Cell Biology
Cell Polarity
Apical-basal asymmetry maintained by Par/Crumbs/Scribble complexes — drives epithelia, neurons, embryonic axes
Cell polarity is the asymmetric organisation of a cell into distinct apical and basolateral (or front-and-back) domains, established and maintained by three antagonistic protein modules — the Par complex, the Crumbs complex, and the Scribble complex. Discovered by Kemphues, Priess, and Strome in 1988 through par mutants in Caenorhabditis elegans, the same machinery polarises epithelial sheets, neuronal axon-dendrite identity, migrating leukocytes, and the embryonic body axes of every metazoan studied. About 80 percent of human cancers are carcinomas, and dismantling cell polarity is a near-universal step in their progression.
- Core modulePar-3 · Par-6 · aPKC · Cdc42
- Apical scaffoldCrumbs · Pals1 · PatJ
- BasolateralScribble · Dlg · Lgl
- First identifiedKemphues 1988 (C. elegans)
- ConservationWorm → fly → mouse → human
- Cancer relevanceLost in ~80 % of carcinomas
Interactive visualization
Press play, or step through manually. The visualization is yours to drive — try it before reading on.
Watch the 60-second explainer
A condensed visual walkthrough — narrated, captioned, under a minute.
Why cell polarity matters
- Epithelia are vectorial. An intestinal enterocyte absorbs nutrients on its apical (luminal) side via SGLT1 and GLUT5 transporters and exports them basolaterally via GLUT2 — the right transporter must reach the right membrane, or absorption fails. Fischer-rat-thyroid (FRT) and MDCK cell-line studies show that more than 200 plasma-membrane proteins partition between the two domains with greater than 90 percent fidelity, all enforced by polarity machinery.
- Tight junctions need polarity. The Par-3/Par-6/aPKC complex localises to the apical-most cell-cell junction, where it specifies the position of claudin-based tight junctions that seal the paracellular barrier. Knocking out Par-3 in mouse mammary epithelium disrupts barrier function and ZO-1 localisation within 48 hours.
- Asymmetric cell division depends on it. The C. elegans one-cell zygote uses cortical PAR asymmetry to position the spindle posteriorly, producing an unequal first cleavage (AB ≈ 60 percent of cytoplasm, P1 ≈ 40 percent). Drosophila neuroblasts use the same module with Inscuteable to bias spindle orientation and distribute fate determinants Numb and Prospero asymmetrically — the foundation of the neural lineage tree.
- Neuronal axon specification. A pyramidal neuron in culture commits to a single axon among 4–5 minor neurites within 24–48 hours of plating, by accumulating Par-3/Par-6/aPKC and PI3K activity at the tip of the chosen neurite. Knockdown causes multiple axons or none — a polarity-driven binary decision.
- Carcinoma progression. Loss of E-cadherin, Crumbs3, or Scribble correlates with poor survival in colorectal, breast, lung, and pancreatic cancer cohorts. The TCGA pan-cancer dataset shows aPKC-iota (PRKCI) amplification in roughly 35 percent of pancreatic ductal adenocarcinomas, where it drives the invasive phenotype.
- Planar polarity orients tissues. Mammalian inner-ear hair cells display a stereotyped V-shaped stereocilia bundle aligned across the cochlea by Vangl2 and Frizzled. Vangl2 mutants (looptail mice) have neural-tube-closure failure and cochlear-orientation defects in 100 percent of homozygotes.
- Immune-cell migration. Neutrophils chase chemotactic gradients (fMLP, IL-8) by polarising Cdc42 + Par-6 to the leading edge and PtdIns(3,4,5)P3 to the front. The same Par module that organises a stationary epithelial cell's apical pole organises a moving leukocyte's leading edge.
Common misconceptions
- Polarity equals geometry. Polarity is biochemical asymmetry, not just shape. A round neutrophil with no morphological cue is biochemically polarised the moment it senses a chemoattractant gradient — Cdc42 activity, PI3K lipids, and Par-6 sort to one face within seconds.
- Cell polarity is the same as planar cell polarity. Apical-basal polarity (perpendicular to the epithelial plane) and PCP (within the plane) use different core modules — Par/Crumbs/Scribble for apical-basal, Frizzled/Vangl/Flamingo for planar. They interact (E-cadherin and tight junctions feed both), but the core proteins differ.
- Par-3 binds Par-6 directly. Par-3 and Par-6 are both PDZ-domain scaffolds, but their interaction in some contexts is bridged by aPKC and Cdc42-GTP rather than direct. The exact binding mode shifts between species — Drosophila Bazooka/Par-6/aPKC versus mammalian Par-3/Par-6/aPKC — and over the cell cycle.
