What is Dr. Zubair Khalid's research focus?

Dr. Zubair Khalid specializes in molecular virology, mRNA vaccine development, and computational biology, with a focus on avian pathogens like IBDV and Avian Reovirus.

Where is Dr. Zubair Khalid currently working?

Dr. Zubair Khalid is a Postdoctoral Research Associate at the University of Maryland (UMD), specifically within the Department of Animal and Avian Sciences.

DNA Synthesis Steps: Replication Explained

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-09-20

DNA synthesis is the process by which a cell copies its genome before division. This article walks through the full process of DNA synthesis in order: how replication starts at defined origins, how the double helix is unwound, how short RNA primers are laid down, how DNA polymerase extends new strands in the 5-prime to 3-prime direction, and how the discontinuous pieces on the lagging strand are joined into one continuous molecule. By the end you will be able to describe each step in dna synthesis, name the enzyme responsible for each step, state the directionality of every reaction, and explain where dna synthesis takes place in a eukaryotic cell.

You do not need a laboratory to follow this. What you need is a working mental model of the DNA double helix (two antiparallel strands, each with a 5-prime phosphate end and a 3-prime hydroxyl end) and a willingness to track which strand is which. Everything else follows from one hard chemical constraint: every known template-directed DNA polymerase can only add a nucleotide to a free 3-prime hydroxyl group. That single fact dictates the entire choreography of the replication fork, including why one strand is made continuously and the other in fragments.

What DNA Synthesis Actually Is

DNA synthesis is template-directed polymerization. A parental strand is read, and a complementary daughter strand is built nucleotide by nucleotide. Each incoming deoxynucleoside triphosphate (dNTP) is selected by base pairing with the template (A with T, G with C), and the polymerase catalyzes a nucleophilic attack in which the 3-prime hydroxyl of the growing chain attacks the alpha phosphate of the incoming dNTP. Pyrophosphate is released, and the chain gains one nucleotide.

Two consequences follow immediately.

First, the reaction is a condensation that consumes energy. The dNTPs themselves carry the activation energy, so no separate ATP input is required at the polymerization step.

Second, the reaction has an obligatory direction. The new strand grows 5-prime to 3-prime, meaning nucleotides are added to the 3-prime end. This is not a convention or a preference. It is a structural property of the polymerase active site, which positions the incoming dNTP against the template while holding the primer terminus. Replicative polymerases extend nucleotide polymers in a 5-prime to 3-prime direction, and this inherent directionality is what forces one strand to be made forwards and the other backwards [1].

That asymmetry is the central organizing principle of replication. The leading strand is synthesized continuously, in the same direction the fork moves. The lagging strand is synthesized discontinuously, in short pieces called Okazaki fragments, as the fork exposes new template in the opposite orientation [2].

Where DNA Synthesis Takes Place

In eukaryotes, nuclear DNA synthesis occurs in the nucleus during S phase of the cell cycle. The genome is duplicated once per cycle, and the S-phase checkpoint coordinates the timing of origin activation so that replication completes without exhausting replisome components [3]. The proteins involved are dozens in number, including the origin recognition complex (ORC) and Cdc6 during initiation, the CMG helicase during unwinding, DNA polymerase alpha-primase for priming, and the leading-strand and lagging-strand polymerases for bulk synthesis [4].

Two qualifications matter.

First, not all DNA synthesis in a eukaryotic cell is nuclear. Mitochondria maintain their own genomes using a distinct machinery assembled from phage-like, bacterial, and eukaryotic components, and mitochondrial replication strategies differ from the nuclear paradigm [5]. The steps described here are the nuclear ones.

Second, DNA synthesis also occurs outside S phase in specialized repair contexts. Break-induced replication (BIR) is a recombination-dependent pathway that repairs single-ended double-strand breaks and supports telomerase-independent telomere maintenance in some cancers. BIR uses a migrating D-loop and a different polymerase deployment than the S-phase replisome [6]. That pathway is a repair topic, not the canonical replication cycle, and it is not what most people mean by dna synthesis steps.

The Steps in DNA Synthesis, In Order

The numbered walkthrough below follows one replication origin from activation through ligation. Read it once for the sequence, then use the enzyme table to fix the players in memory.

