14. Where Shall We Begin? (987;10/15;Ready)

If you wanted a book in the 15th century, it had to be copied by hand. A scribe would start at page one and work page by page until the entire book was finished.

A human genome is also a kind of book. One might therefore imagine that the cell copies each chromosome by starting at one end and progressing to the other.

That wouldn’t work. Human cells must replicate their genomes in about eight hours, but human DNA replication proceeds at only about 50 nucleotides per second. At that speed, copying a typical chromosome with a single replication machine would take more than a month.

Instead, replication begins at many sites within chromosomes called replication origins. Proteins assemble at each active origin, open the double helix there, and recruit two replisomes—multi-protein machines that move outward in opposite directions, copying DNA as they go.

So the cell solves its speed problem not by making each replisome faster, but by putting thousands of them to work at once. It is as if thousands of scribes opened the book simultaneously, each copied a short section, and all stopped when they reached work completed by their neighbors.

Bacterial and yeast replication origins

What defines a replication origin? The answer depends on the species. To provide perspective, we’ll take a quick look at two simpler, single-celled organisms--the bacterium E. coli and the yeast S. cerevisiae--before turning to humans. As we move from bacteria to multicellular eukaryotes, origins become more numerous, less rigidly defined, and more flexibly used.

The situation with E. coli is straightforward. It has a small circular genome (about five million base pairs) and replication begins at one precise, 245-base-pair origin. That origin contains several short DNA sequences recognized by initiator proteins that assemble the machinery needed for replication.

The yeast S. cerevisiae is also a single-cell, but being a eukaryote (like humans) its DNA is wrapped around histone proteins to form chromatin. The yeast genome contains about 12 million base pairs (about 2.5x the size of E. coli's) and is divided among 16 linear chromosomes, not just a single circular one.

Unlike E. coli with its single replication origin, yeast has several hundred potential replication origins, about 200 of which will be used during S phase. The unused ones remain dormant but can be called into action if a nearby replication fork stalls or replication is otherwise delayed.

The yeast cell therefore prepares more starting points than it expects to use. Dormant origins are not wasted machinery; they are reserve capacity positioned throughout the genome before the cell knows where trouble might occur.

Finally, yeast origins are less rigidly defined than the single bacterial origin. They contain a recognizable DNA sequence pattern, but the pattern is more variable and the surrounding chromatin also helps determine whether an origin is used.

Human replication origins

Eukaryotes like yeast and humans need many replication origins because their replisomes move much more slowly than bacterial ones. E. coli copies DNA at roughly 1,000 nucleotides per second, while yeast and human replisomes generally move at about 50--approximately 20 times slower.

Eukaryotic replisomes must also contend with chromatin. Nucleosomes ahead of each replisome are disrupted, while histones and their chemical markings must be reestablished on the newly synthesized DNA behind it.

Starting from many origins allows thousands of slower replisomes to copy different sections of a genome simultaneously. In fact, while the yeast has several hundred potential replication origins, the human genome contains 50,000 to 100,000. Only a fraction—perhaps 15,000 to 50,000—are activated during any particular S phase.

Human origins are not precise sites like the origin of E. coli. Many occupy broad initiation zones spanning thousands or even tens of thousands of base pairs, within which replication can begin at different locations. Nor do they share a single defining DNA sequence. There is no universal sequence in human DNA that simply says, “Start copying here.”

This creates a distinction that is absent in E. coli: one question is what makes a region capable of initiating replication; another is whether that potential origin will actually fire during a particular S phase.

During G1, the cell prepares—or licenses—far more possible starting points than it will use. Only a fraction will fire. Some will remain available as backups, while others will be copied by replication forks arriving from neighboring origins before they have a chance to fire themselves.

No single feature determines which regions are licensed or which licensed origins are activated. Instead, several features influence the probabilities:

1. Chromatin accessibility: Initiation proteins can more readily reach DNA in loosely packed chromatin. Accessible regions are therefore more likely to be licensed and, once licensed, to fire efficiently.

   
   

2. Transcriptional environment: Origins are often found near gene-rich, actively transcribed regions, particularly near promoters where the chromatin is already open. Because patterns of gene activity differ among cell types, patterns of origin use can differ as well.

   
   

3. Epigenetic and nuclear context: Chemical modifications to DNA and histones, along with a region’s location within the nucleus, can help recruit replication proteins and influence whether an origin fires early, late, or not at all.

   
   

These features do not define human origins in the way a particular DNA sequence defines the origin of E. coli. They shift the odds that a region will be prepared and used. A human origin is therefore less a fixed address than an opportunity created by its surroundings.

The underlying genome remains essentially the same, yet the map of where and when replication begins can change as a cell specializes.

This flexible system provides speed, adaptability, and backup starting points, but it raises a new problem: how does the cell prepare thousands of potential origins while ensuring that no section of DNA is copied twice?

In the next chapter, we’ll see how cells license replication origins--preparing a surplus of possible starting points while preventing replicated DNA from being licensed again until the next cell cycle.