In this section, we provide an overview of the major steps in signal transduction, starting with the signaling molecules themselves. We explore the molecular basis for ligand-receptor binding and the chain of events initiated in the target cell by binding of the signal to its receptor, focusing on a few components that are central to many signal transduction pathways.
Signaling Molecules Can Act Locally or at a Distance
As noted above, cells respond to many different types of signals—some originating from outside the organism, some internally generated. Those that are generated internally can be described by how they reach their target. Some signaling molecules are transported long distances by the blood; others have more local effects. In animals, signaling by extracellular molecules can be classified into three types based on the distance over which the signal acts (Figure 1a–c).
Fig1. Types of extracellular signaling. (a–c) Cell-to-cell signaling by extracellular chemicals occurs over distances from a few micrometers in autocrine and paracrine signaling to several meters in endocrine signaling. (d) Proteins attached to the plasma membrane of one cell can interact directly with cell-surface receptors on adjacent cells.
In endocrine signaling, the signaling molecules are synthesized and secreted by signaling cells (for example, those found in endocrine glands), transported through the circulatory system of the organism, and finally act on target cells dis tant from their site of synthesis. The term hormone generally refers to signaling molecules that mediate endocrine signaling. Insulin secreted by the pancreas and epinephrine secreted by the adrenal glands are examples of hormones that travel through the blood and thus mediate endocrine signaling.
In paracrine signaling, the signaling molecules released by a cell affect only those target cells in close proximity. A neuron releasing a neurotransmitter (e.g., acetylcholine) that acts on an adjacent neuron or on a muscle cell (inducing or inhibiting muscle contraction) is an example of paracrine signaling. In addition to neurotransmitters, many of the protein growth factors that regulate development in multicellular organisms act at short range. Some of these proteins bind tightly to com ponents of the extracellular matrix and are unable to signal to adjacent cells; subsequent degradation of these matrix components, triggered by injury or infection, releases these growth factors and enables them to signal. Many of the developmentally important signaling proteins that we discuss in Chapter 16 diffuse away from the signaling cell, forming a concentration gradient and inducing different responses in adjacent cells depending on the concentration of the signaling protein.
In autocrine signaling, cells respond to substances that they themselves release. Some growth factors act in this fashion, and cultured cells often secrete growth factors that stimulate their own growth and proliferation. This type of signaling is particularly characteristic of tumor cells, many of which overproduce and release growth factors that stimulate inappropriate, unregulated self-proliferation, a process that may lead to formation of a tumor.
Many integral membrane proteins located on the cell sur face play important roles as signals (Figure 1d). In some cases, such membrane-bound signaling proteins on one cell directly bind receptors on the surface of an adjacent target cell, often triggering its proliferation or differentiation. In other cases, proteolytic cleavage of a membrane-bound signaling protein releases the extracellular segment, which functions as a soluble signaling molecule.
Some signaling molecules can act at both short and long ranges. For example, epinephrine (also known as adrenaline) functions as a hormone (endocrine signaling) as part of the “fight or flight” response to a sudden danger in the environment, and also as a neurotransmitter (paracrine signaling). Another example, is epidermal growth factor (EGF), which is synthesized as an integral plasma-membrane protein. Membrane-bound EGF can bind to receptors on an adjacent cell. In addition, cleavage by an extracellular protease releases a soluble form of EGF, which can signal in either an autocrine or a paracrine manner.
Receptors Bind Only a Single Type of Hormone or a Group of Closely Related Hormones
Receptor proteins for all hydrophilic extracellular small molecule, peptide, and protein signaling molecules are located on the surface of the target cell. The signaling molecule, or ligand, binds to a site on the extracellular domain of the receptor with high specificity and affinity. Ligand binding de pends on multiple weak, noncovalent forces (i.e., ionic, van der Waals, and hydrophobic interactions) and molecular complementarity between the interacting surfaces of a receptor and ligand. Like an enzyme, each type of receptor binds only a single type of signaling molecule or a group of very closely related ones. For example, the growth hormone receptor binds to growth hormone, but not to other hormones with very similar, though not identical, structures. Similarly, acetylcholine receptors bind only this small molecule and not others that differ from it only slightly in chemical structure, while the insulin receptor binds insulin and related hormones called insulin-like growth factors 1 and 2 (IGF-1 and IGF-2), but no other hormones. The binding specificity of a receptor refers to its ability to bind or not bind closely related substances.
