Systems Biology of Cell Signaling: Recurring Themes and Quantitative Models

Description

How can we understand the complexity of genes, RNAs, and proteins and the associated regulatory networks? One approach is to look for recurring types of dynamical behavior. Mathematical models prove to be useful, especially models coming from theories of biochemical reactions such as ordinary differential equation models. Clever, careful experiments test these models and their basis in specific theories. This textbook aims to provide advanced students with the tools and insights needed to carry out studies of signal transduction drawing on modeling, theory, and experimentation. Early chapters summarize the basic building blocks of signaling systems: binding/dissociation, synthesis/destruction, and activation/inactivation. Subsequent chapters introduce various basic circuit devices: amplifiers, stabilizers, pulse generators, switches, stochastic spike generators, and oscillators. All chapters consistently use approaches and concepts from chemical kinetics and nonlinear dynamics, including rate-balance analysis, phase plane analysis, nullclines, linear stability analysis, stable nodes, saddles, unstable nodes, stable and unstable spirals, and bifurcations. This textbook seeks to provide quantitatively inclined biologists and biologically inclined physicists with the tools and insights needed to apply modeling and theory to interesting biological processes. Key Features: · Full-color illustration program with diagrams to help illuminate the concepts · Enables the reader to apply modeling and theory to the biological processes · Further Reading for each chapter · High-quality figures available for instructors to download Cover Half Title Title Page Copyright Page Table of Contents Detailed Contents Preface Acknowledgements Author CHAPTER 1: Introduction 1.1 SIGNAL TRANSDUCERS ARE CELLULAR COMPONENTS THAT ACT MAINLY BY REGULATING OTHER CELLULAR COMPONENTS 1.2 THE SIGNAL TRANSDUCTION PARTS LIST IS LONG 1.3 SIGNAL TRANSDUCTION IN BACTERIA IS ACCOMPLISHED BY SHORT, (MOSTLY) LINEAR, (MOSTLY) NON-INTERCONNECTED PATHWAYS 1.4 THE EGFR SYSTEM IS DEEP, INTERCONNECTED, AND COMPLICATED 1.5 COMPLICATED SYSTEMS CAN BE SIMPLIFIED BY ASSUMING MODULARITY 1.6 ORDINARY DIFFERENTIAL EQUATIONS PROVIDE A POWERFUL FRAMEWORK FOR UNDERSTANDING MANY SIGNALING PROCESSES 1.7 THEORY CAN HELP HIGHLIGHT THE COMMONALITIES OF DIVERSE BIOLOGICAL PHENOMENA 1.8 SIX BASIC TYPES OF RESPONSE ARE SEEN OVER AND OVER AGAIN IN CELL SIGNALING 1.9 FIVE OR SIX BASIC CIRCUIT MOTIFS ARE SEEN OVER AND OVER AGAIN IN SIGNALING SYSTEMS SUMMARY MOVING FORWARD FURTHER READING CHAPTER 2: Receptors 1: Monomeric Receptors