Regenerative Biology and Medicine

Index

Part I Regenerative Biology

Chapter 1 An Overview of Regenerative Biology

Chapter 2 Repair of Skin by Fibrosis

Chapter 3 Regeneration of Epidermal Structures

Chapter 4 Regeneration of Neural Tissues

Chapter 5 Regeneration of Digestive, Respiratory and Urinary Tissues

Chapter 6 Regeneration of Musculoskeletal Tissues

Chapter 7 Regeneration of Cardiac Muscle and Hematopoietic Tissues

Chapter 8 Regeneration of Appendages

Chapter 1

An Overview of Regenerative Biology

I Introduction

We live in a dangerous world. Each day, our bodies are subjected to injuries ranging from small cuts and bruises that are easily repaired, to more serious traumas and diseases that can cause extensive structural and functional loss. Even in the absence of injury, many cell types normally have a limited life span and must be continually replaced to maintain tissue integrity. The evolutionary process has provided vertebrate animals with two mechanisms for cell replacement and tissue repair. The first is regeneration, a homeostatic process that maintains or restores the original architecture and function of a tissue. The second is fibrosis, in which damaged tissues are invaded by fibroblasts that do not restore the original tissue architecture, but instead patch the damaged area with collagenous scar tissue.

Modern medicine is able to replace body parts through cell, tissue and organ transplantation, but these technologies have drawbacks that can exact a high price on patients in terms of side effects and morbidity. What we really desire are ways to inhibit fibrosis and evoke regeneration in situ where it fails to occur naturally. Achieving this goal requires that we obtain a much deeper understanding of the cellular and molecular mechanisms of regeneration, and how these mechanisms differ from those that lead to fibrosis, than we currently possess. When sufficiently complete, this understanding will drive a revolutionary transformation of medicine.

II A Brief History of Regenerative Biology

Historically, regenerative biology has its roots in observations on the healing of wounds and of appendage regeneration (see Goss, 1991). Severe injuries such as penetrating wounds, multiple fractures, spinal cord compression, or damage to the eyes have occurred throughout the history of the human species, due to conflict and accidents. Testimony to the importance of wounds is found on the walls of Paleolithic caves, which bear the imprints of hands with amputated fingers. Early humans would have noted the kinds of wounds that were survivable and those that were not, and attempted rudimentary means to facilitate their repair. They were also likely to have been aware of the phenomenon of regeneration from observations on food animals such as crustaceans and deer, which can regenerate their legs and antlers, respectively. The fact that trees renew their leaves and seeds on a cyclical basis, and re-grow severed limbs in another direction would also have contributed to an awareness of, and curiosity about, regeneration.

Treatments to aid wound repair preceded by many millennia any knowledge of how the repair was actually accomplished. Methods to facilitate wound healing, including surgical interventions, were central to the medicine of ancient Sumerian, Egyptian, Chinese, Indian, and Inca civilizations (Majno, 1975; Brown, 1992; Falabella, 1998). The cleansing and debridement of wounds was a common practice in ancient cultures and many different vegetable and mineral concoctions were applied topically to treat wounds. The ancient Chinese and Egyptians used honey and wine as antiseptics and the Chinese have used a bread mold to treat minor burns for over 2000 years (Majno, 1975; Fu et al., 2001). Trephination to relieve intracranial pressure was performed by the Incas in the treatment of head wounds (Majno, 1975). A thousand years ago, as shown in Fig. 1.1, the Indian physician, Sushruta, described the use of autogeneic skin transplants to reconstruct severed noses and ears (Majno, 1975).

Figure 1.1 Autotransplant of forehead skin flap to reconstruct the nose, described by Sushruta ~1000 AD. The flap was cut to conform to the outline of a nose, peeled down and twisted 180° at the stalk (arrows). It was then positioned over the nasal area with the dermal side down and held in place with wooden rods until healed.

The Greek and Roman physicians Hippocrates (460–370 BC), Celsus (25 BC–50 AD), and Galen (130–201 AD) contributed greatly to the development of medicine, including the treatment of wounds (Brown, 1992). Part of Galen’s career was spent caring for injured gladiators, which gave him extensive experience in treating wounds to many parts of the body. Galen compiled virtually all that was known about anatomy, physiology, and medical treatment at the time (much of it erroneous) into at least 35 volumes. When the Western Roman Empire collapsed near the end of the 5th century AD, Galen’s texts were translated into Arabic, the language of medicine in the Eastern Roman Empire, and later from Arabic into Latin. These texts were the dominant guides for medical practice until the end of the Middle Ages.

A creative blossoming of biology and medicine took place from the 14th to the 17th centuries, based on more accurate and detailed observations of form and function. Major contributions were the anatomical descriptions of embryonic and adult structure made by Da Vinci (1452–1519), Vesalius (1514–1564), Paracelsus (1493–1541), Fallopius (1523–1562), and Fabricius (1537–1614). Descartes (1596–1650) and Borelli (1608–1679) wrote important texts on physiology, and Harvey (1578–1657) produced his famous treatises on the circulation of the blood and the reproduction of animals. The surgeon Guy de Chruliac (1300–1370) published Wounds and Fractures in 1363, a book that detailed many kinds of wounds and how to treat them. Wilhelm Fabry (1560–1634) described nearly 70 topically applied formulations for treatment of wounds, many of which have been re-examined and found to have true therapeutic value (Kirkpatrick et al., 1996).

The technological and conceptual revolutions that brought forth modern science took place in the 17th century. The technological revolution was the beginning of experimentation in chemistry and the development of the telescope and microscope, with new results and ideas being disseminated by the printing press, which had been invented two centuries earlier. The conceptual revolution began with a challenge to the Aristotelian view of the motion of celestial bodies, which were considered to move in perfectly circular orbits on invisible spheres (Van Doren, 1991). Kepler (1571–1630) showed that the planets moved in elliptical orbits and that the time required for each orbit was proportional to their distance from the sun. Galileo (1564–1630) proved mathematically what Copernicus had concluded a century earlier: that the earth orbits around the sun, not the reverse, which was the doctrinaire religious view of the time. Newton (1642–1727) related forces and motion in his three general laws of motion, now generally referred to as Newtonian mechanics. Descartes (1596–1650) generalized these physical and mathematical relationships into the philosophy of mechanism—that everything in the material universe, including organisms, could be treated as machines that behave according to strict mathematical laws of physics. In 1687, Newton published his Mathematical Principles of Natural Philosophy, in which he formalized four rules of reason to guide scientific investigation.

1. The best explanation for a natural phenomenon is the simplest one that is true and sufficient (William of Ockham’s razor).

2. The same causes should be assigned to the same natural effects (inductive reasoning).

3. Those qualities common to all objects or phenomena within our realm of experience are assumed to be universal (deductive reasoning).

4. The most important rule of all: accept no explanations not directly supported by observation or experiment.

These rules provided a powerful way of answering questions about how things work, changing our view of the world forever.