Binding of ligand to receptor causes a conformational change in the receptor that initiates a sequence of reactions leading to a specific response inside the cell. Organisms have evolved to be able to use a single ligand to stimulate different cells to respond in distinct ways. Different cell types often have different receptors for the same ligand, and activation of each receptor type induces a different intracellular signal transduction pathway. For instance, the surfaces of skeletal muscle cells, heart muscle cells, and the pancreatic acinar cells that produce hydrolytic digestive enzymes each have different types of receptors for acetylcholine. In a skeletal muscle cell, release of acetylcholine from a motor neuron innervating the cell triggers muscle contraction by activating an acetylcholine gated ion channel. In heart muscle, the release of acetylcholine by certain neurons activates a G protein–coupled receptor and slows the rate of contraction and thus the heart rate. Acetylcholine stimulation of pancreatic acinar cells triggers a rise in the concentration of cytosolic Ca2+ that induces secretion of the digestive enzymes stored in secretory granules to facilitate digestion of a meal. Thus the activation by acetylcholine of different types of acetylcholine receptors that are expressed in different cell types leads to different cellular responses.
Alternatively, the same receptor may be found on various cell types in an organism, but binding of a particular ligand to the receptor triggers a different response in each type of cell, given the particular complement of proteins expressed by the cell. The same epinephrine receptor (the β-adrenergic receptor) is found on liver, muscle, and fat (adipose) cells, it stimulates depolymerization of glycogen to glucose in the first two cell types, but hydrolysis and secretion of stored fat in adipose cells. In these ways, the same ligand can induce different cells to respond in a variety of ways, often in a manner that coordinates the overall response of the organism. This property is known as the effector specificity of the receptor-ligand complex.
Protein Kinases and Phosphatases Are Employed in Many Signaling Pathways
Activation of virtually all cell-surface receptors leads directly or indirectly to changes in protein phosphorylation through the activation of protein kinases, enzymes that add phosphate groups to specific residues of target proteins. Some receptors activate protein phosphatases, which remove phosphate groups from specific residues on target proteins. Phosphatases act in concert with kinases to switch the function of various target proteins on or off (Figure 2).
Fig2. Regulation of protein activity by a kinase/phosphatase switch. The cyclic phosphorylation and de-phosphorylation of a protein is a common cellular mechanism for regulating protein activity. In this example, the target, or substrate, protein is inactive (light green) when not phosphorylated and active (dark green) when phosphorylated; some proteins have the opposite pattern. Both the protein kinase and the phosphatase act only on specific amino acids in specific target proteins, and their activities are usually highly regulated.
The human genome encodes about 600 protein kinases and 100 different phosphatases. In general, each protein ki nase phosphorylates specific amino acid residues in a specific set of target, or substrate, proteins whose patterns of expression generally differ in different cell types. Animal cells contain two types of protein kinases: those that add phosphate to the hydroxyl group on tyrosine residues (protein tyrosine kinases) and those that add phosphate to the hydroxyl group on serine or threonine (or both) residues (protein serine/threonine kinases). We will see in this chapter and the next that the catalytic subunits of all known protein kinases have similar three-dimensional structures, including an N-terminal and a C-terminal lobe; highly conserved amino acids cluster around the catalytic site and are essential for binding ATP. All kinases recognize their specific substrates by binding not only to the side chain to be phosphorylated, but also to specific amino acids that surround the phosphorylated residue. Thus one can analyze the amino acid sequences surrounding tyrosine, serine, and threonine residues in a protein and make a good prediction as to which kinases might phosphorylate those residues.
Many proteins are substrates for multiple kinases, each of which usually phosphorylates different amino acids in the protein. Each phosphorylation event has the potential to modify the activity of a particular target protein in different ways, some activating its function, others inhibiting it. An example we will encounter later is glycogen phosphorylase kinase, a key regulatory enzyme in glucose metabolism. In many cases, addition of a phosphate group to an amino acid creates a binding surface that allows a second protein to bind; in the following chapter, we will encounter many examples of such kinase-driven assembly of multiprotein complexes.
Commonly the catalytic activity of a protein kinase itself is modulated by phosphorylation by other kinases, by the binding of other proteins to it, and by changes in the concentrations of various small intracellular signaling molecules and metabolites. Importantly, the activity of all protein kinases is opposed by the activity of protein phosphatases, some of which themselves are regulated by extracellular signals. Thus the activity of a protein in a cell can be a complex function of the activities of the usually multiple kinases and phosphatases that act on it.
