The Notch signaling pathway is one of a small number of cell-communication systems that animal bodies use over and over again, from the earliest stages of embryonic development through adult tissue maintenance and, when things go wrong, in the progression of disease. It works through direct physical contact between neighboring cells, making it fundamentally a short-range conversation rather than a broadcast signal. What makes Notch unusual among signaling pathways is its dual nature in disease: mutations that hyperactivate it drive certain blood cancers, while mutations that disable it promote solid tumors in skin and other tissues. That paradox has made it both a compelling drug target and a frustratingly difficult one.
How the Pathway Works
Notch signaling begins when a ligand protein on one cell’s surface binds to a Notch receptor on a neighboring cell. Mammals have four Notch receptors (Notch1 through Notch4) and five ligands belonging to two families (Delta-like and Jagged). The receptor sits in the cell membrane with most of its bulk sticking out into the space between cells, protected by a structure called the negative regulatory region that keeps it from activating on its own. Activation requires that the ligand-bearing cell physically pull on the receptor’s outer portion after binding it. This mechanical tug, driven by the ligand being drawn back into the sending cell, peels open the protective structure and exposes a hidden cleavage site.
Once that site is exposed, a metalloprotease enzyme cuts the receptor near the membrane surface, shedding the large outer portion. What remains is a short membrane-tethered stub that gets cut again by a second enzyme complex called gamma-secretase, which slices within the membrane itself. Only after this second cleavage does the inner portion of the receptor break free and travel to the nucleus.
1Cell. Notch Signaling Pathway: Key Roles in Development and Disease Experiments applying physical force directly to the receptor’s protective region have confirmed that pulling is what exposes the hidden cut site, supporting the model that ligand endocytosis in the neighboring cell provides the necessary mechanical force.2PubMed Central. Direct observation of proteolytic cleavage at the S2 site upon forced unfolding of the Notch negative regulatory region
Inside the nucleus, the freed Notch fragment does not bind DNA on its own. Instead, it joins a protein already sitting on DNA called CSL (known by different names in different organisms). Before Notch arrives, CSL typically recruits silencing partners that keep target genes turned off. When the Notch fragment docks onto CSL, it displaces those silencers and recruits an activator called Mastermind, forming a three-protein complex that switches target genes on. Structural studies have shown that this assembly induces a substantial shape change in CSL itself, essentially flipping it from a repressor into an activator.3Cell. Structural Basis for the Ternary Complex Formed by CSL, NotchIC, and Mastermind Bound to DNA This on-off toggle quality, where the same DNA-binding protein can either silence or activate depending on who else is present, gives the pathway a crisp switch-like character.
Signals That Bypass the Standard Route
The cleavage-to-nucleus sequence described above is the textbook version, often called canonical Notch signaling. But evidence has accumulated that Notch proteins can also influence cells without following this script. In some contexts, Notch appears to function without needing a ligand to trigger it, or without its released fragment entering the nucleus to turn on genes. One well-studied example involves Notch interacting directly with components of the Wnt/beta-catenin pathway, another major developmental signaling system. In these cases, Notch acts after translation, adjusting the stability or activity of beta-catenin at the protein level rather than through gene regulation.4PubMed Central. Non-canonical Notch signaling: emerging role and mechanism This kind of crosstalk matters because it means blocking just the canonical arm of Notch, say by inhibiting gamma-secretase, might not shut down everything Notch does in a given tissue.
Building an Embryo One Decision at a Time
Notch signaling appears remarkably early in development and keeps showing up at each stage. Its roles fall into a few recurring patterns.
Sorting Neighbors Into Different Fates
One of the most elegant things Notch does is called lateral inhibition. When a group of initially identical cells all begin expressing both Notch receptors and Notch ligands, small random differences get amplified: a cell that happens to express a bit more ligand sends stronger signals to its neighbors, pushing them to adopt one fate while it adopts another. The result is a “salt and pepper” pattern where cells of two different types alternate throughout a tissue.5PubMed Central. Notch-mediated lateral inhibition regulates proneural wave propagation when combined with EGF-mediated reaction diffusion This mechanism is responsible for selecting which cells become neurons and which remain as support cells in the developing nervous system, among many other examples.
