Science One-Pagers

1A receptor that never lets the signal in

G-protein-coupled receptor · PIP₂ · IP₃ and calcium

The claim: a signal molecule binds the outside of the membrane and never crosses it, yet calcium floods the cytosol — because the cell rebuilds the message inside out of its own parts.

The papers that established it

• Northup, Sternweis, Smigel, Schleifer, Ross & Gilman (1980), PNAS 77:6516–20. 10.1073/pnas.77.11.6516 Purified the G protein itself, so the thing in the middle of the diagram stopped being an inference and became a protein you could put in a tube. Gilman shared the 1994 Nobel Prize for this line of work.
• Streb, Irvine, Berridge & Schulz (1983), Nature 306:67–9. 10.1038/306067a0 Added IP₃ directly to pancreatic cells and watched calcium come out of a store that was not the mitochondria. This is the single step the diagram's arrow to the ER is drawing.
• Berridge & Irvine (1984), Nature 312:315–21. 10.1038/312315a0 The paper that named IP₃ a second messenger and set out the split the diagram shows: PIP₂ cut in two, DAG staying in the membrane to work on protein kinase C, IP₃ leaving to move calcium.
• Rasmussen and colleagues, with Kobilka (2011), Nature 477:549–55. 10.1038/nature10361 The receptor and its G protein caught together as a crystal structure. One transmembrane helix swings 14 Å outward. The "shape change" in the diagram has a number attached to it now. Kobilka shared the 2012 Nobel Prize.

Since then: the diagram is still a fair account of this pathway. What it leaves out is that this is one route among several — the same family of receptors can work through adenylyl cyclase and cAMP instead, which is why there is a line about that on the figure.

2Two halves, then several answers at once

Receptor tyrosine kinase · dimerization · phosphotyrosine docking

The claim: two receptor halves sit apart, each binds a signal, they join, each phosphorylates the other using ATP, and several relay proteins dock on the result.

The papers that established it

• Lemmon & Schlessinger (2010), Cell 141:1117–34. 10.1016/j.cell.2010.06.011 The standard review of how this whole receptor family works, by two of the people who worked it out. Worth noting that even in 2010 they call the strategies for getting these receptors to dimerize "surprisingly diverse".

What has changed

• Zuo, Schwartz, Walker, Han, Sternberg & Ferguson (2026), PNAS 123:e2602436123. 10.1073/pnas.2602436123 A cryo-EM structure of a receptor dimer that has formed without any ligand at all. In this picture the two halves are already together, held in a shape that keeps their kinase domains apart; the signal's job is to release that restraint rather than to introduce the halves to each other. The insulin receptor has long been known to be a permanently joined dimer for the same kind of reason.

So is the diagram wrong? No — but it is the classic model, which is what the textbook and the Checkpoint ask for. The honest version is: joining up is the switch for some of these receptors, and changing shape once joined is the switch for others. If you want the one-sentence upgrade: the signal does not always bring the halves together, but it always changes what the joined pair can do.

3Downhill is free. Uphill is paid for.

Gated channel · the sodium-potassium pump · passive and active transport

The claim: two proteins move the same ion through the same membrane, and only one of them costs the cell an ATP.

The paper that established it

• Skou (1957), Biochimica et Biophysica Acta 23:394–401. 10.1016/0006-3002(57)90343-8 Found an ATP-splitting enzyme in crab nerve that only worked when both sodium and potassium were present — which is how you discover a pump. The whole right-hand side of the diagram is this paper. Skou shared the 1997 Nobel Prize for it, forty years later.

Since then: nothing in the diagram has been overturned. The pump's three-out, two-in ratio and its ATP dependence are as Skou's line of work established them.

4Electrons buy a gradient. The gradient buys the ATP.

Electron transport chain · chemiosmosis · ATP synthase

The claim: electrons are passed down a chain of carriers, the energy released is spent pumping protons across a membrane, and the protons falling back through one enzyme is what actually makes the ATP.

The papers that established it

• Mitchell (1961), Nature 191:144–8. 10.1038/191144a0 Five pages proposing that the link between electron transport and ATP is not a chemical intermediate at all but a proton gradient across a membrane. It was widely disbelieved for a decade. Nobel Prize, 1978.
• Jagendorf & Uribe (1966), PNAS 55:170–7. 10.1073/pnas.55.1.170 The experiment that made Mitchell's idea hard to argue with: soak chloroplasts in acid, jump them into base, and ATP appears — in the dark, with no electron transport running. A proton gradient alone was enough.
• Abrahams, Leslie, Lutter & Walker (1994), Nature 370:621–8. 10.1038/370621a0 The structure of the ATP-making head, showing its three catalytic sites caught in three different states — the structural case that the thing rotates. Walker shared the 1997 Nobel Prize.