- Loss-of-polarity causes cancer single-handedly. Loss of Scribble or Lgl in Drosophila imaginal discs causes overgrowth only when combined with cooperating oncogenic signals (e.g. RasV12). In mammals, polarity loss is a permissive condition for malignant transformation but not sufficient on its own — tumours require driver mutations as well.
- Polarity is established once and stays. Polarity is a dynamic steady state. aPKC continuously phosphorylates Lgl and Par-1 to keep them off the apical membrane; protein turnover happens on minute timescales. Withdrawing kinase activity collapses asymmetry within tens of minutes.
- Bacteria don't have polarity. They do. Caulobacter crescentus has a stalked pole and a flagellated pole with completely different protein compositions, established by the master regulator CtrA and oscillating ParA-ParB-like systems. The mechanism is convergent rather than homologous, but the principle of biochemical cortical asymmetry is universal.
How cell polarity works
The central biochemical switch is mutual exclusion enforced by atypical protein kinase C (aPKC). Apical aPKC, scaffolded by Par-6 and Par-3 and activated by Cdc42-GTP, phosphorylates basolateral substrates Lgl, Par-1, and Numb on serines that flank polybasic phosphoinositide-binding regions. Phosphorylation neutralises the basic patch and ejects the substrate from the apical membrane, while basolateral phosphatases (PP1/PP2A) and kinases (Par-1) reciprocally drive Par-3 from the basolateral side. This double-negative logic produces a sharp boundary at the apical-basolateral interface — typically the level of the tight junction in vertebrate epithelia or the adherens junction in flies.
The Crumbs complex sits apical to the Par complex and stabilises it. Crumbs is a transmembrane protein with a tiny intracellular tail (37 residues in human Crumbs1) that binds Pals1 (also called Stardust in flies) and via PatJ to Par-6. Crumbs knockouts (mouse Crumbs3) fail to form proper apical microvilli; over-expression expands the apical domain at the expense of basolateral. The Scribble module (Scribble + Dlg + Lgl) anchors basolateral identity, with Scribble acting as a large LRR-PDZ scaffold platform. In Drosophila, scrib, dlg, and lgl mutants were the first identified neoplastic tumour-suppressor genes, defined by their giant overgrown larval imaginal discs.
Symmetry breaking — going from a uniform cell to a polarised one — uses positive feedback. In the C. elegans zygote, the sperm centrosome locally inhibits cortical RhoA, which weakens actomyosin contraction at the future posterior; cytoplasmic flow then advects PAR proteins, with anterior PARs (Par-3, Par-6, PKC-3) moving with the flow and posterior PARs (Par-1, Par-2) loaded onto the now-relaxed posterior cortex. Once asymmetric, mutual antagonism between the two PAR clusters locks in the pattern within about 15 minutes, ready for the first asymmetric division.
Apical-basal vs planar cell polarity vs front-rear polarity
| Feature | Apical-basal | Planar (PCP) | Front-rear (migrating cell) |
|---|---|---|---|
| Axis | Perpendicular to epithelial plane | Within the epithelial plane | Migration direction |
| Core complex | Par / Crumbs / Scribble | Frizzled / Vangl / Prickle / Dishevelled / Flamingo | Par-6 / aPKC / Cdc42 (re-deployed) |
| Trigger | Cell-cell contacts (E-cadherin), substrate | Long-range Wnt/Frizzled gradient | Chemoattractant gradient (fMLP, IL-8, EGF) |
| Output | Vectorial transport, junction placement | Tissue-wide orientation (bristles, hair cells) | Lamellipodium forward, retraction back |
| Timescale | Minutes to hours | Hours to days | Seconds |
| Cancer link | Carcinoma initiation | Neural-tube defects, hair-cell pathology | Metastatic invasion |
| Model organism showcase | C. elegans zygote, MDCK cells | Drosophila wing, mouse cochlea | Dictyostelium, neutrophil |
Famous experiments
- Kemphues, Priess & Strome 1988. Genetic screen in C. elegans for maternal-effect mutants that lost anterior-posterior asymmetry of the one-cell zygote — defined the par genes (par-1 to par-6) and launched the field.
- Bridges 1916, then Gateff 1978. Mary Bridges first noted Drosophila lethal-giant-larvae mutants; Elisabeth Gateff identified them as the first neoplastic tumour suppressors — disc tissue grew without limit when transplanted into adults.
- Drubin & Nelson 1996. Synthesis paper unifying epithelial, neuronal, and yeast polarity around shared scaffolds and small GTPases — the conceptual frame the field still uses.
- Joberty, Petersen, Macara 2000 + Lin et al 2000. Identified Par-6 as the Cdc42-GTP-binding link to atypical PKC in mammalian cells, completing the molecular bridge from worm genetics to vertebrate epithelia.
