Chapter_15_Solutions

Chapter_15_Solutions - Chapter 15 Mechanisms of Cell...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Chapter 15 Mechanisms of Cell Communication GENERAL PRINCIPLES OF CELL COMMUNICATION DEFINITIONS 151 152 153 154 155 156 157 158 159 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 Guanine nucleotide exchange factor (GEF) Receptor Adaptation or desensitization Interaction domain Gap junction Negative feedback loop NO synthase (NOS) Steroid hormone Paracrine signaling Phosphorylation cascade Second messenger or small intracellular mediator Endocrine cell Signaling cascade Extracellular signal molecule Protein kinase Neurotransmitter GTPase-activating protein (GAP) Scaffold protein Contact-dependent signaling 15 In This Chapter GENERAL PRINCIPLES OF CELL COMMUNICATION A337 SIGNALING THROUGH A346 G-PROTEIN-COUPLED CELL-SURFACE RECEPTORS (GPCRs) AND SMALL INTRACELLULAR MEDIATORS SIGNALING THROUGH A355 ENZYME-COUPLED CELLSURFACE RECEPTORS SIGNALING PATHWAYS A362 DEPENDENT ON REGULATED PROTEOLYSIS OF LATENT GENE REGULATORY PROTEINS SIGNALING IN PLANTS A364 TRUE/FALSE 1520 False. Signaling molecules that bind to cell-surface receptors do not have to cross the plasma membrane; thus, they can be large or small, hydrophilic or hydrophobic. By contrast, signaling molecules that bind to intracellular receptors must be sufficiently small and hydrophobic to diffuse across the plasma membrane. False. The concentration of a neurotransmitter in the synaptic cleft is much higher than the concentration of a circulating hormone in the blood, or of a local mediator in the neighborhood of the signaling cell. The characteristic difference in concentration of signaling molecules is reflected in differences 1521 A337 A338 Chapter 15: Mechanisms of Cell Communication in ligand affinity of the corresponding receptors. Neurotransmitter receptors typically have a lower affinity for their ligands than do hormone receptors and receptors for local mediators. The combination of high neurotransmitter concentration and low affinity receptors allows the neurotransmitter to dissociate rapidly from the receptor to terminate a response. As a rule of thumb, the Kd for binding to a receptor is about equal to the concentration of signaling molecule to which the receptor is exposed to produce a physiological response. 1522 False. Most second messengers, including cyclic AMP, Ca2+, and IP3, are water soluble and diffuse freely through the cytosol; however, second messengers such as diacylglycerol are lipid soluble and diffuse in the plane of the membrane. THOUGHT PROBLEMS 1523 Although succinylcholine binds to the acetylcholine receptor very tightly, it does not trigger the conformational change necessary to open the ion channel and initiate muscle contraction. Succinylcholine prevents normal muscle contraction by competing with acetylcholine for binding to the receptor, thereby blocking its action. For this reason, succinylcholine is referred to as an acetylcholine antagonist. Paracrine signal molecules are kept in the vicinity in several ways: rapid uptake by neighboring cells, destruction by extracellular enzymes, and immobilization by binding to the extracellular matrix. Both types of signaling can occur over long distances: neurons can send action potentials along very long axons (from the spinal cord to the fingers, for example), and hormones are passed through the bloodstream throughout the organism. Neurons secrete large amounts of neurotransmitters into a small, well-defined space at the synapse, yielding a high local concentration. Neurotransmitter receptors, therefore, need to bind to neurotransmitters with only low affinity (high Kd). By contrast, hormones are diluted extensively in the bloodstream, where they circulate at minuscule concentrations; hormone receptors, therefore, generally bind their hormones with extremely high affinity (low Kd). Neuronal signaling is very fast, limited only by the speed of propagation of the action potential and the workings of the synapse. In addition to speed, nerves communicate directly with one or a few cells. Hormonal signaling is slower, limited by blood flow and diffusion over relatively large distances, but it communicates at the same time with all the diverse and widely dispersed target cells in the body. 1524 1525 1526 A. A telephone conversation is analogous to synaptic signaling in the sense that it is a private communication from one person to another, usually some distance away and sometimes very far away. It differs from synaptic signaling because it is (usually) a two-way exchange, whereas synaptic signaling is a one-way communication. B. Talking to people at a cocktail party is analogous to paracrine signaling, which occurs between different cells (individuals) and is locally confined. C. A radio announcement is analogous to an endocrine signal, which is sent out to the whole body (the audience) with only target cells (individuals tuned to the specific radio station) affected by it. D. Talking to yourself is analogous to an autocrine signal, which is a signal that is sent and received by the same cell. 1527 In both cases the signaling pathways themselves are rapid. When the pathway modifies a protein that is already present in the cell, its activity is changed immediately, leading to a rapid response. When the pathway modifies gene expression, there will be a delay corresponding to the time it takes for the mRNA and protein to be made and for the cellular levels of the protein to be GENERAL PRINCIPLES OF CELL COMMUNICATION altered sufficiently to invoke a response, which would usually take an hour or more. 1528 In a multicellular organism, it is important that cells survive only when and where they are needed. Having cells depend on signals from other cells may be a simple way of ensuring this. A misplaced cell, for example, would probably fail to get the necessary survival signals (as its neighbors would be inappropriate) and would therefore kill itself. This strategy can also help regulate cell numbers: if cell type A depends on a survival signal from cell type B, the number of B cells could control the number of A cells by making a limited amount of the survival signal, so that only a certain number of A cells could survive. There is evidence that such a mechanism does operate to help regulate cell numbers--in both developing and adult tissues. Cells with identical receptors can respond differently to the same signal molecule because of differences in the internal machinery to which the receptors are coupled. Even when the entire signaling pathway is the same, cells can respond differently if they express different effector proteins at the ends of the pathways. The concentration of cyclic GMP in the smooth muscle cells lining the blood vessels of the penis is controlled by its rate of synthesis by guanylyl cyclase and its rate of degradation by cyclic GMP phosphodiesterase. The natural signal molecule NO binds to guanylyl cyclase and stimulates its activity, thereby increasing the concentration of cyclic GMP by increasing its rate of synthesis. The drug Viagra binds to cyclic GMP phosphodiesterase and inhibits its activity, thereby increasing the concentration of cyclic GMP by decreasing its rate of degradation. In general, in eucaryotic cells the binding of a single gene regulatory protein to a promoter is not sufficient to activate transcription. Activation of transcription depends on a collection of gene regulatory proteins, of which an activated nuclear receptor is just one. Different target cells can respond differently to an activated nuclear receptor, depending on the other gene regulatory proteins it expresses. The modifications of cholesterol to make steroid hormones increase the hydrophilicity of the molecules by removing the hydrocarbon tail and by introducing polar groups. These modifications make the molecules sufficiently hydrophilic to diffuse from their carrier molecules in the bloodstream to cells, but not so hydrophilic as to prevent their crossing the plasma membrane to enter cells. By contrast, cholesterol is so hydrophobic that it normally spends all its time in the membrane. A lipid that is virtually insoluble in water could not serve as a messenger because it could not move readily from one cell to another via the extracellular fluid. In the case of the steroid receptor, a one-to-one complex of steroid and receptor binds to DNA to activate transcription; thus, there is no amplification between ligand binding and activation of the target gene. By contrast, for ion-channel-coupled receptors, a single ion channel will let through thousands of ions in the time it remains open, which serves as an amplification step in this type of signaling pathway. In the normal signaling pathway PK2 activates PK1. If PK1 is permanently activated, a response is observed independent of the status of PK2. If the order were reversed--that is, PK1 activates PK2--signaling would not occur when PK2 carried an inactivating mutation. If the experimental setup were changed so that PK1 was mutationally inactive and PK2 carried an activating mutation, no response would be observed since PK2 would not be able to activate PK1, which carries the inactivating mutation. A339 1529 1530 1531 1532 1533 1534 1535 A. 4 B. 5 A340 C. D. E. F. G. 1536 Chapter 15: Mechanisms of Cell Communication 2 6 7 1 3 Phosphorylation/dephosphorylation offers a simple, universal solution to the problem of controlling protein activity. In a signaling pathway, the activities of several proteins must be rapidly switched from the off state to the on state, or vice versa. Attaching a negatively charged phosphate to a protein is an effective way to alter its conformation and activity. And it is an easy modification to reverse. It is a universal solution in the sense that one activity-- that of a protein kinase--can be used to attach a phosphate, and a second activity--a protein phosphatase--can be used to remove it. About 2% of the genes in the human genome encode protein kinases, which presumably arose by gene duplication and modification to create appropriate specificity. Because serines, threonines, and tyrosines are common amino acids on the surfaces of proteins, target proteins can evolve to have appropriate phosphorylation sites at places that will alter their conformations. Finally, phosphorylation/dephosphorylation provides a flexible response that can be adjusted to give rapid on/off switches or more long lasting changes. All of these attributes of phosphorylation/dephosphorylation are missing with allosteric regulators. While it is possible, in principle, for small molecules to turn proteins on or off, it is not a universal solution. Specific molecules would have to be `designed' for each target protein, which would require evolution of a metabolic pathway for synthesis and degradation of each regulatory molecule. Even if such a system evolved for one target protein, that specific solution would not help with the evolution of a system for any other target protein. In addition, regulation by binding of small molecules is very sensitive to the concentration of the regulator. For a monomeric target protein, the concentration of a small molecule would have to change by 100-fold to go from 9% bound to 91% bound--a minimal molecular switch (see Problem 3103). Few metabolites in cells vary by such large amounts. GTP-binding proteins are uniformly on when GTP is bound and off when GDP is bound; thus, GEFs turn GTP-binding proteins on and GAPs turn them off. The same is not true for protein kinases and phosphatases. Attachment of a phosphate will turn some target proteins on and others off. Indeed, attachment of a phosphate at one location in a protein can turn it on, while phosphorylation at a different location can turn the same protein off. Thus, while protein kinases throw the molecular switch, it is not always in the same direction. The use of a scaffolding protein to hold the three kinases into a signaling complex increases the speed of signal transmission and eliminates crosstalk between pathways; however, there is relatively little opportunity for amplification of the signal from the receptor to the third kinase. Freely diffusing kinases offer the possibility for greater signal amplification since the first kinase can phosphorylate many molecules of the second kinase, which in turn can phosphorylate many molecules of the third kinase. The speed of signal transmission is likely to be slower, unless the concentration of kinases (and the potential for amplification) is high enough to compensate for their separateness. Finally, free kinases offer the potential for spreading the signal to other signaling pathways and to other parts of the cell. The organization that a cell uses for a particular signaling pathway depends on what the pathway is intended to accomplish. 