo Moreover, we show that beryllium also inhibits cell migration in vitro using axolotl and human fibroblasts. Interestingly, beryllium did not act as an immunostimulatory agent as it does in Anurans and mammals, nor did it affect keratinocyte migration, proliferation or re-epithelialization, suggesting that the effect of beryllium is cell type-specific. While we did not detect an increase in cell death during regeneration in response to beryllium, it did disrupt cell proliferation in mesenchymal cells. Taken together, our data show that normal blastema organogenesis cannot occur without timely infiltration of local fibroblasts and highlights the importance of positional information to instruct pattern formation during regeneration. In contrast, non-blastemal-based skin regeneration can occur despite early inhibition of fibroblast migration and cell proliferation.

Position-specific induction of ectopic limbs in non-regenerating blastemas on axolotl forelimbs

Ectopic retinoic acid (RA) has been hypothesized to reprogram the positional identity of cells in developing and regenerating limbs to a single positional value corresponding to the posterior‐ventral‐proximal (PVPr) position on the limb. We tested this hypothesis by using RA to reprogram the information of blastema cells that were induced to form at different positions around the limb circumference. We observed that RA treatment of blastemas in anterior and dorsal locations, but not posterior and ventral locations, resulted in the induction of complete ectopic limbs. These position‐specific differences in limb induction are probably due to differences in the positional disparity between the RA‐reprogrammed blastema cells and the cells at the periphery of the wound. These observations are consistent with the hypothesis that RA treatment reprograms the information in blastema cells to the PVPr position on the limb, since anterior and dorsal positions have the largest disparity and posterior and ventral have the smallest disparity from the PVPr identity.

Advances in signaling in vertebrate regeneration as a prelude to regenerative medicine

While all animals have evolved strategies to respond to injury and disease, their ability to functionally recover from loss of or damage to organs or appendages varies widely damage to skeletal muscle, but, unlike amphibians and fish, they fail to regenerate heart, lens, retina, or appendages. The relatively young field of regenerative medicine strives to develop therapies aimed at improving regenerative processes in humans and is predicated on >40 years of success with bone marrow transplants. Further progress will be accelerated by implementing knowledge about the molecular mechanisms that regulate regenerative processes in model organisms that naturally possess the ability to regenerate organs and/or appendages. In this review we summarize the current knowledge about the signaling pathways that regulate regeneration of amphibian and fish appendages, fish heart, and mammalian liver and skeletal muscle. While the cellular mechanisms and the cell types involved in regeneration of these systems vary widely, it is evident that shared signals are involved in tissue regeneration. Signals provided by the immune system appear to act as triggers of many regenerative processes. Subsequently, pathways that are best known for their importance in regulating embryonic development, in particular fibroblast growth factor (FGF) and Wnt/beta-catenin signaling (as well as others), are required for progenitor cell formation or activation and for cell proliferation and specification leading to tissue regrowth. Experimental activation of these pathways or interference with signals that inhibit regenerative processes can augment or even trigger regeneration in certain contexts.

Amphibian regeneration and stem cells

Larval and adult urodeles and anuran tadpoles readily regenerate their limbs via a process of histolysis and dedifferentiation of mature cells local to the amputation surface that accumulate under the wound epithelium as a blastema of stem cells. These stem cells require growth and trophic factors from the apical epidermal cap (AEC) and the nerves that re-innervate the blastema for their survival and proliferation. Members of the fibroblast growth factor (FGF) family synthesized by both AEC and nerves, and glial growth factor, substance P, and transferrin of nerves are suspected survival and proliferation factors. Stem cells derived from fibroblasts and muscle cells can transdifferentiate into other cell types during regeneration. The regeneration blastema is a self-organizing system based on positional information inherited from parent limb cells. Retinoids, which act through nuclear receptors, have been used in conjunction with assays for cell adhesivity to show that positional identity of blastema cells is encoded in the cell surface. These molecules are involved in the cell-cell signaling network that re-establishes the original structural pattern of the limb. Other systems of interest that regenerate by histolysis and dedifferentiation of pigmented epithelial cells are the neural retina and lens. Members of the FGF family are also important to the regeneration of these structures. The mechanism of amphibian regeneration by dedifferentiation is of importance to the development of a regenerative medicine, since understanding this mechanism may offer insights into how we might chemically induce the regeneration of mammalian tissues.

