How DNA Replication Actually Works: The Molecular Machine That Copies You

## The Locked Door

Before any DNA can be copied, the two strands have to be separated. They're held together by hydrogen bonds between complementary bases — A pairs with T, G pairs with C — and in the human genome, there are about 3 billion of these base pairs, all wound into a tight right-handed helix with about 10.5 base pairs per turn.

Separating them isn't just a matter of breaking hydrogen bonds. The molecule is supercoiled, compacted, and wound around histone proteins. And the act of unwinding one section of DNA creates tension elsewhere. This is where helicases come in — but helicases, it turns out, create their own problem that requires a second class of enzyme to fix.

## What Helicases Actually Do

Helicases are ring-shaped motor proteins that encircle one strand of DNA and use ATP hydrolysis to physically translocate along it, forcing the strands apart as they move. In bacteria, the primary helicase is DnaB. In eukaryotes, the equivalent is the CMG complex — a much more elaborate assembly built around proteins called CDC45, MCM2-7 (a hexameric ring), and GINS.

The MCM2-7 hexamer is loaded onto DNA during a process called replication origin licensing, which happens in the gap between cell divisions. The helicase is loaded in an inactive form, and then activated at the start of S phase by kinases. This two-step process — load first, activate later — is one of the cell's safeguards against replicating DNA more than once per cell cycle.

Once activated, the helicase unwinds DNA at roughly 500–1000 base pairs per second in bacteria, somewhat slower in eukaryotes. The unwound single-stranded regions are immediately coated by single-stranded DNA binding proteins (SSBPs in bacteria, RPA in eukaryotes) that keep the strands apart and protect them from forming secondary structures.

## The Topoisomerase Problem

Here's the problem that unwinding creates: DNA is a closed circular molecule in bacteria, and even in linear chromosomes, the chromatin structure effectively constrains the ends. When you unwind a helix, the rotational strain has to go somewhere. If it can't dissipate by rotating the entire chromosome (it can't), it accumulates as positive supercoiling ahead of the replication fork.

Positive supercoiling ahead of the fork means the DNA becomes tighter and tighter as replication proceeds. Left unchecked, this would stall the helicase completely within a few kilobases.

The solution is topoisomerases — enzymes that temporarily cut one or both strands of DNA, allow rotation to occur, and then reseal the cut. Type I topoisomerases cut one strand and relieve torsional strain by allowing the intact strand to rotate. Type II topoisomerases (like DNA gyrase in bacteria, or Topo IIα in eukaryotes) cut both strands simultaneously, pass another segment of double-stranded DNA through the break, and reseal it. Type II topoisomerases actively introduce negative supercoils, working against the accumulating positive supercoiling ahead of the fork.

Bacterial DNA gyrase is the target of fluoroquinolone antibiotics — drugs like ciprofloxacin work by trapping gyrase in its DNA-cutting intermediate state, creating lethal double-strand breaks. The reason these antibiotics are selectively toxic to bacteria (and not human cells) is that bacterial gyrase and human Topo IIα are structurally different enough that the drugs bind one but not the other.

## Two Replication Forks, Moving Apart

In most organisms, when a helicase is loaded onto DNA and activated, it splits into two complexes that move in opposite directions from the origin of replication. This creates a "replication bubble" that expands bidirectionally.

Each side of the bubble has its own replication fork — its own leading and lagging strand machinery, its own topoisomerase activity, its own set of RPA-coated single-stranded DNA. The two forks move outward until they either reach the ends of the chromosome or collide with another replication fork coming from a neighboring origin.

In bacteria, there's typically one origin per chromosome. In human cells, there are about 50,000 potential origins, of which ~10,000–30,000 are activated in any given S phase. This is necessary simply because of scale: at the replication rates observed in human cells, a single origin would take about a month to replicate one chromosome. The parallel activation of thousands of origins compresses this to 6–8 hours.

The next chapter goes into the replication fork itself — the full machinery of proteins assembling at each unwound region to actually synthesize new DNA.

Unwinding the Helix: Helicases and the Topoisomerase Problem

// COMMENTS

ON THIS PAGE