In biology, the invention of the compound microscope in the early 1600s made it possible to view biological structure in greater detail than ever before and thus better understand the nature of biological phenomena. This technological leap forward led to development of the science of microscopic anatomy in the 18th century. The comparative anatomist John Hunter (1728–1793) studied healing skin wounds and discovered granulation tissue and the transitional role it played in scar tissue formation (Brown, 1992). The dogma of preformation, which held that the mechanism of ontogeny was the growth of a tiny pre-delineated adult within the egg, was toppled by the studies of C.F. Wolff on chick embryos, which showed that the embryo developed by a series of epigenetic steps that built form out of amorphous substance.

The 19th century saw numerous medical and surgical advances that improved the prospects of recovery from serious injury and disease (Allan, 1977). The development of ether anesthesia made pain-free surgery possible and thus increased the types of surgical interventions that could be made in the human body to treat different conditions. Surgery, however, also increased the possibility of death by systemic bacterial infection. Joseph Lister, a student of Pasteur, introduced the use of dressings soaked in carbolic acid and strict hygienic measures in hospitals to combat sepsis after surgical operations (Brown, 1992). Nineteenth century surgeons revived the techniques of skin grafting described by Sushruta many centuries earlier (Majno, 1975).

The most significant development of the 19th century for the future of biology and medicine, however, was the rise of materialistic biology, which was an outcome of Descarte’s philosophy of mechanism. Until the mid-18th century, it was thought that the properties of life forms were due to an immaterial vital force, not the material chemical and physical forces that regulated the properties of inanimate objects (Coleman, 1977). However, experiments begun by Lavoisier (1743–1794) in the second half of the 18th century and carried on by others into the 19th century showed that life depended on chemical reactions that were reproducible in the laboratory. With the formulation of the cell theory by Schleiden and Schwann in 1838–1839 and the later microscopic observations of Virchow, Remak and others, it became clear that cells are the fundamental units that carry out the chemical reactions of life and that new cells are created by the division of existing cells, thus explaining growth and reproduction.

The 20th century saw an unparalleled explosion of biological and medical knowledge. Major advances were the discovery and production of antibiotics, the development of molecular replacement therapies for diseases such as diabetes, an understanding of the immune system that revealed the antigenic differences between self and non-self, and the development of highly sophisticated imaging and surgical technologies. These advances, coupled with advances in engineering and materials science and the development of immunosuppressive drugs, have given us the ability to transfuse blood and replace damaged and dysfunctional tissues and organs through tissue and organ transplants and implants of bionic devices.

Without question, however, the most fundamental and far-reaching event of 20th century biology was the discovery, in mid-century, that DNA is the hereditary material (Avery et al., 1944; Hershey and Chase, 1952) and that it has a helical structure consisting of two deoxyribose sugar–phosphate backbones held together by complementary base pairs, adenine to thymine and guanine to cytosine (Watson and Crick, 1953). The enormous power of this structure to explain how the genetic material is replicated and mutates, and how information for protein structure is encoded in it and expressed, led to exponential advances in our knowledge of cell, developmental, and evolutionary biology.

Regeneration became a focus of systematic scientific investigation only in the 18th century, though it figured prominently in the ancient Greek myths of the Hydra and Prometheus recounted by Homer and Hesiod (Dinsmore, 1998) and in the medieval accounts of the regeneration of severed human arms and legs (Goss, 1991). Abraham Trembly performed detailed experiments on the regeneration of hydra that made a deep impression on the biologists of the time (Lenhoff and Lenhoff, 1991), while Reaumer and Spallanzani reported observations on the regeneration of limbs in crustaceans and newts, respectively (Skinner and Cook, 1991; Dinsmore, 1991). Studies on limb development and regeneration in the latter part of the 19th and the early part of the 20th centuries made major contributions to the understanding of development. Prior to the 20th century, limb regeneration in amphibians and crustaceans was explained in the context of preformation. The limbs of these animals were presumed to contain multiple copies of preformed appendages, the growth of which was stimulated by amputation. In the early 20th century, regeneration was recognized as a regulative process that restored the whole from the remaining part. A major research aim of Thomas Hunt Morgan (1866–1945), before he turned to genetics, was to explain regeneration in terms of chemical and physical principles (Morgan, 1901). Over a century later, we are still in the process of formulating this explanation and simultaneously advancing the age-old dream of being able to regenerate tissues and organs that do not regenerate spontaneously. Now, in the second decade of the 21st century, rapid advances in multidisciplinary biological science, chemistry and informatics have put that goal within reach.

III Regeneration Occurs at All Levels of Biological Organization

All organisms regenerate, though the degree of regenerative ability varies among species and with level of biological organization within the individual organism (Goss, 1969). Some invertebrates, such as ascidians, coelenterates, flatworms and annelid worms, can regenerate whole organisms from fragments of the body. The roots, stems, and leaves of plants have extensive regenerative capacity, and entire plants can be grown from single cells or small cuttings (Carlson, 2007; Birnbaum and Sanzhez-Alvarado, 2008, for reviews). For example, a single carrot cell can regenerate a whole carrot (Steward et al., 1964). Compared to these life forms, the regenerative capacity of vertebrate tissues, including humans, is limited, but no less vital.

All molecules, cells and tissues experience turnover, and thus must be continually regenerated to maintain the integrity of the organism. On the molecular level, regeneration is ubiquitous. All cells can adjust the balance of protein synthesis and degradation in response to biochemical or mechanical load. For example, cardiomyocytes replace most of their molecules over the course of two weeks and adjust their rate of protein synthesis upwards under a sustained increase in blood pressure, becoming hypertrophied (Gevers, 1984). At the level of the single-cell free-living unicellular protozoans such as the amoeba, Tetrahymena, Stentor and Acetabularia can regenerate themselves after removal of large cytoplasmic fragments, as long as nuclear material is present in the remaining fragment (Goss, 1969). As little as 1/80 of an amoeba is capable of reconstituting a complete amoeba (Vorontsova and Liosner, 1960).

Regeneration is most complex at the tissue level and above. On the one hand, it is thought that some differentiated cells, such as neurons of the human cerebral cortex, are never renewed. On the other hand, we know that the cells of many tissues turn over and that the rate of turnover varies with the tissue. Rapid turnover and replacement is characteristic of epithelial cells of the skin and linings of the digestive, respiratory and urogenital tracts, as well as the myeloid and lymphoid cells of the hematopoietic system. The liver, on the other hand, turns over once a year, and the skeletal system once a decade. This type of regeneration, which involves cell replacement within an intact tissue, can be termed maintenance regeneration. By contrast, massive cell loss caused by injury or disease (i.e., tissue loss) elicits a much more intense local response to restore the missing tissue. This response can be termed injury-induced regeneration. It is important to remember that both forms of regeneration are homeostatic mechanisms (see Galliot and Ghila, 2010, for review).