Step 1. Initiation at the Origin

Replication begins at defined chromosomal positions called origins of replication. In eukaryotes, initiation proceeds in two separable stages: origin licensing and origin firing. Licensing is the loading of the MCM replicative helicase motor onto origin DNA, and firing is its activation. Reconstitution of these steps with purified budding yeast proteins established the mechanistic framework, and more recent work has extended licensing reconstitution to purified human proteins, revealing where metazoan mechanisms deviate from the yeast paradigm [7].

The helicase engine itself is the minichromosome maintenance (MCM) complex, a hexameric ring. Loading occurs as a double hexamer, and the two ring-shaped hexamers are staggered, producing a kinked axial channel through which the DNA passes [8]. In the loaded state, the DNA is not yet melted. Activation requires local melting of fully base-paired DNA so the helicase can gain a foothold and begin processive unwinding. Snapshots of an archaeal MCM ring captured in different degrees of melting show that successive ATP binding at MCM ATPase sites drives discrete, sequential melting steps mediated by a specific MCM aromatic residue. Eukaryotic loaded MCM rings adopt principally two arrangements at the ATPase, one that does not melt DNA and one tuned to melt it, which indicates a shared sequential melting mechanism across archaea and eukaryotes [9].

Once activated, the CMG helicase (Cdc45-MCM-GINS) unwinds the parental duplex and DNA polymerase alpha-primase initiates synthesis on both template strands [10]. One detail from origin-mapping experiments is worth carrying forward: in an origin-dependent yeast system, synthesis was mostly initiated outside the origin sequence, and rightward leading strands were primed to the left of the origin while leftward leading strands were primed to the right. Each leading strand is established from a lagging-strand primer synthesized by the replisome on the opposite side of the origin. Blocking elongation of the primers synthesized left of the origin blocked rightward leading-strand synthesis, which demonstrates that the two replisomes are interdependent for establishing leading-strand synthesis. This arrangement removes any need for dedicated leading-strand priming and gives the lagging-strand polymerase Pol delta a required role in connecting the nascent leading strand to the advancing replisome [10].

Step 2. Unwinding by Helicase

The CMG helicase translocates along the parental duplex and separates the two strands ahead of the polymerases. Several consequences follow.

The DNA ahead of the fork becomes positively supercoiled and must be relieved by topoisomerase activity. The separated single strands are coated by single-strand DNA binding proteins to prevent them from reannealing or forming secondary structures. And the two exposed templates now have opposite polarity with respect to the direction of fork movement, which sets up the leading and lagging strand distinction.

The helicase does not unwind symmetrically in a structural sense. In the loaded MCM double hexamer, the presensor 1 loops of Mcm3, Mcm4, Mcm6, and Mcm7 engage the lagging strand with an approximate step size of one base per subunit, while the Mcm2 and Mcm5 loops do not. The staggered coupling of the two opposing hexamers positions the DNA in front of the two Mcm2-Mcm5 gates, with each strand pressed against one gate, suggesting that lagging-strand extrusion begins in the middle of the double hexamer [8].

Step 3. Priming by Primase

No replicative DNA polymerase can start a chain from nothing. Each new strand must begin with a primer that supplies a free 3-prime hydroxyl, and that primer is RNA.

In eukaryotes, the DNA polymerase alpha-primase complex (Pol alpha/Pri) performs this function. It synthesizes a short RNA primer and then extends it briefly with DNA, creating a hybrid RNA-DNA initiation primer. Pol alpha/Pri initiates synthesis on both the leading and lagging strands [11]. The same division of labor appears in bacteria. In the Firmicutes Bacillus subtilis, DnaG primase and DnaE polymerase act together to synthesize a hybrid RNA/DNA initiation primer on both the leading and lagging strands at the phage SPP1 origin, mirroring what the eukaryotic Pol alpha complex does. DnaE, acting as an RNA-primed DNA polymerase, extends the primer in a reaction modulated by DnaG and a single-strand binding protein, then hands the primer off to PolC, a DNA-primed DNA polymerase that performs the bulk of elongation on both strands [12].

That handoff is the general pattern. Primase lays down a short starter, a low-processivity polymerase extends it a little, and a high-processivity replicative polymerase takes over for the long haul.

