18.11.03 · organismal-bio / development

Organogenesis: inductive interactions, morphogen gradients, and stem cell niches

stub3 tiersLean: nonepending prereqs

Anchor (Master): Gilbert, S. F. — Developmental Biology, 12th ed. (2020)

Intuition Beginner

After gastrulation has produced the three germ layers (ectoderm, mesoderm, endoderm), the embryo begins building its organs. This process is called organogenesis. Organs do not form in isolation -- tissues signal to one another through chemical messages. One group of cells sends a signal that tells neighbouring cells what to become. This is called an inductive interaction.

The classic example: the notochord (a rod of mesoderm along the midline) secretes signalling molecules that instruct the overlying ectoderm to thicken, fold, and close into a neural tube -- the future brain and spinal cord. If the notochord is removed experimentally, no neural tube forms; if an extra notochord is grafted nearby, a second neural tube can be induced.

Limbs grow outward from the body wall as buds. Each limb bud contains a signalling centre at its tip (the apical ectodermal ridge) that keeps the underlying cells proliferating, causing the limb to lengthen. Another signalling centre at the posterior edge (the zone of polarizing activity) secretes a morphogen that patterns the digits: high concentrations produce the pinky side, low concentrations produce the thumb side.

Even after organs form, some tissues retain stem cells -- undifferentiated cells that can divide and produce specialized cell types throughout life. These stem cells reside in protected microenvironments called niches. The bone marrow contains hematopoietic stem cells that continuously produce blood cells. The lining of the intestine is renewed every few days from stem cells at the base of tiny pits called crypts. The brain retains neural stem cells in specific zones that can generate new neurons.

Visual Beginner

Organ system	Key inductive interaction	Major signalling molecules
Neural tube	Notochord induces overlying ectoderm	Shh (ventral), BMP (dorsal)
Limb	AER and ZPA pattern the limb bud	Fgf8 (AER), Shh (ZPA), Wnt7a (dorsal)
Kidney	Ureteric bud and metanephric mesenchyme reciprocally induce	Gdnf/Ret, Wnt11, BMP
Heart	Cardiac crescent fuses at the midline	Nkx2-5, Bmp2, Wnt inhibition
Tooth	Oral epithelium and underlying mesenchyme	BMP4, Fgf8, Shh, Wnt

Worked example Beginner

The formation of the neural tube (neurulation) is the first major organogenesis event in vertebrate embryos and illustrates how inductive interactions build a complex organ step by step:

Neural plate formation. The notochord secretes BMP antagonists (Chordin, Noggin), which permit the overlying ectoderm to adopt neural fate instead of skin. This strip of neural ectoderm thickens to form the neural plate.
Neural fold elevation. The edges of the neural plate rise upward, forming paired neural folds. The cells at the hinge points change shape (elongating and narrowing through apical constriction), causing the flat plate to buckle upward like folding a piece of paper.
Neural tube closure. The neural folds meet at the dorsal midline and fuse. Closure begins in the middle of the embryo and proceeds both anteriorly (toward the brain) and posteriorly (toward the tail). In mammals, the neural tube closes at multiple independent closure sites simultaneously.
Neural crest emigration. As the neural folds fuse, a population of cells at the crest of each fold detaches and migrates away. These neural crest cells are a remarkable multipotent population that gives rise to: peripheral nervous system (sensory ganglia, autonomic ganglia, Schwann cells), facial bones and cartilage, melanocytes, smooth muscle of the great vessels, and adrenal medulla.
Patterning of the tube. Once closed, the neural tube is patterned along two axes. The anterior-posterior axis specifies brain regions (forebrain, midbrain, hindbrain, spinal cord). The dorsal-ventral axis is patterned by opposing morphogen gradients: Shh from the floor plate (ventral) specifies motor neurons and ventral interneurons; BMP from the roof plate (dorsal) specifies sensory interneurons and neural crest.

If the neural tube fails to close completely, the result is a neural tube defect: spina bifida (incomplete closure posteriorly) or anencephaly (incomplete closure anteriorly). Folate supplementation before and during early pregnancy significantly reduces the risk.

Check your understanding Beginner

Exercise (easy, multiple choice).

During neurulation, what structure induces the overlying ectoderm to form the neural plate?

A. The primitive streak B. The notochord C. The neural crest D. The somites

Hint

The neural plate forms directly above a rod of mesodermal cells that runs along the midline of the embryo. What is this rod called?

Answer

B. The notochord. The notochord secretes BMP antagonists (Chordin, Noggin) that block BMP signalling in the overlying ectoderm, permitting it to adopt neural fate. The notochord also secretes Sonic hedgehog (Shh), which later patterns the ventral neural tube. If the notochord is removed before neurulation, no neural tube forms. If an extra notochord is grafted alongside the neural tube, a second set of ventral neurons (including motor neurons) is induced at the graft site.

Exercise (easy, multiple choice).

In the developing limb bud, the apical ectodermal ridge (AER) controls:

A. Which digits form (thumb vs. pinky identity) B. How long the limb grows (proximal-distal outgrowth) C. Whether the limb is dorsal or ventral D. The number of limbs the embryo develops

Hint

The AER sits at the very tip of the growing limb bud and keeps the underlying mesenchyme proliferating. What aspect of limb development does this control?

Answer

B. How long the limb grows (proximal-distal outgrowth). The AER secretes Fgf8 (and other FGFs) that maintain the underlying mesenchyme in a proliferative, undifferentiated state. As long as the AER is present, the limb continues to elongate. If the AER is surgically removed, limb outgrowth stops prematurely, producing a truncated limb. Digit identity (A) is controlled by the zone of polarizing activity (ZPA), which secretes Shh. Dorsal-ventral patterning (C) is controlled by Wnt7a from the dorsal ectoderm.

Exercise (easy, true/false).

Stem cells in the adult body exist only in the bone marrow.

Hint

Stem cells are found wherever tissues need continuous renewal or repair. Think of tissues that constantly replace their cells.

Answer

False. Stem cells reside in multiple niches throughout the body. The bone marrow contains hematopoietic stem cells (blood cell production) and mesenchymal stem cells (bone, cartilage, fat). But stem cells also exist in the intestinal crypt base (Lgr5+ cells renewing the gut lining every 3--5 days), the basal layer of the epidermis (skin renewal), the subventricular zone and subgranular zone of the brain (neural stem cells), hair follicles (bulge stem cells), and the cornea (limbal stem cells). Each niche provides specific signals (Wnt, Notch, BMP, FGF) that maintain the stem cell state and regulate the balance between self-renewal and differentiation.

Exercise (easy, multiple choice).

Neural crest cells give rise to all of the following EXCEPT:

A. Sensory ganglia of peripheral nerves B. Bones and cartilage of the face C. The spinal cord D. Melanocytes (pigment cells)

Hint

Neural crest cells emigrate from the dorsal side of the closing neural tube and migrate throughout the body. The spinal cord is the neural tube itself -- which cells form it?

Answer

C. The spinal cord. The spinal cord is formed from the neural tube, not from neural crest cells. Neural crest cells are a separate population that detaches from the dorsal neural tube during closure and migrates to diverse destinations. They give rise to sensory ganglia (A), facial bones and cartilage (B), melanocytes (D), autonomic ganglia, Schwann cells, adrenal medulla cells, and smooth muscle of the cardiac outflow tract. The extraordinary developmental potential of neural crest cells makes them a key focus of organogenesis research.

Formal definition Intermediate+

Neurulation and neural tube patterning

Primary neurulation (in mammals and birds) converts the neural plate into the neural tube through folding and fusion. The process involves:

Apical constriction: apical actin-myosin cables in neuroepithelial cells contract, narrowing the apical surface and causing the epithelium to bend. Non-muscle myosin II and the actin-binding protein Shroom3 are essential for this process.
Hinge points: three hinge points (median hinge point at the notochord, paired dorsolateral hinge points) serve as fulcra for neural fold elevation.
Closure: the neural folds meet and fuse through cell adhesion molecules (E-cadherin, NCAM) and the surface ectoderm overlying the fusing tube re-forms a continuous layer.

