What Happens During the G1, S, and G2 Phases?

Between one cell division and the next, a cell spends most of its life in a stretch called interphase, which breaks down into three distinct stages: G1, S, and G2. During G1, the cell grows in size and ramps up protein production. During S phase, it copies its entire genome. During G2, it checks the new DNA for errors and stockpiles the molecular machinery needed to physically split in two. Each phase has its own internal logic, its own quality-control checkpoints, and its own way of going wrong when things break down.

G1 Is Where the Cell Decides Whether to Commit

G1 stands for “gap 1,” but calling it a gap undersells what is happening. The cell is actively growing, building proteins, producing new organelles, and reading signals from its environment to decide whether conditions are right for another round of division. During G1, the cell increases in size, synthesizes proteins, and prepares for cell division.¹ Some proteins are made predominantly in this phase; the S-100 protein in cultured glial cells, for example, peaks during G1.²

The single most important event in G1 is the passage through what researchers call the restriction point. Before the restriction point, the cell is still responsive to outside signals: growth factors telling it to proceed, or their absence telling it to pause. After the restriction point, the cell is committed to entering S phase regardless of what happens externally. The molecular gatekeeper here is the Rb protein. In early G1, Rb is mono-phosphorylated by cyclin D paired with CDK4 or CDK6, generating a set of distinct forms that selectively bind to transcription factors.³ Then, at the restriction point, a different enzyme pair, cyclin E with CDK2, inactivates Rb entirely through heavy phosphorylation, releasing the brakes on gene expression needed for DNA replication.

Interestingly, the old textbook picture of this process has turned out to be wrong in an important detail. The standard model said cyclin D and CDK4/6 progressively add more and more phosphate groups to Rb until it was sufficiently inactivated. Researchers have shown that cyclin D-CDK4/6 actually only mono-phosphorylates Rb in living cells, adding a single phosphate group at one of 14 possible sites, and that the progressive inactivation model, while partly correct as a concept, does not apply to Rb itself.⁴ This distinction matters because it implies there is some other, still unidentified target of CDK4/6 that drives the progressive buildup during early G1. The recognition between cyclin D-CDK4/6 and Rb depends on a specific docking interaction at the tail end of the Rb protein; removing just the last 37 amino acids reduced phosphorylation roughly twenty-fold.⁵

The G1/S Checkpoint Catches Damaged DNA Before Replication

If a cell’s DNA is damaged during G1, the last thing you want is for that damage to be faithfully copied. The cell has a safeguard: the tumor suppressor p53. When DNA damage is detected, p53 activates a G1 checkpoint that stops cells just before they begin replicating.⁶ It does this mainly by turning on a gene called p21, which blocks the enzyme complexes the cell needs to start S phase. When researchers knocked out p21 in human cancer cells, the G1 arrest triggered by p53 was completely abolished, confirming that p21 is the critical go-between.⁷

This checkpoint is one of the most frequently disabled in cancer. Mutations in p53 occur in roughly half of all human tumors, and losing this G1 brake means damaged cells replicate their broken DNA, accumulating more mutations with each cycle.

S Phase Copies the Entire Genome

S stands for “synthesis,” and the main job is DNA replication. Human cells carry about 6.4 billion base pairs of DNA, and copying all of it takes several hours. The process does not begin at a single starting point and work linearly; instead, thousands of replication origins fire in a coordinated sequence so the whole genome gets copied on time.⁸

At each origin, the DNA double helix needs to be unwound before it can be copied. The machinery for this is loaded in an inactive state during the previous G1 phase. Once S phase begins, helper proteins called Cdc45 and GINS join the core helicase to form an active complex that pries the two DNA strands apart, allowing the replication machinery to read and copy each strand.⁹ The cell loads the helicase once per origin per cycle, and this “once and only once” rule is critical for preventing any stretch of DNA from being duplicated twice.

