lecture10

The Wayback Machine - https://web.archive.org/web/20070525080552/http://darwin.nmsu.edu:80/~molbio/mcb520/lecture10.html

Cell Signaling

General Principles

Although unicellular organisms such as yeast, posses signaling pathways to communicate with other yeasts, we will deal with intercellular signaling between cells of multicellular organisms, such as animals.
Extracellular, hydrophilic signaling molecules (i.e. ligand), released by exocytosis, are recognized by membrane-bound receptors in the plasma membrane on the target cell. In most cases, this binding event triggers a cascade of intracellular signals (a.k.a., second messengers), that ultimately alter the behavior of the target cell. In some instances, both the ligand and the receptor are membrane-bound (i.e. contact dependent signaling) (Fig 15-1). Some signal molecules are hydrophobic and are thus able to cross the plasma membrane and get to the cytosol, where they bind intracellular receptors (Fig 15-2).

Secreted ligands mediate three forms of signaling (Fig 15-3):

Paracrine signaling: signal acts locally, on cells in the immediate vicinity. Autocrine signaling is a variation of paracrine signaling in which the cells that produce the signal are, themselves, a target for that signal (i.e. contain receptors for the signal they produce) (Fig 15-5). It is a mechanism to encourage a group of cells to adopt a similar behavior.

Endocrine signaling: signal molecules called hormones are released into the blood stream, where they are present at very low concentrations, act on target cells distributed throughout the body. These cells contain high-affinity receptors that allow them to "pull out" the hormone from the blood. Hormone signaling is usually chronic, in other words, acts in the long term.

Synaptic signaling: is a combination of paracrine and endocrine signaling. Neurotransmitter released into synaptic cleft acts on the postsynaptic cell situated nanometers away. However, it may be a form of long-distance signaling because the axon connecting the pre- and postsynaptic cell can be centimeters-long. Furthermore, in general, receptors for neurotransmitters have lower affinities for their ligands than hormone receptors have for theirs. At the synaptic cleft, neurotransmitter concentration is very high, and their signaling acts in the short-term, as it is terminated quickly by hydrolysis or take-up by the presynaptic cell.

Another way to coordinate the activities of neighboring cells is through gap junctions (Fig 15-7). These directly connect the cytoplasms of two cells, via a water-filled channel that is permeable to small second messengers such as Ca²⁺ or cAMP but not macromolecules.
Each cell is programmed to respond to specific combinations of signaling molecules (Fig 15-8).
Different cells can respond different to the same signal either because: (1) each have a different receptor for the same signal, or (2) the intracellular events triggered by the signal is different for each cell, even though the signal binds the same receptor. (Fig 15-9)

The concentration of a molecule can be adjusted quickly, either by de novo synthesis or by degradation, only if the lifetime of the molecule is short (Fig 15-10). Signal molecules, in general, have short half-lives.
Examples of signals that bind to intracellular enzymes or receptors:

NO, nitric oxide, a gas that freely diffuses across membranes, made by NO synthase from arginine. Its very short half-life, it is converted into nitrates and nitrites by oxygen and water, limits its action to just the few cells immediately surrounding the producing cell. In many cells, NO reacts with iron in the active site of guanylyl cyclase, stimulating it to produce cGMP. NO has been implicated in the relaxation of blood vessels (causes dilation) as well as in the activation of some immune cells. Researchers that discovered the physiological role for NO were just awarded the 1998 Nobel Prize in Physiology and Medicine.

Steroid hormones, thyroid hormones, retinoids, and vitamin D are small hydrophobic molecules that despite having different structure and function act by a similar molecular mechanism. They diffuse directly across the plasma membrane and bind to intracellular receptors, which, now in an active state, directly regulate the transcription of specific genes. These transcription factors are structurally related and constitute the steroid receptor superfamily (Fig 15-12). In many cases, the response to these signals takes place in two steps: the direct induction (or repression) of the transcription of a small number of specific genes within ~30 min is the primary response; the products of these genes in turn activate (or repress) other genes and produce a delayed, secondary response (Fig 15-13).

Most cell-surface soluble-signal receptors belong to one of three classes defined by the mechanism they use to transduce ligand-binding into intracellular signaling events. We have already discussed the first group, neurotransmitter-gated ion channels (e.g. AChR). The second group, G-protein-coupled receptors, acts indirectly to regulate the activity of a separate plasma-membrane-bound target protein, which can be an enzyme or an ion channel. The interaction between the receptor and the target protein is mediated by a third protein, called a heterotrimeric GTP-binding regulatory protein (G protein) (Fig 15-14B). The activation of the target protein either alters the concentration of one or more intracellular mediators or alters the ion permeability of the plasma membrane. The intracellular mediators act in turn to alter the behavior of intracellular proteins. All G-protein-linked receptors belong to a large superfamily of homologous, seven-pass transmembrane proteins.