Keeping the Segmentation Clock in Sync
Vertebrate embryos form their body segments, the precursors to vertebrae and ribs, through a rhythmic process. Cells in the tissue that will become segments each run an internal oscillator, a molecular clock that cycles gene expression on and off. For segments to form properly, neighboring cells need to oscillate in step. Notch signaling is what synchronizes them. When Notch signaling fails, individual cells keep oscillating but fall out of phase with their neighbors, producing severe anatomical defects.6PubMed Central. Exome sequencing of head and neck squamous cell carcinoma reveals inactivating mutations in NOTCH1 Experiments in zebrafish have directly demonstrated this synchronization role: in embryos lacking the normal Notch ligand DeltaC, researchers delivered artificial pulses of DeltaC expression using a heat-shock-driven gene. Those pulses restored synchrony among cells and rescued segment formation, with the spacing of segment boundaries tracking the timing between pulses.7Development. Pulses of Notch activation synchronise oscillating somite cells and entrain the zebrafish segmentation clock Careful timing experiments further showed that Notch is needed only in the region where oscillations occur, not in the downstream tissue that reads out the pattern to form actual segment boundaries.8PLOS Genetics. Notch Signalling Synchronizes the Zebrafish Segmentation Clock but Is Not Needed To Create Somite Boundaries
Shaping the Heart
Notch signaling is essential for building a functioning heart. In the developing heart’s inner lining, Notch activity coordinates the formation of heart valves and the muscular ridges (trabeculae) that line the ventricles.9PubMed. Notch signaling in cardiac development and disease It also regulates how heart muscle cells multiply and specialize during chamber development, and is required for the blood vessels that supply the heart itself (the coronary vessels) to form with the correct identity.10PubMed Central. Coordinating tissue interactions: Notch signaling in cardiac development and disease In all these contexts, Notch acts as a coordinator between different tissue layers rather than simply dictating a single cell’s behavior.
Deciding Which Immune Cells to Make
The immune system offers one of the clearest demonstrations of Notch acting as a fate switch. In the bone marrow, a shared progenitor cell can give rise to either T cells or B cells. Notch1 activation pushes that progenitor toward the T cell fate. When researchers engineered bone marrow cells to constitutively activate Notch1, the result was a dramatic expansion of T-lineage cells at the expense of B cells.11Immunity. Constitutive Activation of Notch1 Specifies a T-Cell Phenotype in the Bone Marrow Beyond that initial T-versus-B decision, Notch continues to influence which subtype of T cell a developing cell becomes.12PubMed Central. Notch Signaling in T-Cell Development and T-ALL This tight link between Notch and T cell identity turns out to have direct consequences for cancer, as we will see below.
Maintaining the Gut Lining
Notch’s importance does not end after development. In adult tissues that turn over rapidly, Notch helps maintain the balance between stem cells and mature specialized cells. The intestinal lining is the best-studied example. The gut epithelium replaces itself roughly every five days, driven by stem cells tucked into small pockets called crypts. Notch signaling is required to keep those stem cells in their undifferentiated, self-renewing state.13PubMed Central. Notch regulation of gastrointestinal stem cells
When Notch1 is deleted in the mouse intestine, stem cell markers are profoundly lost within days.14PubMed Central. Notch receptor regulation of intestinal stem cell homeostasis and crypt regeneration And when the two Notch ligands present in crypts, Dll1 and Dll4, are simultaneously knocked out, every dividing progenitor cell converts into a mucus-secreting goblet cell, with complete loss of the stem cell population.15Gastroenterology. Dll1- and Dll4-Mediated Notch Signaling Are Required for Homeostasis of Intestinal Stem Cells In other words, without Notch, the gut stops making the absorptive cells it needs and overproduces goblet cells instead. This finding becomes directly relevant when we consider drugs that block Notch, because the gut is one of the first tissues to suffer.
Notch in Cancer, Both Accelerator and Brake
Here is where the pathway’s complexity really stands out. In some cancers, Notch drives tumor growth. In others, it restrains it. Whether Notch acts as an oncogene or a tumor suppressor depends on the tissue.
