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

## The Scale Problem

E. coli has about 4.6 million base pairs of DNA. Human cells have about 3.2 billion base pairs — roughly 700 times more. E. coli divides every 20–30 minutes under ideal conditions. Human cells in culture divide every 16–24 hours, with S phase (the DNA synthesis phase) taking 6–8 hours.

If a human cell used a single replication origin the way bacteria do, and replicated at the same speed as bacterial polymerases (~1000 bp/second), it would take about 37 days to replicate the genome. Clearly something is different.

## Multiple Origins in Eukaryotes

The solution eukaryotes evolved is parallelism: thousands of replication origins, most of which are activated at the start of S phase, allow the genome to be copied in segments simultaneously.

In humans, there are about 30,000–50,000 potential origins. Not all of them fire in every cell cycle. The ones that do activate are distributed across the chromosomes in a way that ensures full coverage — origins that don't fire early serve as backup for regions not yet replicated when the nearby origin fires.

Origin licensing begins in late mitosis and early G1 phase. The origin recognition complex (ORC) marks potential origins. Then CDC6 and CDT1 recruit MCM2-7 helicase complexes (two per origin, one for each fork direction). This loading is called licensing.

Activation happens at the onset of S phase, when two kinases — CDK2-Cyclin E and DDK (Dbf4-dependent kinase) — phosphorylate the MCM complex and recruit additional factors (CDC45 and GINS) to form the active CMG helicase. Synthesis begins.

## The Re-Replication Block

One strict requirement in eukaryotes: each origin must fire at most once per cell cycle. Re-replication (copying the genome twice before division) would be catastrophic.

The cell enforces this through two mechanisms that together destroy the licensing machinery after firing:
- After origin activation, CDT1 is ubiquitinated and degraded by the proteasome.
- Geminin, an inhibitor that appears in S phase and persists until the next mitosis, binds CDT1 and blocks re-licensing.

This is a one-way valve: CDT1 enables licensing, licensing is destroyed upon firing, and the new inhibitor prevents re-licensing until the next G1. Even if CDK2 and DDK are still active, there's nothing to license.

## Prokaryotic Replication: Simpler but Still Regulated

Bacteria replicate from a single origin (oriC in E. coli) bidirectionally, and the two forks meet at a termination zone roughly opposite the origin. The simpler genome and much faster polymerase make this feasible.

But bacterial replication is still regulated. E. coli prevents premature re-initiation partly through DnaA — the initiator protein that binds to oriC. After firing, the DNA at oriC is methylated by Dam methylase. But newly synthesized DNA is temporarily unmethylated at GATC sequences (the same sites that matter for mismatch repair). SeqA, a protein that sequesters hemimethylated DNA, binds to the newly replicated oriC and physically prevents DnaA from re-binding for several minutes. This is enough time for the cell to divide and separate the two chromosomes before either can initiate again.

## Replication Timing in Eukaryotes

Not all origins in human cells fire at the same time in S phase. The genome has early-replicating and late-replicating regions, and the timing is reproducible and cell-type-specific. Generally, active genes in open chromatin replicate early; heterochromatin and silenced regions replicate late.

This timing has consequences. Late-replicating regions have less time to repair errors before division, and show slightly higher mutation rates. Some tumor suppressor genes reside in late-replicating regions — a property that may contribute to their susceptibility.

The connection between replication timing, chromatin organization, and mutation rate is still an active research area. The pattern is clear; the mechanism that sets it is less so.

The final chapter connects replication failure to cancer in more detail — what goes wrong, which checkpoints fail to catch it, and why chemotherapy so often targets replication.

One Origin vs. Thousands: Why Bacteria and Eukaryotes Replicate Differently

// COMMENTS

ON THIS PAGE