Priming is not a one-time event. On the lagging strand it must be repeated for every Okazaki fragment. Current evidence supports a semi-distributive model in which Pol alpha/Pri acts both associated with the replisome and as a free complex [2]. In budding yeast, both Okazaki fragment initiation and productive firing of replication origins are sensitive to Pol alpha abundance, and both processes are disrupted at similar Pol alpha concentrations. When the replisome adaptor protein Ctf4 is absent or cannot interact with Pol alpha, lagging-strand initiation is impaired at Pol alpha concentrations that still support normal origin firing, which indicates that Pol alpha recruitment for initiation and for ongoing lagging-strand priming are distinctly sensitive to Ctf4 [11].

Step 4. Elongation by DNA Polymerase

Elongation is where the strand asymmetry becomes visible.

The leading strand is extended continuously in the same direction the fork travels. The lagging strand is extended discontinuously, in Okazaki fragments, because the template is oriented so that synthesis must run back toward the fork [2]. The leading strand is synthesized continuously while the lagging strand is copied in small fragments requiring repeated priming [2].

In eukaryotes, the three main replicative polymerases are Pol alpha, Pol delta, and Pol epsilon. Studies in yeasts and human cells using mutant polymerases that incorporate molecular signatures into nascent DNA implicate Pol epsilon in leading-strand synthesis and Pol alpha together with Pol delta in lagging-strand replication [1]. Structural work has shown how the spatial organization of these enzymes around the core helicase supports their strand-specific roles [1].

Processivity is maintained by a sliding clamp. Proliferating cell nuclear antigen (PCNA) encircles the DNA and tethers the polymerase to it. PCNA is loaded onto chromatin, and its availability is a rate-limiting feature of replication. Excessive origin firing saturates chromatin-bound PCNA, which restricts further PCNA loading and lagging-strand synthesis when checkpoint control is lost. PCNA-associated factor 15 (PAF15) binds PCNA specifically on the lagging strand through a high-affinity PIP motif and occupies the DNA-encircling channel, protecting the clamp and its associated enzymes from premature unloading by the ATAD5-RFC complex. Overexpression of PAF15, or forced redistribution of PAF15 to the leading strand, disrupts replisome progression and induces cell death, and Timeless-Claspin mitigates this by blocking PAF15-PCNA binding on the leading strand [3]. The takeaway is that the two strands are not treated identically by the cell even after the polymerases are chosen.

Step 5. Maturation and Ligation of Okazaki Fragments

Each Okazaki fragment ends with a short stretch of RNA from its primer. That RNA must be removed, the gap filled with DNA, and the nick sealed.

The final joining reaction is catalyzed by DNA ligase, which forms a phosphodiester bond between the 3-prime hydroxyl at the end of one fragment and the 5-prime phosphate at the start of the next. Once ligation is complete, the lagging strand is a continuous polynucleotide, chemically indistinguishable from the leading strand.

Okazaki fragment processing depends on the same clamp that supports elongation. PCNA is described as a sliding clamp for DNA polymerase processivity and Okazaki fragment processing [11]. This is one reason lagging-strand synthesis is the more fragile of the two: it requires repeated priming, repeated clamp loading, repeated primer removal, and repeated ligation, and each of those steps is a point of regulation and potential failure.

A Note on the Replication Stress Response

The lagging strand is not just mechanically busier. There is a strong link between lagging-strand replication dynamics and basal activation of the replication stress response (RSR) during an unperturbed S phase. One hypothesis is that the RSR monitors the generation of Okazaki fragments to control DNA synthesis in what has been called the DNA replication control mode of the RSR. In this model, the RSR enforces gradual progression of DNA replication by restricting origin firing, which is necessary to establish the replication program and prevent genomic instability, and it coordinates the replication program by modulating progression to avoid exhausting cellular resources [2].