Enzyme-linked receptors either function directly as enzymes or associate with enzymes (Fig 15-14C). Most are single-pass transmembrane proteins, with the ligand-binding site outside the cell and the catalytic site inside. They are more heterogeneous than the previous two groups, however, most are protein kinases, or are associated with protein kinases that phosphorylate specific sets of proteins in the target cell.

Intracellular relay systems for the above classes of receptors are proteins of two kinds: proteins that become phosphorylated by protein kinases, and proteins that are induced to bind GTP when the signal arrives. In both cases the proteins gain one or more phosphates in their activated state and loose the phosphate when the signal decays (Fig 15-15). These proteins in turn generally cause the phosphorylation of downstream proteins as part of a phosphorylation cascade. Kinases are also of two main types: serine/threonine kinases, which phosphorylate proteins on S and (less often) T residues, and tyrosine kinases, which phosphorylate proteins on Y residues. Often overlooked, protein phosphatases, which remove phosphate groups from proteins, are also major players in these cascades.
The ultimate behavior of a cell is the result of the integration of many signaling cascades. Fig 15-16 presents two possible molecular mechanisms for the integration of 2 signals via phosphorylation cascades.

2. Signaling via G-protein-coupled receptors

When an extracellular ligand binds to a G-protein-linked receptor, the receptor changes its conformation and activates the heterotrimeric G protein that associates with it by causing them to release their bound GDP and replace it with GTP. The switch is turned off when the G protein hydrolyses its own bound GTP, converting it back to GDP. But before that happens, the active G protein has diffused away from the receptor and delivers its message to its downstream target. Most G-protein-linked receptors activate a cascade that leads to changes in the concentration of intracellular second messengers. The two most important of these are cyclic AMP (cAMP) and Ca²⁺ (Fig-15-18).

cAMP is normally found at ~10^-7M in the cytosol of most cells. It is made from ATP by a plasma-membrane-bound enzyme called adenylyl cyclase, and is rapidly and continuously degraded to 5'-AMP by cAMP phosphodiesterases (Fig 15-20). Many extracellular signals control the levels of cAMP by altering the activity of adenylyl cyclase. Many hormones that bind to different receptors act through the activation of adenylyl cyclase (Table 15-1). These various receptors are coupled to adenyl cyclase by the same G protein, which is called stimulatory G protein (G_s) because it causes enzyme activation.
But exactly how does G_s activate adenylyl cyclase? A heterotrimeric G protein has three subunits, a, b, and g. The G_s a chain (a_s) binds and hydrolyze GTP and activates adenylyl cyclase. b and g form a tight complex (bg), which anchors G_s to the inner leaflet of the plasma membrane. In its inactive form G_s exits as a trimer with GDP bound to a_s (Fig 15-23). When stimulated by binding to a ligand-activated receptor, a_s exchanges GDP for GTP, which causes it to dissociate from bg. a_s then binds to adenylyl cyclase activating it to produce cAMP. The lifetime of the active a_s is short: its GTPase activity is also stimulated when bound to adenylyl cyclase. The GTP is quickly hydrolyzed to GDP, which inactivates both a_s and adenylyl cyclase, and allows reassociation of a_s with bg. Thus, the system is reset for another signaling event. Cholera toxin and non-hydrolyzable analogs of GTP (e.g. GTPgS), block the GTPase activity of a_s, thus prolonging the activation of adenylyl cyclase (i.e. system is not shut off).
When the hormone adrenaline (a.k.a. epinephrine) binds b-adrenergic receptors, it activates adenylyl cyclase via G_s. However, when it binds a₂-adrenergic receptors, it inhibits the enzyme. The a₂-adrenergic receptors are coupled to adenylyl cyclase by an inhibitory G protein (G_i). It can contain the same bg complex as G_s, but G_i has a different a chain (a_i), which if activated binds and inhibits the activity of adenylyl cyclase. Note that dissociated bg from G_i also contributes to this inhibition by binding any free a_s available in the vicinity. Pertussis toxin made by the bacterium that causes whooping cough catalyzes the ADP ribosylation of a_i, which prevents it from binding and inhibiting adenylyl cyclase.
Cyclic AMP acts mainly by activating the enzyme cyclic-AMP-dependent protein kinase (A-kinase), which is a Ser/Thr-kinase. The amino acids phosphorylated by A-kinased are marked by the presence of two or more basic amino acids on their amino-terminal side. Substrate for A-kinase vary from cell to cell which accounts for the different effects of cAMP in different cells. In the inactive state, A-kinase is a complex of two regulatory subunits, able to bind cAMP, and two (inactive) catalytic subunits. Upon cAMP binding, the complex dissociates thus activating the catalytic subunits (Fig 15-24).

