9 topics. Each one opens with its key idea, then the detail you need to use it.
1. Reception: three receptor classes, three logics
Key idea: GPCR activates a separate G protein; RTK is its own kinase and must dimerize; ion channels open a pore; steroid receptors sit inside the cell and regulate transcription directly.
Group receptors by where they sit and what they do first. G protein-coupled receptors (GPCRs) are single polypeptides that thread through the membrane seven times. Ligand binding twists the intracellular face so it can grab a G protein, which swaps its GDP for GTP, breaks apart, and moves along the membrane to switch on an effector enzyme. Receptor tyrosine kinases (RTKs) are different: the receptor is itself an enzyme. Ligand binding drives two receptor subunits together, and once their intracellular kinase domains are side by side they phosphorylate each other on tyrosines. Those phosphotyrosines become docking sites where relay proteins attach and get activated. Ligand-gated ion channels are the fastest — the ligand binds, a gate opens, and ions flood down their gradient, which is how a neuron's postsynaptic membrane responds in milliseconds. Intracellular receptors are the exception to the whole scheme: testosterone, estrogen, and thyroid hormone are small and hydrophobic, so they cross the membrane and bind receptors in the cytosol or nucleus, and the receptor-hormone complex acts directly as a transcription factor.
2. The GPCR pathway: G protein, cAMP, PKA
Key idea: Ligand to GPCR to G protein (GDP swapped for GTP) to adenylyl cyclase to cAMP to PKA to phosphorylated targets — and cAMP never touches DNA.
Trace this pathway in order, because the exam asks you to. Epinephrine binds a beta-adrenergic GPCR. The receptor changes shape and acts on a stimulatory G protein (Gs), causing its alpha subunit to release GDP and bind GTP. The GTP-bound alpha subunit detaches and activates adenylyl cyclase, a membrane enzyme that converts ATP into cyclic AMP. cAMP is the second messenger: small, soluble, and free to diffuse through the cytosol carrying the message away from the membrane. Its main target is protein kinase A. Inactive PKA is a four-subunit complex — two regulatory, two catalytic. When cAMP binds the regulatory subunits, the catalytic subunits are released and start phosphorylating targets: metabolic enzymes, ion channels, and transcription factors such as CREB. The pathway shuts itself down in two ways. The alpha subunit hydrolyzes its own GTP back to GDP and reassembles with the other subunits, and phosphodiesterase degrades cAMP to AMP. That second switch is a favorite exam probe. Give a cell a drug that inhibits phosphodiesterase and cAMP piles up, so the response lasts longer and grows stronger — it does not fade.
3. RTKs, dimerization, and Ras
Key idea: No dimerization, no cross-phosphorylation, no signal — and one growth factor binding event launches a branching cascade, not a single line.
Growth factors work through RTKs, and the mechanism hinges on one fact: a lone RTK monomer is inactive. Its kinase domain cannot do anything useful alone. When PDGF or VEGF binds, two receptor subunits pair up, the two intracellular kinase domains come within reach of one another, and each phosphorylates tyrosines on its partner. This cross-phosphorylation both switches the kinase domains on and creates a row of phosphotyrosine docking sites. Relay proteins with the right recognition domains bind there: PI3K, phospholipase C, and adaptors that recruit Ras. Ras is a small G protein tethered to the inner face of the membrane, and it works as a binary switch — GTP-bound is ON, GDP-bound is OFF. Active Ras starts the MAP kinase cascade, a series of kinases each phosphorylating the next, ending in transcription factors that turn on genes for cyclins and other proteins needed to enter S phase. Insulin uses the same architecture through a receptor that is already a permanent dimer, autophosphorylating when insulin binds and signaling through IRS proteins and PI3K to move GLUT4 transporters into the membrane.
4. Second messengers, amplification, and termination
Key idea: PLC splits PIP2 into DAG and IP3; IP3 releases ER calcium; degrading IP3 and pumping calcium back are what make the signal a pulse instead of a permanent state.