- Tepass et al 1990 (Crumbs in Drosophila). Identified Crumbs as essential for establishing the apical membrane in fly embryonic ectoderm — overexpression expanded apical territory, deletion abolished it.
Frequently asked questions
What are the Par, Crumbs, and Scribble complexes?
Three evolutionarily conserved scaffold-and-kinase modules that mutually antagonise each other to define apical and basolateral domains. The Par complex contains Par-3, Par-6, atypical protein kinase C (aPKC), and the Rho-family GTPase Cdc42; it localises just below the apical surface at the tight junction. The Crumbs complex contains Crumbs (a transmembrane protein), Pals1/Stardust, and PatJ; it sits in the apical membrane and recruits Par-3/Par-6/aPKC to refine the apical domain. The Scribble complex contains Scribble, Discs-large (Dlg), and Lethal-giant-larvae (Lgl) and occupies the basolateral domain. Lgl is phosphorylated and excluded by apical aPKC, which is the central biochemical switch that keeps apical and basolateral identities mutually exclusive.
How was cell polarity first discovered?
Kemphues, Priess, and Strome reported in 1988 a screen for maternal-effect mutants in the nematode Caenorhabditis elegans whose one-cell embryos failed to establish anterior-posterior asymmetry. The genes were named par-1 through par-6, for partitioning-defective. par mutants laid eggs that divided symmetrically instead of producing the larger anterior AB blastomere and smaller posterior P1, and they failed to localise germ-line P granules to the posterior. The molecular cloning of par-3, par-6, and pkc-3 (aPKC) over the following decade revealed the same scaffold + atypical PKC complex in flies, frogs, and mammals — a deep conservation that established Par as the universal polarity module.
Why is loss of cell polarity a hallmark of cancer?
Carcinomas, which arise from epithelial cells and account for roughly 80 percent of all human cancers, must dismantle apical-basal polarity to leave their tissue of origin and invade. Loss of Scribble, Dlg, or Lgl was first identified in Drosophila as causing neoplastic overgrowth — the original mutations gave the imaginal discs the disordered tumour-like appearance Mary Bridges named lethal-giant-larvae. In mammals, Scribble is downregulated in many breast and colon carcinomas, and Crumbs3 loss disrupts tight junctions and licenses epithelial-mesenchymal transition. The Par complex is more often hijacked than lost: aPKC is amplified or overactive in 40 to 70 percent of pancreatic and ovarian cancers, where it cooperates with oncogenic Ras to promote invasion.
What is planar cell polarity?
Planar cell polarity (PCP) is asymmetry within the plane of an epithelium — perpendicular to the apical-basal axis — that orients structures across the tissue. Examples include the uniform posterior orientation of bristles on the Drosophila wing, the V-shaped patterning of stereocilia bundles in mammalian inner ear hair cells, and the orientation of cilia beating in airway epithelium. The core PCP module is composed of Frizzled, Vangl (Van Gogh-like), Prickle, Dishevelled, Diego, and the protocadherins Flamingo/Celsr. These proteins partition into mutually exclusive proximal and distal cortical domains, and PCP signalling propagates from cell to cell to coordinate orientation across the field. Disrupting PCP in mice (Vangl1 or Vangl2 mutants) causes neural-tube-closure defects and cochlear hair-cell disorganisation.
How do neurons polarise into axon and dendrites?
Hippocampal pyramidal neurons in culture extend several minor neurites from the soma, and within roughly 24 to 48 hours one of them grows much faster, accumulates Par-3 and Par-6 plus aPKC at its tip, and becomes the axon. The remaining neurites become dendrites and accumulate the dendrite markers MAP2 and the Scribble complex on their basolateral-equivalent membranes. Knocking out Par-3 or aPKC in cultured neurons prevents axon specification or causes multiple axons. The neuronal polarity programme borrows the same Par module deployed in epithelia, but the geometry is reorganised — the axon-distal tip is the equivalent of the apical pole.
What signals establish polarity in the embryo?
External cues feed into the conserved polarity machinery. In the C. elegans one-cell embryo, the sperm centrosome contacts the cortex and locally inhibits actomyosin contraction, breaking symmetry; the resulting cortical flow segregates Par-3/Par-6/PKC-3 to the anterior and Par-1/Par-2 to the posterior in roughly 10 to 15 minutes. In Drosophila and mammalian epithelia, cell-cell contacts via E-cadherin, Nectin, and JAM-A recruit the Par complex to nascent junctions, which then direct apical-basal asymmetry. In single migrating cells (neutrophils, fibroblasts), gradients of chemoattractants activate Cdc42 at the leading edge, which recruits Par-6/aPKC and orients the microtubule-organising centre forward. The same module reads the input that breaks symmetry.