Cell Fate, Potency, and Commitment

Potency is the range of cell types a cell (or its descendants) is capable of producing if placed in every possible environment; fate is the range of cell types it actually produces in its normal, unperturbed position in the embryo. A cell's fate is always a subset of its potency, often a much smaller one. The zygote is totipotent: it and its immediate blastomere descendants can generate every cell type of the embryo proper and the extra-embryonic tissues (placenta, yolk sac). As cleavage proceeds, cells become pluripotent — able to form all cell types of the embryo body but no longer the extra-embryonic lineages (this is the defining property of the inner cell mass and of embryonic stem cells derived from it). Later still, cells become multipotent (a hematopoietic stem cell makes many blood cell types but not muscle or neurons), then unipotent (a spermatogonial stem cell makes only sperm), and finally terminally differentiated, expressing a fixed, stable repertoire of tissue-specific genes. Superimposed on this potency ladder is a second axis: how stably a cell is committed to its current path, tested experimentally rather than just inferred from appearance. Specification (sometimes called determination in the loose sense, though careful authors reserve that word for the next stage) means a cell or tissue, if isolated in a neutral culture medium or transplanted to a neutral part of the embryo, will still differentiate autonomously according to its "default" fate — but it can still be redirected by strong enough signals from a new environment. Determination means the cell's commitment has become stable and heritable through mitosis even in a foreign environment that would normally instruct a completely different fate; transplant a determined cell into the eye, and it still becomes muscle. Specification is a reversible hypothesis the cell is currently entertaining; determination is a decision that has been transcriptionally and epigenetically locked in. Crucially, potency and commitment are read out experimentally, never just by watching normal development. Three classic assays distinguish them: 1. Isolation assay — cut the tissue out and culture it alone; does it self-differentiate (specified) or stay uncommitted? 2. Transplantation assay — move the tissue to a different region of a host embryo at the same stage; does it follow its old fate (determined) or adopt the host's local fate (still labile)? 3. Nuclear transplantation / reprogramming assay — take the nucleus out of a differentiated cell and put it in an enucleated egg (or force expression of reprogramming factors); does the genome support totipotent development again? These three assays, applied across a century of embryology, built the whole conceptual edifice this article describes.

Step 1 — Unequal inheritance sets up the first differences. Even before any cell "decides" anything, the egg cytoplasm is not uniform (see Cytoplasmic Determinants and Egg Polarity). Maternal mRNAs and proteins — in Drosophila, bicoid and nanos mRNA; in ascidians, muscle determinant PEM/macho-1-associated cytoplasm; in C. elegans, the PAR proteins and P-granules — are localized to specific cortical or vegetal regions of the egg. Cleavage divisions partition this cytoplasm unequally among blastomeres. A blastomere that inherits a high dose of a particular determinant begins zygotic transcription with a different transcription-factor complement than its sister that did not. This is autonomous specification: cell fate here depends only on what a cell was handed at division, not on communication with neighbors.

Step 2 — Cell–cell signaling refines and overrides the initial pattern. Beyond the first few cleavages, most embryos (vertebrates especially) rely heavily on conditional specification, i.e., induction. A signaling cell secretes a ligand — a Wnt, a BMP or its antagonist Noggin/Chordin, an FGF, a Nodal-family TGF-β — that diffuses or acts by direct contact to a receiving cell's receptor. Ligand binding activates a receptor kinase or a Frizzled/Smoothened-type receptor, triggering an intracellular cascade (Smad phosphorylation for TGF-β/BMP/Nodal; β-catenin stabilization for Wnt; Ras–MAPK for FGF) that ends with a change in the activity of sequence-specific transcription factors already present or newly synthesized in the nucleus.