1537 1538 1539 A. 3. PH (pleckstrin homology) domains bind phosphorylated inositol phospholipids in the plasma membrane. B. 1. PTB (phosphotyrosine-binding) domains bind phosphotyrosines in target proteins. GENERAL PRINCIPLES OF CELL COMMUNICATION C. 1. SH2 (Src homology 2) domains bind phosphotyrosines in target proteins. D. 2. SH3 (Src homology 3) domains bind proline-rich sequences in target protein. 1540 1. If more than one effector molecule must bind to activate the target molecule, the response will be sharpened in a way that depends on the number of required effector molecules. At low concentrations of the effector most target proteins will have a single effector bound (and therefore be inactive). At increasing concentrations of effector the target proteins with the requisite number of bound effectors will rise sharply, giving a correspondingly sharp increase in the cellular response. 2. If the effector activates one enzyme and inhibits another enzyme that catalyzes the reverse reaction, the forward reaction will respond sharply to a gradual increase in effector concentration. This is a common strategy employed in metabolic pathways involved in energy production and consumption. 3. The above mechanisms give sharp responses, but a true all-or-none response can be generated if the effector molecule triggers a positive feedback loop so that an activated target molecule contributes to its own further activation. If the product of an activated enzyme, for example, binds to the enzyme to activate it, a self-accelerating, all-or-none response will be produced. A341 CALCULATIONS 1541 At a circulating concentration of hormone equal to 1010 M, about 1% of the receptors will have a bound hormone molecule {[RH]/[R]TOT = 1010 M/(1010 M + 108 M) = 0.0099}. Half of the receptors will have a bound hormone molecule when the concentration of hormone equals the Kd; that is, at 108 M {[RH]/[R]TOT = 108 M/(108 M + 108 M) = 0.5}. The relationships between concentration of ligand (hormone, in this case), Kd, and fraction bound are developed in Problem 3103. 1542 A. The concentrations of insulin in each of the unknown samples can be read directly from the standard curve by finding where the measured bound/free ratio intersects the calibration curve and reading the value for unlabeled insulin from the x axis. Sample 1 Sample 2 Sample 3 1.1 pg/mL 6.1 pg/mL 2.7 pg/mL B. The most accurate part of the curve lies between bound/free ratios of about 0.4 to 0.7. Below 0.4 the curve flattens out so that large differences in insulin concentration give small differences in bound/free ratios. C. The assay for human insulin will work even if the antibodies are directed against pig insulin, provided that the antibodies were raised in animals other than humans. As long as there are structural features (antigenic determinants) that are shared between the pig and human insulins--so that the antibodies raised in the pig react with human insulin--the assay will still be valid for measuring human insulin concentrations. The validity of RIA depends on the identical behavior of the antigen in the unknown samples with the antigen in known standards. Since the unknown sample and the standard are human insulins, the assay is valid. You may have wondered why the assay would not work if the antibodies were raised in humans. If antibodies against pig insulin were raised in humans, the antibodies would recognize those portions of pig insulin that are different from human insulin. That is because humans (like other animals) do not normally produce antibodies against their own proteins. As a A342 Chapter 15: Mechanisms of Cell Communication result, the antibodies generated in humans would not bind to human insulin and the assay would not work. Reference: Yalow RS (1978) Radioimmunoassay: a probe for the fine structure of biologic systems. Science 200, 12361245. 1543 A. To calculate the amount of insulin needed per assay, you first have to determine the number of radioactive iodine atoms required for an assay. Given that there is one radioactive iodine atom per insulin molecule, you can then convert the number of radioactive iodine atoms into picograms of insulin per assay. On average, two atoms of radioactive iodine (half-life of 7 days) will give rise to one disintegration in a week. Because each disintegration has a 50% probability of being detected as a count, 2 107 disintegrations per week will be required to reach 1000 counts per minute, the limit of detection of the RIA. disint = 1000 counts 2 disint 60 min 24 hr 7 days min week count hr day week = 2.0 107 disintegrations per week Thus, 1000 cpm corresponds to 4.0 107 atoms of radioactive iodine. The weight of this number of insulin molecules is 0.76 picograms. insulin = 4.0 107 molecules = 0.76 pg B. For optimal sensitivity of RIA, the amounts of the tracer and unknown should be the same. Thus, given the starting assumptions, at optimum sensitivity you will be able to detect 0.76 pg of unlabeled insulin. 1544 A. A cell will contain 100,000 molecules of A and 10,000 molecules of B at these rates of synthesis and average lifetimes. The number of molecules equals the rate of synthesis times the average lifetime. For A, the number of molecules = (1000 molecules/sec)(100 sec). B. After 1 second, the number of A molecules will have increased by 10,000 to a total of about 110,000 molecules per cell--a 10% increase over the number present before the boost in synthesis. The number of B molecules will also increase by 10,000 to a total of about 20,000, which represents a doubling of its concentration. (For simplicity, the breakdown in A and B over a one-second interval can be neglected.) C. Because of its larger proportional increase in the short term, molecule B would be the preferred signal molecule. Note that after a sufficiently long time both molecules would increase by a factor of 10 in response to a 10-fold increase in rate of synthesis. For signaling it is the rapidity of the change that is most critical. This calculation illustrates the surprising principle that the time it takes to switch a signal on is determined by the lifetime of the signal molecule. 12 moles 11,446 g 10 pg g 6 1023 molecules mole DATA HANDLING 1545 A. Only pool 3 has colonies that have bound the radioactive ligand, as shown by the black circles. Thus, one or more of the plasmids in this pool contain the cDNA for the receptor. B. Not all of the colonies on the dish transfected with pool-3 plasmid DNA are radioactive because not all of the transfected cells received the plasmid carrying the receptor cDNA. In a pool of 1000 plasmids, it is expected that only GENERAL PRINCIPLES OF CELL COMMUNICATION a few of the transfected cells will have received a plasmid carrying the receptor cDNA. C. This experiment identifies a pool of plasmids, at least one of which carries the cDNA for the receptor. The usual next step is to subdivide this pool into smaller pools containing, for example, 10 plasmids each and then repeat the transfection. Once again a positive pool is identified by autoradiography. Finally, individual plasmids from the small pool are isolated and tested by transfection and autoradiography. At this final stage a positive autoradiograph identifies a plasmid that carries the receptor cDNA. 1546 The results shown in Figure 155 are consistent with chemical signaling: in particular, the limited range of the signal (experiments A and B); the ability to bend around corners (experiment B); the ability to penetrate a semipermeable membrane but not a glass coverslip (experiments A and C); and the profound influence of a gentle stream of liquid (experiment D). These experiments go a long way toward demonstrating a secreted chemical signal, but the final proof is the isolation of the signal molecule, which was accomplished many years afterward and shown to be cyclic AMP. Reference: Bonner JT & Savage LJ (1947) Evidence for the formation of cell aggregates by chemotaxis in the development of the slime mold Dictyostelium discoideum. J. Exp. Zool. 106, 126. 1547 A. Both cell lines appear to contain a glucocorticoid receptor, based on the dexamethasone-induced increases in CAT activity. When transfected with the construct carrying the glucocorticoid-responsive viral enhancer, cell line 1 showed a 5-fold increase and cell line 2 showed an 80-fold increase in CAT activity in the presence of dexamethasone. B. Cell line 1 differed significantly from cell line 2 in the level of CAT activity detected in the absence of dexamethasone after transfection of the construct containing the glucocorticoid-responsive enhancer. Either of the following explanations for this difference is reasonable. (1) Cell line 1 contains a mutant glucocorticoid receptor that can partially activate the glucocorticoid enhancer in the viral segment in the absence of glucocorticoid. (2) Cell line 1 contains a tissue-specific protein that recognizes and stimulates a second (nonglucocorticoid-responsive) enhancer on the viral segment. C. These two potential explanations make different predictions for the outcome of experiments in which a variety of shorter viral segments are tested for CAT activity in the two cell lines. If the difference between the cell lines were due to different glucocorticoid receptors that recognize the same enhancer, then shorter pieces of the viral segments would all behave in one of two ways: (1) they would not contain the glucocorticoid enhancer and would not be activated by dexamethasone in either cell line, or (2) they would contain the glucocorticoid enhancer and would show the same differential response to dexamethasone as the intact viral segment. On the other hand, if the difference between the cell lines were due to different enhancer elements (glucocorticoid responsive and nonglucocorticoid responsive), then it should be possible to separate the different enhancers onto different DNA segments. Segments that contain only the glucocorticoid enhancer should give identical responses in the two cell lines; segments that contain only the nonglucocorticoid enhancer should give a stimulation in cell line 1 (that is independent of glucocorticoid) but no stimulation in cell line 2. Actual experiments of this kind indicate that the viral segment contains a second enhancer that is specifically activated in cell line 1. Results of experiments such as illustrated in this problem reinforce the idea that genes are controlled in a tissue-specific fashion by the interplay of multiple regulatory factors. Reference: DeFranco D & Yamamoto KR (1986) Two different factors act separately or together to specify functionally distinct activities at a single transcriptional enhancer. Mol. Cell. Biol. 6, 9931001. A343 A344 Chapter 15: Mechanisms of Cell Communication 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. A C G T C G T A A C C C C G T 2 5 2 2 C G C G G A A T C A T 3 3 3 2 A A G G G G T G T T C 2 1 5 3 A C T A A A G T A C C 5 3 1 2 G T T T A G A T G A A 4 0 3 4 G A T G A C T A T G A 4 1 3 3 A A T T A T T T G T A 4 0 1 6 T A G T A G T C C T T C T 1 3 2 5 G A A A A A A A A A A * 0 1 0 A G G G G G G G G G G G 0 0 * 0 G G A A A A A C G A A A 8 1 2 0 A A A A C A T A A T T T 6 1 0 4 A T C C C A C C C C C C C 1 * 0 0 C A A A A A A A A A A C * 1 0 0 A G T G G C G A T G G T 1 1 6 3 G T T T T T A A C T A T 3 1 0 7 T G A T G C A A G C T T 3 2 3 3 G T T G A C A T A T T 3 1 2 5 C T G T G C G C G A T 1 3 4 3 T C T T A T A C A A G 4 2 1 4 G T A T G T G T T A C 2 1 3 5 G C A C C T C T C A T 2 5 1 3 