Secondary neurulation (in the caudal region of mammals) forms the most posterior neural tube without folding: the medullary cord (a solid rod of cells) cavitates to form a lumen.

After closure, the neural tube is patterned along the dorsal-ventral axis by two opposing morphogen gradients:

Signal	Source	Target	Ventral-to-dorsal gradient
Shh	Notochord, then floor plate	Ventral neural tube progenitors	High ventral, low dorsal
BMP (BMP4, BMP7)	Roof plate	Dorsal neural tube progenitors	High dorsal, low ventral
Wnt	Roof plate	Dorsal interneurons	High dorsal, low ventral

Shh acts as a morphogen -- different concentrations specify different neuronal subtypes in the ventral neural tube:

Shh concentration	Progenitor domain	Neuronal fate
Highest	pMN	Motor neurons
High	p3	V3 interneurons
Moderate	p2	V2 interneurons
Low-moderate	p1	V1 interneurons
Low	p0	V0 interneurons

This concentration-dependent specification was demonstrated by Jessell and colleagues (1997--2000) using in vitro assays: explants of naive neural tube exposed to increasing Shh concentrations generated progressively more ventral neuronal types.

The anterior-posterior axis of the neural tube is patterned by a combination of signals: retinoic acid (from the adjacent somites, posteriorizing), FGF (from the caudal region), and anterior signals (FGF8 from the isthmic organizer at the midbrain-hindbrain boundary, Wnt1 from the dorsal midbrain). The isthmic organizer secretes FGF8, which acts as a mitogen and patterning signal for the midbrain and cerebellum.

Limb development

The vertebrate limb is patterned along three axes, each controlled by a distinct signalling centre:

Proximal-distal axis (stylopod-zeugopod-autopod). The apical ectodermal ridge (AER), a thickened ridge of ectoderm at the distal tip of the limb bud, secretes Fgf8 (and Fgf4, Fgf9, Fgf17 in the posterior AER). FGF signals maintain the underlying progress zone mesenchyme in a proliferative, undifferentiated state. Cells that leave the progress zone early (as the limb elongates) form proximal structures (humerus/femur); cells that leave later form distal structures (radius-ulna/tibia-fibula; digits). The AER is maintained by Shh from the ZPA through a positive feedback loop: Shh induces Gremlin in the mesenchyme, and Gremlin (a BMP antagonist) sustains AER integrity and Fgf expression.

The two-signal model (proximal-distal identity) proposes that early limb mesenchyme is exposed to retinoic acid (RA) from the body wall (proximal signal) and FGF from the AER (distal signal). High RA + low FGF specifies stylopod (upper arm/thigh); low RA + high FGF specifies zeugopod (forearm/calf) and autopod (hand/foot). The transcription factors Meis1/2 (proximal) and HoxA/D genes (distal) mediate this response.

Anterior-posterior axis (thumb to pinky). The zone of polarizing activity (ZPA), a cluster of mesenchymal cells at the posterior margin of the limb bud, secretes Sonic hedgehog (Shh). Shh forms a concentration gradient (high posterior, low anterior) that specifies digit identity:

Low Shh: digit 1 (thumb/big toe)
Moderate Shh: digits 2 and 3
High Shh: digits 4 and 5 (pinky/little toe)

Digits 2--5 are specified by the temporal exposure model: cells that experience Shh for a short period become digit 2; longer exposure produces digits 3, then 4. Digit 5 is specified by cells that are autonomously posterior and produce Shh themselves. Grafting an additional ZPA to the anterior of the limb bud produces a mirror-image duplication of digits (e.g., 4-3-2-2-3-4), confirming Shh as the polarizing morphogen.

Dorsal-ventral axis. Wnt7a from the dorsal limb ectoderm activates Lmx1b in the underlying mesenchyme, specifying dorsal fate. The ventral ectoderm expresses Engrailed-1 (En1), which represses Wnt7a ventrally. Loss of Wnt7a causes the limb to adopt a double-ventral (palmar/plantar on both sides) phenotype; loss of En1 causes a double-dorsal phenotype.

Kidney development

The vertebrate kidney develops through three successive forms, each more complex:

Pronephros: the simplest kidney, functional in larval fish and amphibians; vestigial in mammals (a few tubules that regress by week 4 in humans). Consists of a single nephron per segment: a filtration unit (glomerulus) and a tubule.
Mesonephros: an intermediate kidney, functional in embryonic fish, amphibians, and birds; transient in mammals (weeks 4--8 in humans). Composed of multiple tubules along the Wolffian (mesonephric) duct. In male mammals, some mesonephric tubules persist as the efferent ductules of the testis.
Metanephros: the definitive (permanent) kidney in amniotes, beginning at week 5 in humans. The metanephros forms through a reciprocal induction between two tissues:
- The ureteric bud (an outgrowth of the Wolffian duct) invades the metanephric mesenchyme.
- The mesenchyme secretes GDNF (Glial cell line-Derived Neurotrophic Factor), which activates the Ret receptor tyrosine kinase on the ureteric bud, causing it to grow and branch.
- The branching ureteric bud secretes Wnt11, which feeds back to maintain GDNF expression in the mesenchyme -- a positive feedback loop.
- At the tips of each branch, the ureteric bud induces clusters of mesenchymal cells to undergo a mesenchymal-to-epithelial transition (MET), forming renal vesicles that develop into nephrons (the functional units of the kidney, each containing glomerulus, proximal tubule, loop of Henle, and distal tubule).
- The ureteric bud branches form the collecting duct system and the renal pelvis and ureter.

Each human kidney contains approximately 1 million nephrons. Nephron number is determined by the extent of ureteric bud branching during development and is complete by birth (no new nephrons form postnatally in humans). Low nephron endowment is associated with hypertension and chronic kidney disease in adulthood.

Heart development

Cardiac development proceeds through defined stages:

Cardiac crescent (cardiogenic mesoderm): mesodermal cells in the anterior lateral plate mesoderm, specified by BMP (from the endoderm) and Wnt inhibition (from the underlying endoderm and the notochord). The transcription factor Nkx2-5 (a homeobox gene) is the master regulator of cardiac identity.
Linear heart tube: bilateral cardiac progenitor fields migrate medially and fuse at the midline, forming a simple contractile tube. The tube has four regions (arterial to venous): truncus arteriosus, bulbus cordis, ventricle, and atrium. Blood flows in a peristaltic wave from posterior (venous/inflow) to anterior (arterial/outflow).
Cardiac looping: the linear tube bends rightward and posteriorly (dextral looping), establishing the future spatial relationships of the chambers. The direction of looping is determined by the left-right axis (see 18.11.02 pending): left-sided Nodal and Pitx2 drive rightward looping. If the left-right axis is disrupted, looping can be randomized (L-loop instead of D-loop), producing heterotaxy with severe cardiac malformations.
Septation: the heart is partitioned into four chambers by the formation of septa (atrial septum, ventricular septum, outflow tract septum). Atrial septation involves the septum primum and septum secundum, which grow down from the roof of the atrium, leaving the foramen ovale as a one-way shunt that allows blood to bypass the non-functional fetal lungs. The outflow tract septum is formed by neural crest cells that migrate into the cardiac outflow tract and divide it into the aorta and pulmonary trunk. Ventricular septation is completed by the muscular ventricular septum growing upward from the apex and the membranous septum derived from endocardial cushion tissue.
Valve formation: endocardial cushions (swellings of extracellular matrix in the atrioventricular canal and outflow tract) are invaded by mesenchymal cells derived from the endocardium via epithelial-mesenchymal transition (EMT), driven by BMP2 and TGF-beta signalling. These cushions are remodeled into the atrioventricular (mitral and tricuspid) and semilunar (aortic and pulmonary) valves.