Regeneration and asexual reproduction are intimately related in some invertebrates (Birnbaum and Sanzhez-Alvarado, 2008). Coelenterates and flatworms appear to use the same mechanisms for regeneration as they do for reproduction. Thus, planaria and hydra can undergo reproduction by fission, with two new individuals being reconstituted from the fission products. Although vertebrates cannot reproduce by fission, regeneration is nevertheless relevant to their reproduction, since it allows them to survive to reproductive age and thereby propagate the species. The relationship between reproduction and regeneration in vertebrates has been exquisitely summed up in two quotes. The great cell biologist E.B. Wilson wrote in the second edition of his classic book The Cell in Development and Heredity (1925) that …life is a continuum, a never-ending stream of protoplasm in the form of cells, maintained by assimilation, growth and division. The individual is but a passing eddy in the flow who vanishes and leaves no trace, while the general stream of life goes forward. Forty-three years later, in his book Principles of Regeneration (1969) the 20th century master of regenerative biology Richard J. Goss observed that If there were no regeneration there could be no life. If everything regenerated there would be no death. All organisms exist between these two extremes. Other things being equal, they tend toward the latter end of the spectrum, never quite achieving immortality because this would be incompatible with reproduction. In other words, as individuals we use regeneration to locally reverse the second law of thermodynamics for a short time, a struggle we ultimately lose, but one that we win as a species over a longer span of time through reproduction.

IV Mechanisms of Vertebrate Regeneration

We ask four basic questions when exploring how regeneration occurs. The questions are: (1) what is the origin (cell type and anatomical location) of the cells that perform regeneration; (2) what are the characteristics of the cells; (3) what kinds of activities do the cells engage in to accomplish regeneration; and (4) what are the factors that regulate these activities? Only partial answers to these questions have been obtained, but they have revealed four potential mechanisms of tissue regeneration in vertebrates. These are cellular re-growth, regeneration from pre-existing differentiated parent cells, transdifferentiation, and the activation of adult stem cells (Fig. 1.2).

Figure 1.2 The four mechanisms of regeneration. (A) Cellular re-growth. Axon re-growth to muscle target (M) after transecting the axon (Ax) of a motor neuron (MN). Arrow indicates direction of re-growth. (B) Regeneration from parent differentiated cells. (1) Compensatory hyperplasia (CH), in which a differentiated cell proliferates (P) to expand the population. (2) Dedifferentiation/redifferentiation (D/R), in which the cell first reverts to an undifferentiated state (dedifferentiates, D), proliferates (P), and redifferentiates (R) into the parent cell type. (3) Epithelial (E) to mesenchymal (M) and mesenchymal to epithelial transformation, allowing epithelial cells to migrate and proliferate as mesenchymal cells before redifferentiating into epithelial cells again. (C) Transdifferentiation, which may occur directly by initiation of a new pattern of gene expression simultaneously with repression of the old one, or indirectly by dedifferentiation (D) to a more primitive state, then transdifferentiating (T) to a new cell type. (D) Activation of adult stem cells. Adult stem cells (ASC) typically divide to produce a lineage-committed daughter (LC) and another stem cell, a process called self-renewal (SR). The lineage committed cell proliferates (P) to form precursor cells (PC) that differentiate into the terminal cell type. The diagram depicts a unipotent ASC, but many ASCs are multipotent.

A Cellular Re-Growth

Regeneration by cellular re-growth involves the loss of a cytoplasmic part and its subsequent restoration by re-growth from the remainder of the cell. This occurs in vertebrate tissues in which elongated cells or cell processes are bundled together to form the tissue. For example, within skeletal muscle, myofibers regenerate across small gaps after transection by re-sealing the membranes of their cut ends followed by membrane extension and re-fusion. The most spectacular regeneration by cell re-growth in vertebrates occurs in the transected axons of peripheral nerves, where the part of the transected axon that is continuous with the nerve cell body grows back to its target tissue (Yannas, 2001).

B Regeneration from Pre-Existing Parent Cells

This is a versatile form of regeneration that can involve three different mechanisms: compensatory hyperplasia; epithelial to mesenchymal transformation (EMT) and the reverse, mesenchymal to epithelial transformation (MET); and dedifferentiation/redifferentiation.

1 Compensatory Hyperplasia

Compensatory hyperplasia is the proliferation of cells while they maintain their differentiated structure and function. The classic example is liver regeneration (Michalopoulos and DeFrances, 1997; Trembly and Steer, 1998). After partial hepatectomy the hepatocytes of the liver, as well as its non-parenchymal cell types (Kupffer, Ito, bile duct epithelial, and fenestrated epithelial cells), divide while continuing to perform their functions of glucose regulation, synthesis of blood proteins, secretion of bile, and drug metabolism, until the original mass of the liver is restored. Other cell types regenerated by this mechanism are β-cells of the pancreatic islets, cardiomyocytes of the heart, and endothelial cells of regenerating blood vessels.

2 Epithelial to Mesenchymal and Mesenchymal to Epithelial Transformation (EMT/MET)

Epithelial and mesenchymal cells can be transformed into one another, a phenomenon called epithelial to mesenchymal transformation (EMT) and mesenchymal to epithelial transformation (MET). These transformations are prominent mechanisms of development and of pathologies such as fibrosis and cancer (Thiery et al., 2009, for review). EMT allows epithelial cells to migrate, whereas MET restricts migration by establishing an epithelium. Developmental examples of EMT and MET are the transformation of epithelial cells into mesenchymal cells during the morphogenetic movements of gastrulation and the formation of kidney tubules from metanephric mesenchyme, respectively (Gilbert, 2010, for review). EMT in cancer is responsible for metastasis. EMT/MET is also essential for epithelial wound repair, where epithelial cells assume a state in which they can migrate across a wound surface while maintaining loose contacts, then re-form into a tight epithelium. In urodele (salamander) spinal cord regeneration, EMT of ependymal epithelial cells allows a gap in the cord to be bridged by mesenchymal cells. MET then restores the epithelium, the cells of which form end-feet channels for axon re-growth (Chernoff et al., 2003).

3 Dedifferentiation/Redifferentiation (D/R)

Dedifferentiation is a reverse epigenetic reprogramming resulting in the loss of phenotypic specialization and reversion of cells to a less differentiated state that allows them to proliferate and redifferentiate into their parent cell type. Dedifferentiation/redifferentiation is a relatively common mechanism of regeneration in lower vertebrates. Teleost fish regenerate fins and barbels by D/R and certain species of lizards can regenerate tails by this mechanism. The divas of D/R in the vertebrate world, however, are the larval and adult urodele amphibians. These animals use dedifferentiation to regenerate complex structures that adult frogs, birds, reptiles and mammals cannot, including limbs, tails, jaws, lens, spinal cord, neural retina, and intestine (Brockes and Kumar, 2008, for review).

C Transdifferentiation

Transdifferentiation is the conversion of one cell type to another cell type, usually with an intervening dedifferentiation step that confers plasticity on the cell. This plasticity allows local injury factors to direct proliferation and differentiation of the dedifferentiated cells into the cell type that needs to be regenerated. Good examples of this type of regeneration are the regeneration of lens cells from pigmented epithelial cells of the dorsal iris in newts, and the regeneration of cartilage from fibroblasts during urodele limb regeneration. Theoretically, it is also possible that instead of going through a dedifferentiation step, the activity of the set of genes specifying the old cell type is suppressed simultaneously with the activation of suites of genes characteristic of the new cell type.