Regeneration of the urodele limb: a review

Urodele amphibians have been widely used for studies of limb regeneration. In this article, we review studies on blastema cell proliferation and propose a model of blastemal self-organization and patterning. The model is based on local cell interactions that intercalate positional identities within circumferential and proximodistal boundaries that outline the regenerate. The positional identities created by the intercalation process appear to be reflected in the molecular composition of the cell surface. Transcription factors and signaling molecules involved in patterning are discussed within the context of the boundary/intercalation model.

Retinoid antagonists inhibit normal patterning during limb regeneration in the axolotl, Ambystoma mexicanum

Retinoic acid (RA) has been detected in the regenerating limb of the axolotl, and exogenous RA can proximalize, posteriorize, and ventralize blastemal cells. Thus, RA may be an endogenous regulatory factor during limb regeneration. We have investigated whether endogenous retinoids are essential for patterning during axolotl (Ambystoma mexicanum) limb regeneration by using retinoid antagonists that bind to specific RAR (retinoic acid receptor) or RXR (retinoid X receptor) retinoid receptor subtypes. Retinoid antagonists (Ro41-5253, Ro61-8431, LE135, and LE540) were administered to regenerating limbs using implanted silastin blocks loaded with each antagonist. The skeletal pattern of regenerated limbs treated with Ro41-5253 or Ro61-8431 differed only slightly from control limbs. Treatment with LE135 inhibited limb regeneration, while treatment with LE540 allowed relatively normal limb regeneration. When LE135 and LE540 were implanted together, regeneration was not completely inhibited and a hand-like process regenerated. These results demonstrate that interfering with retinoid receptors can modify pattern in the regenerating limb indicating that endogenous retinoids are important during patterning of the regenerating limb.

Leaping lopsided: a review of the current hypotheses regarding etiologies of limb malformations in frogs

Recent progress in the investigation of limb malformations in free-living frogs has underlined the wide range in the types of limb malformations and the apparent spatiotemporal clustering of their occurrence. Here, we review the current understanding of normal and abnormal vertebrate limb development and regeneration and discuss some of the molecular events that may bring about limb malformation. Consideration of the differences between limb development and regeneration in amphibians has led us to the hypothesis that some of the observed limb malformations come about through misdirected regeneration. We report the results of a pilot study that supports this hypothesis. In this study, the distal aspect of the right hindlimb buds of X. laevis tadpoles was amputated at the pre-foot paddle stage. The tadpoles were raised in water from a pond in Minnesota at which 7% of surveyed newly metamorphosed feral frogs had malformations. Six percent (6 of 100) of the right limbs of the tadpoles raised in pond water developed abnormally. One truncated right limb was the only malformation in the control group, which was raised in dechlorinated municipal water. All unamputated limbs developed normally in both groups. Three major factors under consideration for effecting the limb malformations are discussed. These factors include environmental chemicals (primarily agrichemicals), encysted larvae (metacercariae) of trematode parasites, and increased levels of ultraviolet light. Emphasis is placed on the necessary intersection of environmental stressors and developmental events to bring about the specific malformations that are observed in free-living frog populations.

Morphological clues from multilegged frogs: are retinoids to blame?

Morphological analysis was performed on multilegged deformed frogs representing five species from 12 different localities in California, Oregon, Arizona, and New York. The pattern of duplicated limbs was consistent with mechanical perturbation by trematode infestation but not with the effects of retinoids.

Retinoic acid and thyroid hormone may function through similar and competitive pathways in regenerating axolotls

The objective of this study was to determine whether thyroid hormone (TH) would interfere with retinoic acid (RA), which proximalizes axolotl larvae regenerate limb pattern. RA and TH are ligands for members of the steroid hormone thyroid hormone nuclear binding protein superfamily which form functional homodimers, but may also form stable heterodimers with the RXR protein and may recognize identical DNA sequences. TH alone does not affect limb pattern but induces metamorphosis in regenerating animals. Coinjected animals do not metamorphose, and when compared to RA controls regenerate more proximal and in some cases anteroposterior (AP) and dorsoventral (DV) duplicate limb structures. In addition, the tissues that are normally lost or changed during metamorphosis appear to be sensitized resulting in the formation of (1) new dorsal gill lamellae accompanied by bifurcation and broadening of the original gill lamellae, (2) partial resorption of the tail fin, and (3) changes in eye position and snout morphology. Bifurcation of gill lamellae tips, but not the formation of supernumerary gills, is also observed in animals treated with RA alone. These results indicate that the molecular mechanism of RA and TH function through similar and perhaps competitive pathways.