Epithelial-mesenchymal transition (EMT)

EMT is a cellular programme in which epithelial cells lose cell-cell adhesion and apical-basal polarity and acquire a migratory, mesenchymal phenotype. EMT occurs in multiple organogenesis events:

Neural crest delamination from the neural tube
Cardiac endocardial cushion formation (valve development)
Kidney mesenchyme condensation and subsequent MET
Palate fusion

The EMT programme involves downregulation of E-cadherin (by the transcription factors Snail, Slug, Twist) and upregulation of N-cadherin, vimentin, and fibronectin. EMT is reversible: mesenchymal cells can undergo mesenchymal-epithelial transition (MET) during kidney nephron formation and somite epithelialization. Pathological EMT contributes to cancer metastasis and organ fibrosis.

Stem cell niches

Stem cells are maintained in specialized microenvironments (niches) that provide signals balancing self-renewal and differentiation:

Hematopoietic stem cell (HSC) niche (bone marrow). HSCs reside near sinusoidal endothelial cells (vascular niche) and osteoblasts (endosteal niche). Key niche signals:

Signal	Source	Function
SCF (Stem Cell Factor)	Mesenchymal stromal cells, endothelial cells	Activates c-Kit receptor on HSCs; essential for HSC maintenance
CXCL12 (SDF-1)	CX12-abundant reticular (CAR) cells	Retains HSCs in the marrow via CXCR4 receptor
Angiopoietin-1	Osteoblasts	Activates Tie2 on HSCs; promotes quiescence
Thrombopoietin (TPO)	Osteoblasts, liver	Maintains HSC quiescence via Mpl receptor
Osteopontin	Osteoblasts	Negative regulator; limits HSC proliferation

HSCs can be mobilized out of the niche into the peripheral blood by administering G-CSF (granulocyte colony-stimulating factor), which downregulates CXCL12 and disrupts the retention signal. This is the basis for peripheral blood stem cell harvesting for bone marrow transplantation.

Intestinal crypt niche. The intestinal epithelium is the most rapidly renewing tissue in the body, replaced every 3--5 days. Lgr5+ stem cells reside at the crypt base, intercalated between Paneth cells (which serve as the niche):

Wnt signals from Paneth cells and underlying mesenchyme maintain stem cell identity. Lgr5 is a Wnt target gene and a receptor for R-spondin (a Wnt potentiator).
Notch signalling from adjacent cells prevents differentiation toward the secretory lineage; high Notch = absorptive (enterocyte) fate; low Notch = secretory (goblet, Paneth, enteroendocrine) fate.
BMP from the villus mesenchyme promotes differentiation and is antagonized by Noggin from the crypt mesenchyme, creating a Wnt-high/BMP-low crypt zone and a Wnt-low/BMP-high villus zone.
EphB2/EphB3 receptors (Wnt targets) guide cell migration: stem cells and Paneth cells are EphB2-positive and retained at the crypt base; differentiated cells lose EphB expression and are pushed upward.

Neural stem cell niche. In the adult mammalian brain, neural stem cells reside in two regions:

Subventricular zone (SVZ) of the lateral ventricles: astrocyte-like B cells (the stem cells) divide to produce C cells (transit-amplifying), which generate neuroblasts (A cells) that migrate along the rostral migratory stream to the olfactory bulb, where they integrate as interneurons. The SVZ niche includes ependymal cells (which provide Noggin, antagonizing BMP to maintain stemness), the cerebrospinal fluid (source of Shh and other signals), and the vasculature (providing endothelial signals).
Subgranular zone (SGZ) of the hippocampal dentate gyrus: Type 1 radial glia-like cells (the stem cells) produce intermediate progenitors that differentiate into granule neurons. This is one of the few sites of adult neurogenesis in humans, and its function is associated with learning, memory, and mood regulation. Physical exercise, enriched environments, and antidepressants increase SGZ neurogenesis; chronic stress and inflammation decrease it.

Counterexamples to common slips

The notochord does not become the backbone (spine). The vertebrae form from the somites (paraxial mesoderm). The notochord mostly regresses, persisting only as the nucleus pulposus of the intervertebral discs.
Neural crest cells and neural tube cells are not the same. Neural crest cells are a migratory, multipotent population that detaches from the dorsal neural tube; neural tube cells remain in place and form the central nervous system.
Shh is not only a "brain and spinal cord" patterning molecule. Shh also patterns the limb (digit identity), the lung (branching), the tooth (cusp patterning), and hair follicles. It is one of the most pleiotropic signalling molecules in development.
Stem cells are not "undifferentiated cells that can become anything." Adult (tissue-specific) stem cells are multipotent (limited to their tissue of origin), not pluripotent. Hematopoietic stem cells make blood cells, not neurons. Only embryonic stem cells and induced pluripotent stem cells (iPSCs) are pluripotent.
EMT and MET are not developmental oddities. They are fundamental processes that recur throughout organogenesis and, when reactivated inappropriately in adults, drive cancer metastasis (EMT) and organ fibrosis.

Key theorem with proof Intermediate+

Theorem (Shh gradient domain specification in the ventral neural tube). Consider the ventral neural tube as a one-dimensional domain extending from the floor plate (position $x = 0$ , ventralmost) to the midpoint of the tube (position $x = L$ ). Shh is produced by the floor plate and diffuses dorsally with diffusion coefficient $D$ , degraded at rate $μ$ , establishing a steady-state gradient:

[S] (x) = S_{0} e^{- x / λ}, λ = D / μ .

Neuronal subtypes are specified at Shh concentration thresholds $T_{1} > T_{2} > T_{3} > T_{4}$ (where $T_{1}$ is the threshold for pMN motor neuron progenitors, requiring the highest Shh). Each progenitor domain $D_{i}$ occupies the region where $[S] (x) \geq T_{i}$ but $[S] (x) < T_{i - 1}$ (for $i > 1$ ). The width of domain $D_{i}$ is:

Δ x_{i} = λ ln \frac{T _{i - 1}}{T _{i}} .

If Shh source strength $S_{0}$ is halved (e.g., heterozygous Shh mutation), the most dorsal ventral domain ( $D_{4}$ , specified at threshold $T_{4}$ , the lowest threshold) contracts by a distance:

δ x = λ ln 2.

If $δ x > λ ln (S_{0} / T_{4})$ (i.e., if halving reduces the Shh concentration at position $L$ below $T_{4}$ ), the outermost domain is eliminated entirely.

Proof. The boundary of domain $D_{i}$ is at position $x_{i}$ where $[S] (x_{i}) = T_{i}$ :

S_{0} e^{- x_{i} / λ} = T_{i} ⟹ x_{i} = λ ln \frac{S _{0}}{T _{i}} .

The width of domain $D_{i}$ (between thresholds $T_{i - 1}$ and $T_{i}$ ) is:

Δ x_{i} = x_{i - 1} - x_{i} = λ ln \frac{S _{0}}{T _{i - 1}} - λ ln \frac{S _{0}}{T _{i}} = λ ln \frac{T _{i}}{T _{i - 1}} = λ ln \frac{T _{i - 1}}{T _{i}} .

This is independent of $S_{0}$ : the widths of intermediate domains depend only on the threshold ratios, not on the absolute Shh concentration. This is the robustness property of the French flag model applied to neural tube patterning.

When $S_{0}$ is halved to $S_{0} /2$ , the boundary of the outermost domain $D_{4}$ shifts from $x_{4} = λ ln (S_{0} / T_{4})$ to $x_{4}^{'} = λ ln (S_{0} /2 T_{4})$ . The contraction is:

δ x = x_{4} - x_{4}^{'} = λ ln \frac{S _{0}}{T _{4}} - λ ln \frac{S _{0}}{2 T _{4}} = λ ln \frac{2 T _{4}}{T _{4}} = λ ln 2.

If $δ x > x_{4} = λ ln (S_{0} / T_{4})$ , then $x_{4}^{'} < 0$ , meaning the threshold $T_{4}$ is not reached anywhere in the tube: the outermost domain is eliminated. This occurs when $ln 2 > ln (S_{0} / T_{4})$ , i.e., when $S_{0} < 2 T_{4}$ . $□$

Bridge. This result illustrates a general principle of morphogen-driven patterning: intermediate domains are robust to changes in morphogen concentration (their widths depend only on threshold ratios), but the outermost (lowest-threshold) domain is vulnerable to reductions in morphogen production. This explains why heterozygous loss-of-function mutations in morphogen pathways often produce specific, limited patterning defects rather than global failure. In the Shh system, heterozygous Shh mutations in humans cause holoprosencephaly (failure of forebrain cleavage, which requires the highest Shh concentration), while more posterior neural tube domains (requiring lower Shh) are preserved. The same logic applies to digit patterning in the limb: reducing Shh signal preferentially eliminates the most posterior (highest-Shh) digits.