D Activation of Adult Stem Cells (ASCs)

ASCs are sequestered as relatively rare undifferentiated cells within regeneration-competent tissues as they differentiate during embryonic development, where they reside in environmental microniches that maintain their stem cell character. When activated by other niche factors, they can divide symmetrically to give rise to two stem cells or two lineage-committed cells, or asymmetrically to give rise to a lineage-committed cell and another stem cell (Morrison and Spradling, 2008). The production of another stem cell by a stem cell division is termed self-renewal. There are two major classes of ASCs, epithelial and mesenchymal. ASCs have varying degrees of developmental potential, depending on the tissue they serve. Some are multipotential and give rise to several cell types, as in the hematopoietic system, others such as epidermal stem cells appear to be unipotent and give rise only to keratinocytes. Activation of ASCs may take place differently in maintenance vs. injury-induced regeneration. In maintenance regeneration, the stem cells undergo continual but slow division in response to environmental signals, feeding a constant stream of progeny into a new microenvironment that induces them to differentiate, but ramp up stem cell activation and proliferation in response to injury.

1 Epithelial Stem Cells

Epithelia make up 60% of the differentiated tissue types in the body (Slack, 2000) and are derived from the ectodermal and endodermal layers of the gastrula. Epithelial cells are anchored to one another and to a basement membrane by specialized cell junctions. The principal intracellular proteins that distinguish them are keratins. Epithelial cells constitute the epidermis of the skin, and line cavities of organ systems such as the digestive, respiratory, and urinary tracts, and central nervous system. The olfactory and optic nerves are derived from nasal epithelium and the retinal ganglion epithelium, respectively. The hepatocytes of the liver, the acinar cells of the pancreas, and their associated ductule and duct cells are arranged as epithelia. The cardiovascular system is lined with specialized epithelial cells called endothelial cells that are distinguished from epithelial cells by their heart and blood vessel-specific location, morphology and expression of vimentin rather than keratins, as well as specific cell surface antigens.

The stem cells that regenerate the epidermis of the skin and the epithelia of tubular organ systems are located in the basal layers of their respective tissues. The liver and pancreas regenerate by reproduction of parental cells after surgical loss of tissue, but their ductules harbor stem cells for regeneration after injury by other means. Endothelial cells are regenerated by stem cells residing in the bone marrow and the circulation, and by cells in the endothelium itself that have regenerative capacity.

2 Mesenchymal Stem Cells

The remainder of the tissue types of the body is derived from the mesodermal and ectomesodermal (neural crest) layers of the gastrula and neurula. These include the musculoskeletal system, cardiac muscle, smooth muscle, and loose connective tissues associated with these tissues, and the dermis of the skin. Mesenchymal stem cells (MSCs) were originally isolated from bone marrow cell cultures in vitro by Friedenstein et al. (1966) as an adherent population of fibroblast-like cells. These archetypal MSCs are capable of differentiating into osteoblasts, adipocytes, chondroblasts, and skeletal and smooth muscle cells (Pittenger et al., 1999). They co-exist with hematopoietic stem cells (HSCs), the other major population of mesodermally-derived stem cells in the marrow. The cell surface protein profile of marrow MSCs is distinguished from that of HSCs by the absence of the cell differentiation markers CD31, CD34 and CD45, but it has been difficult to define a group of MSC-specific cell surface proteins (Roche et al., 2006). Populations of MSCs that share similar but not identical features with bone marrow MSCs have been isolated from the periosteum and endosteum of bones, tooth pulp, adipose tissue, skeletal muscle, heart, and perivascular tissue (Liu et al., 2009 for review). There is evidence that most MSCs may actually be perivascular pericytes, which would explain why they seem to be found in so many tissues (Caplan, 2008).

E The Regeneration of Many Tissues Involves More than One Mechanism

Most tissues use one of these four mechanisms as their primary means of regeneration, but a substantial number can regenerate by more than one mechanism. Thus, red blood cells and intestinal epithelium regenerate solely by adult stem cells, whereas the liver and pancreas use compensatory hyperplasia to regenerate after surgical injury, but stem cells for chronic chemical injury. Urodele appendages regenerate primarily by dedifferentiation/redifferentiation, but adult stem cells, EMT/MET, cellular re-growth and transdifferentiation also contribute to the regenerative process.

V The Niche Concept in Tissue Repair

The cells of all tissues reside in microenvironments (niches) consisting of an extracellular matrix (ECM) and a variety of soluble molecules, some of which are made locally and others by tissues some distance away and circulating in the blood (Fig. 1.3). The composition and spatial organization of the niche determine the activities of the cells. Both the ECM and soluble molecules can act as signaling molecules (ligands) that bind to receptors embedded in cell plasma membranes or residing in the nucleus of the cells. Ligand binding triggers a conformational change in the receptor that sets off a chain of intracellular reactions resulting in changes in the configuration of the cytoskeleton or a change in the short-term or long-term pattern of gene transcription of the cell.

Figure 1.3 Niche factors regulating cell behavior. Two cells are shown interacting with self or with one another in autocrine (A), paracrine (P) and juxtacrine (J) fashion, via ligands (L) and receptors (R). The cells also receive long-distance paracrine signals via cell extensions (e.g., axons) and endocrine (E) signals via the blood that activate nuclear receptors (NR). In addition, the extracellular matrix (ECM) surrounding the cells or on which the cells rest contains multiple peptide sequences that signal through integrin receptors (IN).

Niche composition has been particularly well studied in wound repair by fibrosis (Barrientos et al., 2008) and regeneration by ASCs (Scadden, 2006; Knoblich, 2008; Morrison and Spradling, 2008; Voog and Jones, 2010, for reviews). Current evidence suggests that many niches direct tissue repair via the recapitulation of developmental programs (Voog and Jones, 2010), but this is not always the case. Anatomically, the niches of ASCs can be widely distributed throughout tissues (e.g., satellite cells of muscle), confined to restricted spaces (e.g., bulge of hair follicle, subventricular zone of the lateral ventricles), or scattered as facultative spaces (hematopoietic stem cells of the bone marrow).

A Niche Signaling Molecules

The signaling molecules of the niche can be classified as paracrine, autocrine, juxtacrine, and endocrine. Paracrine signals are secreted by cells and diffuse over short ranges to bind with receptors on neighboring cells. Signals can be propagated from cell to cell in this way. Many (but not all) paracrine signals can be grouped into four families on the basis of their structure. These are the fibroblast growth factor (FGF) family, the Hedgehog family, the Wnt family, and the TGF-β superfamily, consisting of the TGF-β family, the bone morphogenetic protein (BMP) family, the Nodal proteins, the Vg1 family, and several other proteins (Gilbert, 2010). An autocrine signal is one that binds to receptors on the surface of the cell that produces it. Juxtacrine signaling involves contact between cells, in which a ligand on one cell surface binds to a receptor on the other. Endocrine signals circulate in the blood and bind to nuclear receptors. Some paracrine signals, such as retinoic acid (RA), also bind to nuclear receptors (Deuster, 2008).