Step 3 — Transcription factors reorganize the chromatin landscape. The transcription factors switched on by determinants or induction do not act on naked DNA; they act on chromatin. Lineage-specifying factors recruit chromatin remodelers and histone-modifying enzymes: Polycomb repressive complexes deposit H3K27me3 to silence alternative-fate genes; Trithorax-group complexes and active-mark writers (H3K4me3, H3K27ac) keep the chosen program open. DNA methylation by DNMT3A/B can further lock silenced loci. This is the molecular substance of "restriction" — genes for unchosen fates are not deleted, but their promoters and enhancers become progressively less accessible, often bound by repressive complexes that must be actively (and normally are not, under physiological conditions) removed to reactivate them.

Step 4 — Commitment becomes self-sustaining. Many fate decisions are stabilized by transcriptional cross-repression: two lineage-determining factors (e.g., GATA1 versus PU.1 in blood progenitors, or Cdx2 versus Oct4/Nanog in the first mammalian lineage split) each repress the other's gene, creating a bistable switch. Once tipped past a threshold, the cell's own transcriptional network holds it in the new state even if the original inducing signal disappears — this converts a transient instructive signal into a heritable, mitotically stable decision, i.e., converts specification into determination.

Step 5 — Potency is read out, not simply observed. Because chromatin changes are reversible in principle, the only rigorous test of "has this cell's potency actually narrowed, or does it just look that way" is intervention: dissociate/culture, transplant heterotopically, or replace the nucleus. Nuclear transplantation experiments (below) show that even in a terminally differentiated frog intestinal cell, the DNA retains all instructions; what has changed is which of those instructions the egg cytoplasm's reprogramming factors, or the cell's own transcription factors and chromatin marks, will permit to run.

Roux's needle experiment (1888) — the first (misleading) evidence for mosaic development. Wilhelm Roux killed one of the two blastomeres of a frog embryo at the two-cell stage with a hot needle, leaving the dead cell attached. The surviving blastomere developed into a half-embryo — roughly one side of a normal larva, with a corresponding half neural tube. Roux concluded that each blastomere is intrinsically committed at the two-cell stage to make only "its half" of the embryo — an early argument for strict, autonomous mosaic development and full determination by the first cleavage.

Driesch's isolation experiment (1891) — the sea urchin overturns Roux. Hans Driesch separated the first two blastomeres of a sea urchin embryo completely, by vigorously shaking them apart in seawater (rather than leaving a dead cell attached). Each isolated blastomere, freed of any influence from its sister, developed into a complete, if initially small, pluteus larva. This proved each blastomere retained the potency to form the entire embryo — the opposite conclusion from Roux's. The key methodological lesson: Roux's dead cell had remained physically attached and continued to exert an influence (a mechanical/positional constraint) on its living neighbor, artifactually restricting what the survivor did. True potency can only be judged after full isolation. Driesch's result established that sea urchin blastomeres are regulative at this stage — their fate is not yet fixed, only specified provisionally — and it is Driesch's, not Roux's, conclusion that generalizes as the correct account of totipotency at the two-cell stage in most animals examined since.

Spemann's constriction experiments (1901–1903) — specification is stage- and axis-dependent. Hans Spemann used a fine hair loop to constrict newt eggs at the first cleavage, sometimes along the future dorsoventral axis, sometimes perpendicular to it. When the constriction happened to separate the gray crescent cytoplasm (the region marking the future dorsal side) evenly between the two halves, both halves developed into complete, if often twinned, embryos. When the gray crescent was confined entirely to one half, only that half formed a complete embryo; the other half became an unorganized "belly piece" of ventral tissue with no axial structures. This showed that potency for forming a complete body axis is not distributed uniformly in the egg, but depends on inheriting the gray-crescent cytoplasm — direct evidence linking cytoplasmic localization to developmental potency, and setting the stage for Spemann and Hilde Mangold's 1924 organizer transplantation experiments (the classic demonstration of induction, treated in the Fate Map and gastrulation literature).