Exercises Intermediate+

Exercise 1 (easy, multiple choice).

In the developing limb bud, which signalling centre controls anterior-posterior patterning (digit identity)?

A. Apical ectodermal ridge (AER) B. Zone of polarizing activity (ZPA) C. Dorsal ectoderm D. Progress zone

Hint

The ZPA sits at the posterior margin and secretes a morphogen that forms a gradient across the limb. Which digits require the highest concentration?

Answer

B. Zone of polarizing activity (ZPA). The ZPA secretes Sonic hedgehog (Shh), which forms a gradient from posterior (high) to anterior (low). High Shh specifies posterior digits (digits 4 and 5); low Shh specifies anterior digits (digits 1 and 2). The AER (A) controls proximal-distal outgrowth via FGF signals. The dorsal ectoderm (C) controls dorsal-ventral patterning via Wnt7a. The progress zone (D) is the region of proliferating mesenchyme maintained by the AER.

Exercise 2 (easy, multiple choice).

During kidney development, the ureteric bud is induced to grow and branch by which signal from the metanephric mesenchyme?

A. Sonic hedgehog (Shh) B. Bone morphogenetic protein (BMP) C. Glial cell line-Derived Neurotrophic Factor (GDNF) D. Fibroblast growth factor (FGF)

Hint

The ureteric bud expresses a receptor tyrosine kinase called Ret. What ligand activates Ret to drive branching?

Answer

C. GDNF. The metanephric mesenchyme secretes GDNF, which activates the Ret receptor on the ureteric bud epithelium. Ret signalling drives ureteric bud outgrowth and branching. Knockout of either GDNF or Ret in mice results in renal agenesis (complete absence of kidneys). The ureteric bud reciprocally induces the mesenchyme to form nephrons via Wnt signals. This reciprocal induction is one of the best-characterized examples of an inductive tissue interaction.

Exercise 3 (medium, short answer).

Describe the reciprocal induction between the ureteric bud and the metanephric mesenchyme during kidney development. What happens to each tissue if the other is removed?

Hint

This is a two-way interaction: each tissue signals to the other. What does the mesenchyme need from the bud to form nephrons, and what does the bud need from the mesenchyme to branch?

Answer

The ureteric bud and metanephric mesenchyme engage in a reciprocal (two-way) induction:

Mesenchyme induces the bud: The metanephric mesenchyme secretes GDNF, which activates Ret on the ureteric bud, driving its outgrowth from the Wolffian duct and its subsequent branching. If the mesenchyme is removed, the ureteric bud fails to branch and regresses.
Bud induces the mesenchyme: The branching ureteric bud secretes Wnt signals (particularly Wnt9b and Wnt4) that induce the adjacent mesenchymal cells to condense and undergo mesenchymal-to-epithelial transition (MET), forming renal vesicles that develop into nephrons. If the ureteric bud is removed, the metanephric mesenchyme fails to form nephrons and undergoes apoptosis (it "dies" without the inductive signal).

This reciprocal dependence means that neither tissue can develop independently -- kidney formation is an emergent property of the interaction between the two.

Exercise 4 (medium, short answer).

Explain how the dorsal-ventral axis of the neural tube is patterned by opposing morphogen gradients. What neuronal types are specified at each extreme?

Hint

Two signals come from opposite sides of the neural tube. One source is ventral (floor plate) and the other is dorsal (roof plate). What cell types does each produce?

Answer

The neural tube is patterned dorsoventrally by two opposing morphogen gradients:

Ventral signal: Sonic hedgehog (Shh) is produced by the notochord and then by the floor plate (the ventralmost cells of the neural tube). Shh concentration is highest ventrally and declines dorsally. High Shh specifies motor neurons (in the pMN domain) and ventral interneurons (V0, V1, V2, V3).
Dorsal signal: BMP (BMP4, BMP7) and Wnt are produced by the roof plate (the dorsalmost cells). BMP concentration is highest dorsally and declines ventrally. High BMP specifies dorsal interneurons (D1--D6) and neural crest cells at the dorsal margin.

The intersection of these two gradients creates a smooth transition from ventral (motor) to dorsal (sensory) neuronal subtypes. At each position along the DV axis, the ratio of Shh to BMP signalling determines the progenitor domain identity. This dual-gradient system is remarkably robust: removing one gradient causes partial respecification but does not completely eliminate patterning, because the remaining gradient still provides positional information.

Exercise 5 (hard, short answer).

A mouse embryo carries a conditional knockout of the gene encoding the Shh receptor Smoothened (Smo) specifically in the neural tube. Predict the effect on ventral neural tube patterning. Would any motor neurons form? Explain.

Hint

If the neural tube cannot receive Shh signal, what happens to the ventral progenitor domains? Consider whether any Shh-independent mechanisms could compensate.

Answer

Loss of Smoothened in the neural tube eliminates all Shh signal transduction. The predicted effect is the complete loss of ventral progenitor domains and their neuronal derivatives:

No motor neurons form. The pMN domain requires the highest Shh concentration and is the first to be lost.
No V3 interneurons (p3 domain). The next-highest Shh threshold is also eliminated.
V0, V1, V2 interneurons are also lost or severely reduced, as even low Shh signalling is required for ventral identity.
The ventral neural tube adopts a dorsalized phenotype: dorsal interneuron markers (Pax7, Pax6) expand into the ventral region, and the floor plate fails to form.

No motor neurons would form, because motor neuron specification requires direct Shh signalling to activate the transcription factor Olig2 (pMN domain) and subsequently Hb9 (Mnx1) for motor neuron differentiation. There is no known Shh-independent pathway for motor neuron generation in the spinal cord.

This phenotype is essentially identical to the Shh null mutant, confirming that Smoothened is the essential transducer of Shh in neural tube patterning. Interestingly, in the Shh null mouse, some ventral interneurons (V0, V1) still form, suggesting a Shh-independent contribution from earlier notochord-derived signals or from retinoic acid.

Exercise 6 (hard, symbolic).

In the limb bud, Shh is produced at the ZPA (position $x = 0$ , posterior margin) and diffuses anteriorly with decay length $λ$ . Digit identity is specified by the total temporal exposure to Shh, not just the instantaneous concentration. If a cell at position $x$ is exposed to Shh for time $t$ (where $t$ increases as the limb bud grows), the cumulative dose is:

D (x, t) = \int_{0}^{t} [S] (x, τ) d τ

Assuming $[S] (x, τ) = S_{0} e^{- x / λ}$ is approximately constant during the exposure window (the limb bud does not change shape significantly during digit specification), the cumulative dose is $D (x, t) = S_{0} t e^{- x / λ}$ .

Digit 2 is specified at cumulative dose threshold $D_{2}$ , digit 3 at $D_{3}$ , and digit 4 at $D_{4}$ , where $D_{4} > D_{3} > D_{2}$ .

(a) Derive the spatial positions $x_{2}, x_{3}, x_{4}$ of the digit boundaries as functions of $S_{0}, t, λ, D_{2}, D_{3}, D_{4}$ .

(b) A mutation causes the ZPA to produce half the normal Shh ( $S_{0} \to S_{0} /2$ ) but the limb bud grows for twice as long ( $t \to 2 t$ ) to compensate. Do the digit boundaries shift? Explain.

Hint

For (a), set $D (x_{i}, t) = D_{i}$ and solve for $x_{i}$ . For (b), check whether the product $S_{0} \cdot t$ changes.