Short amino acid sequences of cell adhesive ECM components also act as signals by binding to cell surface receptors called integrins (Hynes, 2002). For example, maintenance of HSCs in a quiescent state relies on their adherence to osteoblasts of the bone marrow stroma through N-cadherin of adherens junctions and fibronectin-binding integrins (α4β1, α5β1) (Zhang et al., 2003). Blocking this adhesion inhibits hematopoiesis in long-term bone marrow cultures (Whetton and Graham, 1999). Mechanical properties of the ECM such as porosity, stiffness, elasticity, tension and compression play key roles in cell behavior as well (Ingber, 2006; Engler et al., 2006; Guilak et al., 2009). Whether or not cells interact with each other and the ECM in two dimensions (as in tissue culture monolayers) or three dimensions also influences cell behavior. Solid tissues, for example, behave much more normally if cultured in three-dimensional scaffolds. Figure 1.4 illustrates examples of stem cell regulation by niche factors.

Figure 1.4 Maintenance of (A) hematopoietic stem cell (HSC), and (B) epithelial stem cell (EpSC) niches. The HSC niche is maintained by paracrine signals between HSCs (double arrows) and from capillaries (Cp, single arrows) and by juxtacrine contact (bars) of HSCs with stromal cells such as osteoblasts (OB). Epithelial cells are maintained in their niche by juxtacrine contact (bars) between themselves and with their basement membrane (BM) and by paracrine signals between the EpSCs and capillaries.

Cells often need to signal one another at long range by juxtacrine or paracrine means. Such signaling is achieved by structural extensions of the signaling cell to the target cell, for example the extension of motor axons from spinal neuron cell bodies to muscles and the extension of filopodia from ingressing mesenchymal cells to the ectoderm in gastrulating sea urchin embryos (Rorth, 2003).

B Signal Transduction Pathways

Six main signaling pathways used by cells have been identified that fall into two groups based on receptor structure (Fig. 1.5). In the first group, ligands bind to monomeric receptors. This group includes the Notch, Wnt, and Hedgehog pathways. Notch is a juxtacrine pathway, whereas Wnt and Hedgehog are paracrine pathways. In the second group, ligands bind to dimeric receptors. This group includes the receptor tyrosine kinase (RTK), JAK-STAT, and TGF-β pathways, all of which are paracrine. In general, these pathways use variations on a common theme to signal to the interior of the cell. The change in receptor conformation by a ligand gives the cytoplasmic domain of the receptor kinase activity that initiates a chain of phosphorylation reactions by other kinases using ATP that alters the cytoskeleton or activates/suppresses transcription factors (Gilbert, 2010). Many kinases also regulate intracellular signaling by endocytosis of receptors or ligands (Polo and Di Fiore, 2006). There are two intracellular destinations for endosomes. In one, the receptor is delivered to compartments involved in signaling, and in the other the receptor is delivered to a degradative compartment, thereby keeping the intensity and duration of the signal within a normal range. In addition, there is an apoptotic pathway to eliminate unwanted or unneeded cells, and an autophagic pathway for intracellular remodeling.

Figure 1.5 Simplified sketches of six major signaling pathways used for development and regeneration. Descriptions in text. The signal in the Notch pathway actively maintains cells as stem cells whereas the signals in the other pathways activate cells to divide asymmetrically. (A) Notch pathway. L=ligand, N=notch protein, ICD=intracellular domain, P=presenelin, HAT=histone acetyltransferases, RBP-Jκ=DNA binding protein. The ICD, HAT and RBP-Jκ form a complex that binds to the regulatory regions of differentiation genes to block their transcription. The blockade is maintained by the protein NRP-1, which prevents the translation of Numb (Nu) mRNA. If translated, Numb inhibits the removal of the ICD from Notch and the transcription-inhibiting complex is not formed, activating the cell. (B) Wnt pathway. Fz and LRP6=co-receptors for Wnt. PAR-1=protein kinase that phosphorylates (P) the Disheveled protein (D). GSK-3=glycogen synthase kinase-3, Ax=axin, APC=adenomatous polyposis coli protein, β-C=beta catenin. Lef and Tcf=transcription factors that in the absence of Wnt signaling inhibit transcription of genes that activate stem cells. (C) Hedgehog pathway. Ptc=patched protein, Smo=smoothened protein, Gli=transcription factor that is the counterpart of cubitus interruptus in Drosophila . (D) RTK pathway. GF=growth factor, P=phosphate, AP=adaptor protein that recognizes phosphorylated receptor and activates the Ras protein, leading to a chain of phosphorylation reactions that ends in ERK, a mitogen activated protein kinase (MAPK) that phosphorylates transcription factors (TFs). (E) TGF-β (Smad) pathway. P=phosphate, S=Smad proteins. The type I and II receptors are each dimers. (F) JAK-STAT pathway. P=phosphate. The STAT proteins form transcriptional complexes that are homodimers (H) or heterodimers (HT).

1 Notch Pathway

The Notch receptor is a major mediator of stem cell self-renewal and fate determination during embryonic development and regeneration (Lai, 2004). Vertebrate cells have four different Notch receptors (Lundkvist and Lendhal, 2001). Notch is a transmembrane protein that is signaled (activated) by the membrane-bound ligands Delta, Jagged and Serrate on neighboring cells. Endocytosis and membrane trafficking regulate ligand and receptor availability at the cell surface. Endocytosis of bound ligand is thought to generate a mechanical force that causes a conformational change in Notch that allows its extracellular domain to be cleaved off by a metalloprotease called ADAM (Kopan and Ilagan, 2009). Next, the presenelin enzyme cleaves off the Notch intracellular domain (NICD) (Lecourtis and Schweisguth, 1998; Schroeter et al., 1998). The NICD translocates to the nucleus, where it interacts with the DNA-binding protein RBP-Jκ and the histone acetyltransferases p300 and PCAF (Wallberg et al., 2002) to activate target genes whose products act as transcriptional repressors of genes encoding products that promote cell differentiation (Nakamura et al., 2000).

The activity of the NICD is inhibited by the Numb protein (Jan and Jan, 1998). Asymmetric localization of Numb in one daughter cell during the division of an ASC thus leads to lineage commitment of that cell, whereas the other daughter remains as a stem cell. Expression of the Numb protein is regulated at the translational level by Nrp-1 (Musashi-1 in Drosophila), a RNA-binding protein (Sakakibara et al., 1996). Nrp-1 binds numb mRNA, preventing it from being translated (Imai et al., 2001). Thus in the absence of Nrp-1, Numb will be made and localized, leading to lineage commitment.

2 Canonical Wnt Pathway

An important set of transcriptional factors that maintains ASCs in a quiescent state is the Tcf/Lef family, downstream effectors of the canonical Wnt signaling pathway. The family consists of four members, Tcf-1-3 and Lef-1. These transcription factors are widely expressed during embryonic development and in stem cells, where in combination with Groucho-related proteins, they serve as transcriptional repressors in the absence of a Wnt signal (Cavallo et al., 1998; Roose et al., 1998; Brantjes et al., 2001). In the absence of Wnt signaling, β-catenin is continually degraded by a destruction complex consisting of two tumor suppressor proteins axin and adenomatous polyposis coli (APC) that act as the scaffold for the complex, and two kinases, CK1 and GSK-3 that continually phosphorylate β-catenin (Clevers, 2006). Phosphorylated β-catenin is ubiquinated and targeted for degradation by the proteasome.