and Ligands 2.1 THE β[sub(2)]-ADRENERGIC RECEPTOR CAN FUNCTION AS A MONOMERIC RECEPTOR THAT BINDS MONOMERIC LIGANDS 2.2 EXPERIMENTS SHOW THE RECEPTOR’S EQUILIBRIUM AND DYNAMICAL BEHAVIORS 2.3 A SIMPLE BINDING-DISSOCIATION MODEL EXPLAINS THE HYPERBOLIC EQUILIBRIUM RESPONSE 2.4 A SEMILOG PLOT EXPANDS THE RANGE BUT DISTORTS THE GRADED CHARACTER OF THE RESPONSE 2.5 THE SYSTEM APPROACHES EQUILIBRIUM EXPONENTIALLY 2.6 INCREASING THE ASSOCIATION RATE DECREASES t[sub(1/2)]; SO DOES INCREASING THE DISSOCIATION RATE 2.7 GOING UP IS FASTER THAN COMING DOWN 2.8 THE DISSOCIATION RATE CONSTANT k−[sub(1)] DETERMINES THE HALF-LIFE AND MEAN LIFETIME OF A LIGAND–RECEPTOR COMPLEX 2.9 PARTIAL AGONISTS, ANTAGONISTS, AND INVERSE AGONISTS CAN BE EXPLAINED BY ASSUMING THAT BINDING AND ACTIVATION OCCUR IN DISTINCT STEPS SUMMARY FURTHER READING CHAPTER 3: Receptors 2: Multimeric Receptors and Cooperativity INTRODUCTION 3.1 THE HILL EQUATION IS A SIMPLE EXPRESSION FOR THE EQUILIBRIUM BINDING OF LIGAND MOLECULES TO AN OLIGOMERIC RECEPTOR 3.2 THE HILL EXPONENT IS A MEASURE OF HOW SWITCH-LIKE A SIGMOIDAL RESPONSE IS 3.3 THE HILL EQUATION ACCOUNTS FOR HEMOGLOBIN’S OXYGEN BINDING PRETTY WELL, BUT THE ASSUMPTIONS UNDERPINNING THE MODEL ARE DUBIOUS 3.4 THE MORE-PLAUSIBLE MONOD–WYMAN– CHANGEUX (MWC) MODEL YIELDS SIGMOIDAL BINDING CURVES 3.5 THE MWC MODEL ACCOUNTS FOR THE BINDING OF OXYGEN TO HEMOGLOBIN, BUT NOT THE BINDING OF EGF TO THE EGFR 3.6 THE KNF MODEL CAN ACCOUNT FOR EITHER ULTRASENSITIVE OR SUBSENSITIVE BINDING 3.7 RESPONSE SENSITIVITY IS CUSTOMARILY DEFINED IN FOLD-CHANGE TERMS 3.8 THE RELATIONSHIP BETWEEN BINDING AND ACTIVATION YIELDS A VARIETY OF POSSIBLE RESPONSES SUMMARY FURTHER READING CHAPTER 4: Downstream Signaling 1: Stoichiometric Regulation STOICHIOMETRIC REGULATION INSIDE THE CELL 4.1 IN THE HIGH-AFFINITY LIMIT, DOES A HYPERBOLIC RESPONSE MAKE INTUITIVE SENSE? 4.2 THE EQUILIBRIUM RESPONSE CHANGES FROM HYPERBOLIC TO LINEAR WHEN DEPLETION OF THE UPSTREAM REGULATOR IS NOT NEGLIGIBLE 4.3 THE DYNAMICAL RESPONSE IS SIMILAR EVEN WHEN THE DEPLETION OF THE UPSTREAM REGULATOR IS NOT NEGLIGIBLE 4.4 LIGAND DEPLETION PLUS NEGATIVE COOPERATIVITY CAN PRODUCE A THRESHOLD 4.5 STOICHIOMETRIC REGULATORS MUST SOMETIMES COMPETE WITH STOICHIOMETRIC INHIBITORS SUMMARY FURTHER READING CHAPTER 5: Downstream Signaling 2: Covalent Modification 5.1 A MASS ACTION PHOSPHORYLATION– DEPHOSPHORYLATION CYCLE YIELDS A MICHAELIAN STEADY-STATE RESPONSE WITH EXPONENTIAL APPROACH TO THE STEADY STATE 5.2 THE STEADY-STATE RESPONSE OF A PHOSPHORYLATION–DEPHOSPHORYLATION REACTION WITH MICHAELIS–MENTEN KINETICS CAN BE ULTRASENSITIVE 5.3 RATE–BALANCE PLOTS ARE MUCH LIKE THE ECONOMIST’S SUPPLY-AND-DEMAND PLOTS 