T Cell Leukemia
The most striking oncogenic role for Notch is in T cell acute lymphoblastic leukemia (T-ALL). More than half of human T-ALL cases carry activating mutations in NOTCH1, making it the most commonly mutated gene in this blood cancer.16PubMed. Activating mutations of NOTCH1 in human T cell acute lymphoblastic leukemia These mutations hit two main spots on the receptor: the region that normally keeps the receptor locked shut (causing it to activate without proper ligand contact) and the tail region that controls how quickly the active form is destroyed (letting it persist longer than it should). The discovery that NOTCH1 mutations span all major molecular subtypes of T-ALL generated intense interest in therapies targeting the pathway.17PubMed Central. The role of NOTCH1 signaling in T-ALL Notch1 signaling has also been implicated in promoting the spread of breast cancer cells by activating a program called epithelial-to-mesenchymal transition, a process by which tumor cells become more mobile and invasive.18PubMed Central. Notch1 signaling regulates the epithelial-mesenchymal transition and invasion of breast cancer in a Slug-dependent manner
Squamous Cell Carcinomas
The opposite situation holds in many solid tumors. In skin squamous cell carcinomas, roughly three-quarters of tumors carry mutations that disable NOTCH1 or NOTCH2, including nonsense mutations and frameshifts that truncate the protein so it cannot signal at all.19PubMed Central. Loss-of-function mutations in Notch receptors in cutaneous and lung squamous cell carcinoma Similarly, in head and neck squamous cell carcinoma, about 40% of NOTCH1 mutations are predicted to truncate the protein, consistent with Notch acting as a tumor suppressor rather than an oncogene in these tissues.6PubMed Central. Exome sequencing of head and neck squamous cell carcinoma reveals inactivating mutations in NOTCH1 This dual character means that a therapy designed to block Notch in a T-ALL patient could theoretically promote skin cancer, and vice versa. It is one reason drug development in this space has been so cautious.
Inherited Diseases Linked to Notch
Several genetic disorders are caused by mutations in specific Notch pathway components, offering natural experiments that reveal which tissues depend most heavily on which family members.
Alagille syndrome is a multisystem disorder that most often involves the liver, heart, skeleton, eyes, and face. The majority of cases, about 94%, are caused by mutations in JAG1, the gene encoding the Notch ligand Jagged1.20PubMed Central. NOTCH2 mutations cause Alagille syndrome, a heterogeneous disorder of the notch signaling pathway In the remaining patients without JAG1 mutations, some carry mutations in NOTCH2, the receptor that Jagged1 typically activates in these tissues. Patients with NOTCH2 mutations universally have liver involvement and show similar rates of eye and kidney problems as JAG1 patients, but they tend to have less cardiac involvement, fewer vertebral abnormalities, and a lower likelihood of the characteristic facial features associated with the syndrome.21PubMed Central. NOTCH2 mutations in Alagille syndrome This pattern makes sense: a defective ligand prevents signaling through multiple receptors in multiple tissues, while a defective receptor only affects signaling in those tissues where that particular receptor is the relevant one.
CADASIL (cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy) is a hereditary small-vessel disease caused by mutations in NOTCH3. It leads to recurrent strokes, progressive cognitive decline, and migraine with aura, typically beginning in mid-adulthood. NOTCH3 is the primary Notch receptor in vascular smooth muscle cells, and mutations cause the accumulation of abnormal receptor fragments around small blood vessels in the brain.22PubMed Central. NOTCH3 and CADASIL syndrome: a genetic and structural overview CADASIL is considered rare but is likely underdiagnosed, since its symptoms overlap with other causes of stroke and dementia.
Why Targeting Notch With Drugs Has Been Difficult
Given Notch’s involvement in so many cancers and diseases, the obvious question is why there is not already a successful Notch-blocking drug on the market. The main obstacle is that Notch is doing essential work in healthy adult tissues at the same time you want to shut it down in a tumor.
The first generation of Notch-targeted drugs were gamma-secretase inhibitors (GSIs), which block the second cleavage step needed to release the Notch fragment. The problem became apparent quickly in preclinical testing: GSIs caused severe intestinal toxicity. Treated animals developed goblet cell metaplasia, essentially the same runaway goblet-cell conversion seen in genetic experiments that knock out Notch in the gut, along with crypt cell death and tissue damage.23Toxicological Sciences. Modulation of Notch Processing by γ-Secretase Inhibitors Causes Intestinal Goblet Cell Metaplasia and Induction of Genes Known to Specify Gut Secretory Lineage Differentiation These gastrointestinal effects were confirmed across different GSI compounds and animal species, including in nonclinical safety testing of the GSI avagacestat in dogs.24Toxicological Sciences. Nonclinical Safety Assessment of the γ-Secretase Inhibitor Avagacestat The toxicity is essentially a pharmacological version of the genetic knockout experiments: block Notch in the gut and stem cells are lost, goblet cells take over, and the tissue falls apart.