Wnt proteins, of which 15 have been identified, bind to the Frizzled (Fz) receptor and its co-receptor LRP5/6 (Wehrli et al., 2000). This interaction phosphorylates LRP5/6, leading to loss of the ability of the β-catenin destruction complex to phosphorylate β-catenin, which is now stabilized and accumulates in the nucleus, where it displaces Groucho proteins and synergizes with Lef and Tcf to activate transcription. Inhibition of the β-catenin destruction complex is via both inhibition of GSK3 by the Disheveled protein, which is phosphorylated by the PAR-1 kinase, and by the recruitment of axin away from the complex by LRP5/6 (Clevers, 2006; Bilic et al., 2007). Canonical Wnt signaling is inhibited by a number of antagonists that mimic the molecular structure of Fz and also by the Dickkopf (DKK) proteins, which bind directly to LRP-5/6 (Bafico et al., 2001).

Dsh may have a role in a non-canonical Wnt signaling pathway through its ability to interact with Rho GTPase, an enzyme that can activate kinases to phosphorylate cytoskeletal proteins and alter cell shape, polarity and motility (Bejsovec, 2005; Katoh, 2005). A second non-canonical Wnt pathway works through Fz by its activation of a phospholipase enzyme that leads to the release of stored calcium from the endoplasmic reticulum (Kohn and Moon, 2005). Calcium activates many proteins involved in a wide variety of biological functions.

3 Hedgehog Pathway

The hedgehog family of signaling molecules has three members, sonic hedgehog (Shh), Indian hedgehog (Ihh), and desert hedgehog (Dsh). These proteins are important in stem cell self-renewal, proliferation and fate specification, and tissue patterning (Zhang et al., 2001; Reya et al., 2001; Lai et al., 2003). Unlike the Notch ligands or the Wnts, the hedgehog signal acts to inhibit the action of its receptor, Patched. Patched is bound to, and inhibits, the actual signal transducer for hedgehog, Smoothened. When hedgehog binds to Patched, Smoothened is activated. To be active, hedgehog must be cleaved to form an amino-terminal peptide that is esterified at its C-terminus to a cholesterol molecule (Porter et al., 1996; Taipale and Beachy, 2001).

Smoothened is phosphorylated and, in Drosophila, releases the Cubitus interruptis (Ci) protein that is tethered to microtubules. Ci is a transcription factor that acts as a repressor or activator, depending on whether it is cleaved. In the absence of hedgehog signal, the carboxy-terminal domain of Ci is cleaved off and moves to the nucleus, where it acts as a transcriptional repressor. In the presence of hedgehog signal, the entire Ci molecule is released and translocated to the nucleus, where it acts as a transcriptional activator (Van den Heuvel, 2001; Lum and Beachy, 2004). Different sets of genes are activated depending on the concentration of Ci. In vertebrates, three different proteins, Gli 1–3, have evolved to effect transcriptional repression or activation in the hedgehog pathway (Stecca and Ruiz i Altaba, 2002). In the absence of hedgehog signal, the carboxy termini of Gli 2 and 3 are enzymatically cleaved off so that transcription cannot be activated. The carboxy terminus of Gli-1 is not removed, but intact Gli-1 cannot activate transcription by itself. In the presence of hedgehog signal, the enzymes that cleave off the carboxy terminus of Gli 2 are inhibited by Smoothened. Gli-2 and Gli 1 and 2 now act together as transcriptional activators, while Gli-3 acts as a transcriptional repressor (Wang et al., 2000; Aza-Blanc et al., 2000). The hedgehog and Wnt pathways share certain similarities, suggesting that they may be sister pathways that synergize to affect stem cell proliferation (Lum and Beachy, 2004).

4 Receptor Tyrosine Kinase (RTK) Pathway

The receptor tyrosine kinase (RTK) pathway is used by a wide variety of growth factors such as fibroblast growth factors (FGFs), platelet derived growth factor (PDGF), epidermal growth factors (EGFs), vascular endothelial growth factor (VEGF), and stem cell factor (SCF) (Schlessinger, 2000). These ligands bind to specific RTKs. RTKs are transmembrane proteins that are dimerized by their ligands and undergoes a conformational change that results in the autophosphorylation of specific tyrosines on the cytoplasmic domain of the receptor. An adaptor protein recognizes one of these tyrosine residues, leading to the activation of G proteins such as Ras, which activate a cascade of kinase phosphorylations. The last member of this cascade, phosphorylated extracellular signal-regulated kinase (ERK, also called mitogen-activated protein kinase, or MAP), enters the nucleus where it phosphorylates and activates transcription factors.

5 JAK-STAT Pathway

The JAK-STAT pathway (JAK=janus kinases; STAT=signal transducers and activators of transcription) is activated by many cytokines and growth factors (including FGFs) that bind to receptors lacking the intrinsic tyrosine kinase activity of the RTK pathway (Aaronson and Horvath, 2002). Binding of the ligand dimerizes the receptors and induces a conformational change allowing JAK proteins to bind to their intracellular domains and phosphorylate tyrosine residues, converting the receptor into a tyrosine kinase. The activated receptors now phosphorylate STAT proteins, allowing them to form homo- and heterodimers and move rapidly into the nucleus, where they associate with other proteins to form transcriptional complexes. In mammals, there are four JAK genes and seven STAT genes, which provide diversity of receptor activation and transcriptional binding.

6 Transforming Growth Factor Beta (TGF-β) Pathway

The TGF-β family of growth factors consists of two subfamilies, the TGF-β/Activin/Nodal subfamily and the bone morphogenetic protein (BMP)/growth and differentiation factor (GDF)/Muellerian inhibiting substance (MIS) subfamily (Shi and Massague, 2003). These signaling molecules signal through serine-threonine kinase receptors that consist of two transmembrane proteins known as the type I and II receptors (Attisano and Wrana, 2002). In vertebrates, there are seven different type I receptors and five type II receptors. Different type I and II receptors form heterodimers, depending on which ligand they bind. Upon binding ligand, the type II receptor phosphorylates the type I receptor, activating its kinase domain. Depending on the ligand, the activated receptor then phosphorylates different classes of Smad proteins.

There are eight Smads that fall into three classes. The receptor Smads (R-Smads, 1, 2, 3, 5, 8) are the only ones directly phosphorylated by the receptors. They accumulate in the nucleus where they associate with a Co-Smad (Smad 4) and other proteins into complexes that activate or repress transcription. Smads 6 and 7 compete for binding sites with the other Smads and are inhibitory. Three of the type I receptors transduce TGF-β-like signals by phosphorylating Smads 2 and 3, whereas the other four type I receptors activate Smads 1, 5 and 8 to mediate BMP signals.