Where the picture got more complicated

• Schägger & Pfeiffer (2000), EMBO Journal 19:1777–83. 10.1093/emboj/19.8.1777 Showed the complexes are not scattered at random in the membrane: they assemble into larger assemblies, which the authors named the respirasome.
• Letts, Fiedorczuk, Degliesposti, Skehel & Sazanov (2019), Molecular Cell 75:1131–46. 10.1016/j.molcel.2019.07.022 Structures of one of those assemblies, and a surprise: trapping the mobile carrier inside it slowed the chain down, which argues against the idea that the assemblies exist to hand electrons along a private channel.

So is the diagram wrong? No, and in a useful way: the 2019 work argues against the complexes passing electrons hand-to-hand in a sealed assembly, which is exactly what the diagram declines to draw. Mobile carriers diffusing between separate complexes is still the right picture for this course. What the diagram leaves out is that the complexes are packed together in the membrane rather than floating free — true, and not something that changes the mechanism.

5One message, written out and then read aloud

Transcription · splicing · translation

The claim: a gene is copied into RNA inside the nucleus, the copy is processed and leaves, and a ribosome reads it three bases at a time, with each tRNA bringing one matching amino acid.

The papers that established it

• Hoagland, Stephenson, Scott, Hecht & Zamecnik (1958), Journal of Biological Chemistry 231:241–57. PubMed ID 13538965. Found the small RNA that carries an amino acid — the physical thing Crick had argued must exist as an "adaptor". This is the tRNA in the diagram.
• Brenner, Jacob & Meselson (1961), Nature 190:576–81. 10.1038/190576a0 Showed that an unstable RNA, not the ribosome itself, carries the information from gene to ribosome. The mRNA strand in the diagram is this paper.
• Nirenberg & Matthaei (1961), PNAS 47:1588–602. 10.1073/pnas.47.10.1588 Fed a cell-free system an RNA made only of uracil and got a protein made only of phenylalanine. The first codon ever read. Nobel Prize, 1968.
• Crick (1970), Nature 227:561–3. 10.1038/227561a0 Crick restating what the central dogma actually claimed, because people had been misquoting it for twelve years. Short, and worth reading for that alone.
• Berget, Moore & Sharp (1977), PNAS 74:3171–5. 10.1073/pnas.74.8.3171 Found that an mRNA did not match its gene in one continuous stretch: pieces had been cut out. Introns. Sharp shared the 1993 Nobel Prize, with Richard Roberts, whose group found the same thing independently.

What has changed

• Reimer, Mimoso, Adelman & Neugebauer (2021), Molecular Cell 81:998–1012. 10.1016/j.molcel.2020.12.018 Sequenced individual RNA molecules while they were still being made, and caught splicing happening when the polymerase was only 75 to 300 bases past the splice site — often while the next exon was still being transcribed.

So is the diagram wrong? Its order is a simplification. The diagram shows a finished copy and then a tidy splicing step; in a real nucleus, most splicing happens while the RNA is still being written, with the finished end of the molecule already being cut and joined before the far end exists. The parts are all correct and in the right relationship — they just overlap in time more than a still picture can show. This is the one thing an animation of this figure could fix for free.

6Cut both with one enzyme and the ends fit

Restriction enzymes · sticky ends · recombinant plasmids

The claim: cutting a plasmid and a piece of donor DNA with the same enzyme leaves overhangs that pair, ligase seals the join, and bacteria carrying the result can be selected on an antibiotic plate.

The papers that established it

• Mertz & Davis (1972), PNAS 69:3370–4. 10.1073/pnas.69.11.3370 Showed that the EcoRI enzyme cuts the two strands at staggered positions, leaving short single-stranded ends that are identical and complementary on every fragment — so any two pieces cut with it can be joined. The whole idea of the diagram is in this one result.
• Cohen, Chang, Boyer & Helling (1973), PNAS 70:3240–4. 10.1073/pnas.70.11.3240 Did it: built a new plasmid in a test tube from pieces of two others, put it into E. coli, and showed the bacteria now carried working genes from both parents. Four pages. It is the start of genetic engineering.

What labs do now

• Gibson, Young, Chuang, Venter, Hutchison & Smith (2009), Nature Methods 6:343–5. 10.1038/nmeth.1318 One tube, one temperature, three enzymes, and overlapping pieces join themselves — no restriction sites required. Most cloning today is done this way or by something like it.

So is the diagram out of date? It is the classic method, and it is still the one worth learning first, because the idea it teaches — make the ends match before you try to join them — is what every newer method is also doing, just more conveniently. Gibson assembly makes matching ends with an exonuclease instead of a restriction enzyme. Same idea, fewer constraints.