Gene expression during amphibian limb regeneration

Limb regeneration in adult urodeles is an important phenomenon that poses fundamental questions both in biology and in medicine. In this review, we focus on recent advances in the characterization of the regeneration blastema at cellular and molecular levels and on the current understanding of the molecular basis of limb regeneration and its relationship to development. In particular, we discuss (i) the spatiotemporal distribution of genes and gene products in the mesenchyme and wound epidermis of the regenerating limb, (ii) how growth is controlled in the regeneration blastema, and (iii) molecules that are likely to be involved in patterning the regenerating limb such as homeobox genes and retinoids.

Retinoic acid changes the proximodistal developmental competence and affinity of distal cells in the developing chick limb bud

In the developing chick limb bud, retinoic acid (RA) has a striking effect on anteroposterior axis formation, resulting in a duplicated pattern of digits. There is no evidence, however, that RA affects proximodistal axis formation in the developing chick limb bud, although RA induces proximodistal duplication in regenerating amphibian limbs. We describe a series of investigations on the effect of RA on the proximodistal axis in the chick limb bud. A RA-containing bead applied to the anterior margin of the chick limb bud at stage 20 induces the anteroposterior duplication of autopodial structures at the wrist level. We found that the RA-treated tissue has the ability to form more proximal structures. When a tissue graft from the RA-treated anterior region was implanted into a stage 17 wing bud (in which the stylopod is developing in the progress zone), the graft produced a humerus, radius-ulna, and digits. When the graft was implanted into a stage 19 wing bud (in which the zeugopod is developing in the progress zone), a zeugopod and digits were formed. These results were associated with changes in the expression of Hox-A genes in the RA-treated grafts, whose domains were reorganized to be similar to those in host tissues 24 h after grafting. When a small graft of RA-treated tissue was implanted into the apex of a stage 19 wing bud, the cells were found in the zeugopod and autopod, whereas cells of control fragments were found only in the autopod region. In vitro, distal cells from different stage limb buds are known to segregate from each other. However, RA-treated stage 24 distal cells did not sort out from stage 20 distal cells and mixed homogeneously. These results suggest that RA induces distal cells to adopt "younger" properties which render them susceptible to forming more proximal patterns under the direction of host signals. The effects of RA on proximodistal patterns in developing chick limb buds appear to differ from its effects on proximodistal patterns in regenerating urodele limbs because RA can induce the proximodistal duplication in situ in the regenerating limbs.

Amphibian limb regeneration: rebuilding a complex structure

The ability to regenerate complex structures is widespread in metazoan phylogeny, but among vertebrates the urodele amphibians are exceptional. Adult urodeles can regenerate their limbs by local formation of a mesenchymal growth zone or blastema. The generation of blastemal cells depends not only on the local extracellular environment after amputation or wounding but also on the ability to reenter the cell cycle from the differentiated state. The blastema replaces structures appropriate to its proximodistal position. Axial identity is probably encoded as a graded property that controls cellular growth and movement through local cell interactions. The molecular basis is not understood, but proximodistal identity in newt blastemal cells may be respecified by signaling through a retinoic acid receptor isoform. The possibility of inducing a blastema on a mammalian limb cannot be discounted, although the molecular constraints are becoming clearer as we understand more about the mechanisms of urodele regeneration.

Endogenous retinoids in the zebrafish embryo and adult

Retinoic acid and its isoforms are considered to be endogenous compounds which regulate embryonic development. In the work reported here we have determined which retinoids are present in zebrafish embryos and how their levels change throughout development and into adulthood. All-trans-RA is present and its level does not change significantly during embryogenesis. We failed to detect other retinoic acid isomers such as 9-cis-RA and 4-oxo-RA, but we did observe a rapid rise in the level of didehydroretinol after gastrulation. The most striking result is that the zebrafish embryo, like Xenopus and tunicates, contains a vast excess of t-retinal whereas the embryos of higher vertebrates have an excess of t-retinol. However, as the zebrafish grows, the levels of t-retinol rise so that by adulthood t-retinol and t-retinal concentrations are more equivalent, indicating a changing pattern of retinoid metabolism with growth. To examine the significance of the use of t-retinal as a precursor of t-RA we treated embryos with disulphiram, an inhibitor of retinaldehyde dehydrogenase. This resulted in embryos with an undulating notochord and correspondingly abnormal somites and ventral floor plate. In contrast to this effect, 4-methylpyrazole, which inhibits alcohol dehydrogenases, had no effect on development. This effect of disulphiram suggests that t-RA may be involved in the establishment of the anteroposterior axis of the embryo.