5.4 RATE-BALANCE ANALYSIS EXPLAINS THE MICHAELIAN STEADY-STATE RESPONSE 5.5 THE DYNAMICS OF THE SYSTEM CAN ALSO BE UNDERSTOOD FROM THE RATE–BALANCE PLOT 5.6 RATE-BALANCE ANALYSIS HELPS EXPLAIN ZERO-ORDER ULTRASENSITIVITY 5.7 DOES ZERO-ORDER ULTRASENSITIVITY OCCUR IN VIVO? 5.8 THE TEMPORAL DYNAMICS OF A MULTISTEP ACTIVATION PROCESS TELLS YOU THE NUMBER OF PARTIALLY RATE-DETERMINING STEPS 5.9 ASSUMING MASS ACTION KINETICS, STEADY-STATE MULTISITE PHOSPHORYLATION IS DESCRIBED BY A KNF-TYPE EQUATION 5.10 PRIMING CAN IMPART POSITIVE COOPERATIVITY ON MULTISITE PHOSPHORYLATION 5.11 DISTRIBUTIVE MULTISITE PHOSPHORYLATION IMPROVES SIGNALING SPECIFICITY 5.12 INESSENTIAL PHOSPHORYLATION SITES CAN CONTRIBUTE TO THE ULTRASENSITIVITY 5.13 INESSENTIAL BINDING SITES CAN CONTRIBUTE TO ULTRASENSITIVE RECEPTOR ACTIVATION 5.14 VARIATION: COHERENT FEED-FORWARD REGULATION 5.15 VARIATION: RECIPROCAL REGULATION SUMMARY FURTHER READING CHAPTER 6: Downstream Signaling 3: Regulated Production or Destruction 6.1 STIMULATED PRODUCTION YIELDS A LINEAR STEADY-STATE RESPONSE WITH EXPONENTIAL APPROACH TO THE STEADY STATE 6.2 THE STABILITY OF THE STEADY STATE CAN BE QUANTIFIED BY THE EXPONENT IN THE EXPONENTIAL APPROACH EQUATION 6.3 SATURATING THE BACK REACTION BUILDS A THRESHOLD INTO THE STEADY-STATE RESPONSE 6.4 ZERO-ORDER DEGRADATION MAKES DRUG DOSING DICEY SUMMARY FURTHER READING CHAPTER 7: Cascades and Amplification INTRODUCTION 7.1 CASCADES CAN DELIVER SIGNALS FASTER THAN SINGLE SIGNAL TRANSDUCERS 7.2 A CASCADE OF MICHAELIAN RESPONSES LEADS TO SIGNAL DEGRADATION 7.3 FOLD-SENSITIVITY DECREASES AS A SIGNAL DESCENDS A CASCADE OF MICHAELIAN RESPONSES 7.4 ULTRASENSITIVITY CAN RESTORE OR INCREASE THE DECISIVENESS OF A SIGNAL 7.5 IN XENOPUS OOCYTE EXTRACTS, RESPONSES GET MORE ULTRASENSITIVE AS THE MAPK CASCADE IS DESCENDED SUMMARY FURTHER READING CHAPTER 8: Bistability 1: Systems with One Time-Dependent Variable 8.1 CELL FATE INDUCTION IS TYPICALLY ALL-OR-NONE AND IRREVERSIBLE IN CHARACTER 8.2 XENOPUS OOCYTE MATURATION IS AN ALL-OR-NONE, IRREVERSIBLE PROCESS 8.3 THE RESPONSE OF ERK2 IS ALL-OR-NONE IN CHARACTER 8.4 THERE IS POSITIVE FEEDBACK IN THE OOCYTE’S MAPK CASCADE 8.5 THE RESPONSE OF ERK2 TO PROGESTERONE IS NORMALLY IRREVERSIBLE 8.6 THE MOS/ERK2 SYSTEM CAN BE REDUCED TO A MODEL WITH A SINGLE TIMEDEPENDENT VARIABLE BECAUSE OF A SEPARATION OF TIME SCALES 8.7 RATE-BALANCE ANALYSIS SHOWS WHAT IS REQUIRED FOR A BISTABLE RESPONSE 8.8 INCREASING THE PROGESTERONE CONCENTRATION PUSHES THE SYSTEM THROUGH A SADDLE-NODE BIFURCATION 8.9 TWEAKING THE MODEL CAN CHANGE AN IRREVERSIBLE RESPONSE TO A HYSTERETIC ONE 8.10 THE DYNAMICS OF THE SYSTEM CAN BE INFERRED FROM THE RATE–BALANCE PLOT 8.11 THE