A C C C T G G T A G T 2 3 3 3 C A C T C A T C G C T 2 5 1 3 T A A G A T G C A A T 5 1 2 3 A A A A G A T A T A A 8 0 1 2 A A A A C A C G G T A A 6 2 2 1 A T A A T T C C G A A T 4 2 1 4 A C A T C A C A A A C 6 4 0 1 A C G T C G A A A C A G T 4 2 3 2 Figure 1538 Frequencies of G, C, A, and T at each of the positions in the bound sequences (Answer 1548). An asterisk (*) and shading indicates nucleotides that are present in 10 or more of the sequences. The consensus sequence shows those nucleotides that are represented in more than half the sequences in this set of 11; whenever, two nucleotides are listed, the lower one was also prevalent. consensus 1548 A. The glucocorticoid response elements are located primarily between nucleotides 240 and 50 on the LTR. The electrophoretic pattern of the starting mixture shows that all the fragments from the mutant LTRs are equally represented initially. In the electrophoretic pattern of the bound mixture, the fragments are unequally represented: the fragment from the 240 deletion is nearly equal to the fragment from the complete LTR; the fragments from the 50 and +50 deletions are totally absent; and the fragments from deletions 202 and 137 are intermediate. The apparent gradient of binding suggests that several glucocorticoid response elements are located between nucleotides 240 and 50. This suggestion was confirmed by experiments like the ones summarized in part B. B. A consensus sequence for glucocorticoid receptor binding can be obtained by listing the frequencies of G, C, A, and T at each position in the bound sequences, as shown in Figure 1538. Individual nucleotides that are present in more than half the sequences are shown immediately below the frequencies. Two consensus glucocorticoid-binding sequences (AGAA/TCAGT/A and AGAACA) have been reported in the literature. It turns out that the common nucleotides that are outside the central group in Figure 1538 are not part of the true binding site, as additional experiments have shown. References: Payvar F, DeFranco D, Firestone GL, Edgar B, Wrange O, Okret S, Gustafsson J-A & Yamomoto KR (1983) Sequence-specific binding of glucocorticoid receptor to MMTV DNA at sites within and upstream of the transcribed region. Cell 35, 381392. Scheidereit C, Geisse S, Westphal HM & Beato M (1983) The glucocorticoid receptor binds to defined nucleotide sequences near the promoter of mouse mammary tumor virus. Nature 304, 749752. 1549 A. The cycloheximide-induced alteration of the puffing pattern is due to its effect on protein synthesis. The result indicates that newly synthesized proteins are required to turn off the early puffs and to turn on the late puffs. Presumably, the proteins are synthesized from the early puffs. The shut-off of transcription from the intermolt puffs is insensitive to cycloheximide treatment. This suggests that the receptorecdysone complex turns off these puffs directly. B. The immediate regression of the early puffs upon ecdysone removal indicates that the ecdysonereceptor complex is required continuously to keep the genes turned on. The premature activation of the late puffs under these conditions was unexpected. If activation of the late puffs depended only on products of the early puffs, then they should be turned on at the same time (or even delayed due to a lower level of early product). The premature activation suggests that the receptorecdysone complex actually functions as an GENERAL PRINCIPLES OF CELL COMMUNICATION INTERMOLT GENES zero time OFF OFF EARLY GENES LATE GENES A345 Figure 1539 Diagram relating ecdysonereceptor (ER) binding to the pattern of gene activity (Answer 1549). ER 1 hour OFF ER ER OFF ER 5 hours OFF ER ER ER 8 hours OFF OFF inhibitor, delaying activation until the concentrations of the presumptive early-puff products reach some critical level. Removal of ecdysone allows the puffs to be induced at lower concentrations of early products. C. These experimental observations are summarized in the diagram shown in Figure 1539. The ecdysonereceptor complex binds to regulatory regions of intermolt, early, and late puffs. Binding at intermolt puffs turns them off, binding at early puffs turns them on, and binding at late puffs keeps them off. Products from one or more early puffs bind at the regulatory regions of early and late puffs, ultimately turning off the early puffs and turning on the late puffs. Reference: Ashburner M, Chihara C, Meltzer P & Richards G (1973) Temporal control of puffing activity in polytene chromosomes. Cold Spring Harbor Symp. Quant. Biol. 38, 655662. 1550 The analysis of individual frog oocytes shows clearly that the response to progesterone is all-or-none, with no oocytes having a partially activated MAP kinase. Thus the graded response in the population results from an all-ornone response in individual oocytes, with different mixtures of fully mature or immature oocytes giving rise to intermediate levels of MAP kinase activation (Figure 1540). It is not so clear why individual oocytes respond differently to different concentrations of progesterone, although there is significant variability among oocytes in terms of age and size (and presumably in the number of progesterone receptors and the concentrations of components of the MAP kinase signaling module and downstream targets). Whether a graded response in a population of cells indicates a graded response in each cell or a mixture of all-or-none responses is a question that arises in many contexts in biology. Reference: Ferrell JE & Machleder EM (1998) The biochemical basis of an allor-none cell fate switch in Xenopus oocytes. Science 280, 895898. 1551 A. The difference in binding of CGP-12177 and dihydroalprenolol to extracts of isoproterenol-treated cells suggests that the ligand-binding sites on some receptors are not directly exposed in the lysate but instead are enclosed by membrane. Dihydroalprenolol, being hydrophobic, can cross membranes and thus bind to all receptors. By contrast, CGP-12177, which is hydrophilic, cannot cross membranes and thus can bind only to exposed receptors. GRADED RESPONSE ALL-OR-NONE RESPONSE active MAP kinase (%) 100 50 0 0.001 0.01 0.1 1 10 progesterone (mM) Figure 1540 Graded or all-or-none responses in individual oocytes that give rise to a graded response in the population (Answer 1550). A346 Chapter 15: Mechanisms of Cell Communication These results suggest that there are two populations of vesicles in the cell lysates: one with receptors facing outward and the other with receptors facing inward. The two populations can be separated on sucrose-density gradients. The presence of 5-nucleotidase in one population but not in the other indicates that the CGP-12177-binding vesicles formed from fragments of the plasma membrane. B. The parallel between CGP-12177 binding and hormone-dependent adenylyl cyclase activity suggests that the two are related; that is, the receptors that bind to CGP-12177 are the same ones that can activate adenylyl cyclase. Finding that CGP-12177 binding is specifically associated with vesicles formed from plasma membrane fragments supports this suggestion. The other population of vesicles presumably represents internal vesicles formed by endocytosis, which would yield vesicles with the ligand-binding sites in the interior. These experiments suggest that isoproterenol-induced desensitization results from internalization of the receptors, making them unavailable for interaction with hormone and separating them from the proteins that couple them to adenylyl cyclase. Reference: Hertel C, Muller P, Portenier M & Staehelin M (1983) Determination of the desensitization of b-adrenergic receptors by [3H]CGP-12177. Biochem. J. 216, 669674. 1552 A. If phosphorylation of the two subunits occurs independently and at equal rates, four different types of receptor will exist: nonphosphorylated receptor, receptor phosphorylated on the g subunit, receptor phosphorylated on the d subunit, and receptor phosphorylated on both subunits. At 0.8 mole phosphate/mole receptor, each subunit would be 40% phosphorylated and 60% nonphosphorylated. Thus the ratio of the various receptor forms would be 36% with no phosphate (0.6 0.6), 24% with only the g subunit phosphorylated (0.6 0.4), 24% with only the d subunit phosphorylated (0.6 0.4), and 16% with both subunits phosphorylated (0.4 0.4). At 1.2 mole phosphate/mole receptor, the ratios would be: 16% with no phosphate, 24% with the g subunit phosphorylated, 24% with the d subunit phosphorylated, and 36% with both subunits phosphorylated. B. These experiments suggest that desensitization requires only one phosphate per receptor and that phosphorylation of either the g or the d subunit is sufficient for desensitization. For both preparations, the fraction that behaved like the untreated receptor matched best the fraction calculated to carry no phosphate: 36% versus 36% at 0.8 mole phosphate/mole receptor and 18% versus 16% at 1.2 mole phosphate/mole receptor. This result suggests that phosphorylation of either subunit is sufficient to promote desensitization. If a specific subunit were required to be phosphorylated, then the expected fractions behaving like the untreated receptor would have been 60% (24% + 36%) at 0.8 mole phosphate/mole receptor and 40% (24% + 16%) at 1.2 mole phosphate/mole receptor. Reference: Huganir RL, Delcour AH, Greengard P & Hess GP (1986) Phosphorylation of the nicotine acetycholine receptor regulates its rate of desensitization. Nature 321, 774776. SIGNALING THROUGH G-PROTEIN-COUPLED CELLSURFACE RECEPTORS (GPCRs) AND SMALL INTRACELLULAR MEDIATORS DEFINITIONS 1553 1554 Stimulatory G protein (Gs) Trimeric GTP-binding protein (G protein) SIGNALING THROUGH GPCRS AND SMALL INTRACELLULAR MEDIATORS 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 Calmodulin Cyclic AMP phosphodiesterase G-protein-coupled receptor (GPCR) GPCR kinase (GRK) Ryanodine receptor Phospholipase C-b (PLCb) Regulator of G protein signaling (RGS) Inositol 1,4,5-trisphosphate (IP3) Cyclic AMP-dependent protein kinase A (PKA) Protein kinase C (PKC) Rhodopsin Ca2+/calmodulin-dependent kinase (CaM-kinase) CRE-binding (CREB) protein A347 TRUE/FALSE 1568 1569 False. It is the substrates for PKA, not PKA itself, that differ in different cell types. True. The activity of a population of protein molecules whose activity is regulated by phosphorylation depends on the percentage of the molecules that are phosphorylated, which in turn depends on the relative rates of phosphate addition and removal. True. Nuclear receptors with a bound ligand bind to DNA sequences in the genome to activate (or inhibit) a specific gene; thus, there is a one-to-one correspondence between the signal and the response. By contrast, signaling pathways that involve enzymes or ion channels can significantly amplify a signal. One activated protein kinase, for example, can phosphorylate many molecules of its target protein. 