Distribution of cellular retinoic acid-binding proteins I and II in the chick embryo and their relationship to teratogenesis

The distribution of cellular retinoic acid-binding proteins I and II (CRABP I and II) during the first 6 days of chick development has been investigated using immunoblotting. Since retinoic acid (RA) is teratogenic to some parts of the embryo, stimulatory to other parts, and has no effect on others it may be that the distribution of cytoplasmic proteins such as CRABP I and II plays some role in this differential activity. Neither protein is expressed in the day 2 embryo, but from day 3 onwards both proteins are expressed and CRABP I is in considerable excess over CRABP II. Within the day 4 embryo there is some significant variation in the distribution according to tissue type. Neural tissues, neural crest derivatives, and limb buds most strongly express CRABP I whilst other tissues contain only moderate levels, and heart and epidermis do not express CRABP I at all. CRABP II has a widespread distribution, although at a lower level than CRABP I, with the exception of somites and ectoderm which do not express it at all. In the limb buds, there is a significant variation in CRABP I levels across the anteroposterior axis which suggests that these two CRABPs may have different functions during development. The relationship of these distributions in the embryo to the role of endogenous RA and the teratogenic effects of RA is discussed.

Pattern duplication by retinoic acid treatment in the regenerating limbs of Korean salamander larvae, Hynobius leechii, correlates well with the extent of dedifferentiation

In the regenerating limbs of Korean salamanders, Hynobius leechii, retinoic acid (RA) induces duplication of skeletal structures in the proximodistal (PD) axis and often in the transverse axes. In the present study, the stage-dependent effects of RA for the duplication of limb skeletal structures at two amputation levels, the distal stylopodium and the distal zeugopodium, were studied using larval limbs of Korean salamanders. The results showed that the mean level of proximalization (MLP) by RA treatment increased during the stages of dedifferentiation and early bud formation while the MLP declined thereafter in both amputation levels. The decline of the MLP at the later stages of regeneration was due to the high frequency of hypomorphic regeneration or blocked regeneration. When the effects of RA treatment at two amputation levels were compared, the overall trends were similar but the actual timing was delayed for 2-4 days in the proximal level of amputation. Furthermore, the peak level of proximalization was achieved earlier and the peak level remained longer in the distal stylopodial level of amputation compared to the distal zeugopodial level of amputation. Since the histological observations revealed that the dedifferentiation period was also extended up to 2-4 days in the proximal level of amputation, the acid phosphatase activity during the course of regeneration was measured to look for a quantitative relationship between the enzyme activity and the states of dedifferentiation. The results show that the level and the duration of acid phosphatase activity in the upper arm regenerates are both higher and longer than those in the lower arm regenerates. Furthermore, RA treatment caused an increase in acid phosphatase activity. Thus our results suggest that the state of dedifferentiation might be closely linked to the extent of proximalization of regenerating limbs by RA treatment.

The effect of vitamin A (retinoids) on pattern formation implies a uniformity of developmental mechanisms throughout the animal kingdom

Retinoids are low molecular weight, lipophilic derivatives of vitamin A which have a profound effect upon the development of a diverse array of animals. Here, I review these effects on Invertebrates: a colonial hydroid, a colonial ascidian, and Vertebrates: the regenerating amphibian limb, the developing chick limb bud, the regenerating amphibian tail, the anteroposterior axis of the early embryo, the developing chick embryo skin. There is a striking uniformity of effect of retinoids on pattern formation when applied to these diverse organisms. The majority react by being posteriorized in their development, although additional effects can also be seen. Several hypotheses which can explain these results are discussed along with the deduction that they lead to: retinoids may be components of a universal developmental mechanism or they may simply act in a similar way to alter a universal developmental mechanism. In either case the experimental analysis of retinoid effects on development has important implications for the evolution of developmental mechanisms.