VELOCITY VECTOR FIELD CAN BE REPRESENTED AS A POTENTIAL LANDSCAPE SUMMARY FURTHER READING CHAPTER 9: Bistability 2: Systems with Two Time-Dependent Variables 9.1 TWO-VARIABLE POSITIVE FEEDBACK AND DOUBLE-NEGATIVE FEEDBACK LOOPS CAN FUNCTION AS BISTABLE SWITCHES 9.2 LINEAR STABILITY ANALYSIS EXPLAINS THE DYNAMICS OF THE SYSTEM NEAR EACH OF THE STEADY STATES 9.3 TO APPLY LINEAR STABILITY ANALYSIS TO A TWO-VARIABLE SYSTEM, WE CALCULATE EIGENVECTORS AND EIGENVALUES 9.4 THE SYSTEM CAN CHANGE BETWEEN STATES VIA A SADDLE-NODE BIFURCATION 9.5 DOUBLE-NEGATIVE FEEDBACK PLUS ULTRASENSITIVITY CAN YIELD BISTABILITY 9.6 PERFECT SYMMETRY CAN PRODUCE A PITCHFORK BIFURCATION 9.7 IN THE ABSENCE OF PERFECT SYMMETRY, A PITCHFORK BIFURCATION MORPHS INTO A SADDLE-NODE BIFURCATION SUMMARY FURTHER READING CHAPTER 10: Transcritical Bifurcations in Phase Separation and Infectious Disease INTRODUCTION 10.1 LIQUID–LIQUID PHASE SEPARATION CAN PRODUCE DISCRETE FUNCTIONAL DOMAINS THAT LACK MEMBRANES 10.2 PHASE SEPARATION CAN BE MODELED BY A SINGLE RATE EQUATION WITH POSITIVE FEEDBACK AND A TRANSCRITICAL BIFURCATION 10.3 THE TIME COURSE OF DROPLET FORMATION IS SIGMOIDAL 10.4 THE SAME PRINCIPLES UNDERPIN THE FORMATION OF PHOSPHOLIPID VESICLES 10.5 THE SIR (SUSCEPTIBLE-INFECTEDRECOVERED) MODEL EXPLAINS WHY INFECTIOUS DISEASES SOMETIMES SPREAD EXPLOSIVELY 10.6 THE SIR MODEL PREDICTS EXPONENTIAL GROWTH FOLLOWED BY EXPONENTIAL DECAY 10.7 THE BASIC REPRODUCTION NUMBER R[sub(0)] DETERMINES WHETHER AN INFECTION WILL GROW EXPONENTIALLY 10.8 THE PROPORTION OF THE POPULATION THAT WILL ULTIMATELY BECOME INFECTED DEPENDS ON R[sub(0)] 10.9 MANIPULATING R[sub(0)] CAN DELAY AN EPIDEMIC, DECREASE THE PEAK, AND DIMINISH THE FINAL NUMBER OF INFECTED INDIVIDUALS SUMMARY FURTHER READING CHAPTER 11: Negative Feedback 1: Stability and Speed INTRODUCTION 11.1 NEGATIVE FEEDBACK CAN INCREASE THE STABILITY OF A STEADY STATE 11.2 NEGATIVE FEEDBACK CAN ALLOW A SYSTEM TO RESPOND MORE QUICKLY FURTHER READING CHAPTER 12: Negative Feedback 2: Adaptation INTRODUCTION 12.1 BACTERIA FIND FOOD SOURCES THROUGH A BIASED RANDOM WALK 12.2 BACTERIA SUPPRESS TUMBLING IN RESPONSE TO CHEMOATTRACTANTS AND THEN ADAPT PERFECTLY 12.3 A PLAUSIBLE NEGATIVE FEEDBACK MODEL CAN ACCOUNT FOR PERFECT ADAPTATION 12.4 THE RESPONSE OF THE ERK MAP KINASES TO MITOGENIC SIGNALS IS TYPICALLY TRANSITORY 12.5 DELAYED NEGATIVE FEEDBACK CAN YIELD NEAR-PERFECT ADAPTATION 12.6 ULTRASENSITIVITY IN THE FEEDBACK LOOP IMPROVES THE SYSTEM’S ADAPTATION 12.7 INDUCTION OF IMMEDIATE-EARLY GENE PRODUCTS IS NOT REQUIRED FOR ERK INACTIVATION IN MANY CELL TYPES SUMMARY FURTHER READING CHAPTER 13: Adaptation 2 : Incoherent Feedforward Regulation and State-Dependent