1570 THOUGHT PROBLEMS 1571 The mutant G protein would be constantly active. Each time the a subunit hydrolyzed GTP to GDP, the GDP would spontaneously dissociate, allowing GTP to bind and reactivate the a subunit. Normally, GDP is tightly bound by the a subunit, which keeps the G protein in its inactive state until release of GDP is stimulated by interaction with an appropriate GPCR. Each time a G protein picked up a nonhydrolyzable analog of GTP it would be locked into its active form, from which it could not escape by the usual route of GTP hydrolysis. In the absence of adrenaline, most of the G protein would be in the GDP-bound form, which is released only slowly in the absence of stimulation by an activated receptor. Thus, you might expect a slow activation of G protein and a correspondingly slow increase in glycogen breakdown. In the presence of adrenaline, GDP would be rapidly released and nonhydrolyzable GTP would be bound. A brief exposure to adrenaline normally would stimulate glycogen breakdown for a short time, until adrenaline was removed and the signaling pathway was turned off. However, in the presence of a nonhydrolyzable analog of GTP the pathway would remain on even after adrenaline was removed. Thus, the nonhydrolyzable analog would cause a prolonged response to a pulse of adrenaline. RGS proteins are GAPs that have a critical role in shutting off G protein responses in animals and yeasts. They stimulate the GTPase activity of G 1572 1573 A348 Chapter 15: Mechanisms of Cell Communication proteins, converting them to their inactive (GDP-bound) form, and thereby limit the duration of a response. adenine ring 1574 The `cyclic' in cyclic AMP refers to the ring of atoms formed by the phosphorus atom, its two oxygen atoms, and the carbons at the 3, 4, and 5 positions of the ribose sugar (Figure 1541). The ball-and-stick representations above and below the chemical formula give a more realistic representation of the chemical structure. The six-member phosphodiester ring is fused to the five-member ribose ring, forming a fairly planar structure that resembles the adenine ring in size and shape. In the more common representation (center) the phosphodiester ring looks very strained, but in reality it's not. Rapid breakdown keeps cyclic AMP levels low, typically about M. An extracellular signal can increase this concentration more than 20-fold in seconds. The lower the cyclic AMP level, the larger the proportional increase achieved upon activation of adenylyl cyclase, which makes new cyclic AMP. By analogy, if you have $100 in the bank, and you deposit another $100, you have doubled your wealth. If you have only $10 to start with and you deposit $100, you have increased your wealth 10-fold, a much larger proportional increase resulting from the same deposit. Since b-adrenergic receptors are coupled to adenylyl cyclase through a G protein, a reasonable guess would be GTP. Activation of the receptor by isoproterenol would stimulate the G protein to exchange bound GDP for free GTP In the absence of free GTP the G protein would not be activated and . thus it would not activate adenylyl cyclase. The requirement for GTP in the receptor-mediated activation of adenylyl cyclase was one of the original clues that led to the discovery of G proteins. Any mutation that generated a regulatory subunit incapable of binding to the catalytic subunit would produce a permanently active PKA. When the catalytic subunit is not bound to the regulatory subunit, it is active. Two general types of mutation in the regulatory subunit could produce a permanently inactive PKA. A regulatory subunit that was altered so that it could bind the catalytic subunit, but not bind cyclic AMP, would not release the catalytic subunit, rendering PKA permanently inactive. Similarly, a mutant regulatory subunit that could bind cyclic AMP, but not undergo the conformational change needed to release the catalytic subunit, would permanently inactivate PKA. Both glycogen and triglyceride breakdown depend on cyclic AMP-dependent signaling pathways. By inhibiting cyclic AMP phosphodiesterase, caffeine increases cyclic AMP levels to promote glycogen breakdown in muscle and liver and to promote triglyceride breakdown in fat cells. The effects on glycogen are not so important as the effects on triglycerides. By breaking down triglycerides to release fatty acids earlier in the race, runners decrease their dependence on carbohydrate oxidation at that stage, thereby preserving carbohydrates for use throughout the race. IP3-triggered Ca2+ responses are terminated in two ways: by adding or removing phosphates from IP3 and by pumping Ca2+ outside of the cell or into the ER. Because the intracellular concentration of Ca2+ is so low, an influx of relatively few Ca2+ ions leads to large changes in its cytosolic concentration. Thus, a 10-fold increase in Ca2+ can be achieved by raising the concentration of Ca2+ into the micromolar range, which would require entry of relatively few ions into the cytosol. By contrast, many more Na+ ions (104 more) would need to enter the cytosol to change its concentration by the same amount. In muscle, more than a 10-fold change in Ca2+ can be achieved in microseconds by releasing Ca2+ from the intracellular stores of the sarcoplasmic reticulum, a task that would be difficult to accomplish if changes in the millimolar range were required. 107 phosphodiester ring N N 5 CH2 O O 4 H 3 O H NH2 1575 N O H 1 H 2 OH N P O 1576 ribose ring adenine ring phosphodiester ring 1577 Figure 1541 Chemical formula and balland-stick representations of cyclic AMP (Answer 1574). The ball-and-stick are rotated 90 relative to each other to illustrate the similarities between the adenine ring and the fused ribose and phosphodiester rings. 1578 1579 1580 SIGNALING THROUGH GPCRS AND SMALL INTRACELLULAR MEDIATORS 1581 EGTA, by chelating Ca2+, would be expected to interfere with signaling pathways that use Ca2+ as a second messenger. Glucagon triggers glycogen breakdown in liver via a cyclic AMP pathway and thus would not be affected by EGTA. By contrast, vasopressin signals glycogen breakdown via a Ca2+ pathway and would be blocked by injection of EGTA. The time the catalytic kinase subunit spends in its active conformation depends on the extent to which its regulatory subunits are modified. Each modification by phosphorylation or by Ca2+ binding nudges the equilibrium toward the active conformation of the kinase subunit; that is, each modification increases the time spent in the active state. By summing the inputs from multiple pathways in this way, phosphorylase kinase integrates the signals that control glycogen breakdown. When CaM-kinase II is exposed to Ca2+/calmodulin, it becomes an active protein kinase and phosphorylates itself (among other targets). In its phosphorylated state CaM-kinase II remains active even in the absence of Ca2+, thereby prolonging the duration of the kinase activity beyond that of the initial activating Ca2+ signal. Its self-phosphorylation allows it to `remember' its exposure to Ca2+/calmodulin. It finally `forgets' when a protein phosphatase removes the phosphate, shutting off its activity. A349 1582 1583 1584 A. An inhibitor of cyclic GMP phosphodiesterase would prevent the reduction in cyclic GMP that normally occurs in response to light activation of rhodopsin. High levels of cyclic GMP would keep the Na+ channels open, preventing the membrane hyperpolarization that is essential for the visual response. B. A nonhydrolyzable analog of GTP would lead to prolonged activation of transducin in response to activated rhodopsin. Continued activation of transducin would keep cyclic GMP phosphodiesterase high, which would lead in turn to a protracted decrease in cyclic GMP, a prolonged hyperpolarization of the membrane, and an extended visual response. C. An inhibitor of rhodopsin-specific kinase would prolong the visual response by increasing the signaling lifetime of the activated form of rhodopsin. Normally, rhodopsin-specific kinase adds a phosphate to the cytoplasmic tail of rhodopsin, inhibiting the interaction of activated rhodopsin with transducin. 1585 Because these patients recover abnormally slowly from a flash of bright light, it is likely that they are defective in the return of activated rhodopsin to its inactive state. This recovery process begins with the phosphorylation of the cytosolic tail of rhodopsin by rhodopsin-specific kinase. Phosphorylated rhodopsin is then bound by arrestin. Additional reactions remove the phosphate and replace the all-trans-retinal with 11-cis-retinal, finally regenerating a rhodopsin molecule that is ready for another cycle of phototransduction. Thus far, patients with Oguchi's disease have been found to have defects in the gene for rhodopsin-specific kinase or in the gene for arrestin. The signal is amplified at four points in the pathway. (1) A single, activated b-adrenergic receptor can activate many copies of the G protein, acting as a GEF to promote GDP release and GTP binding. Adenylyl cyclase, by contrast, is activated stoichiometrically; that is, adenylyl cyclase is active only so long as the activated Ga subunit is bound. (2) While active, a single molecule of adenylyl cyclase can convert many molecules of ATP to cyclic AMP. The next step--activation of PKA--is stoichiometric: four cyclic AMP molecules activate two PKA molecules. (3) A single molecule of PKA can add phosphates to many molecules of phosphorylase kinase. (4) A single molecule of phosphorylase kinase can add phosphates to many molecules of glycogen phosphorylase. Glycogen phosphorylase represents the end of the signaling pathway and the beginning of the biochemical pathway for utilization of the energy stored in glycogen (thus, its ability to cleave many molecules of glucose from glycogen is not considered an amplification step in the signaling pathway). 1586 A350 1587 Chapter 15: Mechanisms of Cell Communication The b-adrenergic receptor is turned off directly by the conformational change that occurs when adrenaline is no longer bound. The Ga subunit becomes inactive when it hydrolyzes the attached GTP to GDP, which allows it to reassociate with the bg subunits. Adenylyl cyclase becomes inactive as soon as Ga dissociates. Cyclic AMP is constantly being converted to AMP by cyclic AMP phosphodiesterase. In the absence of its continued synthesis, cyclic AMP quickly returns to its pre-stimulated level. At low concentrations, cyclic AMP dissociates from the regulatory subunits of PKA, which rebind the catalytic subunits to turn off PKA. In the absence of ongoing phosphorylation of phosphorylase kinase, a protein phosphatase removes the phosphates, turning off phosphorylase kinase. Similarly, a protein phosphatase quickly removes phosphates from glycogen phosphorylase, thereby turning it off. 40 bound (cpm/mg 103) 30 20 10 0 0 2 4 6 total specific 3H-alprenolol nonspecific 8 (nM) 10 3H-alprenolol CALCULATIONS 1588 A. The specific binding curve is obtained by subtracting the nonspecific curve from the total. As illustrated in Figure 1542, the specific binding curve reaches a plateau above 4 nM alprenolol. Thus, the b-adrenergic receptors are saturated with alprenolol above this concentration. B. There are 1500 b-adrenergic receptors per frog erythrocyte. Since one alprenolol binds per receptor, the number of bound alprenolol molecules is equal to the number of receptors. At saturation, 20,000 cpm of alprenolol binds per mg of erythrocyte membrane (Figure 1542). Thus, the amount of bound alprenolol is 20 103 cpm mmol 6 1020 molecules mg bound 13 alprenolol = 8 108 erythrocyte mg 10 cpm mmol = 1500 molecules per erythrocyte Since one molecule of alprenolol binds per b-adrenergic receptor, there are 1500 b-adrenergic receptors per erythrocyte. Reference: Lefkowitz RJ, Limbird LE, Mukherjee C & Caron MG (1976) The b-adrenergic receptor and adenylate cyclase. Biochim. Biophys. Acta 457, 139. 1589 A. Since 5.5 mmol of GppNp were bound per mole of total rhodopsin and 0.0011% of the rhodopsin was activated, 5.5 mmol GppNp were bound per 0.011 mmol of activated rhodopsin. Thus 500 molecules (5.5 mmol/ 0.011 mmol) of transducin were activated per activated rhodopsin. This experiment indicates that the first stage of amplification during visual excitation is achieved through the effect of activated rhodopsin on transducin. The calculation implies that there is very little additional amplification in the interaction of transducin with cyclic GMP phosphodiesterase. B. These relative binding affinities indicate that transducinGDP binds to activated rhodopsin and transducinGTP is released, permitting activated rhodopsin to interact with many transducin molecules, as expected by the deduced amplification mechanism. Furthermore, the tight binding of transducinGTP to cyclic GMP phosphodiesterase suggests that activation of phosphodiesterase is stoichiometric; that is, one molecule of activated transducin activates one molecule of cyclic GMP phosphodiesterase. References: Fung BK-K & Stryer L (1980) Photolyzed rhodopsin catalyzes the exchange of GTP for bound GDP in retinal rod outer segments. Proc. Natl Acad. Sci. USA 77, 25002504. Fung BK-K, Hurley JB & Stryer L (1981) Flow of information in the light-triggered cyclic nucleotide cascade of vision. Proc. Natl Acad. Sci. USA 78, 152156. Figure 1542 Specific binding of alprenolol to frog erythrocyte membranes (Answer 1588). SIGNALING THROUGH GPCRS AND SMALL INTRACELLULAR MEDIATORS A351 DATA HANDLING 1590 A. While all the mutations cause a sterile phenotype, loss of the a-subunit gene causes proliferation to arrest even in the absence of the a-factor pheromone. Since arrested proliferation is one of the normal responses to the binding of the a-factor pheromone to its receptor, it must be that the bg subunits, rendered free by the absence of the a subunit, normally transmit the mating signal to the downstream effector molecules. This interpretation is reinforced by the phenotypes of the double mutants, which no longer show the arrestedproliferation phenotype characteristic of the simple a-subunit deletion. (Note that the simple deletion mutants of the a-subunit described here cannot be isolated in reality because they do not proliferate under normal conditions. The phenotype of cells missing a-subunit activity was determined in other ways. Can you guess how?) B. 1. An a subunit that can bind GTP but not hydrolyze it would be expected to exhibit an arrested-proliferation, sterile phenotype in the presence or absence of the a-factor pheromone. An a subunit with bound GTP releases the bg subunit, which would trigger the downstream pathway leading to proliferation arrest. 