Test of a model for the effects of retinoic acid on urodele limb regeneration

Previous studies have shown that in axolotls (Ambystoma mexicanum), retinoic acid (RA) treatment evokes pattern completion in limb regenerates derived from anterior and dorsal half zeugopodia (lower arms and legs), but causes regenerative failure in posterior and ventral half zeugopodia. Pattern completion in anterior and dorsal half limbs may be explained by postulating that intercalary regeneration occurs in the antero-posterior (AP) and dorsoventral (DV) axes between blastema cells that are posteriorized (anterior half limb) or ventralized (dorsal half limb) by RA, and circumferential anterodorsal cells that remain unaffected by RA and thus maintain their original positional identities. The contrasting regenerative failure of RA-treated posterior and ventral half zeugopodia may likewise be explained by postulating that all the blastema cells in the posterior half are posteriorized, and all the cells in the ventral half are ventralized by RA, thus eliminating differentials in transverse positional identity essential for blastema formation and outgrowth. To test these postulates we grafted blastemas derived from limbs halved in the AP and DV axes of control and RA-treated animals to untreated whole limb stumps and analyzed the patterns of supernumerary (SN) regeneration. The site or location of SN formation will demonstrate (1) whether RA has posteriorized and ventralized the positional identity of the blastema cells and (2) if blastema cells in the periphery of the anterodorsal quadrant of the limb are resistant to these RA-induced changes in positional identity.(ABSTRACT TRUNCATED AT 250 WORDS)

Delta retinoic acid receptor isoform delta 1 is distinguished by its exceptional N-terminal sequence and abundance in the limb regeneration blastema

In amphibian limb regeneration memory for position in the proximal-distal axis can be respecified by retinoic acid. The favoured candidates to mediate this effect are the retinoic acid receptors (RARs) and of the RARs identified in the regeneration blastema, the delta receptor is the most abundant. The presence in blastemal mesenchyme of at least two delta receptor isoforms, delta 1 and delta 2, alternatively spliced at the A-B junction, was demonstrated in expression studies and by PCR cloning. The delta 1 receptor is abundant in regenerative structures such as the limb and tail, whereas the delta 2 and alpha receptors show a more uniform pattern of expression across adult newt tissues. Full-length cloning of the delta 1 receptor established the presence of an unusually long open reading frame and N-terminal sequence that appears unique among vertebrate retinoic acid receptors. Transient transfection of expression constructs into COS cells followed by Western blotting confirmed the existence of at least three potential initiation sites for delta 1 translation. The possibility that delta 1 RAR expression may specify positional memory directly was tested in RNase protection experiments. delta 1 receptor message is increased on amputation, but does not exhibit a pronounced differential distribution along the proximal-distal axis in normal and regenerating limbs, nor does it show a persistent alteration in expression levels following a dose of retinoic acid sufficient to respecify position. The possibility that the morphogenetic effects of RA may be mediated through receptor interactions is raised by the finding that single mesenchymal blastemal cells in culture can express multiple RAR subtypes (delta 1 and alpha) and isoforms (delta 1 and delta 2).

Differential and stage dependent effects of retinoic acid on chondrogenesis and synthesis of extracellular matrix macromolecules in chick craniofacial mesenchyme in vitro

Retinoic acid (RA) is well known to be a potent teratogen and induces a variety of facial defects in vivo, but at concentration levels lower than those that cause facial defects, RA seems to play an important role in normal facial development. In a previous study, we demonstrated the ability of RA to stimulate chondrogenesis in vitro in HH stage 23/24 chick mandibular (MND) but not frontonasal (FNP) mesenchyme cultured in a serum-free medium. The present study furthers these results by examining the effects of RA on chondrogenesis of chick facial mesenchyme at earlier embryonic stages and the effects on cell proliferation and synthesis of specific extracellular matrix macromolecules at stage 23/24. MND and FNP cells were cultured as micromasses for 4 days in defined media. As described previously, chondrogenesis in stage 23/24 MND cells was significantly enhanced by concentrations of RA of 0.1-1 ng/ml; however, at all earlier stages examined (18 to 22) RA at these concentrations had no significant effect. Higher concentrations of the retinoid inhibited chondrogenesis in MND cultures from all stages tested. Cells of the FNP from all stages displayed no significant change in chondrogenesis below 1 ng/ml RA and a dose dependent inhibition at higher concentrations. Thus RA's promotional effects in the face are not only tissue specific (MND), but also stage-dependent (HH 23/24). The specific effects of RA on matrix production and cell proliferation of stage 23/24 MND and FNP cells was examined by analysis of 35S sulfate, 3H thymidine and 3H proline incorporation. Analysis of 35S sulfate incorporation into sulfated proteoglycans confirmed that concentrations of RA of 0.1-1 ng/ml stimulated cartilage matrix production in MND but not FNP cultures. Above this level of RA, 35S sulfate incorporation was reduced in both. Likewise, 3H proline incorporation into collagenous protein, and to a lesser extent non-collagenous proteins, was stimulated by low levels of RA in MND, but not FNP cultures. Higher concentrations of the retinoid in either MND or FNP cultures did not lower collagen production, undoubtedly due to stimulation of non-chondrogenic cells within the population. This indicates that levels of RA as high as 100 ng/ml cause phenotypic change rather than cell death. This last point is corroborated by the analysis of 3H thymidine uptake in the cultures which was only transiently modified in most. The data indicate that cell proliferation occurred even in the presence of high RA levels.