Answer

(a) Setting $D (x_{i}, t) = D_{i}$ :

S_{0} t e^{- x_{i} / λ} = D_{i} ⟹ e^{- x_{i} / λ} = \frac{D _{i}}{S _{0} t} ⟹ x_{i} = λ ln \frac{S _{0} t}{D _{i}}

The digit boundaries are at:

x_{2} = λ ln \frac{S _{0} t}{D _{2}}, x_{3} = λ ln \frac{S _{0} t}{D _{3}}, x_{4} = λ ln \frac{S _{0} t}{D _{4}}

The inter-digit spacing is:

x_{2} - x_{3} = λ ln \frac{D _{3}}{D _{2}}, x_{3} - x_{4} = λ ln \frac{D _{4}}{D _{3}}

These spacings depend only on the threshold ratios, not on $S_{0}$ or $t$ individually.

(b) The cumulative dose depends on the product $S_{0} \cdot t$ . Halving $S_{0}$ and doubling $t$ leaves $S_{0} \cdot t$ unchanged:

D^{'} (x, 2 t) = \frac{S _{0}}{2} \cdot 2 t \cdot e^{- x / λ} = S_{0} t e^{- x / λ} = D (x, t)

The digit boundaries do not shift. The temporal compensation exactly cancels the reduction in morphogen concentration. This result illustrates a general principle: in the temporal exposure model of digit specification, it is the time-integrated dose that matters, not the instantaneous concentration. This provides robustness: a weaker signal sustained over a longer period produces the same pattern as a stronger signal over a shorter period. Experimental evidence supports this: in the chick limb, experimentally reducing Shh levels but extending the exposure time produces normal digit patterns.

Exercise 7 (medium, short answer).

Cardiac looping transforms a linear heart tube into a curved tube with the future spatial arrangement of the chambers. What determines the direction of looping, and what are the consequences if the direction is reversed?

Hint

Looping direction is controlled by the left-right axis, which is established during gastrulation. What happens if the LR axis is disrupted?

Answer

The direction of cardiac looping is determined by the left-right (LR) axis, which is established during gastrulation by nodal flow and asymmetric Nodal/Pitx2 signalling (see 18.11.02 pending). Left-sided Pitx2 expression drives dextral (rightward) looping -- the normal D-loop. The result is that the future right ventricle is positioned to the right of the future left ventricle, and the aortic arch is correctly aligned.

If the LR axis is reversed (situs inversus) or randomized (heterotaxy), cardiac looping can proceed in the opposite direction (L-loop). An L-loop places the morphological right ventricle to the left of the morphological left ventricle, producing ventricular inversion. This is often accompanied by transposition of the great arteries (TGA), in which the aorta arises from the right ventricle and the pulmonary artery from the left ventricle -- a lethal condition if untreated, because the systemic and pulmonary circulations are in parallel rather than in series. Heterotaxy patients with randomized situs have a high incidence of complex congenital heart disease (up to 50--90%, depending on the specific laterality defect).

Exercise 8 (medium, short answer).

Describe the signals that maintain Lgr5+ intestinal stem cells in the crypt. What happens to the intestinal epithelium if Wnt signalling is abolished?

Hint

Lgr5 is a Wnt target gene and a marker of active intestinal stem cells. What does this tell you about the role of Wnt? What tissue provides the Wnt signal?

Answer

Lgr5+ intestinal stem cells are maintained by a Wnt-high, BMP-low environment in the crypt base:

Wnt from Paneth cells and the underlying mesenchymal cells activates beta-catenin signalling in the stem cells, maintaining their proliferative capacity and stem cell identity. Lgr5 itself is a direct Wnt target and encodes a receptor for R-spondin (a potent Wnt potentiator). R-spondin binding to Lgr5 enhances Wnt signalling by preventing the degradation of Wnt receptors.
Notch signalling from neighbouring cells prevents premature secretory differentiation, keeping stem cells in the absorptive progenitor pool.
BMP antagonism: Noggin from the crypt mesenchyme blocks BMP signals, which would otherwise promote differentiation. This creates a BMP-low zone in the crypt.

If Wnt signalling is abolished (e.g., conditional knockout of beta-catenin or TCF4 in the intestinal epithelium), the consequences are severe and rapid:

Lgr5+ stem cells are lost within 1--2 days.
Crypt structures degenerate: no new cells are produced.
The villi (finger-like projections) cannot be replenished and the epithelium atrophies.
The mouse dies within days due to intestinal failure.

Conversely, constitutive activation of Wnt signalling (e.g., APC mutation, which prevents beta-catenin degradation) drives hyperproliferation and is the initiating event in most colorectal cancers (familial adenomatous polyposis).

Advanced treatment Master

Congenital malformations of organogenesis

Disruption of specific organogenesis pathways produces characteristic birth defects:

Neural tube defects. Failure of neural tube closure affects ~1 in 1,000 pregnancies worldwide.

Spina bifida (incomplete posterior neural tube closure): ranges from spina bifida occulta (hidden, asymptomatic, a gap in the vertebral arch) to spina bifida aperta (myelomeningocele, the meninges and spinal cord protrude through the defect, causing paralysis, bladder/bowel dysfunction). Folate supplementation reduces incidence by ~70%, but a significant fraction is folate-resistant, implicating genetic factors (MTHFR polymorphisms, Planar Cell Polarity genes).
Anencephaly (failure of anterior neural tube closure): the cranial neural tube remains open, and the forebrain degenerates. Incompatible with prolonged survival. More common in female fetuses.
Craniorachischisis: the entire neural tube fails to close (the most severe NTD), caused by mutations in PCP pathway genes (Vangl2, Celsr1, Scrbi).

Cardiac malformations. Congenital heart disease (CHD) affects ~1 in 100 live births.

Tetralogy of Fallot: the most common cyanotic CHD, comprising four features: (1) pulmonary stenosis (narrowing of the outflow to the lungs), (2) ventricular septal defect (VSD, a hole between the ventricles), (3) overriding aorta (the aorta is positioned over the VSD rather than over the left ventricle), and (4) right ventricular hypertrophy (thickening of the right ventricle due to increased workload). The embryological basis is malalignment of the conal septum during outflow tract septation, often caused by neural crest migration defects. Associated with chromosome 22q11 deletion (DiGeorge syndrome), which affects the TBX1 gene required for pharyngeal arch and cardiac outflow tract development.
Atrial septal defects (ASD) and ventricular septal defects (VSD): incomplete septation of the atria or ventricles. VSD is the most common CHD (~30% of all CHD). Most small VSDs close spontaneously.
Transposition of the great arteries (TGA): the aorta arises from the right ventricle and the pulmonary artery from the left ventricle, producing parallel (rather than serial) systemic and pulmonary circulations. Requires immediate prostaglandin E1 to maintain ductal patency until surgical correction (arterial switch operation).

Limb malformations.

Polydactyly (extra digits): caused by ectopic Shh expression in the anterior limb bud (preaxial polydactyly) or expansion of the Shh domain (postaxial polydactyly). In mice, loss of the Shh inhibitor Gremlin causes polydactyly via unregulated Shh expansion. In humans, mutations in the ZRS (Zone of Polarizing Activity Regulatory Sequence -- a long-range Shh enhancer in intron 5 of Lmbr1) cause preaxial polydactyly by driving ectopic Shh expression in the anterior limb bud.
Syndactyly (fused digits): failure of interdigital apoptosis, a programmed cell death process that sculpts the digits. BMP signalling in the interdigital mesenchyme triggers apoptosis; loss of BMP signalling (e.g., Bmp7 knockout in mice) produces syndactyly. Apoptosis is mediated by BMP-induced expression of the pro-apoptotic factors Bax and caspase activation.
Limb reduction defects: failure of limb bud initiation (tetra-amelia, loss of Wnt3 or Fgf10) or premature loss of the AER (oligodactyly, loss of Fgf8 or Gremlin).

Thalidomide: mechanism of teratogenicity

Thalidomide, prescribed as a sedative and anti-emetic in the late 1950s and early 1960s, caused an estimated 10,000--20,000 cases of severe birth defects, primarily limb reduction defects (phocomelia -- "seal limbs," with hands/feet attached close to the trunk) and defects of the ears, eyes, heart, and kidneys. The teratogenic window was narrow: days 20--36 post-fertilization in humans, corresponding to the period of limb bud outgrowth and AER function.