Newer approaches have tried to be more selective. Antibodies that target individual Notch receptors or specific ligands avoid the broad blockade of all Notch signaling, and intermittent dosing schedules have been tested to allow gut recovery between doses. But gastrointestinal side effects remain the most persistent challenge across multiple GSIs and Notch-targeting antibodies tested in the clinic.
Synthetic Notch Receptors and the Next Generation of Cell Therapy
Rather than trying to block Notch, some researchers have taken the pathway’s elegant design and repurposed it. Synthetic Notch, or synNotch, receptors are engineered proteins that borrow Notch’s mechanical activation principle but are wired to custom outputs. The external portion can be designed to recognize essentially any target molecule on a cell surface, and instead of releasing a natural Notch fragment, cleavage releases an engineered transcription factor that turns on whatever gene the researcher chooses.
This has proven especially useful in improving CAR-T cell therapy, where engineered immune cells are directed to attack tumors. A persistent problem with CAR-T cells is that tumor-associated targets are rarely exclusive to cancer cells, so the T cells sometimes attack healthy tissue too. SynNotch receptors function as logic gates: a T cell is engineered so that the synNotch receptor must first detect one antigen (found preferentially on the tumor) before the T cell is even allowed to produce the CAR that targets a second antigen. This “if A, then attack B” logic dramatically improves specificity.25PubMed Central. SynNotch CAR-T cell, when synthetic biology and immunology meet again
In glioblastoma, one of the most aggressive brain cancers, synNotch-CAR T cells have been designed to recognize a tumor-specific mutation as the priming signal, then locally produce a CAR targeting a second, more broadly expressed antigen. This allows thorough tumor killing while confining T cell activity to the tumor site.26PubMed Central. SynNotch-CAR T cells overcome challenges of specificity, heterogeneity, and persistence in treating glioblastoma The approach addresses a specific frustration with glioblastoma: tumors are genetically heterogeneous, so targeting any single antigen inevitably misses some cells. By using a two-step system, synNotch designs can cast a wider net for killing while keeping a tight leash on where the killing occurs.
Crosstalk With Other Pathways
Notch rarely acts alone in any biological context. Its effects consistently depend on what other signals a cell is receiving at the same time, and two pathways it interacts with particularly frequently are Wnt and Hedgehog. Notch and Wnt are co-required in a striking range of processes across the animal kingdom, from patterning the sea urchin embryo to maintaining the mammalian gut lining and skin. In many of these contexts, the two pathways have opposing effects: Wnt promotes stem cell self-renewal while Notch pushes cells toward specific fates, or vice versa, creating a balanced system where disrupting either pathway tips the scale. Aberrant signaling through both pathways together is implicated in several cancers, suggesting they can cooperate in disease as well as in normal biology.27PubMed Central. Wnt-Notch signalling crosstalk in development and disease
This integration with other pathways partly explains why Notch can have such different effects in different tissues. The same Notch signal can promote cell division in one context and halt it in another, depending on which other pathways are active. It also means that simple models of “Notch on equals growth” or “Notch off equals differentiation” tend to break down outside the specific tissue where they were first described. Researchers studying Notch in one organ system frequently learn this the hard way when their findings fail to generalize.
An Ancient and Deeply Conserved System
Notch signaling is not a recent evolutionary invention. Comparative studies across animal genomes indicate that a functional Notch pathway was present in the last common ancestor of all animals, and some of the protein domains used by Notch receptors and ligands trace back even further into single-celled eukaryote evolution. The pathway assembled over time from a mix of ancient protein modules and newly evolved components specific to multicellular life. This deep conservation explains why the same pathway can pattern tissues as different as fruit fly wings and human hearts: the fundamental logic of contact-dependent neighbor signaling was established before any of those structures existed, and evolution has repeatedly deployed the same toolkit for new purposes.