On the generation of form by the continuous interactions between cells and their extracellular matrix

The central issue of this essay is the problem of how multicellular organisms develop and maintain the complex architecture and intricate shape of tissues and organs. The concepts pattern formation, morphogenesis and differentiation are defined and discussed suggesting a distinction between processes that underlie uniformity (e.g. basic body plans) and those underlying inter- and intra-species variation. The initial stage of limb bone development--the formation of the mesenchymal condensation--is described in detail. On the basis of these data and many additional example from other developmental systems, the central role of continuous cell-ECM interactions in the generation of form is deduced. Evidence is provided as to the leading role of the mesenchymal-fibroblast-like cells in sculpturing tissue and organ architecture. It is proposed that a group of cells within their ECM, rather than the single cell, is the functional unit relevant to the generation of form. The continuous cell-ECM interactions lead to the generation of form not by a detailed obligate pathway, but rather by a process of 'selective stabilization' (Kirschner & Mitchison, 1986), i.e. a gradual organization into more stable structures, where existing structural configuration serve to increase the likelihood of certain configurations and reduce that of others. Data are quoted to support the notion that even cell division does not erase all the structural information imprinted in the cell. The role of the metazoan genome in morphogenesis is discussed in the light of the process of selective stabilization.

Mouse limb bud cells respond to retinoic acid in vitro with reduced growth

Retinoic acid (RA) has dramatic effects on the pattern of developing and regenerating vertebrate limbs. These effects are considered to result from RA-induced changes in the positional identity of limb cells, and involve the formation of extra structures. Whether the growth required to form the supernumerary parts of the pattern is a primary effect of RA treatment or a secondary effect that follows after a change in positional identity is not at present known. In this paper we have investigated the effects of RA treatment on the growth of cells from anterior and posterior halves of mouse limb buds in vitro. We observed that under our culture conditions, limb bud cells treated with 1 nM to 1 microM RA (0.3 ng/ml to 300 ng/ml) continue to grow but do so at a significantly slower rate than control cultures. There is a maximum inhibition of growth (50% of controls) between 10 nM and 100 nM RA, which corresponds to the measured range of concentrations of RA in vivo. Our observation of a significant decrease in growth rate over a wide range of RA concentrations is consistent with comparable reports of growth inhibition for a large number of other cell types in vitro as well as with the observation that exogenous RA inhibits blastemal growth in amphibians during the period of exposure to RA. We propose that the effects of RA on growth, either enhancement in vivo or reduction in vitro, can be seen as consequences of the ability of RA to alter positional identity.(ABSTRACT TRUNCATED AT 250 WORDS)

Retinoic acid enhances monoclonal antibody WE3 reactivity in the regenerate epithelium of the adult newt

Monoclonal antibody (mAb) WE3 recognizes an antigen that is developmentally expressed in the wound epithelium during adult newt limb regeneration. Experiments were designed to determine whether retinoic acid (RA), dissolved in dimethyl sulfoxide (DMSO) and administered by intraperitoneal injection, would enhance the temporal appearance of the WE3 antigen. RA given on days 1 or 4 after amputation, when the WE3 antigen is not yet detectable, resulted in moderate reactivity to mAb 2 days after injection and strong reactivity throughout the wound epithelium 4 days after injection. DMSO alone had no enhancing effect. RA also caused limb skin epidermis to exhibit reactivity to mAb WE3, initially near the amputation level, but then also more proximally. By 4 and 6 days after RA injection, epidermis of the flank, eye lid, and unamputated hind limbs also became strongly reactive to mAb WE3. Outer layers of skin epidermis were shed, resulting in an epidermis only one or two cells thick. Epidermis of newts given DMSO alone remained non-reactive to mAb WE3. When RA was given on days 7 and 10 after amputation, when a low level of mAb WE3 reactivity is already present in the wound epithelium, a considerable enhancement of mAb WE3 reactivity occurred through the next few days. No such enhancement was seen with DMSO alone. RA also greatly increased mAb WE3 reactivity in the wound epithelium of denervated limbs, in which case the wound epithelial reactivity to mAb WE3 is normally low. Retinol palmitate also increased mAb WE3 reactivity. The results raise the possibility that the WE3 antigen is a component of most if not all retinoid target tissues in newts.