2. An a subunit that cannot be myristoylated would not be properly localized to the membrane. Since the bg subunit is localized to the membrane by a lipid group attached to the g subunit, the cytosolic a subunit would not interact properly with the membrane-bound bg subunit. As a result, the bg subunit would trigger the downstream pathway, leading to an arrested-proliferation, sterile phenotype in the presence or absence of the a-factor pheromone. 3. An a subunit that could not bind to the pheromone receptor could not `read' the signal from an a-factor-stimulated receptor and would therefore be sterile. It would remain in its GDP-bound form, and thus, remain complexed with the bg subunits. Such a mutant would be expected to display a normal-proliferation phenotype in the absence and presence of the a-factor pheromone. Reference: Kurjan I (1992) Pheromone response in yeast. Annu. Rev. Biochem. 61, 10971129. 1591 A. Your experiments reproduce the paradoxical result you observed in brain slices. Cells transfected with the dopamine receptor, type II adenylyl cyclase, and as* have high levels of cyclic AMP that are increased further by stimulating the dopamine receptor with quinpirole. Yet quinpirole has no effect on cyclic AMP levels in cells that are missing type II adenylyl cyclase or as*. B. Pertussis toxin eliminates the enhanced synthesis of cyclic AMP by quinpirole. Pertussis toxin is an enzyme that ADP ribosylates ai so that Gi can no longer interact with its receptors. In the absence of this interaction the ai subunit cannot exchange GTP for bound GDP and cannot dissociate from the bg subunit. Thus pertussis toxin blocks the enhancing signal from the quinpirole-activated dopamine receptor. C. The simplest interpretation of your results is that type II adenylyl cyclase is maximally activated by a combination of as* and bg subunits. The bg subunit is released from Gi upon activation of the dopamine receptor by quinpirole binding. The actual molecular explanation for the enhancing effects of bg subunits is unknown. Since free bg subunits have no effect on cyclic AMP levels in the absence of as*, they do not stimulate type II adenylyl cyclase unless it has already bound an as subunit. One possible explanation is that the binding of an as to type II adenylyl cyclase causes a conformational change in the cyclase that exposes a binding site for the bg subunit. Binding of the bg subunit to that site could help lock the type II adenylyl cyclase into an active state, leading to enhanced or prolonged synthesis of cyclic AMP. A352 Chapter 15: Mechanisms of Cell Communication D. Expression of the cDNA for the a subunit of transducin eliminates the enhanced synthesis of cyclic AMP by quinpirole. Like pertussis toxin, the a subunit of transducin blocks the signal from the quinpirole-activated dopamine receptor; however, its effects are specific for the bg subunit. This was an important control experiment in the original studies because it pinpointed the bg subunit as the key component of Gi responsible for enhanced cyclic AMP levels in quinpirole-treated cells. With the results from pertussis toxin alone, it could be argued that the ai subunit was having some anomalous enhancing effect in the presence of as*. References: Federman AD, Conklin BR, Schrader KA, Reed RR & Bourne HR (1992) Hormonal stimulation of adenylyl cyclase through Gi protein bg subunits. Nature 356, 159161. Tang W-J & Gilman AG (1991) Type-specific regulation of adenylyl cyclase by G protein bg subunits. Science 254, 15001503. 1592 A. If PKA were essential in hamster cells, it would have been impossible to isolate mutants that lack the enzyme, since they would not survive. Thus PKA is not essential to these hamster cells (nor is it essential in several other cell lines in which such mutants have been isolated). (Note that you cannot conclude that PKA is dispensible for the organism. There are many examples of enzyme defects that have minimal consequences for cells in culture but that severely affect the intact organism. B. Mutations that eliminate the catalytic subunit would be unresponsive to high levels of cyclic AMP and therefore resistant to cyclic AMP. These mutations would be recessive: in the presence of the wild-type catalytic subunit, as in the hybrids formed by fusion between the PKA-negative and the parental cells, cyclic AMP responsiveness would be restored. (It might seem that mutations that eliminate the regulatory subunits would be the same. Without the regulatory subunits, however, PKA would be active at all times, which is a lethal condition--excess PKA activity is the reason that these cells are killed by high levels of cyclic AMP. Thus, cell lines without regulatory subunits would not have been isolated in the first place.) Dominant mutations are somewhat more difficult to explain. In general, dominance indicates an altered activity rather than lack of activity. Dominant mutations have been found in both the regulatory and catalytic subunits. A possible dominant mutation in the regulatory subunit is one that increases its affinity for the catalytic subunit. Such a mutation might plausibly respond only to high levels of cyclic AMP (required to displace the tightly bound regulatory subunit), and it would be dominant because the mutant subunits would bind the catalytic subunits at the low cyclic AMP concentrations that would displace the normal regulatory subunits. Dominant mutations in the catalytic subunit are more difficult to explain. One possibility is a mutant catalytic subunit that binds its regulatory subunits more tightly, giving an altered responsiveness to cyclic AMP. Its dominance might be understood if a combination of a mutant catalytic subunit with a normal catalytic subunit rendered the heterodimer mutantlike in its binding to the regulatory subunits. If this were the case, an even mixture of mutant and normal catalytic subunits would be expected to have only onequarter the normal PKA activity (only one-quarter of the catalytic dimers would have two wild-type subunits). C. These experimental results support the contention that all the cyclic AMP inhibitory effects on cell proliferation are mediated through PKA. These experiments do not address the possibility that cyclic AMP-binding proteins play other critical roles in this hamster cell line. Any role that does not contribute to the inhibition of cell proliferation would have been missed in this assay. Reference: Gottesman MM (1985) Genetics of cyclic-AMP-dependent protein kinases. In Molecular Cell Genetics (MM Gottesman ed), pp 711743. New York: Wiley. SIGNALING THROUGH GPCRS AND SMALL INTRACELLULAR MEDIATORS 1593 A. The concentration of cyclic AMP is a balance between its synthesis (by adenylyl cyclase) and its breakdown (by cyclic AMP phosphodiesterase). Since dunce flies are missing one form of cyclic AMP phosphodiesterase, they hydrolyze cyclic AMP more slowly than usual, which results in an elevated level. B. Homozygous duplications would be expected to produce twice the activity of normal flies, and homozygous deletions would be expected to give no activity. The results with the duplications and deletions of the dunce gene are contrary to this expectation because there are two different genes coding for two distinct cyclic AMP phosphodiesterases in flies, as revealed by the sucrosegradient analysis in Figure 1521. Since the activities are about equal in normal flies, a doubling of one activity increases the total activity to about 1.5 times normal and removal of one activity lowers the total to half normal. C. Since caffeine is an inhibitor of cyclic AMP phosphodiesterase, it would be expected to raise the level of cyclic AMP in flies. If the increased level of cyclic AMP were responsible for the learning defect associated with the dunce mutation, caffeine would be expected to impair learning ability. Experiments show that caffeine does indeed interfere with learning in flies. This result strongly supports the idea that the learning defect in dunce mutants is caused by the loss of phosphodiesterase activity and not, for example, by a regulatory defect that shuts off phosphodiesterase and independently interferes with a learning pathway. It is worth pointing out the element of luck in this story. It would have been very difficult for the scientists working on dunce to deduce the biochemical nature of the gene product (though its sequence would have revealed its function ultimately); similarly, the scientists interested in cyclic nucleotide deficiencies had not noticed a learning defect. The story emerged much more rapidly than normal because the two groups happened to work at the same institution. References: Byers D, Davis RL & Kiger JA (1981) Defect in cyclic AMP phosphodiesterase due to the dunce mutation of learning in Drosophila melanogaster. Nature 289, 7981. Chen C-N, Denome S & Davis RL (1986) Molecular analysis of cDNA clones and the corresponding genomic coding sequences of the Drosophila dunce+ gene, the structural gene for cyclic AMP phosphodiesterase. Proc. Natl Acad. Sci. USA 83, 93139317. 1594 A. Observations 1, 2, and 3 are expected under both the proposed pathways for arachidonic acid production. Observations 4 and 5 distinguish between the pathways. If arachidonic acid arose by cleavage from the diacylglycerol produced along with IP3, then its production should have paralleled that of IP3 in both experiments; that is, arachidonic acid production should have been blocked by treatment with the inhibitor of phospholipase C (since no diacylglycerol would have been produced), and it should have been unaffected by pertussis toxin, the inhibitor of Gi (since diacylglycerol production was unaffected). B. Observations 1 and 2 indicate that arachidonic acid production is dependent on the a1-adrenergic receptor. The stimulation of arachidonic acid production by GTPgS suggests that a G protein is involved in the pathway; the inhibition of arachidonic acid production by pertussis toxin confirms that the G protein is Gi. However, the toxin-sensitive G protein is not the same one that activates phospholipase C, since pertussis toxin does not affect phospholipase-C activity. The final link between the G protein and phospholipase A2 is not defined by these experiments. Reference: Burch RM., Luini A & Axelrod J (1986) Phospholipase A2 and phospholipase C are activated by distinct GTP-binding proteins in response to a1-adrenergic stimulation in FRTL5 thyroid cells. Proc. Natl Acad. Sci. USA 83, 72017205. A353 A354 Chapter 15: Mechanisms of Cell Communication STIMULUS TRANSDUCTION INTRACELLULAR MESSENGERS ACTIVATED PROTEIN KINASES myosin light-chain kinase PHOSPHORYLATED TARGET PROTEINS myosin light chains EFFECTS ON THE CELL shape change Figure 1543 Overall pathway for platelet activation (Answer 1595). Ca2+ receptor IP3 thrombin, collagen phospholipase C membrane phospholipids exocytosis diacylglycerol PKC 40-kd protein serotonin release plasma membrane 1595 A. The activities of Ca2+ and diacylglycerol suggest that the normal sequence of events involves phospholipase C. Collagen fibers and thrombin stimulate a receptor on the surface of the platelet, which in turn activates phospholipase C, presumably through a G protein. Phospholipase C cleaves phosphatidylinositol bisphosphate to produce IP3 and diacylglycerol. IP3 mobilizes internal Ca2+ stores, activating myosin light-chain kinase, which phosphorylates the myosin light chain. This branch of the pathway can be stimulated by the calcium ionophore. Diacylglycerol activates PKC, which phosphorylates the 40-kd protein. This branch of the pathway can be stimulated directly by diacylglycerol. These two individual pathways interact to stimulate serotonin release. The overall pathway for platelet activation is diagrammed in Figure 1543. B. Secretion of serotonin evidently requires both Ca2+ and diacylglycerol, since neither alone causes any secretion (Figure 1543B). These experiments imply that the 40-kd protein is involved, although direct proof is lacking. Ca2+ is thought to be more directly involved in secretion, enabling the fusion of membranes that is required for exocytosis. Reference: Nishizuka Y (1983) Calcium, phospholipid turnover and transmembrane signaling. Phil. Trans. R. Soc. Lond. B 302, 101112. 