7 Apoptosis and Autophagy

Apoptosis is cell suicide, in which defined biochemical pathways lead to the enzymatic self-digestion of cell components without inflammation. This defined program distinguishes apoptosis from necrosis, the global disintegration of a cell due to trauma with accompanying inflammation. Apoptotic cells shrink and develop cell surface blebs, degrade their DNA, and break up into small membrane-bound fragments that first display find me signals and then an eat me phosphatidylserine signal that binds to macrophage and dendritic cell surfaces (Zitvogel et al., 2010; Nagata et al., 2010). These phagocytic cells then engulf and digest the fragments, while simultaneously producing anti-inflammatory cytokines (Ravichandran, 2003; Green, 2005). Embryonic and regenerating adult tissues rely on apoptosis to regulate cell number and turnover and to sculpt the shape of developing or regenerating organs, as well as to eliminate cells that can harm the organism, such as virally infected cells, cancer cells, immune cells that can cause autoimmunity, and cells with damaged DNA (Nagata et al., 2010; Gilbert, 2010).

Whether or not a cell decides to undergo apoptosis depends on the balance between anti-apoptotic signals and pro-apoptotic signals made by other cells (Zitvogel et al., 2010; Nagata et al., 2010). There are two sets of apoptotic pathways, one triggered by internal cell signals (intrinsic pathway), and one by signals external to the cell (extrinsic pathway) (Fig. 1.6). The intrinsic pathway is triggered in the absence of anti-apoptic signals (growth factors, adhesive signals), or when the cell suffers internal damage to its DNA, or accumulates mis-folded proteins, and leads to cell death by compromising mitochondrial function. Healthy mitochondria display the anti-apoptotic protein Bcl-2 on their surface, but two other proteins Bad and Bax, now bind to Bcl-2 blocking its action and form pores in the mitochondrial membrane, allowing cytochrome c to leak out. Cytochrome c binds to apoptotic protease activating factor-1 (APAF-1) to form an apoptosome complex. Apoptosomes bind and activate caspase 9, which activates caspase-3 and -7 (Shi, 2004). An example of a cell type that is programmed to die in the absence of an anti-apoptotic signal is the red blood cell, which requires erythropoietin for survival. Two major signals of the external pathway that directly trigger cell death are FasL and tumor necrosis factor (TNF). FasL is produced by cytotoxic T-cells and induces target cells to undergo apoptosis by binding to its receptor Fas. TNF is a growth factor involved in inflammation and triggers apoptosis by binding to the TNF receptor on target cells. In the external pathway, signals are transmitted to the cytoplasm that activate caspase-8, which like caspase-9, leads to the activation of caspase-3 and -7. These caspases digest the cell into fragments that send signals to macrophages to find the fragments and engulf them (Fig. 1.7).

Figure 1.6 Basic scheme of apoptosis vs. survival signals. Survival signals maintain mitochondria intact through Bcl-2. Cells undergo apoptosis through two pathways. The intrinsic pathway (black) is triggered in the absence of survival signals, by DNA damage, or by cell stress that results in an overload of misfolded proteins that cannot be cleared. The BAD and BAX proteins inhibit Bcl-2, causing cytochrome C to leak from damaged mitochondrial membranes. Cytochrome C leakage results in a chain of reactions ending in the activation of the executioner caspases 3 and 7. The extrinsic pathway (red) is activated by direct death signals such as TNF. These signals activate caspase 3 and 9 via caspase 8.

Figure 1.7 Engulfment of apoptotic cells by macrophages. Apoptotic cells and fragments give off find me signals that attract macrophages, and eat me signals that trigger engulfment by the macrophages. The engulfed cells or fragments are endocytosed and become part of lysosomes, which contain enzymes that digest proteins, DNA, fats and carbohydrates that are recycled to provide amino acids and nucleotides to other cells. Reproduced with permission from Nagata et al., Autoimmunity and the clearance of dead cells, Cell 140:619-630. Copyright 2010, Elsevier.

The life and death genes are remarkably conserved across vast phylogenetic distances and in fact were first discovered in C. elegans (Ellis and Horvitz, 1986). Interestingly, the fibroblasts of caspase-3 and -7 knockout mice are resistant to both mitochondrial and direct death-receptor-mediated apoptosis, suggesting that these enzymes are important for events of apoptosis earlier than the final acts of destruction (Lakhani et al., 2006). TUNEL (Tdt-mediated dUTP-biotin nick end labeling) or antibodies to caspase 3 are used to reveal cells undergoing apoptosis in tissue sections.

Autophagy is a different kind of cellular self-digestion that does not result in cell death, but in the degradation of cell organelles and proteins, and their re-cycling into new molecules (Klionsky, 2007; Levine and Kroemer, 2008; Mizushima et al., 2008). Cell components to be recycled are packaged directly into lysosomes (microautophagy) or first into autophagosomes that then fuse with lysosomes (macroautophagy). In the presence of growth factors such as insulin-like growth factor (IGF) and abundant nutrients, the autophagy pathway is inhibited by TOR (target of rapamycin) kinase. IGF acts through RTKs to activate TOR via class I PI3K/Akt pathways (Lum et al., 2005). In yeast, TOR inhibits about 20 downstream genes that carry out autophagy (Mizushima and Klionsky, 2007). A low basal level of autophagy is involved in the molecular regeneration of all cells, but is upregulated during periods of high energy demand such as starvation, growth factor withdrawal, or the need to eliminate damaging cellular components during oxidative stress, infection, or accumulation of protein aggregates (Levine and Kroemer, 2008). Autophagy may have its greatest significance for regeneration and fibrosis in the structural remodeling of cells that takes place during processes such as remodeling of granulation tissue, EMT/MET, and dedifferentiation.

VI Approaches to the Study of Regeneration

A Animal Models

Many animal (and plant) models are used to study regeneration. Invertebrates such as planaria and hydra have long been favorite research models for whole body regeneration (Birnbaum and Sanzhez-Alvarado, 2008). Among vertebrates, which are the focus of this book, various species of fish, larval and adult urodele salamanders, and anuran tadpoles and adults are used to study the regeneration of appendages, heart, and central nervous system tissues. Regeneration by adult stem cells has been investigated largely in mammalian models, but studies on Drosophila are now increasingly contributing to our knowledge of stem cell biology.

B Determining the Cellular Origins of Regenerated Tissues

The most fundamental question of the four that are addressed in our quest to understand regenerative mechanisms is the origin of the cells that carry out regeneration. Determining the origins of cells involved in maintenance or injury-induced regeneration requires marking them in some way and showing that the marked cells do or do not give rise to the regenerated tissue. Natural markers such as pigment or nuclear ploidy, or artificial markers such as DNA labeling of proliferating cells with ³H-thymidine or BrdU, or labeling groups of cells with cell membrane or cytoplasmic dyes such as DiI have been used to follow cells after grafting the marked tissue to an unmarked host and tracking the donor cells in a regenerative situation (Fig. 1.8).