The stimulation of glycogen breakdown by cAMP in skeletal muscle cells provides a classical example of cAMP acting via phosphorylation (Fig 15-25). Adrenaline binds b-adrenergic receptors on the sarcolemma, [cAMP] raises, A-kinase is activated an it phosphorylates two other enzymes: (1) phosphorylase kinase, which in turn phosphrylates glycogen phosphorylase which release glucose 1-phosphate from glycogen. (2) A-kinase also phosphorylates glycogen synthase thus inactivating it. So, both glycogen breakdown and synthesis are activated and inhibited, respectively, at the same time.
In some cells, an increase in [cAMP] activates the transcription of specific genes. For example, in cells that release somastostatin, cAMP turns on the gene encoding this hormone. In its regulatory region the somastostation gene contains a cAMP response element (CRE), that is also found in other genes activated by cAMP. CRE is the binding site for CREB (CRE-binding) protein, which is activated by A-kinase following phosphorylation of a specific S residue.
Protein phosphatases I, IIA, IIB and IIC, catalyze the desphosphorylation of Ser-P thus ensuring that the effect of the cAMP is transient. Except for IIC, all of these phosphatases are composed of a homologous catalytic subunit complexed with one regulatory subunit. Phosphatase I is responsible for the dephosphorylation of many A-kinase target proteins. For example, it dephosphorylates (and inactivates) CREB. It switches glycogen metabolism from breaking down mode to synthesizing mode by dephosphorylating all 3 proteins activated by A-kinase. In adrenaline-stimulated muscle cells, though, the activity of phosphatase I is suppressed by the A-kinase-phosphorylation of a specific phosphatase inhibitor protein (Fig 15-26). This is an additional way by which A-kinase activates the breakdown of glycogen.
It is now recognized that Ca²⁺ is one of the most important second messengers in eucaryotic cells. We have already seen how it triggers the release of transmitter in synapses and of certain secreted proteins in some immune cells. We have also seen that in a resting state, cytosolic [Ca²⁺] is kept very low and that there are several carriers whose function is to keep this concentration that low after it raises due to some signaling event (Fig 15-27). There is another, more ubiquitous, Ca²⁺ signaling pathway in which the extracellular signal that binds to a G-protein-coupled receptor induces the release of calcium from intracellular stores (i.e. ER) via the generation of inositol triphosphate (IP₃), which directly triggers the opening of Ca²⁺channels in the ER membrane (Fig 15-28).

An activated receptor stimulates another heterotrimeric G-protein called G_q, which in turn activates a phosphoinositide-specific phopholipase C called phopholipase C-b. This enzyme quickly cleaves phosphoinositol biphosphate (PIP₂), which is a minor inositol phosphate that accumulates in the inner layer of the plasma membrane of most animal cells. Two products that induce distinct downstream signaling events are generated: IP₃ and diacylglycerol (DAG) (Fig 15-30, -33).

IP₃ diffuses to the cytosol and binds IP₃-gated-Ca²⁺-release channels in the ER membrane. These channels display a Ca²⁺-dependent positive feedback mechanism by which the initial released Ca²⁺ increases the Ca²⁺ released from the ER (Ca²⁺-induced- Ca2+ release). The IP₃ signaling is terminated by its dephosphorylation and by the clearing of [Ca²⁺]_i by the carriers mentioned above.
DAG has two potential signaling roles: (1) it can be further cleaved to release arachidonic acid, which either can act as a messenger or be use in the synthesis of other signaling molecules. (2) Most importantly, DAG activates protein kinase C (C-kinase of PKC), which in turn phosphorylates selected proteins in the target cell. PKC is normally in the cytosol, but it is translocated to the inner layer of the plasma membrane by a mechanism dependent in the rise of intracellular calcium. Once there, Ca2+, DAG, and phosphatidylserine activate the kinase. There are at least 8 PKC isoforms which are particularly abundant in the brain where they modify the excitability of neurons by phosphorylating ion channels. In many cells C-kinase activation leads to increase in the transcription of specific genes, two examples of which are shown in Fig 15-32.