1596 A. The difference in the behavior of myosin light-chain kinase when purified in the presence and absence of protease inhibitors suggests that the regulatory domain of the kinase is sensitive to cleavage by proteases in the cell extract. Activation of the kinase by proteolytic removal of the regulatory domain suggests that the domain in some way covers the active site. This regulatory `flap' moves out of the way upon Ca2+/calmodulin binding. This theme of a regulatory flap covering an active site is fairly common. B. Upon entry into the cytosol, Ca2+ binds to calmodulin, causing a change in its conformation. The Ca2+/calmodulin complex binds to the regulatory domain of myosin light-chain kinase, exposing the active site of the enzyme. The activated kinase then phosphorylates myosin light chains, altering their conformation so that they can bind actin and initiate contraction. As you might predict, contraction is terminated through the action of phosphatases that remove the phosphate from the myosin light chains. Reference: Adelstein RS & Klee CB (1980) Smooth muscle myosin lightchain kinase. In Calcium and Cell Function (WY Cheung ed), Vol. 1, pp 167182. New York: Academic Press. 1597 A. The complete G protein does not activate the K+ channels in the absence of acetylcholine presumably because, like other trimeric G proteins, the active portion is inhibited by one of the subunits. The ability of the Gbg subunit to open the K+ channel in the absence of acetylcholine and GTP suggests that it is the active portion of the G protein. This is different from the active component (Ga) of G proteins triggered by GPCRs that activate adenylyl cyclase. SIGNALING THROUGH ENZYME-COUPLED CELL-SURFACE RECEPTORS acetylcholine A355 K+ channel acetylcholine receptor a GDP acetylcholine Pi acetylcholine receptor a GTP bg bg K+ channel Figure 1544 Diagram illustrating the activation of K+ channels in heart by acetylcholine (Answer 1597). receptor a GDP GDP K+ channel GTP bg K+ B. The opening of the K+ channel in the presence of GppNp and absence of acetylcholine may seem somewhat surprising, since the release of GDP and the binding GTP by G proteins are normally stimulated by an activated receptor. Even in the absence of an activated receptor, G proteins exchange their bound nucleotides with nucleotides in the cytoplasm. Exchange is slow, and any bound GTP is quickly hydrolyzed in the absence of an activated receptor, thereby keeping the channel closed. The K+ channels open slowly when GppNp is present because, each time a GDP is released and a GppNp is bound, the G protein is locked into an active form. Over the course of a minute, enough G protein is activated in this way to open the K+ channels in the absence of acetylcholine. C. A scheme for the G-protein-mediated activation of K+ channels by acetylcholine is shown in Figure 1544. References: Logothetis DE, Kurachi Y, Galper J, Neer EJ & Clapham DE (1987) The bg subunits of GTP-binding proteins activate the muscarinic K+ channel in heart. Nature 325, 321326. Reuveny E, Slesinger PA, Inglese J, Morales JM, Iniguez-Lluhi JA, Lefkowitz RJ, Bourne HR, Jan YN & Jan LY (1994) Activation of the cloned muscarinic potassium channel by G protein bg subunits. Nature 370, 143146. SIGNALING THROUGH ENZYME-COUPLED CELLSURFACE RECEPTORS DEFINITIONS 1598 1599 Ephrins Transforming growth factor-b (TGFb) superfamily 15100 Receptor tyrosine kinase (RTK) 15101 Ras 15102 Rho family 15103 Bacterial chemotaxis 15104 Focal adhesion kinase (FAK) 15105 Phosphoinositide 3-kinase (PI 3-kinase) 15106 Enzyme-coupled receptor 15107 Tyrosine-kinase-associated receptor A356 Chapter 15: Mechanisms of Cell Communication 15108 MAP kinase cascade (MAP kinase module) 15109 JAKSTAT signaling pathway 15110 Pleckstrin homology (PH) domain 15111 SH2 domain 15112 TOR or mTOR TRUE/FALSE 15113 False. Ligand binding usually causes a receptor tyrosine kinase to assemble into dimers, which activates the kinase domains because of their proximity. The receptors then phosphorylate themselves to initiate the intracellular signaling cascade. In some cases, the insulin receptor for example, the receptor exists as a dimer and ligand binding is thought to rearrange their receptor chains, causing the kinase domains to come together. 15114 False. PI 3-kinase phosphorylates inositol head groups at a position (number 3 on the inositol ring) that is not phosphorylated in IP3. (IP3 carries phosphates at the 1, 4, and 5 positions on the inositol ring.) Phosphorylation at this site serves an entirely different function; it creates inositol head groups that can serve as docking sites for intracellular signaling proteins. 15115 True. Protein tyrosine phosphatases, unlike serine/threonine protein phosphatases, remove phosphate groups only from selected phosphotyrosines on a subset of tyrosine-phosphorylated proteins. THOUGHT PROBLEMS 15116 The added antibody is likely to activate the receptor tyrosine kinase. Because antibodies carry two identical binding sites, they bind to two receptor tyrosine kinase molecules, allowing them to phosphorylate each other and activate the signaling pathway. Experiments of this type were the first to demonstrate that receptor dimerization is the critical step in activation of most receptor tyrosine kinases. It should be noted that not all antibodies that cross-link receptors activate them. Presumably these antibodies crosslink the receptors in a way that does not properly juxtapose the kinase domains and target sites. 15117 Ephrins and ephrin receptors are both membrane-bound proteins. An ephrin on one cell binds to an ephrin receptor on a second cell, triggering the ephrin receptor to initiate a signaling pathway in the second cell. Binding to the ephrin receptor also activates the ephrin so that it initiates a signaling pathway in the signaling cell. A ligandreceptor interaction that initiates signaling pathways in both interacting cells is called bidirectional signaling. 15118 A. The mutant receptor tyrosine kinase will be inactive for signaling because it cannot bind its ligand in the absence of an extracellular domain. The mutant form of the kinase will have no effect on normal signaling mediated by the cell's own receptor tyrosine kinases; ligands will bind to them normally, causing their dimerization and activation (Figure 1545A). B. The mutant receptor tyrosine kinase lacking an intracellular domain will also be inactive for signaling because it is missing its kinase domain and phosphorylation sites. Because it retains its ability to bind to the ligand, it will interfere with signaling by the cell's normal receptor tyrosine kinase. Because the mutant kinase is expressed at considerably higher levels than the cell's kinase, the mutant receptor will account for most of the cell's binding of ligand. Even when a normal receptor manages to bind a ligand, it will usually be dimerized with an inactive mutant receptor (Figure 1545B). Under these conditions the normal receptor will remain inactive because it SIGNALING THROUGH ENZYME-COUPLED CELL-SURFACE RECEPTORS A357 (A) Figure 1545 Effects of mutant receptors on receptor tyrosine kinase signaling (Answer 15118). (A) Mutant receptors lacking their extracellular domain. (B) Mutant receptors lacking their intracellular domain. K K K K K K K K K K P K P K (B) K K K K K K cannot cross-phosphorylate the mutant partner, nor can its mutant partner phosphorylate it. This is called a `dominant-negative' effect. 15119 A. Biotin-tagged GST-SH2 proteins could not be used for detecting proteins that bind to SH2 domains in the same way that GST-SH3 proteins can. That is because SH2 domains bind only to target proteins that carry a phosphotyrosine--a modification that doesn't occur normally in bacteria. One way to make the screen work would be to incubate the filters first with a protein tyrosine kinase and ATP. Alternatively, a protein tyrosine might be engineered into E. coli so that it could be turned on at the same time the cDNA library was expressed. B. The main difference between protein interactions with short sequences and the subunitsubunit interactions in multisubunit enzymes lies in their stability and reversibility. Both types of interaction depend largely on the total number and aggregate strength of the weak bonds involved in their formation. Large contact surfaces such as those found among subunits in multisubunit enzymes make for very stable structures, whereas most of the examples of short-sequence recognition are more transient, and in some cases conditional, as in the interaction of SH2 domains with phosphotyrosinecontaining proteins. Reference: Ren R, Mayer BJ, Cicchetti P & Baltimore D (1993) Identification of a ten-amino-acid proline-rich SH3 binding site. Science 259, 11571161. 15120 You would expect to see several differences. (1) You would expect a high background of Ras activity in the absence of an extracellular signal because Ras cannot be turned off efficiently. Since Ras activity depends on the balance between its binding to GTP and its GAP-enhanced hydrolysis of GTP, the balance would be somewhat more in favor of the GTP-bound (active) form than normal. (2) As some Ras molecules will already be in their GTPbound form, Ras activity in response to an extracellular signal would be greater than normal, but would saturate when all Ras molecules were converted to the GTP-bound form. (3) The response to signal would be less rapid because the signal-dependent increase in GTP-bound Ras would occur over an elevated background of preexisting GTP-bound Ras. (4) The response would be expected to be more prolonged than normal and to persist for a while even after the extracellular signal was removed because of the slower rate of conversion of GTP-bound Ras to its inactive GDP-bound form. 15121 Activation in both cases depends on proteins that catalyze GDP/GTP exchange on the G protein or Ras protein. Whereas the GPCRs perform this function directly for G proteins, enzyme-coupled receptors assemble multiple adaptor proteins into a signaling complex when the receptors are activated by phosphorylation and one of these recruits a Ras-activating protein that fulfills this function for Ras. Inactivation of G proteins and Ras proteins is also similar. Ras is turned off by a GAP that promotes hydrolysis of GTP. A358 Chapter 15: Mechanisms of Cell Communication Similarly, the ability of Ga subunits to hydrolyze GTP, which is intrinsically higher than that of Ras, is also stimulated by their interactions with downstream targets such as adenylyl cyclase. 