The molecular mechanism, elucidated over 50 years later, involves:

Anti-angiogenic activity: Thalidomide and its metabolites inhibit FGF2 and VEGF signalling in the developing limb bud vasculature. The limb bud requires rapidly expanding blood vessels for nutrient delivery and paracrine signals from endothelial cells. Thalidomide-induced vessel destruction in the limb bud mesenchyme causes cell death in the progress zone, truncating limb outgrowth.
Cereblon (CRBN) binding: Thalidomide binds to Cereblon, a substrate receptor of the CRL4 E3 ubiquitin ligase complex. This binding redirects the ligase to ubiquitinate specific protein targets (including SALL4, a transcription factor essential for limb development), targeting them for proteasomal degradation. Loss of SALL4 causes Duane-radial ray syndrome (a congenital limb malformation syndrome), directly linking the thalidomide-CRBN mechanism to limb defects. This was identified by Handa,Tokunaga, and colleagues (Ito et al., 2010; Matyskiela et al., 2018).
Species specificity: thalidomide is teratogenic in humans, rabbits, and primates but NOT in mice or rats -- a discrepancy that contributed to its initial acceptance as "safe" based on rodent testing. The species difference arises from sequence differences in the Cereblon binding pocket that affect thalidomide binding affinity.

Thalidomide has been reintroduced for multiple myeloma, leprosy, and other conditions, but its use is strictly controlled under risk evaluation and mitigation strategies (REMS), requiring pregnancy prevention programmes.

Organoid models

Organoids are three-dimensional, self-organizing structures derived from pluripotent stem cells (embryonic stem cells or iPSCs) or adult stem cells that recapitulate key features of organ architecture and function in vitro:

Brain organoids. Lancaster et al. (2013) generated cerebral organoids from human iPSCs by embedding stem cell aggregates in Matrigel and culturing in spinning bioreactors. These organoids develop discrete brain regions (forebrain, midbrain, hindbrain, choroid plexus) with layered cortical structures, radial glia, and functional neurons. Brain organoids have been used to model microcephaly (reduced brain size, caused by premature neuronal differentiation depleting the progenitor pool) and lissencephaly (smooth brain, caused by defects in neuronal migration).

Kidney organoids. Morizane et al. (2015) and Takasato et al. (2015) simultaneously reported protocols for generating kidney organoids from human iPSCs via intermediate mesoderm induction (using CHIR99021 for Wnt activation, followed by FGF9). The resulting organoids contain nephron-like structures (glomeruli with podocytes, proximal tubules, loops of Henle) and collecting ducts, organized in a pattern resembling the developing kidney. Kidney organoids are being used to model polycystic kidney disease, drug nephrotoxicity, and to test strategies for kidney regeneration.

Intestinal organoids. Sato et al. (2009) demonstrated that single Lgr5+ intestinal stem cells, embedded in Matrigel and cultured with EGF, Noggin, and R-spondin (ENR medium), can generate organoids that recapitulate the crypt-villus architecture with all intestinal cell types (enterocytes, goblet cells, Paneth cells, enteroendocrine cells). Intestinal organoids are now used clinically: patient-derived intestinal organoids are expanded in vitro and transplanted to repair damaged intestinal epithelium (the first organoid-based therapy to enter clinical trials).

Limitations of organoid models. Current organoids lack vascularization (limiting growth to ~1 mm diameter without perfusion), immune cells, and the full complexity of in vivo tissue interactions (neural innervation, hormonal inputs). They also exhibit batch-to-batch variability and incomplete maturation (fetal-like gene expression profiles). Strategies to overcome these limitations include co-culture with endothelial cells, transplantation into host animals for vascularization, and microfluidic "organ-on-chip" systems that provide mechanical cues.

Fetal surgery and regenerative medicine

Fetal surgery exploits the remarkable regenerative capacity of the fetus, which can heal without scarring due to a distinct wound healing programme (high hyaluronic acid, low TGF-beta1, high TGF-beta3). Indications include:

Myelomeningocele (spina bifida) repair: the MOMS trial (Adzick et al., 2011) demonstrated that prenatal surgical closure of the spinal defect before 26 weeks' gestation significantly improved outcomes compared to postnatal repair: reduced need for ventriculoperitoneal shunting (36% vs. 82%), improved motor function, and improved cognitive scores. The mechanism involves preventing ongoing spinal cord damage from amniotic fluid exposure and mechanical trauma during gestation.
Twin-twin transfusion syndrome (TTTS): laser photocoagulation of placental anastomoses in monochorionic twins, performed fetoscopically at 16--26 weeks.
Congenital diaphragmatic hernia: fetoscopic endoluminal tracheal occlusion (FETO) to promote lung growth by preventing egress of fetal lung fluid.
Lower urinary tract obstruction: vesicoamniotic shunting to decompress the bladder and prevent renal dysplasia.

Regenerative medicine approaches to organogenesis-related conditions include:

iPSC-derived cell therapies: retinal pigment epithelium (RPE) for age-related macular degeneration (in clinical trials), dopaminergic neurons for Parkinson's disease (in clinical trials), and cardiomyocytes for heart failure.
Decellularized organ scaffolds: whole organs (heart, liver, lung, kidney) are stripped of cells using detergents, leaving the extracellular matrix scaffold, which is then re-seeded with patient-derived stem cells. The goal is to generate patient-specific transplantable organs.
Bioartificial organs: combinations of biomaterial scaffolds, stem cells, and bioreactors to build functional tissue constructs. The most advanced is the bioartificial bladder (Atala et al., 2006), which has been implanted in patients with neurogenic bladder.

Somitogenesis: the clock-and-wavefront model

The segmentation clock is a molecular oscillator that drives the periodic formation of somites from the presomitic mesoderm (PSM). The clock-and-wavefront model (Cooke and Zeeman, 1976) proposes that:

The clock: a gene expression oscillator in the PSM, with a period of ~90 minutes (chick), ~120 minutes (mouse), or ~5 hours (human). The oscillator is driven by negative feedback loops in the Notch, Wnt, and FGF signalling pathways:
- Notch oscillator: Hes/Her (Hairy/Enhancer of split-related) transcription factors repress their own transcription. Hes7 mRNA is produced, translated into Hes7 protein, which enters the nucleus and represses Hes7 transcription. When Hes7 protein degrades (half-life ~20 minutes), transcription resumes. This produces oscillating Hes7 expression.
- Wnt oscillator: Axin2, Dkk1 oscillate in anti-phase to Notch targets.
- FGF oscillator: Lfng (Lunatic fringe) oscillates, modulating Notch receptor sensitivity.
The wavefront: a gradient of FGF8 (high posterior, low anterior) in the PSM sets the determination front. Cells posterior to the front remain oscillating and unspecified; cells anterior to the front (where FGF8 drops below a threshold) stop oscillating and are determined as somite precursors. The wavefront sweeps posteriorly as the embryo elongates.
Somite boundary formation: when the wavefront intersects a specific phase of the clock cycle (e.g., the peak of Hes7 expression), Mesp2 is activated, initiating the segment boundary. Mesp2 activates Ripply2, which represses Tbx6, defining the anterior boundary of the new somite.

Mathematical model. The oscillation period $T$ and the wavefront velocity $v$ determine somite length $S$ :

S = v \cdot T

If the wavefront moves at $v = 100 μ m/hr$ and the clock period is $T = 2$ hours, each somite is $200 μ m$ long. Mutations that slow the clock (increase $T$ ) produce fewer, larger somites (e.g., Hes7 null mice have fewer, larger vertebrae). Mutations that accelerate the wavefront (increase $v$ ) produce longer somites.

The segmentation clock is one of the best-characterized biological oscillators and connects directly to the vertebral column: each somite becomes one vertebra plus its associated ribs, muscles, and dermis. Congenital scoliosis (vertebral segmentation defects) can result from mutations in segmentation clock genes (HES7, LFNG, MESP2, RIPPLY2).

