The Molecule This Whole Thread Was Pointing At

Since Class 6, this thread has kept arriving at the same word without ever stopping to explain it. A frog was alive because of processes happening inside its cells. Cells turned out to be built from a nucleus holding chromosomes. Chromosomes turned out to be made of chromatin, DNA wound around proteins. Reproduction turned out to be, at bottom, an act of DNA copying itself. Mitosis and meiosis both turned out to be, underneath all the machinery of spindle fibres and centromeres, processes built entirely around distributing copies of DNA correctly. Every single time, DNA has been the thing actually doing the work, and every single time, this thread has moved on without asking what DNA actually is, structurally, as a molecule, or how a cell manages to copy something roughly two metres long with enough accuracy that none of the rest of it would work otherwise. A Swiss physician named Friedrich Miescher first isolated this substance from cell nuclei in 1869 and named it nuclein, but its actual three-dimensional structure stayed a mystery for more than eighty years afterward, simply because no one had a way to see a molecule that small. This chapter is where that gap finally closes.

DNA is a long polymer, meaning a chain built from many repeating smaller units, in this case units called nucleotides. Each nucleotide is itself built from three parts: a nitrogenous base, a five-carbon sugar called deoxyribose, and a phosphate group. There are two families of nitrogenous base: purines, which are adenine and guanine, and pyrimidines, which are cytosine and thymine, with a fourth pyrimidine, uracil, appearing only in DNA's close chemical relative, RNA, in thymine's place. A base attaches to its sugar, and a phosphate group attaches to that, and the resulting nucleotides link together, phosphate to sugar to phosphate to sugar, forming a long backbone with the individual bases sticking out sideways from it like rungs waiting to be completed. This chain has a direction: one end carries a free phosphate group and is called the 5' end, the other carries a free sugar hydroxyl group and is called the 3' end. A length of DNA can be enormous. A virus called phi X174 carries 5,386 nucleotides; the bacterium Escherichia coli carries about 4.6 million base pairs; a single set of human chromosomes carries roughly 3.3 billion. Whatever the length, the underlying chemistry is the same repeating pattern, over and over.

A biochemist named Erwin Chargaff had noticed something curious about DNA from many different organisms: whatever the total amount, the quantity of adenine always closely matched the quantity of thymine, and the quantity of guanine always closely matched the quantity of cytosine. In 1953, James Watson and Francis Crick, working from X-ray diffraction images of DNA produced by Rosalind Franklin and Maurice Wilkins, used Chargaff's ratios as a critical clue and proposed a structure that explained them perfectly: the double helix. Franklin's own diffraction images, especially one now known simply as Photo 51, gave the clearest experimental evidence yet of DNA's helical shape and dimensions, evidence Watson and Crick relied on directly in building their model, work for which Franklin herself, having died in 1958, was never able to share in the Nobel Prize eventually awarded for it. The model itself has several defining features. It is built from two polynucleotide chains, not one, with the sugar-phosphate backbones on the outside and the bases pointing inward toward each other. The two chains run in opposite directions, one 5' to 3', the other 3' to 5', a relationship called antiparallel. The bases from each strand pair up across the middle through hydrogen bonds, and they do not pair randomly: adenine always pairs with thymine, held together by two hydrogen bonds, while guanine always pairs with cytosine, held together by three. Because a purine, the larger kind of base, always ends up paired with a pyrimidine, the smaller kind, the distance across the helix stays essentially constant all the way along its length. The whole structure coils in a right-handed spiral, completing one full turn roughly every ten base pairs, a rise of about 3.4 nanometres per turn.

A typical mammalian cell's DNA, stretched out end to end, measures roughly 2.2 metres, packed somehow into a nucleus only about a millionth of a metre across. That packing is not random stuffing, it is careful, structured engineering, and it directly extends the chromosome structure covered earlier in this thread. Eukaryotic cells rely on a set of small, positively charged proteins called histones, rich in the amino acids lysine and arginine, which naturally carry a positive electrical charge. Since DNA itself is negatively charged, along its phosphate backbone, it wraps snugly around a cluster of eight histone proteins, a histone octamer, forming a repeating structural unit called a nucleosome, each one built around roughly 200 base pairs of DNA. Under an electron microscope, a stretch of chromatin studded with nucleosomes looks like a string with beads spaced along it, an image directly visible in electron micrographs and sometimes called exactly that, the beads-on-a-string structure. Those beaded strings coil further still, packed down with the help of additional non-histone proteins into increasingly compact chromatin fibres, and it is only this fully condensed form that becomes visible as the distinct, countable chromosomes described earlier in this thread, appearing specifically at metaphase during cell division. Not all chromatin stays equally packed all the time, either. Regions that remain loosely packed, staining lightly under a microscope, are called euchromatin, and these are the actively transcribed, functionally busy regions. Regions packed more densely, staining darker, are called heterochromatin, largely inactive.

