DNA copies itself during S phase (synthesis phase) of interphase, which comes before mitosis actually begins. In a typical human cell that divides every 24 hours, DNA replication takes about 8 hours and is fully complete well before the cell enters mitosis. By the time mitosis starts, every chromosome has already been duplicated into two identical sister chromatids joined at their center.
This is a common point of confusion. Mitosis is the dramatic finale where chromosomes line up and split apart, but the quiet work of copying all 6 billion base pairs of human DNA happens hours earlier, during a preparatory period called interphase.
Where S Phase Fits in the Cell Cycle
The cell cycle has two major parts: interphase (when the cell grows and prepares) and mitosis (when it physically divides). Interphase itself breaks down into three stages that happen in order:
- G1 phase (~11 hours): The cell grows, produces proteins, and carries out its normal functions. Late in G1, the cell commits to dividing and sets up the replication program.
- S phase (~8 hours): DNA replication happens here. Every chromosome is copied, producing two identical sister chromatids held together by protein rings.
- G2 phase (~4 hours): The cell checks its copied DNA for errors and prepares the machinery needed for division.
Mitosis itself lasts only about 1 hour. That means roughly 95% of the cell cycle is spent in interphase, with DNA replication consuming about a third of that time.
How DNA Gets Copied
During S phase, the double helix doesn’t unzip all at once. Replication starts at thousands of points along each chromosome, with regions of loosely packed DNA copying first and tightly packed regions copying later. A coordinated team of proteins handles the job.
First, an enzyme called helicase pries apart the two strands of the double helix, moving along the DNA and separating it at rates of up to 1,000 base pairs per second. Single-strand binding proteins immediately coat the exposed strands to keep them from snapping back together. Another enzyme relieves the tension that builds up ahead of the opening point, preventing the DNA from tangling into knots.
Once the strands are separated, DNA polymerase reads each strand and builds a matching copy, adding one nucleotide at a time. But there’s a catch: polymerase can only build in one direction. On one strand, it runs smoothly alongside the helicase. On the other strand, it has to work in short segments going the opposite way. A different enzyme lays down tiny RNA starter sequences so polymerase has something to grab onto, and once each segment is complete, yet another enzyme (DNA ligase) stitches the fragments together into one continuous strand.
The result is two complete, identical DNA molecules where there was one before. Each consists of one original strand and one newly built strand.
Built-In Error Correction
DNA polymerase doesn’t just build new DNA. It also proofreads its own work in real time. The enzyme has a separate site that detects mismatched base pairs at the growing tip of the new strand. When it finds one, it reverses direction, clips off the incorrect nucleotides, and resumes building with the correct ones. This proofreading function is the primary guardian of copying accuracy.
Even with proofreading, some errors slip through. A second layer of defense, called mismatch repair, scans the newly copied DNA after polymerase has moved on and fixes remaining mistakes. Together, these systems keep the error rate extraordinarily low, roughly one mistake per billion base pairs copied.
The G2 Checkpoint: No Errors Allowed Into Mitosis
After S phase finishes, the cell doesn’t immediately rush into division. G2 phase acts as a holding room where the cell verifies that replication is complete and the DNA is intact. If damage is detected, a signaling cascade called the DNA damage checkpoint halts the cell cycle. This system works by blocking the activation of a key protein that triggers entry into mitosis. The cell stays parked in G2 until repairs are finished.
This checkpoint is why cells almost never enter mitosis with incompletely copied or damaged DNA. The consequences of skipping this step would be catastrophic: broken chromosomes, lost genetic information, and potentially cancerous mutations passed to daughter cells.
What Happens to Copied DNA During Mitosis
During S phase and through G2, DNA stays in a loosely spread-out form inside the nucleus. This relaxed state is necessary because the replication machinery needs physical access to the strands. You couldn’t copy tightly coiled DNA any more than you could photocopy a crumpled piece of paper.
When mitosis begins, everything changes. The copied DNA condenses dramatically, coiling and folding until each chromosome becomes a compact, visible structure. The two copies of each chromosome, called sister chromatids, remain physically connected at a region near their center by protein rings that were loaded during replication. This connection is critical: it keeps the two copies paired so the cell can line them up at its middle and then pull one copy to each side. Only at the final moment of separation does an enzyme cut through those protein rings, releasing the sister chromatids to opposite ends of the cell.
So while mitosis gets the visual credit for cell division, the essential act of duplicating the genome happened hours earlier, quietly, during S phase. Mitosis is the distribution event. S phase is where the actual copying takes place.