Branching morphogenesis

Branching morphogenesis generates the complex tree-like structures of the lung, kidney, mammary gland, salivary gland, and prostate. The kidney ureteric bud and the lung bud undergo stereotyped branching programs that can be described mathematically:

Dichotomous branching model. Each terminal branch tip bifurcates into two daughter branches. After $n$ generations of branching, the number of terminal tips is $2^{n}$ . In the mouse lung, ~13 generations of branching produce ~8,000 terminal branches over 5 days of development (Metzger et al., 2008).

Mathematical models of branching. The tip of a growing branch can be modelled as an interface moving into the mesenchyme, driven by epithelial proliferation (stimulated by FGF10 from the mesenchyme) and shaped by mechanical forces and ECM remodeling. A minimal reaction-diffusion model:

Epithelium at the branch tip expresses FGFR2b, receiving FGF10 signal from the mesenchyme.
FGF10 is produced in a punctate pattern by the mesenchyme, creating "attraction spots" that direct branch outgrowth.
BMP4 at the branch tip acts as a lateral inhibitor: it suppresses branching in nearby regions, ensuring that only the tip branches and lateral regions remain quiescent.
SHH from the epithelium feeds back to restrict FGF10 expression in the mesenchyme, preventing excessive branching.

The interplay of these signals generates the characteristic branching pattern. Computational models (Cellier et al., 2017; Menshykau et al., 2019) reproduce the three observed branching modes: terminal bifurcation, lateral branching, and trifurcation.

Organ size control: the Hippo pathway

The Hippo pathway is the master regulator of organ size, controlling cell proliferation and apoptosis in developing and adult tissues. The core kinase cascade:

MST1/2 (Hippo homologs) phosphorylate and activate LATS1/2 kinases.
LATS1/2 phosphorylate YAP (Yes-associated protein) and TAZ (WWTR1).
Phosphorylated YAP/TAZ are retained in the cytoplasm and degraded.
Unphosphorylated YAP/TAZ translocate to the nucleus, bind TEAD transcription factors, and activate genes promoting proliferation (Cyclin D1, c-Myc) and inhibiting apoptosis (Birc5/Survivin).

When Hippo signalling is active, YAP/TAZ are phosphorylated and excluded from the nucleus, cell proliferation is limited, and organ growth stops. When Hippo is inactive, YAP/TAZ enter the nucleus and drive proliferation.

Loss of Hippo pathway components (e.g., liver-specific knockout of Mst1/2 or Lats1/2, or constitutive nuclear YAP) causes massive organ overgrowth: mouse livers grow to 4--5 times normal size due to uncontrolled hepatocyte proliferation. This establishes that the Hippo pathway is the "brake" on organ size.

Upstream inputs to the Hippo pathway include:

Cell-cell contact: dense monolayers activate Hippo via E-cadherin-mediated adhesion and the Merlin (NF2) tumour suppressor.
Cell polarity: Crumbs, Scribble, and Par complexes regulate LATS activity.
Mechanical cues: extracellular matrix stiffness, cell geometry, and cytoskeletal tension modulate Hippo signalling through Rho GTPase and the stress fiber sensor Amot (Angiomotin).
Growth factor signals: GPCR ligands (LPA, S1P) inhibit LATS, promoting YAP nuclear localization.

The Hippo pathway is a key link between organogenesis and cancer: YAP/TAZ are oncogenes when constitutively activated, and Hippo pathway inactivation (loss of NF2/Merlin, LATS1/2 mutation) contributes to multiple cancer types including hepatocellular carcinoma, mesothelioma, and breast cancer.

Connections Master

Embryology and morphogenesis 18.11.01. Organogenesis extends the morphogenetic cell behaviours (invagination, EMT, convergent extension) introduced in 18.11.01 into specific organ-forming events. The morphogen gradient models (French flag, reaction-diffusion) are applied here to specific systems: Shh in the neural tube and limb, FGF in the limb bud and kidney.
Gastrulation and axis formation 18.11.02 pending. The germ layers (ectoderm, mesoderm, endoderm) and body axes (AP, DV, LR) established during gastrulation provide the scaffolding for organogenesis. The left-right axis (Nodal, Pitx2) determines cardiac looping direction. The DV axis (BMP gradient) patterns the neural tube. The AP axis (Hox genes) specifies regional identity along the body. All organogenesis events occur within the coordinate system established by gastrulation.
Cell signalling 17.05.01. Organogenesis is fundamentally a signalling problem: every inductive interaction involves ligand-receptor binding, signal transduction cascades, and transcriptional responses. The major signalling pathways (Shh, BMP/TGF-beta, Wnt/beta-catenin, FGF, Notch, Hippo) recur in multiple organogenesis events, illustrating the principle of conserved toolkits: the same pathways are redeployed in different contexts (Shh patterns the neural tube, the limb, the lung, and the tooth) with different outcomes depending on the responding cell's competence and positional history.
Gene regulation 17.06.01. The transcription factors that execute organogenesis programmes (Nkx2-5 for the heart, Pax6 for the eye, Hox genes for segment identity, Olig2 for motor neurons) are downstream readouts of signalling pathways. Combinatorial codes of transcription factors specify cell identity: a motor neuron is defined by the combination of Olig2 + Nkx6.1 + Hb9, not by any single factor.
Stem cells and regeneration 18.11.04. The stem cell niches described in this unit (hematopoietic, intestinal, neural) connect directly to the broader treatment of stem cell biology and regenerative medicine. The principles of niche signalling (Wnt for self-renewal, BMP for differentiation, Notch for lineage choice) are shared across all tissue-specific stem cell populations.
Immunology 18.10.01. The hematopoietic stem cell niche in the bone marrow produces all blood cells, including the immune cells (B cells, T cells, NK cells, macrophages, dendritic cells) covered in the immunology units. Thymic organogenesis (from the third pharyngeal pouch endoderm) creates the organ required for T cell maturation.
Cancer biology. Many organogenesis signalling pathways are hijacked in cancer: Wnt (colorectal cancer), Shh (basal cell carcinoma, medulloblastoma), Notch (T-ALL), Hippo/YAP (hepatocellular carcinoma, mesothelioma). The concept of "cancer as development gone wrong" reflects the shared molecular toolkit.
Evolution and natural selection 19.03.01. The conservation of organogenesis pathways across vertebrates (the same Shh-BMP gradient patterns the neural tube in fish, frogs, birds, and mammals) is a powerful example of deep homology. Evolutionary changes in organ morphology (limb loss in snakes, webbed feet in ducks, echolocation in bats) arise primarily from changes in the timing, level, or spatial domain of organogenesis gene expression -- cis-regulatory evolution rather than protein-coding changes.

Historical & philosophical context Master

The study of organogenesis has deep roots in experimental embryology. Wilhelm Roux (1888) killed one of the two cells in a frog embryo at the two-cell stage and observed that the surviving cell developed into a half-embryo, concluding that development is mosaic (each cell has a predetermined fate). Hans Driesch (1892) separated sea urchin blastomeres and found that each developed into a complete (but smaller) larva, concluding that development is regulative (cells can adjust their fate). This mosaic vs. regulative debate was partially resolved by the discovery of inductive interactions: cells are not predetermined but are specified by signals from their neighbours, making development both regulative (responsive to context) and directed (by specific signals).

The concept of induction was established by Spemann (1901), who showed that the lens of the eye forms only when the optic vesicle contacts the overlying ectoderm. This was extended to limb development by Saunders (1948), who identified the AER, and by Tickle et al. (1975), who demonstrated that grafting the ZPA to the anterior limb bud produces mirror-image digit duplications, establishing the concept of a polarizing morphogen. The molecular identity of the ZPA signal was revealed as Shh by Riddle et al. (1993), who showed that Shh is expressed specifically in the ZPA and that ectopic Shh expression in the anterior limb bud produces the same mirror-image duplication.