Briggs and King (1952) and Gurdon (1962) — nuclear transplantation tests whether determination is reversible at the level of the genome itself. Robert Briggs and Thomas King transplanted nuclei from early frog (Rana pipiens) blastula cells into enucleated unfertilized eggs; a substantial fraction developed into normal tadpoles, showing blastula nuclei remain totipotent. When they used nuclei from later gastrula and neurula endoderm cells, success rates dropped sharply, and John Gurdon later showed this decline is not because the DNA has become permanently unable to support development. Gurdon transplanted nuclei from fully differentiated intestinal epithelial cells of feeding Xenopus laevis tadpoles (identified unambiguously by a genetic marker, a single nucleolus mutant, so the resulting tadpoles could be proven to derive from the donor nucleus and not from a residual host nucleus) into enucleated eggs. A minority — roughly a few percent in the original experiments, improved by serial re-transplantation — developed into fertile adult frogs, genetically identical to the nucleus donor. This is direct, rigorous proof that a fully differentiated somatic cell's nucleus retains the entire genetic instruction set needed to build a whole animal: restriction of potency is a change in gene expression and chromatin state, not loss of genetic material. Gurdon shared the 2012 Nobel Prize in Physiology or Medicine with Shinya Yamanaka, whose 2006 discovery that four transcription factors (Oct4, Sox2, Klf4, c-Myc) can reprogram adult fibroblasts into induced pluripotent stem cells extended the same conclusion: potency loss is reversible by the right combination of transcription factors, without any need for an egg's cytoplasm at all. Together these experiments define the historical arc of the concept: Roux's flawed evidence for early mosaic determination, corrected by Driesch's rigorous isolation assay showing regulative totipotency; Spemann's demonstration that potency is nonuniformly distributed by cytoplasmic localization; and Briggs–King/Gurdon/Yamanaka proving that even terminal differentiation is, at the genomic level, a reversible restriction of expression rather than an irreversible loss of information.

Example 1 — Reading a transplantation result to classify specification vs. determination. Suppose at gastrula stage in a salamander embryo, presumptive epidermis is grafted into the region of the neural plate (a heterotopic, same-stage transplant), and the graft develops into neural tissue matching its new surroundings. The same graft performed several hours later, at the neural fold stage, instead develops into epidermis in its new neural location, forming a patch of skin embedded within the neural tube. How do we classify the tissue's state at each stage? At the gastrula stage, the presumptive epidermis followed its new location's fate — this means it was not yet committed; it was still uncommitted or, at most, weakly specified and easily overridden. Since it responded to its environment rather than its origin, it was not determined at gastrula stage. At the neural fold stage, the same tissue followed its original fate despite being placed in a foreign, strongly neuralizing environment — this is the operational definition of determination: fate has become stable and autonomous, independent of the surrounding signals that would otherwise instruct a different path. The experiment also tells us, quantitatively, the timing of the transition: determination for this tissue occurred sometime between gastrulation and neural fold stage, and a series of transplants at intermediate times (a "determination time-course") could narrow that window further, e.g., to a specific number of hours post-fertilization or a specific number of cell cycles.

Example 2 — Estimating clonal potency loss using a chimera/marker experiment. Imagine an experiment analogous to classic mouse chimera studies: cells from an 8-cell-stage embryo (genetically marked with a lineage tracer, e.g. a constitutively expressed fluorescent reporter) are injected into a host blastocyst. You observe the following contributions of marked cells to different tissues in the resulting chimeric pups:

From this table you can directly infer the potency status of each source, without ever having watched its normal fate: the 8-cell blastomere is still totipotent (contributes to both embryonic and extra-embryonic lineages), the inner cell mass cell has already narrowed to pluripotent (embryonic lineages only, placenta lost as an option), and the trophectoderm cell has narrowed in the opposite direction to a much more restricted, extra-embryonic-only potency. Note the asymmetry: potency restriction is not always a simple funnel toward "fewer total fates" measured on one axis — the trophectoderm/inner-cell-mass split is the very first lineage decision in the mammalian embryo, and it partitions totipotency into two mutually exclusive, non-overlapping restricted potencies (Cdx2-high trophectoderm versus Oct4/Nanog-high inner cell mass), each committed by the mutual cross-repression mechanism described above.