15122 In order for activation of Ras to depend on inactivation of a GAP, both the GAP and the GEF would need to be active in the absence of the signal. In this way the GEF would constantly load GTP onto Ras and the GAP would keep the concentration of RasGTP low by constantly inducing GTP hydrolysis to return Ras to its GDP-bound state. Under these conditions inactivation of the GAP would result in a rapid increase in the RasGTP level, allowing rapid signaling. Although this would be a perfectly effective way to regulate the level of active Ras, it would be wasteful of energy. In order to keep Ras in its inactive state, GTP would be constantly hydrolyzed to GDP, which would then need to be reconverted to GTP (by ATP)--a drain on cellular energy metabolism. Regulation by activation of a GEF avoids this problem. Although avoiding constant GTP hydrolysis is a rational explanation, eucaryotic cells are notoriously profligate in their energy expenditures. At several points in energy metabolism, for example, they operate so-called `futile' cycles that hydrolyze ATP as a means for rapid regulation of the flux through metabolic pathways. Thus, it could be that constant hydrolysis of GTP by a Ras GEF and GAP would not unduly tax the cell's energy budget. Perhaps the cell's method of regulating Ras by controlling the activity of a GEF is simply an evolutionary happenstance. 15123 The small molecule would not be effective in treating cancers with mutationally activated Ras proteins. Activated Ras causes problems because it signals independently of any upstream influences. Thus, preventing receptors from dimerizing would have no effect on activity of the mutant Ras. 15124 Several changes in surface architecture are apparent in the phosphorylated form of MAP-kinase, perhaps most noticeably at 10, 12, and 4 o'clock in the structure shown in Figure 1524. It is difficult, if not impossible, to guess where ATP and target proteins would bind, even if amino acid side chains were shown. Additional information from the crystal structure of activated MAP-kinase with ATP bound shows that the active protein binds ATP at the 10 o'clock position and its target proteins, at the 4 o'clock position (Figure 1546). 15125 Replacing the extracellular domain of a receptorlike tyrosine phosphatase with the binding domain for a defined receptor (for example, the EGF receptor) converts the protein into a receptor for a known ligand (in this case, EGF). Since receptorlike tyrosine phosphatases are single-pass transmembrane proteins, they are likely to be regulated by dimerization, as the EGF receptor is. If this assumption is correct, then dimerization via the grafted ATP Y T YP TP INACTIVE ACTIVE Figure 1546 Nonphosphorylated (inactive) and phosphorylated (active) MAP kinase with bound ATP (Answer 15124). Arrow in the active form indicate the binding site for ATP. Open arrows indicate changes in the conformation of the so-called activation loop in the inactive and active states. SIGNALING THROUGH ENZYME-COUPLED CELL-SURFACE RECEPTORS EGF binding domain will bring the cytoplasmic domains of the hybrid protein together. If proximity is the triggering event, then it shouldn't matter how they are dimerized. For single-pass membrane receptors, making hybrids of this sort offers a general solution to studying their signaling pathways without first knowing the ligand involved. A359 DATA HANDLING 15126 A. If individual receptor tyrosine kinases phosphorylated themselves, only the band corresponding to the normal receptor (with a functional kinase domain and phosphorylation sites) would appear on the autoradiograph in Figure 1525C. B. If receptor tyrosine kinases can phosphorylate each other, then not only would normal receptors pair with each other and be labeled, but mutant receptor 2 (with a dead kinase domain but functional phosphorylation sites) should be labeled whenever it is paired with a receptor that has an active kinase domain (the normal receptor and mutant receptor 3). C. The results in Figure 1525C support the cross-phosphorylation model for autophosphorylation. Receptor 2 is labeled when paired with either receptor 1 or receptor 3 (see Figure 1525C). You might wonder why receptor 1 was labeled when expressed along with receptor 2. In such a mixture receptor 1 would be expected to pair with itself roughly half the time, allowing cross-phosphorylation. References: Honegger AM, Kris RM, Ullrich A & Schlessinger J (1989) Evidence that autophosphorylation of solubilized receptors for epidermal growth factor is mediated by intermolecular cross-phosphorylation. Proc. Natl Acad. Sci. USA 86, 925929. Honegger AM, Schmidt A, Ullrich A & Schlessinger J (1990) Evidence for epidermal growth factor (EGF)-induced intermolecular autophosphorylation of the EGF receptors in living cells. Mol. Cell. Biol. 10, 40354044. 15127 PDGF receptors that bind only PI 3-kinase (PI3K) or PLCg stimulate DNA synthesis to about 70% of the normal value (see Figure 1527, lanes 2 and 5); thus, both PI3K and PLCg mediate signaling pathways that increase DNA synthesis. These two pathways must be somewhat redundant because receptors that bind both PI3K and PLCg (but nothing else) give the same 70% response as either alone (see lane 6). GAP-mediated signaling seems to inhibit the mitogenic response, since receptors with intact PLCg and GAPbinding sites stimulate DNA synthesis to a significantly lesser extent (see lane 7) than receptors with just the PLCg-binding site (see lane 5). PTP appears to play no role in PDGF stimulation of DNA synthesis, since it does not stimulate DNA synthesis alone (see lane 4) or alter the response when paired with a PLCg-binding site (see lane 8). References: Valius M & Kazlauskas A (1993) Phospholipase C-g1 and phosphatidylinositol 3 kinase are the downstream mediators of the PDGF receptor's mitogenic signal. Cell 73, 321334. Valius M, Secrist JP & Kazlauskas A. (1995) The GTPase-activating protein of Ras suppresses platelet-derived growth factor b receptor signaling by silencing phospholipase C-g1. Mol. Cell. Biol. 15, 30583071. 15128 The very steep response curve for activation of MAPK converts it into a molecular switch. Thus, MAPK goes from inactive to active over a very narrow range of input stimulus. This kind of behavior keeps the cascade turned off below a threshold concentration of the input signal, yet delivers a maximum response once that threshold is exceeded. Reference: Huang C-Y F & Ferrell JE (1996) Ultrasensitivity in the mitogenactivated protein kinase cascade. Proc. Natl Acad. Sci. USA 93, 1007810083. A360 Chapter 15: Mechanisms of Cell Communication 15129 If scaffold proteins linked the kinases, the activation curves for MAPKK and MAPK would resemble more closely that of MAPKKK; that is, MAPK would behave less like a molecular switch. The gain in speed and precision of signal transmission and the avoidance of cross-talk between pathways evidently compensate for the loss of signal amplification and switchlike behavior, since cells use scaffold proteins to organize many different MAP kinase cascades. The (evolutionary) choice of scaffold or independent components for a particular MAP kinase cascade presumably reflects the functional consequences of the signaling pathway. 15130 The data in Figure 1529 suggest that Akt phosphorylates itself at serine 473. The kinase-dead mutant, Akt-K179M, is phosphorylated correctly at threonine 308 but not at serine 473. In addition, phosphorylation at serine 473 depends on phosphorylation of threonine 308, as shown by the results with Akt-T308A; thus, it is unlikely that PDK1 carries out this second phosphorylation. Reference: Toker A & Newton AC (2000) Akt/protein kinase B is regulated by autophosphorylation at the hypothetical PDK-2 site. J. Biol. Chem. 275, 82718274. 15131 A. Many transcription factors are phosphorylated in response to cell stimulation: in some instances, this results in tighter binding to DNA, whereas in others it creates an acidic activation domain that promotes transcription. In this case phosphorylation seems to be necessary for DNA binding because anti-phosphotyrosine antibodies and phosphatase treatment inhibit DNA binding. B. Free phosphotyrosine will bind to the SH2 domain of the transcription factor. If simple occupancy of the SH2 domain by a phosphotyrosine were all that was required, free phosphotyrosine would be expected to activate the transcription factor instead of inhibiting it. This implies that the SH2 domain must bind to the phosphotyrosine on the transcription factor. By interfering with this interaction, free phosphotyrosine interferes with the factor's ability to bind to DNA. C. Tyrosine phosphorylation of the transcription factor could promote its dimerization in two general ways. It could be that in the absence of phosphorylation the heptad repeats are masked by the tertiary structure of the protein and that phosphorylation and intramolecular binding to the SH2 domain causes a conformational change that exposes the heptad repeats and allows dimerization (Figure 1547A). Alternatively, it could be that the heptad repeats do not promote a strong enough interaction for formation of a stable dimer and that phosphorylation and intermolecular binding to the SH2 domain is required (Figure 1547B). The dyad symmetry of the DNA sequence element suggests that it is composed of two half-sites for binding. The formation of the dimer allows the transcription factor to interact with both half-sites simultaneously, which greatly increases the strength of binding. Reference: Sadowski HB, Shuai K, Darnell JE & Gilman MZ (1993) A common nuclear signal transduction pathway activated by growth factor and cytokine receptors. Science 261, 17391744. 15132 It is relatively straightforward to match the pattern of intact receptors with the pattern of responses to the attractants. For example, Tsr is missing only in strain 2, and only strain 2 does not respond to serine. Thus, Tsr mediates the response to serine. Similarly, Tar mediates the response to aspartate, and Tap mediates the response to the dipeptide Pro-Gly. If one of these receptors mediates the response to ribose, then it must be Trg since it alone is present in all strains and all strains chemotax to ribose. The patterns of response are abbreviated in the names given to the chemotaxis receptors: Tsr stands for taxis for serine and certain repellents; Tar, for taxis for aspartate and certain repellents; and Trg, for taxis for ribose SIGNALING THROUGH ENZYME-COUPLED CELL-SURFACE RECEPTORS (A) A361 heptad repeats SH2 SH3 ATP Figure 1547 Two general ways for tyrosine phosphorylation to promote dimerization of the transcription factor (Answer 15131). P SH2 SH2 (B) heptad repeats ATP SH3 SH2 P SH2 P SH2 and galactose. Tap was originally named taxis-associated protein because its function was unknown but its gene was clustered with a group of other chemotactic genes; fortuitously, the name works well for taxis for peptides. Reference: Manson MD, Blank V, Brade G & Higgins CF (1986) Peptide chemotaxis in E. coli involves the Tap signal transducer and the dipeptide permease. Nature 321, 253258. 15133 A. The two cloned receptors, normal and truncated, both carry out signal transduction like the receptors in wild-type bacteria. Upon addition of aspartate, all three kinds of bacteria immediately suppress changes in direction of rotation. Thus, the presence of the attractant (aspartate) in the medium is communicated to the flagella in all three kinds of bacteria. B. The adaptive properties of bacteria containing the cloned receptors are very different from wild-type bacteria. Wild-type bacteria return to their normal rate of tumbling (reversal of direction of rotation) within 4 minutes. Bacteria with the cloned normal receptor return to the normal rate of tumbling only after about 60 minutes, and bacteria with the truncated receptor do not begin to tumble even after 1 hour (indeed, they do not tumble even after 3 hours, if the experiment is continued that long). Thus bacteria with the cloned normal receptor adapt more slowly than wild-type bacteria, whereas bacteria with the cloned truncated receptor evidently do not adapt. C. The alterations in adaptation in bacteria with the cloned receptors suggest differences in the methylation rates or extents. The inability of the truncated receptor to be methylated provides a molecular basis for the inability of bacteria to adapt to a high level of aspartate. The molecular basis for the difference between wild-type bacteria and bacteria with the cloned normal receptor is more subtle. The difference in time of adaptation between the two kinds of bacteria is about 15-fold (4 minutes versus 60 minutes), which is about the same as the difference in numbers of receptors per cell. Thus a reasonable explanation is that it takes the receptor methylase 15 times longer to methylate the more abundant normal receptor. Reference: Russo AF & Koshland DE (1983) Separation of signal transduction and adaptation functions of the aspartate receptor in bacterial sensing. Science 220, 10161020. P A362 Chapter 15: Mechanisms of Cell Communication SIGNALING PATHWAYS DEPENDENT ON REGULATED PROTEOLYSIS OF LATENT GENE REGULATORY PROTEINS DEFINITIONS 15134 Notch 15135 Wnt proteins 15136 Wnt/b-catenin pathway 15137 Hedgehog proteins 15138 Cubitus interruptus (Ci) 15139 NFkB proteins TRUE/FALSE 15140 False. Although some signaling pathways activate latent gene regulatory proteins by regulated proteolysis, others control their activity by phosphorylation. 