Calmodulin (Fig 15-34): abundant cytosolic protein found in all eucaryotic cells. 150 amino acids in size with four Ca²⁺ -binding sites. The two sites towards the C-terminal end have a tenfold higher affinity for Ca2+ that the two sites towards the N-terminus. Ca²⁺/calmodulin has no enzyme activity but acts by binding to other proteins, specially Ca²⁺/calmodulin-dependent protein kinases (CaM-kinases).

CaM-kinases, are Ser/Thr-kinases responsible for most of the effects of Ca²⁺ in animal cells. CaM- kinase II, is the best studied multifunctional CaM-kinase, highly expressed in the brain. At active catecholaminergic synapses, concentrated CaM-kinase II activates tyrosine hydroxylase, which is the rate-limiting enzyme in the synthesis of catecholamines. CaM-kinase II autophosphorylates itself which allows it to be active even after cytosolic Ca²⁺ has returned to background levels and calmodulin has dissociated from the enzyme. Thus, CaM-kinase II acts as a "molecular memory-device" of a prior Ca2+ pulse (Fig 15-35). The Ca²⁺ - and cAMP-dependent pathways can interact (or crosstalk).
Heterotrimeric G-proteins can regulate the gating properties of ion channels in excitable cells. They do it by either direct binding, or by regulating channel phosphorylation, or by causing an increase or a decrease in the production of cytoplasmic cyclic nucleotides. Cyclic-nucleotide-gated ion channels are especially important in olfaction and vision.

Olfaction: when stimulated by odorant binding, G protein-coupled olfactory receptors activate an olfactory specific G protein (G_olf), which in turn activates adenylyl cyclase; the increase in [cAMP] opens cAMP-gated cation channels, which allow Na⁺ influx that depolarizes the cell and initiates a nerve impulse that eventually reach the brain.

Vision: transduction mechanism especially well known for rod photoreceptors (sensory neurons that account for monochromatic vision in dim light, Fig 15-39). Photons activate rhodopsin (a seven-pass transmembrane protein) in the disc membrane, which then binds the heterotrimeric G protein transducin (G_t), causing the a_t to dissociate and activate cyclic GMP phosphodiesterase, which hydrolyses cGMP thus decreasing its level in the cytosol. As a result, cGMP dissociates from plasma membrane Na⁺ channels, leading to their closure, which leads to membrane hyperpolarization that is the signal transmitted to the brain (Fig 15-40). Ca²⁺ is a second component in this signaling. The cGMP-gated Na⁺ channels are also permeable to Ca²⁺, thereby their closure leads to a decrease in cytosolic Ca²⁺, which activates the enzyme guanylyl cyclase to make cGMP and eventually reset the photoreceptor. The Ca2+-dependent activation of guanylyl cyclase is mediated by the protein recoverin which is active in the Ca²⁺-unbound state.
The various heterotrimeric G proteins that we have discussed are summarized in Table 15-3.

Extracellular signals are greatly amplified by the use of second messengers and enzymatic cascades (Fig 15-41, -42).

Cooperativity (Fig 15-44, -45) and positive feedback (Fig 15-46) are two mechanisms by which cells sharpen their response.

3. Signaling via enzyme-linked cell surface receptors.

Enzyme-linked receptors have themselves an intrinsic enzyme activity or associate with a protein(s) that does. Usually have only one transmembrane segment. There are 5 classes: (1) receptor guanylyl cyclases, (2) receptor tyrosine kinases (RTKs), (3) tyrosine-kinase-associated receptors, (4) receptor tyrosine phosphatases, and (5) receptor serine/threonine kinases.

The atrial natriuretic peptide (ANP) receptor is an example of a receptor guanylyl cyclase. The binding of ANP activates the cyclase to produce cGMP, which in turn binds to and activates a cyclic GMP-dependent protein kinase (G-kinase) which phosphorylates specific S/T residues in target proteins.
Receptors for most growth factors, i.e. epidermal growth factor (EGF), platelet-derived growth factor (PDGF), nerve growth factor (NGF), etc, are receptor protein kinases (Fig 15-47). The first event in the signaling mediated via these proteins is receptor autophosphorylation. Ligand binding induces receptor dimerization, which activates the intracellular tyrosine kinase domains in each monomer, which in turn cross-phosphorylate themselves. These phosphorylated Y residues serve as high affinity binding sites for a number of intracellular signaling proteins. Many of these proteins in turn become tyrosine phosphorylated and thus activated themselves. Different receptor tyrosine kinases bind different combinations of these signaling proteins thereby activating different cellular responses. Most proteins that bind phosphotyrosine residues on activated receptors usually share two highly conserved noncatalytic domains called SH2 and SH3 for Src homology regions 2 and 3, because they were first found in the Src kinase (Fig 15-49). Some of these proteins are made only of SH2 or SH3 domains and are thus called adaptors (see Drk protein, Fig 15-53).