Second messengers are small non-protein molecules that spread a signal fast because they diffuse. cAMP is one. Calcium signaling is the other major branch, and it runs through different machinery. A GPCR or RTK activates phospholipase C, which cleaves the membrane phospholipid PIP2 into two products: DAG, which stays in the membrane and activates protein kinase C, and IP3, which diffuses to the endoplasmic reticulum and opens IP3-gated calcium channels. Cytosolic calcium then rises about ten- to a hundred-fold, from roughly 0.1 micromolar at rest to around 1 micromolar, and higher still right beside an open channel. Do not confuse that rise with the ten-thousand-fold number you may have seen. That figure is the standing gradient a resting cell maintains between its cytosol and the ER lumen or the fluid outside, both near 1 millimolar. Storing calcium against such a steep gradient is exactly why opening a channel floods the cytosol in a fraction of a second. Calcium-binding proteins like calmodulin translate that spike into enzyme activation. Now the part students skip. IP3 5-phosphatase strips a phosphate from IP3 to make IP2, which does not bind IP3 receptors at all, so the channels close. SERCA pumps calcium back into the ER. Together these end the signal within seconds of the receptor going quiet. That is the design principle: the response tracks the stimulus only because degradation runs constantly in the background. Inhibit a terminating enzyme and you extend and amplify the response.
5. Communication across distance, contact, and induction
Key idea: Gap junctions and plasmodesmata are analogous structures that both create direct cytoplasmic continuity; autocrine, paracrine, synaptic, and endocrine differ in how far the ligand travels; and differentiation is one genome read differently because each cell sits in a different signal environment.
Classify signaling by how far the message travels. Direct contact is the shortest range. Animal cells connect through gap junctions built from connexin proteins — six connexins form a connexon, and two aligned connexons make a pore that lets ions and molecules under roughly 1,000 daltons pass straight between cytoplasms. Plant cells do the equivalent with plasmodesmata, membrane-lined channels punched through the cell wall with a strand of ER running down the middle. Different structures, same job: shared cytoplasm, so neighbors coordinate without anything crossing the extracellular space. Cardiac muscle depends on this — gap junctions let depolarization sweep from cell to cell so the heart contracts as a unit, and losing connexin function causes uncoordinated, arrhythmic contraction. Beyond direct contact: autocrine signaling acts back on the cell that released the ligand, paracrine signals diffuse to nearby cells, synaptic signaling fires neurotransmitters across a narrow cleft, and endocrine hormones travel through the bloodstream to distant targets. Bacteria use quorum sensing, secreting autoinducers whose concentration reports population density; once it crosses a threshold, the whole population switches on genes for biofilm formation, virulence, or bioluminescence.
Signals also decide what a cell becomes. Every cell in your body carries the same genome, so a neuron and a liver cell differ only in which genes they read. That choice comes from the signals each cell sits in, and this is cell differentiation. In induction, one group of cells secretes a paracrine signal that changes the fate of its neighbors. Often the signal spreads out as a morphogen gradient. Cells near the source sit in a high concentration, cells farther away sit in a low one, and different concentrations switch on different genes, so position in the embryo becomes identity. The mechanism is the transduction you already know. Ligand binds receptor, a cascade fires, and the last step activates or inactivates transcription factors. Those transcription factors turn on a set of genes that includes still more transcription factors, which lock the choice in and keep it stable through later divisions. Differentiation never removes DNA. A muscle cell still carries the genes for hemoglobin; it just never transcribes them.
6. Feedback: holding a set point or running away from it
Key idea: Negative feedback returns a variable to a set point (insulin and glucagon on blood glucose); positive feedback amplifies a change until an outside event ends it (clotting, labor, fruit ripening).
Pathways rarely run in a straight line. The output usually loops back and changes the input, and there are only two directions that loop can go.
Negative feedback pushes a variable back toward a set point. Blood glucose is the standard case. After a meal, glucose rises, beta cells in the pancreas release insulin, and insulin makes muscle and liver cells take glucose in and store it as glycogen. Glucose falls, so insulin release falls with it. Drop too low and alpha cells release glucagon instead. The liver breaks glycogen back down and glucose climbs again. Two hormones, opposite directions, one set point — a fasting level near 70 to 100 mg/dL. The same logic runs inside pathways, where an end product often inhibits the enzyme that started its own production. Homeostasis is built almost entirely out of loops like this.
Positive feedback does the opposite. The output makes more of itself, so a small change grows until something outside the loop ends it. A cut blood vessel shows it cleanly. Activated platelets stick to the damaged wall and release ADP and thromboxane A2, which activate more platelets nearby, which release more of the same signal. The plug builds in seconds. Childbirth runs the same loop: the baby's head stretches the cervix, stretch receptors trigger oxytocin release, oxytocin drives stronger uterine contractions, and those contractions push harder on the cervix. Delivery is what stops it. Fruit does it too — ethylene from one ripening apple triggers ripening, and more ethylene, in the apples around it.