GTP-Binding Proteins Are Frequently Used in Signal Transduction Pathways as On/Off Switches
Many cellular processes utilize members of the GTPase superfamily of proteins, which are found in all prokaryotic and eukaryotic cells. All of these GTP-binding switch proteins exist in two forms (Figure 3): (1) an active (“on”) form with bound GTP (guanosine triphosphate), which modulates the activity of specific target proteins, and (2) an inactive (“off”) form with bound GDP (guanosine diphosphate), which cannot affect the activity of target proteins. Members of the GTPase superfamily switch between GTP bound “on” and GDP-bound “off” forms. These proteins are evolutionarily ancient, as evidenced by their widespread functions in protein synthesis [examples include the roles of the eIF2 initiation factor and the EF1α and EF2 proteins in protein elongation; the transport of proteins between the nucleus and the cytoplasm (Ran); the formation of transport vesicles (Sar proteins, and their fusion with target membranes (Rab proteins); and rearrangements of the actin cytoskeleton (Rho, Rac, and Cdc42 proteins].
Fig3. GTPase switch proteins cycle between active and inactive forms. The switch protein is active when it has bound GTP and inactive when it has bound GDP. Conversion of the active into the inactive form by hydrolysis of the bound GTP is accelerated by GAPs (GTPase-activating proteins), RGSs (regulators of G protein signaling), and other types of proteins. Reactivation is promoted by GEFs (guanine nucleotide exchange factors), which catalyze the dissociation of the bound GDP and its replacement by GTP.
Here we focus on members of this superfamily that function in signal transduction pathways, in which conversion between the active GTP-bound and inactive GDP-bound states is tightly regulated. Conversion of the inactive to the active state is usually triggered by a signal (e.g., a hormone binding to a receptor) and is mediated by a guanine nucleotide exchange factor (GEF), which causes the release of GDP from the switch protein. Subsequent binding of GTP, favored by its high intracellular concentration relative to that of GDP, induces a conformational change to the active form. The principal conformational changes involve two highly conserved segments of the GTP-binding protein, termed switch I and switch II, that allow the protein to bind to and activate downstream signaling proteins (Figure 4). Conversion of the active form back to the inactive form is mediated by a GTPase, which is often part of the switch protein itself and which slowly hydrolyzes the bound GTP to GDP and Pi, thus altering the conformation of the switch I and switch II segments so that they are unable to bind to the target effector protein.
Fig4. Switching mechanism of monomeric G proteins. The ability of a G protein to interact with other proteins and thus transduce a signal differs between the GTP-bound “on” state and GDP-bound “off” state. (a) In the active “on” state, two domains, termed switch I (green) and switch II (blue), are bound to the terminal γ phosphate of GTP through interactions with the backbone amide groups of conserved threonine and glycine residues. When bound to GTP in this way, the two switch domains are in a conformation such that they can bind to and thus activate specific downstream effector proteins. (b) Removal of the γ phosphate by GTPase-catalyzed hydrolysis causes switch I and switch II to relax into a different conformation, the inactive “off” state; in this state, they are unable to bind to effector proteins. The three-dimensional models shown here represent both conformations of Ras, a monomeric G protein. A similar spring-loaded mechanism switches the alpha subunit in heterotrimeric G proteins between the active and inactive conformations by movement of three, rather than two, switch segments. [Part (a) data from E. F. Pai et al., 1990, EMBO J. 9:2351-2359, PDB ID 5p21. Part (b) data from M. V. Milburn et al., 1990, Science 247:939-945, PDB ID 4q21.]
The rate of GTP hydrolysis regulates the length of time the switch protein remains in the active conformation and is able to signal its downstream target proteins: the slower the rate of GTP hydrolysis, the longer the protein remains in the active state. The rate of GTP hydrolysis is often modulated by other proteins. For instance, both GTPase-activating proteins (GAPs) and regulators of G protein signaling (RGSs) accelerate GTP hydrolysis (see Figure 3). Many regulators of G protein activity are themselves controlled by extra cellular signals.
Two large classes of GTPase switch proteins are used in signaling. Heterotrimeric G proteins directly bind to and are activated by certain cell-surface receptors. G protein–coupled receptors function as guanine nucleotide exchange factors (GEFs), activating the heterotrimeric G protein to which they are coupled by triggering its release of GDP and binding of GTP. Monomeric (often called low-molecular-weight) G proteins, including Ras and various Ras-like proteins such as Ran and Sar, do not directly bind to receptors, but play crucial roles in many pathways that regulate cell division and cell motility, as is evidenced by the fact that mutations in genes encoding these G proteins frequently lead to cancer.
Intracellular “Second Messengers” Transmit Signals from Many Receptors
The binding of ligands (“first messengers”) to many cell surface receptors leads to a short-lived increase (or decrease) in the concentration of certain nonprotein, low-molecular weight intracellular signaling molecules termed second messengers. These molecules, in turn, bind to proteins, modifying their activity.