None of this was obvious in advance. For decades after Miescher first isolated nuclein, most scientists actually suspected protein, not DNA, was the genetic material, since proteins seemed to offer far more structural variety. The first real crack in that assumption came from a British medical officer named Frederick Griffith in 1928, working with two strains of the pneumonia-causing bacterium Streptococcus pneumoniae. One strain, called S, grew smooth, shiny colonies, wrapped in a protective coat, and killed mice it infected. The other, called R, grew rough colonies, lacked that coat, and was harmless. Griffith found that heat-killed S bacteria, on their own, no longer killed mice, exactly as expected. But when he injected mice with a mixture of heat-killed S bacteria and live, harmless R bacteria, something unexpected happened: the mice died, and living S bacteria, fully virulent, could be recovered from their bodies afterward. Somehow, some substance from the dead S bacteria had transformed the living, harmless R bacteria into the deadly S form, permanently. Griffith called this unidentified substance the transforming principle, but his experiments could not say what it actually was chemically.

It took another fifteen years to pin down the chemical identity of Griffith's transforming principle. Between 1933 and 1944, Oswald Avery, Colin MacLeod and Maclyn McCarty systematically purified the different biochemicals inside heat-killed S bacteria, proteins, DNA, RNA, separately, then tested each one to see which alone could transform live R bacteria into the deadly S form. Only the purified DNA worked. Digesting their sample with enzymes that break down protein, or enzymes that break down RNA, made no difference to the transformation at all, but digesting it with an enzyme that specifically breaks down DNA stopped the transformation completely. That was about as strong a piece of chemical evidence as could be asked for, yet, remarkably, not every biologist at the time was convinced, protein still had many defenders. The final, unambiguous proof came from Alfred Hershey and Martha Chase in 1952, using bacteriophages, viruses that infect bacteria by injecting their genetic material into the bacterial cell while leaving their outer protein coat behind. Hershey and Chase grew separate batches of these viruses, one batch labelled with radioactive phosphorus, which is found in DNA but not protein, the other labelled with radioactive sulfur, which is found in protein but not DNA. After letting each batch infect fresh bacteria, they violently agitated the mixture in a blender to shake loose anything that had stayed outside the bacterial cells, then spun the mixture in a centrifuge to separate the bacteria from everything else. The radioactive phosphorus, meaning the DNA, turned up inside the bacteria. The radioactive sulfur, meaning the protein, did not. DNA, and only DNA, had entered the cell and gone on to direct the production of new viruses. The question was settled.

For any molecule to function as genetic material, it has to satisfy a handful of demands: it must be able to make accurate copies of itself, it must be chemically and structurally stable enough to survive intact, it must allow occasional, slow changes since without any variation at all evolution has nothing to work with, and it must be able to actually express itself as observable traits. Both DNA and RNA can technically satisfy all of these, since both rely on the same base-pairing logic to guide their own copying, which is exactly why some viruses today still use RNA rather than DNA as their genetic material. But DNA has a decisive edge in stability. Every RNA nucleotide carries an extra reactive hydroxyl group that DNA nucleotides lack, making RNA chemically more fragile and easily broken down, and RNA is also capable of acting as a catalyst for chemical reactions, a property that makes it more reactive still. DNA, being double-stranded and built from two mutually complementary strands, gained an additional advantage: damage to one strand can be repaired using the other, undamaged strand as a reference copy. There is a genuinely striking idea behind why this comparison even matters: a large body of evidence now points to RNA, not DNA, as the very first genetic material to exist, with essential life processes originally evolving around it before DNA, chemically more stable, evolved afterward specifically to take over as life's long-term information archive, leaving RNA with its still-vital, more dynamic messenger role instead.