DNA Replication Comes Paired with Histone Production

Copying the DNA is only half the challenge. Every stretch of new DNA needs to be immediately wrapped around histone proteins and packaged into chromatin, the organized structure that keeps the genome manageable. The cell tightly couples DNA synthesis with histone synthesis during S phase so that freshly made DNA is promptly assembled into chromatin.¹⁰ If histone supply falls behind, the consequences are immediate: the rate of DNA synthesis drops shortly after replication begins, and S phase drags out.¹¹

This coupling also matters for preserving the cell’s identity. Histones carry chemical tags, known as epigenetic marks, that tell the cell which genes to keep active and which to keep silent. When replication forks move through the DNA, the existing chromatin is disrupted and must be rebuilt on the new copies. Researchers have proposed that components of the replication machinery itself help recruit the enzymes responsible for re-establishing those marks on newly deposited histones.¹² Understanding how parental epigenetic patterns get copied onto fresh histones in a DNA-sequence-specific way remains an active area of research.¹³

The Intra-S Checkpoint Watches for Trouble During Replication

Replication is inherently risky. The machinery can stall at difficult-to-copy sequences, collide with other processes happening on the DNA, or encounter damage left over from environmental insults. Rather than wait until the end of S phase to assess the damage, the cell runs a surveillance system in real time. A kinase called ATR monitors replication forks as they work and responds to signs of stress by controlling checkpoint signaling, origin firing, and fork stability.¹⁴

When a fork stalls, ATR teams up with a downstream kinase called CHK1 to slow the overall pace of replication and stabilize the stuck fork until the problem is resolved. A protein called ASPM helps load the signaling components onto the stalled fork to get this checkpoint activated.¹⁵ One fascinating recent finding is that this response can be local rather than global. When researchers introduced a single, specific replication-fork barrier in human cells, ATR activated locally at the stalled site without triggering a cell-wide alarm, and the phosphorylation signal marking the trouble spot vanished when ATR was inhibited.¹⁶ So the cell can handle isolated problems at specific spots without slamming the brakes on replication everywhere.

G2 Prepares the Cell for Division

After S phase finishes, the cell enters G2 (gap 2), a period focused on preparing for mitosis. The cell continues to grow, manufactures the proteins needed for the physical act of chromosome separation, and runs a final quality check on its freshly copied DNA. One key task is accumulating cyclin B1, a protein that partners with CDK1 to trigger entry into mitosis. When the protein Mad2 was depleted in experiments, cells entered mitosis with only about 70% of the normal cyclin B1 levels, illustrating how carefully this accumulation is regulated.¹⁷

G2 is also when organelles finish scaling up. Mitochondrial biogenesis, for instance, is coordinated with the cell cycle: mitochondrial DNA replication is synchronized with nuclear DNA replication in S phase, but some mitochondrial proteins continue to be made through G2. The mitochondrial membrane potential does not fully mature until G2.¹⁸ This makes sense: the cell is about to split into two daughters, and each one needs a functional set of energy-producing mitochondria.

The G2/M Checkpoint Provides a Final Safety Gate

If DNA damage is detected in G2, the cell must not enter mitosis, because dividing with broken chromosomes would be catastrophic. The G2 checkpoint enforces this by keeping CDK1 inactive. CDK1 needs to be free of an inhibitory phosphate group on a specific residue (tyrosine-15) to fire. The checkpoint works through a double-lock mechanism: it simultaneously boosts the activity of Wee1, the kinase that adds the inhibitory phosphate, and suppresses Cdc25, the phosphatase that removes it.¹⁹ Two locks are harder to pick than one, and this redundancy helps ensure the cell stays arrested while repairs are underway.