15141 True. Notch carries both its functions--cell-surface receptor and latent gene regulator--in one polypeptide chain. When activated by a ligand such as Delta, its cytoplasmic tail is cleaved off, enters the nucleus, and activates gene expression. 15142 True. Activated NFkB increases expression of the IkBa gene, and IkBa then binds to NFkB and inactivates it, thereby shutting off the response. If the initial activating signal persists, then additional cycles of NFkB activation and inactivation may follow. THOUGHT PROBLEMS 15143 The components of the signaling pathways, in order of their action, are listed below along with their functions. Notch Signaling C. Delta J. Notch L. Presenilin M. Rbpsuh Wnt Signaling O. Wnt E. Frizzled I. LRP D. Dishevelled F. GSK3 A. b-Catenin Hedgehog Signaling G. Hedgehog H. iHog K. Patched N. Smoothened B. Ci 3. Extracellular signal protein 6 and 9. Receptor protein and latent gene regulatory protein 8. Protease 4. Gene regulatory protein 3. Extracellular signal protein 9. Receptor protein 9. Receptor protein 11. Scaffold protein 7. Modulator protein 6. Latent gene regulatory protein 3. Extracellular signal protein 9. Receptor protein 9. Receptor protein 10. Relay protein 6. Latent gene regulatory protein 15144 The extracellular fragments of APP aggregate to form amyloid plaques outside the cells. The amyloid plaques are thought to interfere with nerve function, leading to the characteristic loss of mental acuity that is typical of Alzheimer's disease. SIGNALING PATHWAYS DEPENDENT ON REGULATED PROTEOLYSIS 15145 Cells of flies with the heterozygous DshD/+ genotype probably make just half the normal amount of Dishevelled. Thus, underexpression of Dishevelled corrects the multihair phenotype generated by the overexpression of Frizzled. This relationship suggests that Frizzled acts upstream of Dishevelled; it is easy to imagine how underexpression of a downstream component could correct overexpression of an upstream component. All this makes sense, as Frizzled is a Wnt receptor and Dishevelled is an intracellular signaling protine. However, if you knew nothing of the functions of Dishevelled and Frizzled, with only the genetic interactions as a guide, it would be possible to imagine more complex relationships (involving other unknown components) with Dishevelled acting upstream of Frizzled that could account for the phenotypes given in this problem. See if you can design such a pathway. Reference: Winter CG, Wang B, Ballew A, Royou A, Karess R, Axelrod JD & Luo L (2001) Drosophila Rho-associated kinase (Drok) links Frizzled-mediated planar cell polarity signaling to the actin cytoskeleton. Cell 105, 8191. 15146 The Apc gene is a tumor suppressor gene. Its normal function is to inhibit bcatenin by helping to hold it in the cytosol until a proper signal has been received. When both copies of the Apc gene are inactivated, b-catenin is free to enter the nucleus in the absence of any signal, leading to uncontrolled stimulation of its target genes. 15147 1. The latent gene regulatory protein is attached to the membrane as part of the covalent structure of a transmembrane protein. When a valid signal is received, the regulatory protein is cleaved and enters the nucleus. Notch is an example. 2. The latent gene regulatory protein is actively degraded in the cytosol. When a valid signal is received, the protein is stabilized against degradation, allowing it to enter the nucleus. b-Catenin is an example. 3. The latent gene regulatory protein is anchored to a cytosolic structure and released in response to an appropriate signal. Cubitus interruptus is an example. 4. The latent gene regulatory protein is bound to a protein that holds it in an inactive form. Upon receipt of an appropriate signal, the inhibitory protein is modified so that the gene regulatory protein is released in an active form and transported into the nucleus. NFkB is an example. A363 DATA HANDLING 15148 The slower migrating forms of b-catenin are due to ubiquitylation, not phosphorylation. Because the protein phosphatase had no effect on the slower migrating forms of b-catenin that arose in the presence on ALLN (see Figure 1533, compare lanes 2 and 4), the difference in migration cannot be due to phosphorylation. By contrast, when His-tagged ubiquitylated proteins were first purified from cells treated with ALLN, the slower migrating forms of bcatenin were specifically detected (see lane 6). Reference: Aberle H, Bauer A, Stappert J, Kispert A & Kemler R (1997) bCatenin is a target for the ubiquitin-proteasome pathway. EMBO J. 16, 37973804. 15149 These results indicate that phosphorylation of b-catenin sensitizes it for degradation in proteasomes. If phosphorylation were irrelevant to degradation or if it protected against degradation, slower migrating, ubiquitylated forms of b-catenin should have been present in cell lines that were unable to phosphorylate b-catenin. 15150 A. Although the Hedgehog precursor protein was purified from the bacteria in which it was expressed, it is unlikely to be 100% pure (no purified protein ever is). Incubation over a wide range of concentrations of the protein A364 Chapter 15: Mechanisms of Cell Communication argues against a contaminating protease. If the cleavage were a bimolecular reaction between the precursor protein and the contaminating protease, the rate of reaction should be exquisitely sensitive to concentration. Each 4-fold dilution of the contaminating protease would slow the rate of reaction 4fold. But the dilution also lowers the concentration of substrate, which would also lower the rate. The absence of any effect of dilution on the rate of reaction makes it extremely likely that the precursor protein is cleaving itself. Moreover, the protease must reside primarily, if not entirely, in the Cterminal segment since so little of the N-terminal segment was included in the construct. B. The lack of effect of dilution also indicates that the reaction must be intramolecular. If it were not--if precursor molecules only cleaved other precursor molecules--then the rate of reaction would slow with increasing dilution. Thus, the precursor cleaves itself in an autoproteolytic reaction. Reference: Porter JA, von Kessler DP, Ekker SC, Young KE, Lee JJ, Moses K & Beachy PA (1995) The product of hedgehog autoproteolytic cleavage active in local and long-range signaling. Nature 374, 363366. 15151 The N-terminus of the Hedgehog precursor protein remains associated with the cells when cleaved naturally, but it is secreted when it is synthesized from the truncated construct. These data do not define the nature of the cell association. As a part of its cleavage mechanism, the N-terminal fragment could, for example, become associated with a component of the cell, either inside the cell or on the membrane; alternatively, it could be trapped in an intracellular compartment. The actual explanation is very surprising; the cleavage mechanism uses a membrane cholesterol molecule to complete the cleavage, leaving the N-terminal fragment attached to the membrane via a covalent linkage between glycine 257 and cholesterol. Reference: Porter JA, von Kessler DP, Ekker SC, Young KE, Lee JJ, Moses K & Beachy PA (1995) The product of hedgehog autoproteolytic cleavage active in local and long-range signaling. Nature 374, 363366. 15152 A. Overexpression of constructs encoding the full-length Hedgehog and the Nterminal segment both caused a dramatic increase in the level of Wnt expression (see Figure 1536, embryos 2 and 4). Thus, the N-terminal segment must contain the portion of Hedgehog that is important in signaling Wnt expression. B. Although all the cells of the embryo overexpress Hedgehog, the target receptors through which it acts are localized to cells in the stripes. In the absence of the appropriate receptor, Hedgehog cannot elicit a cellular response. C. The stripes of Wnt expression in the absence of Hedgehog overexpression arise as a result of expression from the flies' normal Hedgehog gene. Reference: Porter JA, von Kessler DP, Ekker SC, Young KE, Lee JJ, Moses K & Beachy PA (1995) The product of hedgehog autoproteolytic cleavage active in local and long-range signaling. Nature 374, 363366. SIGNALING IN PLANTS DEFINITIONS 15153 Phytochrome 15154 Ethylene 15155 Growth regulator, plant hormone 15156 Cryptochrome SIGNALING IN PLANTS 15157 Auxin 15158 Leucine-rich repeat (LRR) receptor kinase A365 TRUE/FALSE 15159 False. Although there is some overlap in the cellcell communication molecules used in plants and animals, there are many significant differences. For example, plants don't use the nuclear receptor family, Ras, JAK, STAT, TGFb, Notch, Wnt, or Hedgehog proteins. 15160 True. When the gravity vector is changed, vesicles filled with the auxin efflux transporter fuse with the plasma membrane so that the transporters are correctly positioned to pump auxin toward the side of the root that points downward. THOUGHT PROBLEMS 15161 The similarities in signaling mechanisms between animals and fungi support the phylogenetic tree in which fungi branched from the animal lineage after plants and animals separated (see Figure 1537B). This branching order is supported by a wide variety of other data, including genomic sequence comparisons. 15162 If the basic mechanisms of cell communication arose in response to multicellularity, then fungi must have separated from the animal lineage after multicellularity evolved. This reasoning would suggest that unicellular fungi may have been derived from multicellular precursors. Not very long ago-- and to great surprise--it was shown that Saccharomyces cerevisiae will form multicellular filamentous forms. Many members of the fungal kingdom have this ability, termed dimorphism, to switch between two morphological forms: a cellular form and a multicellular invasive form. Reference: Madhuri HD & Fink GR (1998) The control of filamentous differentiation and virulence in fungi. Trends Cell Biol. 8, 348353. 15163 Systemic growth regulators have specific effects in those cells that express their receptors. This situation is no different from that of hormone action in animals: steroid hormones, for example, circulate throughout the body but have specific effects in cells that express appropriate receptors. DATA HANDLING 15164 A. The accepted explanation for the ability of an antisense RNA to block expression of the normal gene is that the two RNAs--the antisense RNA and the normal RNA--hybridize to make a double-stranded RNA that cannot be translated. This would effectively block synthesis of the ACC synthase enzyme and prevent the formation of ethylene. But this may not be the true mechanism. In some plants a phenomenon called RIPing pairs duplicated sequences in meiosis and introduces mutations into both. Thus, it may be that the normal ACC synthase gene is inactivated in your transgenic tomatoes. B. In all likelihood! Reference: Oeller PW, Min-Wong L, Taylor LP, Pike DA, Theologis A (1991) Reversible inhibition of tomato fruit senescence by antisense RNA. Science 254, 437439. ...
View Full Document

Ask a homework question - tutors are online