When Watson and Crick first published their double helix model in 1953, they included one line that turned out to be remarkably understated: it has not escaped our notice, they wrote, that the specific base pairing we have proposed immediately suggests a possible copying mechanism for the genetic material. They were right. Because adenine only ever pairs with thymine, and guanine only ever pairs with cytosine, each single strand of the double helix already contains, implicitly, the complete information needed to rebuild its exact partner. If the two strands of a DNA molecule separate, and each one then serves as a template guiding the assembly of a brand new complementary strand alongside it, the result is two complete double helices where there was previously one, and each of those two new molecules ends up built from one original, parental strand plus one freshly synthesised strand. This scheme was given a name: semiconservative replication, semiconservative because each new molecule conserves, keeps intact, exactly half of the original.

A proposal, however elegant, still needs experimental proof, and that proof came from Matthew Meselson and Franklin Stahl in 1958, in an experiment often held up as one of the most beautifully designed in the whole history of biology. They first grew E. coli bacteria for many generations in a medium containing only a heavy isotope of nitrogen, 15N, so that every new DNA molecule the bacteria built during that time incorporated the heavier isotope throughout, making that DNA slightly denser than normal. This heavy DNA could be distinguished from ordinary DNA by spinning it in a centrifuge against a caesium chloride density gradient, where denser molecules settle further down. Meselson and Stahl then transferred these bacteria into a fresh medium containing only ordinary, lighter nitrogen, 14N, and extracted DNA samples at set intervals afterward as the bacteria kept dividing. After exactly one generation, they found something decisive: the DNA was neither fully heavy nor fully light, it settled at a single intermediate, hybrid density, precisely what semiconservative replication predicts, one old heavy strand paired with one newly built light strand in every molecule. After a second generation, the DNA split into two distinct bands, half at that same hybrid density, half fully light, exactly as predicted if the hybrid molecules from generation one had themselves each separated and built new light partner strands. A separate experiment on faba bean chromosomes by Taylor and colleagues, that same year, using radioactively labelled building blocks instead of density differences, reached the identical conclusion in a plant's actual chromosomes: replication really is semiconservative.

Knowing that DNA replicates semiconservatively still leaves the question of how, mechanically, a cell actually pulls this off, fast enough and accurately enough to matter. The main working enzyme is called DNA polymerase, which builds a new strand by reading an existing DNA strand as its template. These enzymes are extraordinarily fast: E. coli, with about 4.6 million base pairs of DNA total, completes replicating its entire genome in around eighteen minutes, meaning DNA polymerase is adding roughly two thousand new nucleotides every single second, while still making very few mistakes. Copying an entire DNA molecule in one go is not physically practical either, separating the full length of the double helix at once would require far too much energy, so instead replication proceeds through a small local opening in the helix called the replication fork, which moves steadily along the molecule as copying proceeds. A further complication arises because DNA polymerase can only build a new strand in one direction, from the 5' end toward the 3' end. Since the two template strands run in opposite directions to each other, one new strand can be built continuously, following the fork as it opens, while the other has to be built in short, separate fragments, working backward, which are afterward stitched together into one continuous strand by a second enzyme called DNA ligase. Replication does not begin at a random point either, it starts from a specific, defined location on the DNA called the origin of replication. And in eukaryotic cells, this entire process is not left to happen whenever, it is tightly scheduled: DNA replication happens specifically during S phase of the cell cycle, precisely the phase this thread's Class 11 chapter already identified as the point where DNA amount doubles while chromosome number stays the same, coordinated tightly with the rest of the division machinery so that a complete, accurate genome is ready exactly when mitosis or meiosis needs it.

Class 6 opened this thread by asking what actually separates a living frog from a car that also happens to move: nutrition, growth, response, reproduction, a short list of shared processes. Every class since has pushed that same question one layer deeper. Class 9 found cells and their organelles doing the work. Class 11 found those organelles duplicating themselves and splitting with remarkable precision through mitosis and meiosis. And this chapter has finally found the molecule underneath all of it: a double helix of paired bases, packed by histone proteins into chromatin and then into chromosomes, copied through a semiconservative process so fast and so accurate that an entire bacterial genome can be duplicated correctly in about eighteen minutes. A frog is alive, in the end, because trillions of its cells are each running this same basic operation, copying a molecule built from four repeating bases, faithfully enough that a tadpole reliably grows into a frog and not something else, yet imperfectly enough, across countless generations, that frogs today are not identical to the very first frog that ever lived. That combination, extraordinary fidelity paired with just enough imperfection, is not a side detail of biology. Once this thread's Class 10 chapter introduced reproduction and Class 11 explained cell division, this chapter's molecule, DNA, turns out to have been the answer running underneath every single class in this progression, all the way back to that very first question about a frog.