The Replication Fork as a Decision Path

The diagram below traces the main workflow from origin activation to two fully synthesized daughter duplexes.

flowchart TD
    A[Origin licensing loads MCM double hexamer] --> B[Origin firing melts DNA]
    B --> C[CMG helicase unwinds duplex]
    C --> D[Pol alpha primase lays RNA primer]
    D --> E{Which template orientation}
    E --> F[Leading strand continuous synthesis]
    E --> G[Lagging strand discontinuous synthesis]
    G --> H[Repeated priming each Okazaki fragment]
    H --> I[Pol delta extends fragments]
    I --> J[Primer removal and gap fill]
    J --> K[DNA ligase seals nicks]
    F --> L[Two daughter duplexes]
    K --> L

Table of Enzymes, Functions, and Directionality

Enzyme or factor	Function	Directionality or polarity	Location in eukaryotes
ORC and Cdc6	Recognize origins and load the MCM helicase during licensing	Not a polymerase, no chain direction	Nucleus, before S phase
MCM2-7 and CMG helicase	Melt origin DNA and unwind the duplex ahead of the fork	Translocates along DNA, separates strands	Nucleus, S phase
DNA polymerase alpha with primase (Pol alpha/Pri)	Synthesize short RNA primers and extend them briefly with DNA on both strands	Synthesizes 5-prime to 3-prime	Nucleus, S phase
DNA polymerase epsilon (Pol epsilon)	Bulk leading-strand synthesis	Synthesizes 5-prime to 3-prime, continuous	Nucleus, S phase
DNA polymerase delta (Pol delta)	Bulk lagging-strand synthesis and Okazaki fragment extension	Synthesizes 5-prime to 3-prime, discontinuous	Nucleus, S phase
PCNA	Sliding clamp that tethers polymerases and supports Okazaki fragment processing	Not a polymerase, encircles DNA	Nucleus, chromatin-bound
PAF15	Binds PCNA on the lagging strand and protects the clamp from premature unloading	Strand-specific, lagging strand	Nucleus, chromatin-bound
Single-strand binding protein	Coats exposed single-stranded template	Not a polymerase	Nucleus, at the fork
Topoisomerase	Relieves supercoiling ahead of the fork	Not a polymerase	Nucleus, at the fork
DNA ligase	Seals nicks between Okazaki fragments	Joins 3-prime hydroxyl to 5-prime phosphate	Nucleus, S phase
DnaG primase and DnaE (bacterial model)	Synthesize a hybrid RNA/DNA initiation primer	Synthesizes 5-prime to 3-prime	Cytoplasm, bacteria
PolC (bacterial model)	Bulk elongation on both leading and lagging strands	Synthesizes 5-prime to 3-prime	Cytoplasm, bacteria

The bacterial entries are included because the Firmicutes system illustrates the primer handoff cleanly: DnaE extends the RNA-primed initiation primer and hands it off to PolC, which then performs the bulk of chain elongation at both the leading and lagging strands, with DnaG and single-strand binding proteins modulating both steps [12].

How to Check You Have the Model Right

Three self-tests separate a correct mental model from a memorized list.

Test one. Ask why the lagging strand cannot simply be made continuously. The answer is directionality. Polymerases extend only 5-prime to 3-prime, and the two template strands are antiparallel, so only one of them presents a template that can be read continuously in the direction of fork movement [1].

Test two. Ask what would happen if primase were depleted. Origin firing and Okazaki fragment initiation would both suffer, and lagging-strand initiation would be impaired before origin firing under conditions where the Ctf4 interaction with Pol alpha is disrupted [11].

Test three. Ask which strand PAF15 protects. It binds PCNA specifically on the lagging strand, and forcing it onto the leading strand disrupts replisome progression [3].

Common Mistakes and Limitations

Confusing replication with transcription. Both processes synthesize nucleic acid from a template, and both use polymerases that work 5-prime to 3-prime. They differ in product and in template handling. Replication copies the entire genome using both strands as templates and produces DNA. Transcription uses only one strand of a gene as template and produces RNA, and it does not require a primer in the same way replication does. The enzymes, the initiation logic, and the cellular timing are different. If you find yourself describing RNA polymerase in a dna synthesis steps answer, you have drifted.

Confusing the leading strand with the template strand. The leading strand is defined by the direction of synthesis relative to fork movement, not by which parental strand it reads. The same physical parental strand can serve as leading-strand template for one fork and lagging-strand template for a fork moving the other way.

Assuming the primer is DNA. The primer is RNA, or a short RNA-DNA hybrid. This is why primer removal is a required step in Okazaki fragment maturation and why ligation cannot occur until the RNA is gone and the gap is filled.