The checkpoint kinases Chk1 and Chk2 are the upstream enforcers that inactivate the Cdc25 phosphatases when DNA is damaged or incompletely replicated.²⁰ When DNA damage occurs during G2, Cdc25A is destroyed through the proteasome, and experimentally forcing Cdc25A back overrides the G2 arrest altogether.²¹

How Cell Size Feeds into the Decision to Divide

A question that runs underneath all three phases is: how does the cell know it is big enough to divide? If cells split before reaching an adequate size, the daughters would be too small to function. The current thinking involves “sizer” mechanisms that link a cell’s physical size to its progression through interphase. One proposal is that the cell monitors translational activity: the bigger the cell, the more ribosomes it has, and the more of certain rate-limiting proteins it produces. Once a threshold amount of a key protein accumulates, the cell is cleared to move forward.²² CDK4 appears to play a distinct role in setting the target cell size for entering S phase, beyond its well-known role in pushing the cell through the restriction point.²³

Mathematical models of the cell cycle capture this with two regimes: a “sizer” phase early on, where the time to progress depends on how small the cell starts, followed by a “timer” phase, where progression happens on a more fixed schedule regardless of size.²⁴ This explains a common observation: cells born small take longer to get through G1, but once they reach the restriction point, the remaining phases run on a more predictable clock.

Nutrient Sensing Across the Phases

Cells do not just monitor their own size; they also pay close attention to nutrient availability. A central player here is mTORC1, a signaling hub that integrates information about amino acids and energy. Researchers have found that mTORC1 activity oscillates throughout the cell cycle, peaking in S and G2 and falling to its lowest point in mitosis and G1. Strikingly, mTORC1 promotes progression through S and G2, not just through G1 as was previously thought.²⁵ This means that starvation does not just prevent a cell from entering the cycle; it can stall a cell that is already midway through replication.

When Cells Exit the Cycle Entirely

Not every cell that finishes mitosis re-enters G1. Many cells leave the cycle and enter a resting state called G0. This exit can be reversible (quiescence) or permanent (senescence or terminal differentiation). Quiescent cells are alive but metabolically quiet and can re-enter the cycle when stimulated. Senescent cells are also stuck in G0, but they keep producing proteins at high levels and cannot be coaxed back. One way to tell the two apart is by looking at markers of active protein synthesis: senescent cells express RPS6, while truly quiescent cells do not.²⁶

The distinction between quiescence and senescence has real medical consequences. Senescent cells accumulate with aging and secrete inflammatory signals that contribute to age-related diseases. Quiescent cells, in contrast, are the body’s reserve: stem cells in the bone marrow, for instance, sit in G0 until called upon. Understanding how a cell chooses one path over the other is an active frontier.

Embryonic Cells That Skip the Gap Phases

The three-phase structure of interphase is not universal. In the early embryos of many animal species, cells cycle with stunning speed by eliminating G1 and G2 altogether, shuttling directly between S phase and mitosis.²⁷ These rapid divisions chop a large fertilized egg into progressively smaller cells without any growth between divisions. Only later, at a stage called the midblastula transition, do gap phases appear and the cell cycle slows to something resembling the adult pattern. This tells us that G1 and G2 are not structurally required for a cell to divide; they exist to give the cell time to grow, assess its environment, and repair damage, luxuries an early embryo full of maternal supplies can temporarily do without.

Cancer Drugs That Target Interphase

Because cancer cells divide relentlessly, the molecular switches that govern interphase have become drug targets. CDK4/6 inhibitors, now widely used in breast cancer treatment, work by blocking the early G1 machinery. Their primary effect is preventing Rb phosphorylation, which keeps the restriction point locked and halts the cell in G1.²⁸ But these drugs also alter cancer cell biology in other ways: they can trigger senescence, boost anti-tumor immune responses, and shift the cell’s metabolic state, effects that researchers are working to exploit in combination therapies.

Efforts to visualize where individual cancer cells sit in the cycle have also advanced. A fluorescent reporter system called FUCCI labels cells different colors depending on their cycle stage, and newer tools can now map each cell to a continuous position within the cycle rather than a broad bin. Intravital imaging with FUCCI has revealed that the majority of cancer cells within solid tumors are actually not cycling and are instead sitting in a quiescent or G0-like state, which helps explain why chemotherapy drugs that target dividing cells leave many tumor cells untouched.²⁹ An updated approach called CALIPERS can assign each nucleus a percentage-completion value along the cycle, enabling researchers to organize cell behavior by cycle progression rather than simply by elapsed time.³⁰ These tools are changing how researchers study drug resistance, tumor heterogeneity, and the timing of cell-cycle re-entry after treatment.