Figure 1.8 Tracing the contribution of cells to regenerated tissues by grafting labeled tissues in place of unlabeled tissues. In the example shown here, muscle from a donor salamander has been labeled in several different ways and then grafted in place of the recipient muscle. HUM: humerus. The grafted muscle can come from animals with a different ploidy than the normal 2N recipient (3N, 1N), or it can be labeled during development with tritiated thymidine (3HT) or a cell membrane dye such as PKH26. The limb is then amputated and regenerates. Labeled cells in the regenerated tissues indicate a contribution from the graft.

More sophisticated genetic marking of cells has considerably refined our ability to look at the origins of regenerated tissues. Two kinds of transgenic marking methods have been developed (Fig. 1.9). In the first method, called non-conditional genetic marking (Fig. 1.9A), zygotes are either injected with a construct consisting of a reporter gene driven by the promoter of a gene that is constitutive (on in every cell), or a construct driven by the promoter of a gene that is normally activated only in one specific cell type during development. In the first case, the reporter will be active in all cells. Specific cells can be tracked after injury by grafting them to an unmarked host. In the second case the reporter will be active only in the cells that express the phenotype-specific gene activity, allowing the marked cells to be followed during maintenance or injury-induced regeneration of the tissue containing them.

Figure 1.9 Genetic marking of cells with constructs delivered by viral vectors. (A) Unconditional marking. A construct consisting of a constitutive promoter (CnP) or a cell-specific promoter (CSP) and a GFP reporter gene is injected into a mouse zygote to make a transgenic mouse. In the first case, all cells express the GFP transgene and labeling specificity is achieved by grafting specific tissues to an unlabeled recipient. In the second case, only the cell type expressing a gene specific for that cell type will be labeled because only that cell type will have the transcription factors that activate the promoter of that gene. The movements and fates of the labeled cells can then be tracked as long as the gene for the label is on. (B) Conditional (inducible) marking. This scheme allows the investigator to mark specific adult cell types at any time prior to an injury to the tissue containing that cell type and determine the origin of regenerated cells. The method requires crossing two sets of mice, one of which carries an inducer construct consisting of a cell-specific promoter (CSP), Cre , and an estrogen receptor gene (ER), and the other which carries a reporter construct consisting of a constitutive promoter (CnP) and a reporter gene (in this example GFP) separated by a DNA sequence with a Lox P site at each end. This DNA sequence is commonly referred to as a floxed stop cassette (FSC). The FSC prevents the reporter gene from being transcribed. When the mice are crossed, some of the progeny carry both constructs. When a tamoxifen (estrogen) pulse is administered to these progeny, the Cre/ER fusion protein is now able to enter the nucleus, where it deletes the FSC, allowing the promoter to drive expression of the reporter gene.

In the second method, called induced or conditional genetic marking (Fig. 1.9B), transgenic animals are made in which the expression of a reporter gene can be activated by a stimulus at any time during development or regeneration, using a Cre/lox DNA recombination system (see Sauer and Henderson, 1988; Branda and Dymecki, 2004). Cre (Causes Recombination) is a recombinase expressed in bacteriophage that cuts DNA at lox P sequences. Two sets of transgenic animals are made. The first set carries a Cre/estrogen receptor fusion gene driven by the promoter of a gene whose activity defines a particular cell phenotype (such as insulin in β-cells of the pancreas). This construct makes a Cre/estrogen receptor fusion protein only in that cell type. The second set carries a construct consisting of a constitutive ubiquitous promoter and a reporter gene separated by a sequence flanked by lox-P sites (a floxed stop cassette, FSC). Crossing these mice yields progeny carrying both constructs. The Cre portion of the fusion protein can excise the FSC of the second construct at the lox P sites, thus recombining the constitutive promoter with the reporter gene and activating it. However, to translocate to the nucleus and carry out this recombination, the Cre/estrogen fusion protein must first be activated, which is accomplished by administration of a pulse of tamoxifen to the animal. The Cre/ER fusion protein now enters the nucleus, where the Cre recombinase excises the FLC, recombining the constitutive promoter and reporter gene, marking only those cells expressing the fusion protein. Conditional genetic marking is now in wide use to differentiate the origin of regenerated tissue from stem cells or differentiated cells.

C Analyzing the Niche Regulation of Cell Activities

The cell interactions involved in niche regulation can be studied in vitro by culturing cells on ECM of different compositions, in the presence of different combinations of soluble molecules, or in the presence of other cell types that have a close anatomical association with the test cells. The niche can be studied in vivo by using antibodies against ligands and/or receptors of a specific cell type, grafting cells into foreign niches, altering the spatial relationships between tissues, and creating knockouts or knockdowns of genes in cells that are suspected of contributing to niche function.

D Cell Imaging and Identification

Whatever our experimental strategies, we need to be able to recognize and isolate cell types in various states of differentiation, from stem cells to terminally differentiated cells. This has been done for over a century by the use of fixatives to preserve tissues, sectioning them into thin slices, and staining the slices on slides with dyes that react with the various molecular constituents of the cells and ECM to give different colors. The cells of tissues stained in this way reveal morphologies and internal cell structure specific to each cell phenotype and tissue. Today, these histological techniques are complemented by molecular techniques such as antibody staining of proteins that are biomarkers for cell type and differentiation state, and polymerase chain reaction to amplify transcripts of genes whose activity defines cell type and differentiation state. Advances in imaging such as confocal microscopy and computerized tomography have accompanied the development of these methods, and various separatory methods have been developed for sorting out different populations of cells from mixtures.

E Comparative Analysis of Regeneration and Fibrosis

Comparative analyses are useful for understanding what genetic circuits are common to regeneration-competent tissues and for understanding the molecular differences between regeneration-competence and deficiency. For example, a direct strategy to identify the molecular differences between regeneration and fibrosis is to compare and contrast the transcriptomes and proteomes of regeneration-competent versus deficient tissues. Three experimental models can be used for these comparisons that are exemplified in studies of limb regeneration (Fig. 1.10).

1. Compare wild-type tissues to the same tissues that have a genetic variation conferring a gain or loss of regenerative capacity.

2. Compare the same tissues at developmental stages when they are capable of regeneration versus stages when they are not (for example, fetal vs. adult).

3. Compare the same tissue between two species, one of which regenerates the tissue and the other does not.

Figure 1.10 Comparative models of regeneration-competent tissues vs. regeneration-deficient tissues. Regenerating amphibian limbs have examples of all three models. (A) Wild type vs. mutant gain or loss of function. Left, regeneration-competent white axolotl which by 6 weeks post-amputation (white line) through the upper forelimb has regenerated the amputated parts. Right, by contrast the short-toe mutant, which is characterized by stubby digits and shortened long bones, has lost the ability to regenerate, as shown by the small blob of tissue that has formed at the amputation site by 6 weeks. (B) Change in regenerative competence with development. Left, a Xenopus tadpole hindlimb amputated through the tarsal region at stage 53 has regenerated perfectly at 60 days post-amputation. Right, a froglet amputated through the tarsus at stage 60 is regenerating a cartilage spike 60 days post-amputation. (C) Species differences in regenerative competence. Left, adult axolotl regenerates perfectly after amputation through the radius/ulna of the forelimb. Right, a Xenopus froglet amputated through the same region of the forelimb regenerates