The replication of our cellular DNA occurs from an internal point called the origin of replication.

Double-stranded DNA (dsDNA) is partially unwound or unzipped by a DNA helicase and the result is the foprmation of a replication bubble.

At this bubble a repliction complex (seen as a yellow ball in the figure below) forms and DNA duplication proceeds in both directions creating a replication fork.

An RNA polymerase or Primase creates a short RNA oligonucleotide within the bubble called a primer. Primase is a part of the primosome which consists of 16 protiens. The primosome is a part of a protein complex called the replisome which is involved in the simultaneous synthesis of both strands of DNA at the replication fork. The replisome also contains the DNA helicase, single-stranded binding protein, DNA polymerase III, DNA polymerase I and DNA ligase.

Primers are approximately 10 nucleotides in length. DNA polymerase III recognises and binds to this primer and starts to add complementary nucleotides as it reads along the parental DNA strand form the onwards. When the polymerase reads an "A", it adds a "T". When it reads a "G", it adds a "C". When it reads a "T" it adds and "A" and for a "C" it adds a "G". All DNA polymerases need a primer before they can start creating a polynucleotide.

The DNA polymerase always reads along the parental strand in a 3' to 5' direction, therefore the new or "nascent" DNA strand grows in a 5' to 3' direction. The DNA polymerase directs this directional growth by forming hydrogen bonds between the newly added bases and the exdisting bases on the parental strand as well as by catalysing the addition of a phosphodiester bond between the 5' phosphate group of a new nucleotide and the 3' hydroxyl that is free on the nascent strand.

However, only one of the two nascent strands can grow contiguously in a 5' to 3' direction. For the reasons described above, all DNA strands grow in a 5' to 3' direction, however this does not seem to be possible on the lagging strand. The discovery of Okazaki fragments provided an explanation and the idea of discontiguous DNA replication. The fragments are primed as the replication fork progresses, and there is space available for a new RNA primer to be made on the lagging strand. The result is a series of DNA fragments approximately 1000 nucleotides long which butt up against each other but are not joined together. The length of each fragment reflects the rate of RNA primer construction along the strand since a new primer is laid down every 1000 nucleitides.

The Okazaki fragments must then be joined together. This is performed by two enzymes, DNA polymerase I and DNA ligase. The leading strand grows in the 5' to 3' direction of the replication fork's movement while the lagging strand grows in the opposite direction to the rpelciation fork

DNA polymerase I can function as a polymerase, a 3' to 5' exonuclease and a 5' to 3' exonuclease. The 5' to 3' exonuclease function destroys the RNA primer and the polymerase activity replaces the primer with DNA. DNA ligase creates phosphodiester boinds to join, or ligate the fragments together into a continuous strand. So the DNA Okazaki fragments still grow in a 5' to 3' directuion, but overall the strand grows in a 3' to 5' direction as each fragment is ligated to the next one.

The result of replication is the formation of two new double-stranded DNA molecules.

And don't forget, these processes are happening all the time in human cells, utilising many replication forks and replisomes all without producing large numbers of errors. When you take the three billion base pair human genome as an example, this is an impressive process.