Inactivation INTRODUCTION 13.1 RECEPTOR TYROSINE KINASE ACTIVATION IS FOLLOWED BY TRANSITORY RAS ACTIVATION 13.2 THE SEQUENTIAL RECRUITMENT OF SOS AND GAP TO THE EGFR CAN BE VIEWED AS INCOHERENT FEEDFORWARD REGULATION 13.3 INCOHERENT FEEDFORWARD SYSTEMS CAN YIELD PERFECT ADAPTATION 13.4 STRICT ORDERING OF SOS AND GAP BINDING TO THE EGFR IS NOT REQUIRED FOR PERFECT ADAPTATION 13.5 THE VOLTAGE-SENSITIVE SODIUM CHANNEL ALSO UNDERGOES SEQUENTIAL ACTIVATION AND INACTIVATION 13.6 EGFR INTERNALIZATION CAN BE VIEWED AS STATE-DEPENDENT INACTIVATION 13.7 GPCR SIGNALING IS SWITCHED FROM G-PROTEINS TO β-ARRESTIN VIA A MECHANISM AKIN TO STATE-DEPENDENT INACTIVATION SUMMARY FURTHER READING CHAPTER 14: Negative Feedback 3: Oscillations INTRODUCTION 14.1 BIOLOGICAL OSCILLATIONS CONTROL MYRIAD ASPECTS OF LIFE AND OPERATE OVER A TEN-BILLION-FOLD RANGE OF TIME SCALES 14.2 THE GOODWIN OSCILLATOR IS BUILT UPON A THREE-TIER CASCADE WITH HIGHLY ULTRASENSITIVE NEGATIVE FEEDBACK 14.3 LINEAR STABILITY ANALYSIS YIELDS A PAIR OF COMPLEX EIGENVALUES 14.4 OSCILLATIONS ARE BORN AND EXTINGUISHED AT HOPF BIFURCATIONS 14.5 SIMPLE HARMONIC OSCILLATORS ARE NOT LIMIT CYCLE OSCILLATORS SUMMARY FURTHER READING CHAPTER 15: Relaxation Oscillators INTRODUCTION 15.1 THE XENOPUS EMBRYONIC CELL CYCLE IS DRIVEN BY A RELIABLE BIOCHEMICAL OSCILLATOR 15.2 THE CELL CYCLE OSCILLATOR INCLUDES A NEGATIVE FEEDBACK LOOP AND A BISTABLE TRIGGER 15.3 A SIMPLIFIED MODEL CAPTURES THE BASIC DYNAMICS OF THE CELL CYCLE OSCILLATOR 15.4 THE CELL CYCLE MODEL HAS A SINGLE UNSTABLE STEADY STATE 15.5 TUNING THE OSCILLATOR CHANGES THE PERIOD MORE THAN THE AMPLITUDE 15.6 PHASE PLANE ANALYSIS SHOWS WHY THE HOPF BIFURCATIONS OCCUR WHERE THEY DO 15.7 INTERLINKED POSITIVE AND DOUBLE-NEGATIVE FEEDBACK LOOPS CAN MAKE THE MITOTIC TRIGGER MORE ALL-OR-NONE AND MORE ROBUST 15.8 THE FITZHUGH–NAGUMO MODEL ACCOUNTS FOR THE ELECTRICAL OSCILLATIONS OF THE SINOATRIAL NODE 15.9 THE FITZHUGH–NAGUMO MODEL CONSISTS OF A QUICK BISTABLE SWITCH AND A SLOWER NEGATIVE FEEDBACK LOOP 15.10 THE CELL CYCLE OSCILLATOR AND THE FITZHUGH–NAGUMO OSCILLATOR SHARE THE SAME SYSTEMS-LEVEL LOGIC 15.11 DEPLETION CAN TAKE THE PLACE OF NEGATIVE FEEDBACK IN A RELAXATION OSCILLATOR SUMMARY FURTHER READING CHAPTER 16: Excitability INTRODUCTION 16.1 THE RECEPTOR TYROSINE KINASE/MAP KINASE SYSTEM INCLUDES MULTIPLE POSITIVE AND NEGATIVE FEEDBACK LOOPS 16.2 EXCITABLE RESPONSES CAN BE GENERATED BY A FAST POSITIVE FEEDBACK LOOP COUPLED TO A SLOW NEGATIVE FEEDBACK LOOP 16.3 NOISE CAN CAUSE AN EXCITABLE SYSTEM TO FIRE SPORADICALLY SUMMARY FURTHER READING CHAPTER 17: Wrap-Up 17.1 THE BUILDING BLOCKS 17.2 MOTIFS 17.3 SIGNAL PROCESSORS 17.4 NONLINEAR DYNAMICS Glossary Index