John Gurdon began his scientific life inauspiciously: a school report famously judged his ambition to become a scientist "a completely ridiculous idea," ranking him last of 250 boys in his year in biology. As a doctoral student at Oxford under Michael Fischberg, Gurdon took up a technically formidable problem that Briggs and King's American work had left unresolved: could the nucleus of a fully differentiated, feeding-stage cell — not merely an early blastula cell — still direct complete development? Using Xenopus laevis, whose eggs are large and robust enough for repeated micromanipulation, Gurdon devised a rigorous genetic marking system (a mutant strain with only one nucleolus per cell instead of two) so that any tadpole arising from a transplanted nucleus could be traced unambiguously to the donor, ruling out the recurring criticism that residual host nuclear material, not the donor nucleus, was responsible for successful clones. His 1958–1962 experiments transplanting nuclei from intestinal epithelial cells of swimming tadpoles produced a small but real yield of nuclear-transfer animals that developed to adulthood and were fertile — a result achieved for a fully differentiated vertebrate cell type for the first time. The work was initially received with considerable skepticism (some colleagues suspected contamination or misidentified donor tissue), and it took further refinements — serial nuclear transfer, in which nuclei from partially developed clones were re-transplanted into fresh eggs to boost efficiency — to fully convince the field. Fifty years later, his conceptual heir Shinya Yamanaka showed the egg cytoplasm was not even strictly necessary: a defined cocktail of four transcription factors could achieve comparable reprogramming from a cultured fibroblast. The two men, whose experiments were separated by half a century and used entirely different techniques, received the Nobel Prize jointly in 2012 for converging on the same fundamental conclusion: differentiation restricts gene expression, not genetic content.

Regenerative medicine and iPSC technology. Because Gurdon and Yamanaka proved determination is experimentally reversible, patient-derived skin or blood cells can now be reprogrammed into induced pluripotent stem cells and then re-differentiated into cardiomyocytes, dopaminergic neurons, or retinal pigment epithelium for autologous transplantation, disease modeling, and drug screening — an entire clinical pipeline built on the principle that potency loss is chromatin-level, not genome-level.

Reproductive cloning and livestock/conservation biotechnology. Somatic cell nuclear transfer, the direct technical descendant of Briggs–King and Gurdon's method, produced Dolly the sheep in 1996 and is now used to propagate elite livestock genetics and to attempt rescue of critically endangered species (e.g., attempts with the northern white rhinoceros using stored somatic cells and related species as egg donors/surrogates).

Cancer biology. Many cancers can be understood partly as a failure or reversal of normal restriction — dedifferentiation, aberrant reactivation of stem-like transcriptional programs, and epigenetic derepression of silenced loci recur as themes in tumor progression, and Polycomb/DNA-methylation pathways central to normal determination are frequently found mutated or dysregulated in tumors.

Assisted reproduction and preimplantation diagnostics. Understanding exactly when mammalian blastomeres are still totipotent underlies preimplantation genetic testing, where a single cell can be biopsied from an early human embryo for genetic screening without compromising the embryo's development, because at that stage remaining blastomeres retain sufficient potency to regulate and complete a full embryo.

Teratoma and germ-cell tumor pathology. Pluripotent cells that escape normal fate restriction (e.g., germ cells that retain pluripotency inappropriately) can form teratomas containing disorganized derivatives of multiple germ layers — a clinical, pathological demonstration, in reverse, of just how much potency a cell can still express when its normal restrictive signals fail.

Ask the tutor: How would you design a transplantation experiment to pinpoint the exact developmental stage at which a specific tissue in an embryo becomes irreversibly determined?