The Ras proteins help relay signals from RTKs to the nucleus to stimulate cell proliferation of differentiation. Ras proteins contain a covalent prenyl group (a lipid) which anchor them to the cytoplasmic layer of the plasma membrane. Ras proteins are monomeric GDP/GTP-binding GTPases. They are active in the GTP-bound state and inactive in the GDP-bound state (Fig 15-50). Two classes of signaling proteins regulate Ras activity: GTPase-activating proteins (GAPs) and Guanine Nucleotide Releasing Proteins (GNRPs) a.k.a GEFs (guanine nucleotide exchange factors, which promote the loss of GDP and the subsequent uptake of GTP by Ras. In principle, RTKs could activate Ras either by activating a GNRP or by inhibiting a GAP. Activated RTKs bind GAPs directly, but bind GNRP only indirectly. In most cases, activation of Ras by RTKs is mediated by the indirect activation of a GNRP.

Fig 15-53: early cell signaling events in R7 development in Drosophila: an example of Ras activation by a RTK. Signal: Boss (bride-of-sevenless), plasma-membrane-bound protein is R8. Sev (sevenless): RTK in R7 plasma membrane. Sos (son-of-sevenless), a GNRP that activates Ras, following its binding to Drk, (downstream of receptor kinases), an adaptor protein that binds as well to activated Sev (via a SH2 domain). Once activated, Ras relays the signal downstream by activating a S/T phosphorylation cascade that is highly conserved in eucariotes, a crucial component of which is the family of mitogen-activated protein (MAP) kinases. MAP-kinase full activation requires phosphorylation of both a T and a Y that are separated by a single amino acid in the protein. This is catalyzed by MAP-kinase-kinase, which itself is activated by a S/T MAP-kinase-kinase-kinase that is thought to be activated by the binding of Ras-GTP (Fig 15-54). Activated MAP kinase then phosphorylates other kinases or transcription factors, which ultimately activate transcription for a specific set of genes. Immediate early genes are a set of genes (usually transcription factors) activated by the above pathway minutes after stimulation of cells with a growth factor.
Tyrosine-kinase-associated receptors, such as cytokine receptors, antigen receptors, and the growth hormone and prolactin receptors, function in much the same way as RTKs except that their kinase domain is encoded by a separate gene and is noncovalently associated with the receptor. Dimerization has also been implicated in receptor activation although, sometimes, more than two subunits may oligomerize as for the IL-2 receptor (Fig 15-55, -56). The kinases associated with these receptors either belong to the Src family (Src, Yes, Fgr, Fyn, Lck, Lyn, Hck and Bin) or to the Janus/STAT family (JAK1 and 2, Tyk2).

CD45, found on the surface of white blood cell, is an example of a receptor tyrosine phosphatase. When cross-linked by extracellular antibodies, its phosphatase domain is activated to remove phosphotyrosines of specific target protein like the Lck kinase.
The TGF-b superfamily of growth factors, which include TGF-b1 to 5, activins and Bone Morphogenetic Proteins (BMPs), as important roles in the proliferation and differentiation of many vertebrate cells. Its receptors constitute an example of receptor S/T kinases.
Some of the Y-specific and S/T-specific proteins kinases that we have discussed are reviewed in Fig 15-57.

4. Target-cell Adaptation.

Adaptation or desensitization is the process of adjustment of sensitivity to a signal by a cell. It is achieved via a negative feedback that occurs with different kinetics. Slow adaptation occurs by gradual removal of receptors from the plasma membrane by receptor-mediated endocytosis and is also known as receptor down-regulation. Rapid adaptation frequently involves ligand-induced phosphorylation of the receptors. The best example is the desensitization of the b₂-adrenergic receptor, which once bound to adrenaline is phosphorylated by a b₂-adrenergic kinase, which allows the binding of b-arrestin, which blocks the ability of the receptor to activate G_s. There are other more complicated mechanisms of adaptation that we won't deal with in this course.

Sep	MAY	Sep
	25
2006	2007	2008