On the exam the tell is direction. Ask one question: does the response cancel the original change, or add to it? Cancel means negative. Add means positive.
7. The cell cycle and its checkpoints
Key idea: CDKs need cyclins to work, cyclin levels oscillate, and the G1 checkpoint is where growth-factor signaling, p53, and Rb decide whether division happens at all.
Interphase is not resting. G1 is growth and normal metabolism, S is DNA replication (each chromosome becomes two sister chromatids joined by cohesin), and G2 is more growth plus preparation for division. M phase — mitosis plus cytokinesis — is the short part. Cells that exit the cycle enter G0, where most neurons spend their lives. The engine is a set of cyclin-dependent kinases, which are present all the time but inactive until they bind a cyclin. Cyclin concentrations rise and fall on a schedule, so CDK activity comes in waves. The G1 checkpoint, sometimes called the restriction point, is the most consequential: it asks whether the cell is big enough, whether nutrients are available, whether DNA is undamaged, and whether growth factors are signaling. Pass it and the cell is committed to a full round of division. The braking machinery works here too. Rb protein grips the transcription factor E2F and holds it inactive; cyclin D-CDK4/6 phosphorylate Rb, E2F is released, and S-phase genes turn on. If DNA is damaged, p53 accumulates and switches on p21, which inhibits cyclin-CDK complexes and stalls the cell in G1 for repair.
8. Mitosis, microtubules, and cytokinesis
Key idea: Microtubules move chromosomes, actin divides the cytoplasm in animal cells, and nuclear envelope breakdown happens in prometaphase to give the spindle access.
Get the order and the machinery right; this is where perturbation questions live. In prophase, chromatin condenses into visible chromosomes and the mitotic spindle starts forming from centrosomes. In prometaphase the nuclear envelope breaks down, and it does so because CDK1-cyclin B phosphorylates the lamin proteins of the nuclear lamina, causing that meshwork to fall apart. This matters mechanically: until the envelope is gone, spindle microtubules cannot reach the chromosomes. Kinetochore microtubules then attach to the kinetochore on each sister chromatid. Metaphase is the alignment of chromosomes at the cell equator with sisters attached to opposite poles. Anaphase begins when the anaphase-promoting complex tags securin for destruction, freeing the enzyme separase to cleave cohesin so sisters are pulled apart. Telophase reverses prophase — the envelope reassembles and chromosomes decondense. Cytokinesis uses a different cytoskeletal polymer entirely. Animal cells pinch in with a contractile ring of actin and myosin, producing a cleavage furrow; plant cells cannot pinch through a cell wall, so Golgi vesicles line up at the middle and fuse into a cell plate that becomes new wall. Two classic drugs test this split: colchicine blocks tubulin polymerization and arrests cells in prometaphase or metaphase because the spindle assembly checkpoint is never satisfied, while cytochalasin B blocks actin polymerization and lets mitosis finish but leaves cytokinesis to fail.
9. Cancer and apoptosis: two ways the cycle ends
Key idea: Oncogenes are hyperactive accelerators needing only one mutated copy; tumor suppressors are brakes that must lose both copies; apoptosis is ordered, non-inflammatory removal that development and damage control both depend on.
Cancer arises from two opposite kinds of mutation. Proto-oncogenes normally encode positive regulators — growth factors, their receptors, Ras, cyclins. A gain-of-function mutation converts one into an oncogene that pushes cell division constantly. The classic case is a Ras protein that has lost the ability to hydrolyze its bound GTP: it is stuck in the ON conformation, driving the MAP kinase cascade with no growth factor present at all. Tumor suppressors are the opposite. p53 and Rb are negative regulators, and cancer generally requires loss of function in both copies. p53 is mutated in roughly half of human cancers, which removes the DNA-damage brake at the G1 checkpoint. Apoptosis, or programmed cell death, is the deliberate demolition option. In the intrinsic pathway, mitochondria release cytochrome c from the intermembrane space into the cytosol, where it binds Apaf-1 to build an apoptosome that activates caspase-9, which activates executioner caspases that cut cellular proteins and chop DNA. The cell shrinks, blebs, and breaks into apoptotic bodies that neighbors engulf, so nothing leaks and no inflammation follows. This sculpts tissues during development — it removes the webbing between fingers and toes and prunes surplus neurons — and it clears cells whose DNA damage is beyond repair.