Treating ligation as trivial. Ligation is the step that makes the lagging strand continuous. Without it, the genome would be a collection of fragments, and the next round of replication would encounter nicks.

Overgeneralizing the eukaryotic polymerase assignment. Pol epsilon is implicated in leading-strand synthesis and Pol alpha with Pol delta in lagging-strand replication, but polymerase usage is flexible in challenging situations such as replication initiation or encounters with replication-blocking adducts [1]. The clean assignment is the normal case, not an invariant law.

Assuming one origin per chromosome. Eukaryotic genomes use many origins, and origin firing is temporally regulated. Excessive origin firing saturates chromatin-bound PCNA and restricts further lagging-strand synthesis when checkpoint control is lost [3].

Forgetting that replication stress is normal. Basal activation of the replication stress response occurs during an unperturbed S phase and is linked to lagging-strand dynamics [2]. A pathway being active does not automatically mean something is wrong.

Ignoring the repair connection. Errors that escape polymerase proofreading are handled by mismatch repair. In Escherichia coli, initiation of mismatch repair involves mismatch recognition by MutS, binding of MutL, and activation of the endonuclease MutH, which incises DNA at a hemi-methylated GATC site. MutS exists in a dimer-tetramer equilibrium, and tetramers mediate more efficient incision than dimers, with a higher preference for nicking the GATC site close to the mismatch, consistent with a fourfold decrease in diffusion constant for the tetramer relative to the dimer [13]. This is a separate pathway from replication, but it exists because replication is not perfectly accurate.

Assuming all DNA synthesis looks the same. Mitochondrial replication in humans has been modeled with asynchronous strand-displacement and RITOLS mechanisms as well as a synchronous strand-coupled model, and the data supporting and contradicting each model mean there is likely no single mitochondrial replication mechanism [5]. Poxviruses replicate in the host cytoplasm and encode most or all of the enzymes needed, and directional deep sequencing of RNA-primed nascent strands identified origins at the switching points between continuous and discontinuous synthesis, supporting a leading-and-lagging model for vaccinia [14]. The nuclear eukaryotic steps are the canonical answer, not the only possible one.

Limitations. The descriptions here are the consensus framework. Individual experimental systems differ in detail, and any specific clinical or diagnostic question about replication in a patient sample requires professional interpretation.

Frequently Asked Questions

What are the steps of DNA synthesis in order?

Initiation at the origin, unwinding by helicase, priming by primase, elongation by DNA polymerase, and ligation of Okazaki fragments by DNA ligase. In eukaryotes these steps occur in the nucleus during S phase.

Where does DNA synthesis take place?

Nuclear DNA synthesis takes place in the nucleus during S phase of the cell cycle. Mitochondria replicate their own genomes with a separate machinery, and some repair pathways synthesize DNA outside S phase.

Why is the lagging strand synthesized discontinuously?

Because polymerases extend only 5-prime to 3-prime and the two template strands are antiparallel. Only one template can be read continuously in the direction of fork movement, so the other is copied in short Okazaki fragments that are later joined.

Which enzyme unwinds the DNA double helix during replication?

The CMG helicase, built around the hexameric MCM2-7 ring. MCM loading occurs during licensing, and activation during firing melts the DNA so processive unwinding can begin.

What is the role of primase in DNA synthesis?

Primase synthesizes the short RNA primer that supplies the free 3-prime hydroxyl every DNA polymerase requires. In eukaryotes this is done by the DNA polymerase alpha-primase complex on both strands.

Which polymerase makes the leading strand in eukaryotes?

Pol epsilon is implicated in leading-strand synthesis, while Pol alpha and Pol delta are implicated in lagging-strand replication. Polymerase deployment becomes more flexible during initiation and at replication-blocking lesions.

Why are Okazaki fragments joined by ligase?

Each fragment ends with RNA primer that must be removed and replaced with DNA, leaving a nick. DNA ligase seals that nick by forming a phosphodiester bond, making the lagging strand continuous.

Does DNA synthesis happen in both directions from an origin?

Yes. Activated origins form bidirectional forks, and the two replisomes are interdependent for establishing leading-strand synthesis, with each leading strand primed by the replisome on the opposite side of the origin.