One second messenger used in virtually all metazoan cells is calcium (Ca2+) ions. We noted in Chapter 11 that the concentration of free Ca2+ in the cytosol is kept very low (~10−7 M) in part by ATP-powered pumps that continually transport Ca2+ out of the cell or into the endoplasmic reticulum (ER). The cytosolic Ca2+ concentration can be increased from tenfold to a hundredfold by a signal-induced release of Ca2+ from the ER lumen or the extracellular environment by the opening of calcium channels in the respective membranes; this change can be detected by fluorescent dyes introduced into the cell. In muscle, a signal-induced rise in cytosolic Ca2+ triggers contraction. In endocrine cells, a similar increase in Ca2+ induces exocytosis of secretory vesicles containing hormones, which are thus released into the circulation. In neurons, an increase in cytosolic Ca2+ leads to the exocytosis of neurotransmitter containing vesicles. In all cells, such a rise in cytosolic Ca2+ is sensed by Ca2+-binding proteins, particularly those of the EF hand family, such as calmodulin, all of which contain the helix-loop-helix motif. The binding of Ca2+ to calmodulin and other EF hand proteins causes a conformational change that permits those proteins to bind various target proteins, thereby switching their activities on or off. Often a rise in Ca2+ is localized to specific regions of the cytosol, allowing a process—such as exocytosis of a secretory vesicle—to occur there without affecting other processes elsewhere in the cell.
Another nearly universal second messenger is cyclic adenosine monophosphate (cAMP). In many eukaryotic cells, a rise in cAMP triggers the activation of a particular protein kinase, protein kinase A, that in turn phosphorylates specific target proteins to induce specific changes in cell metabolism. In some cells, cAMP regulates the activity of certain ion channels. The structures of cAMP and three other common second messengers are shown in Figure 5. Later in this chapter, we examine the specific roles of second messengers in signaling pathways activated by various G protein–coupled receptors.
Fig5. Four common intracellular second messengers. The major direct effect or effects of each compound are indicated below its structural formula. Calcium ions (Ca2+) and several membrane bound phosphatidylinositol derivatives also act as second messengers.
Signal Transduction Pathways Can Amplify the Effects of Extracellular Signals
Multiple signal transduction proteins are frequently combined to form a signal transduction pathway, allowing multiple target proteins to be activated (or inhibited) by a single type of cell-surface receptor. Figure 6 depicts a typical signal transduction pathway downstream of many G protein–coupled receptors. Here, binding of a hormone triggers a conformational change in the receptor, leading to activation of a G protein by catalyzing the exchange of GTP for GDP. The activated G protein binds to and activates an enzyme that synthesizes a second messenger; this small molecule binds and activates a protein kinase that, in turn, phosphorylates and thus changes the activity of one or more target proteins. One specific example of this multiprotein pathway—the regulation of glycogen metabolism in the liver by the hormone epinephrine—, and the molecules involved in that pathway are listed in parentheses in Figure6.
Fig6. A signal transduction pathway involving a G protein, a second messenger, a protein kinase, and several target proteins. The figure depicts a generalized signal transduction pathway; in parentheses are listed the molecules involved in the specific signaling pathway. Binding of the hormone to its cell-surface receptor activation of a G protein 2 by the receptor functioning as a GEF and causing loss of GDP and binding of GTP. The active G protein binds to and activates an enzyme 3 that synthesizes a second messenger to and activates a protein kinase 5. The kinase, in turn, phosphorylates, and thus changes the activity of, one or more target proteins. These proteins can either be cytosolic proteins 6a that induce changes in cellular function, metabolism, or movement or transcription factors in gene expression.
One important advantage of a cascade of proteins in a signal transduction pathway is that it facilitates amplification of an extracellular signal. Activation of a single cell surface receptor protein can result in an increase of perhaps thousands of cAMP molecules or Ca2+ ions in the cytosol. Each of these molecules, in turn, by activating its target protein kinase or other signal transduction protein, can affect the activity of multiple downstream proteins. In many signal transduction pathways, amplification is necessary because cell-surface receptors are typically low-abundance proteins, present in only a thousand or so copies per cell, while the cellular responses induced by the binding of a relatively small number of hormones to the available receptors often require generation of tens or hundreds of thousands of activated effector molecules per cell. In the case of G protein–coupled hormone receptors, signal amplification is possible in part because a single receptor can activate multiple G proteins during the time a hormone remains bound, each of