The Shh gradient in the neural tube was characterized by Jessell and colleagues (Yamada et al., 1991; Roelink et al., 1994; Briscoe et al., 2000), who performed the key experiment of culturing neural tube explants at different Shh concentrations and demonstrating the concentration-dependent specification of ventral neuronal subtypes. This was one of the most rigorous demonstrations of a morphogen acting as a graded signal in vertebrate development and provided a template for analysing morphogen action in other systems.

The concept of the stem cell niche emerged from the work of Schofield (1978), who proposed that hematopoietic stem cells are maintained by a specialized microenvironment rather than being intrinsically self-renewing. This was dramatically confirmed by the identification of the intestinal stem cell niche (Barker et al., 2007, identifying Lgr5+ crypt base stem cells; Sato et al., 2009, demonstrating single-cell organoid formation) and the hematopoietic niche (Calvi et al., 2003, identifying osteoblasts as niche cells; Ding et al., 2012, identifying the perivascular niche). The niche concept has transformed regenerative medicine: rather than transplanting stem cells alone, the field now focuses on providing the correct niche signals to support engraftment and function.

The thalidomide tragedy (1957--1962) had a profound impact on drug regulation and developmental biology. It led directly to the establishment of the Kefauver-Harris Amendment (1962) to the US Food, Drug, and Cosmetic Act, requiring proof of efficacy and safety before drug approval, and the creation of the modern clinical trial system. For developmental biology, thalidomide provided a powerful tool for studying limb development: the drug's specific effect on limb bud vasculature and the discovery of its target (Cereblon, half a century later) illustrated how teratogens can reveal developmental mechanisms. The story also serves as a cautionary tale about species differences in drug response and the limitations of rodent safety testing for human teratogenicity.

Organoid technology represents a convergence of stem cell biology, developmental biology, and tissue engineering. The first organoid (intestinal, from Sato et al., 2009) was made possible by the identification of the minimal niche signals (EGF, Noggin, R-spondin) required to maintain intestinal stem cells in vitro. This "minimal niche" approach has been extended to brain, kidney, liver, lung, pancreas, and retinal organoids. Organoids raise philosophical questions about what constitutes an "organ" -- they are not fully functional organs (they lack vasculature, innervation, immune cells, and systemic connections), yet they self-organize through the same developmental programmes that build organs in vivo. They are best understood as developmental models that recapitulate organogenesis in a simplified, accessible format.

The Hippo pathway, discovered in Drosophila through genetic screens for tissue overgrowth mutants (Harvey et al., 2003; Wu et al., 2003; Dong et al., 2007), illustrates a recurring theme in developmental biology: genes discovered through their role in growth control in flies turn out to be conserved regulators of organ size in mammals and are frequently mutated in human cancer. The pathway connects cellular behaviours (proliferation, apoptosis, mechanosensing) to the emergent property of organ size, a problem that sits at the boundary between developmental biology, systems biology, and physics.

Bibliography Master

Wolpert, L., "Positional information and the spatial pattern of cellular differentiation", J. Theor. Biol. 25 (1969), 1--47.
Saunders, J. W., "The proximo-distal sequence of origin of the parts of the chick wing and the role of the ectoderm", J. Exp. Zool. 108 (1948), 363--403.
Tickle, C., Summerbell, D. & Wolpert, L., "Positional signalling and specification of digits in chick limb morphogenesis", Nature 254 (1975), 199--202.
Riddle, R. D., Johnson, R. L., Laufer, E. & Tabin, C., "Sonic hedgehog mediates the polarizing activity of the ZPA", Cell 75 (1993), 1401--1416.
Roelink, H., Porter, J. A., Chiang, C., Tanabe, Y., Chang, D. T., Beachy, P. A. & Jessell, T. M., "Floor plate and motor neuron induction by vhh-1, a vertebrate homolog of hedgehog expressed by the notochord", Cell 76 (1994), 761--775.
Briscoe, J., Pierani, A., Jessell, T. M. & Ericson, J., "A homeodomain protein code specifies progenitor cell identity and neuronal fate in the ventral neural tube", Cell 101 (2000), 435--447.
Cooke, J. & Zeeman, E. C., "A clock and wavefront model for control of the number of repeated structures during animal morphogenesis", J. Theor. Biol. 58 (1976), 455--476.
Sato, T., Vries, R. G., Snippert, H. J., van de Wetering, M., Barker, N., Stange, D. E., van Es, J. H., Abo, A., Kujala, P., Peters, P. J. & Clevers, H., "Single Lgr5 stem cells build crypt-villus structures in vitro without a mesenchymal niche", Nature 459 (2009), 262--265.
Barker, N., van Es, J. H., Kuipers, J., Kujala, P., van den Born, M., Cozijnsen, M., Haegebarth, A., Korving, J., Begthel, H., Peters, P. J. & Clevers, H., "Identification of stem cells in small intestine and colon by marker gene Lgr5", Nature 449 (2007), 1003--1007.
Lancaster, M. A., Renner, M., Martin, C. A., Wenzel, D., Bicknell, L. S., Hurles, M. E., Homfray, T., Penninger, J. M., Jackson, A. P. & Knoblich, J. A., "Cerebral organoids model human brain development and microcephaly", Nature 501 (2013), 373--379.
Takasato, M., Er, P. X., Chiu, H. S., Maier, B., Baillie, G. J., Ferguson, C., Parton, R. G., Wolvetang, E. J., Roost, M. S., Lopes, S. M. & Little, M. H., "Kidney organoids from human iPS cells contain multiple lineages and model human nephrogenesis", Nature 526 (2015), 564--568.
Ito, T., Ando, H., Suzuki, T., Ogura, T., Hotta, K., Imamura, Y., Yamaguchi, Y. & Handa, H., "Identification of a primary target of thalidomide teratogenicity", Science 327 (2010), 1345--1350.
Matyskiela, M. E., Couto, S., Zheng, X., Lu, G., Hui, J., Stamp, K., Drew, C., Ren, Y., Wang, M., Carpenter, A., Lee, C. W., Clayton, T., Fang, W., Lu, C. C., Riley, M., Abdubek, P., Bhatt, K., Hartke, J., Gascoigne, K., Cathers, B. E. & Chamberlain, P. P., "SALL4 mediates teratogenicity as a thalidomide-dependent cereblon substrate", Nat. Chem. Biol. 14 (2018), 981--987.
Adzick, N. S., Thom, E. A., Spong, C. Y., Brock, J. W., Burrows, P. K., Johnson, M. P., Howell, L. J., Farrell, J. A., Dabrowiak, M. E., Sutton, L. N., Gupta, N., Tulipan, N. B., D'Alton, M. E. & Farmer, D. L., "A randomized trial of prenatal versus postnatal repair of myelomeningocele", N. Engl. J. Med. 364 (2011), 993--1004.
Gilbert, S. F., Developmental Biology, 12th ed. (Sinauer, 2020).
Sadava, D., Hillis, D. M., Heller, H. C. & Berenbaum, M. R., Life: The Science of Biology, 13th ed. (Sinauer, 2023).
Sherwood, L., Human Physiology: From Cells to Systems, 9th ed. (Cengage, 2016).
Harvey, K. F., Pfleger, C. M. & Hariharan, I. K., "The Drosophila Mst ortholog, hippo, restricts growth and cell proliferation and promotes apoptosis", Cell 114 (2003), 457--467.
Dong, J., Feldmann, G., Huang, J., Wu, S., Zhang, N., Comerford, S. A., Gayyed, M. F., Anders, R. A., Maitra, A. & Pan, D., "Elucidation of a universal size-control mechanism in Drosophila and mammals", Cell 130 (2007), 1120--1133.
Schofield, R., "The relationship between the spleen colony-forming cell and the haemopoietic stem cell", Blood Cells 4 (1978), 7--25.
Metzger, R. J., Klein, O. D., Martin, G. R. & Krasnow, M. A., "The branching programme of mouse lung development", Nature 453 (2008), 745--750.
Atala, A., Bauer, S. B., Soker, S., Yoo, J. J. & Retik, A. B., "Tissue-engineered autologous bladders for patients needing cystoplasty", Lancet 367 (2006), 1241--1246.