Developmental Biology: Chapter 9 — Plant development

🔬 Key Illustrated Scientific Figure

Full-Screen Zoom

Click to enlarge on this page

9.1 Why the logic changes #

9.1.1 Walls, fixed cells, open growth #

Plant cells are walled, and they do not migrate. There is no neural crest, no gastrulation by ingression, and no way to build an organ by moving cells into a cleft. Shape changes because the plane of division is oriented and because expansion is anisotropic. Cortical microtubules orient cellulose microfibrils in the wall, and the cell expands more easily at right angles to those fibrils. The cytoskeletal polymers are Cell Biology. The consequence is developmental: morphogenesis is differential growth of cells that stay where they were born.

The body is also not finished in the embryo. Meristems, stem-cell populations at the shoot and root tips and in lateral positions, persist through life and add organs after germination: leaves, side roots, branches, flowers. Embryogenesis builds a seedling axis. Almost everything a botanist recognises as the plant is post-embryonic. Animal embryos, by comparison, lay down most organs before feeding, even when those organs later grow. The flowering-plant body described here is the sporophyte. Alternation of generations is real and is not needed for the patterning below.

Many somatic plant cells can be returned to a meristematic state. That plasticity is the practical face of genomic equivalence in a kingdom that never set a germline aside as early and as irreversibly as a fly does. The signals that push a cell toward root or shoot in experimental regeneration are auxin and cytokinin. How those hormones are perceived — TIR1 and the AUX/IAA proteins for auxin, the two-component receptors for cytokinin — is Plant Physiology. These notes use the hormones as named inputs and do not re-teach the receptors.

9.2 The Arabidopsis embryo #

9.2.1 From an asymmetric zygote to a seedling axis #

The zygote divides asymmetrically. The small apical daughter produces most of the embryo. The large basal daughter produces the suspensor, a file of cells that anchors the embryo and pushes it into the endosperm, and whose uppermost cell, the hypophysis, is later recruited into the embryo as the founder of the quiescent centre and the columella. Stages worth knowing by shape are the octant, the dermatogen stage at which a protoderm is set aside, the globular embryo, the heart stage at which two cotyledons make the embryo bilaterally symmetric, the torpedo stage, and the mature embryo with bent cotyledons.

The seedling axis, from top to bottom, is: shoot apical meristem, between the cotyledons; the cotyledons themselves; the hypocotyl; and the embryonic root, with its root apical meristem and a root cap. That is the apical–basal pattern. Radially, the same embryo has an epidermis from the protoderm, a ground tissue of cortex and endodermis, and a central vascular stele from the procambium. No germ layers, no archenteron. The geometry is a polarised cylinder with a stem-cell population at each pole, not a gastrula.

Polar auxin transport is what orients the cylinder. PIN efflux carriers send auxin directionally. Early flow accumulates auxin in the apical cell. Later the flow reverses its developmental emphasis and an auxin maximum forms at the hypophysis, which is required to specify the root pole. The gnom mutant, defective in a guanine nucleotide exchange factor needed to localise PIN proteins, cannot set that polarity and produces a ball-like seedling without a proper shoot–root axis. The monopteros mutant, defective in the auxin response factor ARF5, lacks the embryonic root and hypocotyl. Both phenotypes are failures of an auxin-response axis, not of a Hox cluster plants do not have. WOX-family homeobox genes, a plant family distinct from animal Hox genes, mark the early apical and basal lineages. WOX2 is apical; WOX8 and WOX9 are basal. They are the embryo’s first transcriptional asymmetry downstream of the division.

9.3 Meristems #

9.3.1 WUSCHEL and CLAVATA #

The shoot apical meristem is a dome. The central zone holds the stem cells. The peripheral zone is where leaf and flower primordia are initiated. The rib zone beneath produces internal stem tissue. Under the stem cells, a small organising centre expresses WUSCHEL (WUS), a homeobox transcription factor. WUS is required to keep the central zone from differentiating. It acts non-cell-autonomously: the protein moves through plasmodesmata into the cells above and promotes stem-cell identity there.

Those stem cells secrete the peptide CLAVATA3 (CLV3). CLV3 is received by the receptor kinase CLAVATA1, with CLAVATA2 and related receptors as partners, and the signal represses WUS. WUS promotes CLV3. The loop is negative feedback. Too much WUS would enlarge the stem-cell pool, which would raise CLV3, which would cut WUS back. Loss of CLV function is a meristem that cannot hear the brake: the dome enlarges, primordia are too many, and the silique is club-shaped, which is what clavata means. Loss of WUS is a meristem that runs out: a few organs form, then the stem cells are gone and the shoot stops. The logic is a niche maintained by a signal from an organiser, which is familiar from animal stem cells. The molecules are not Notch or Wnt. Homology of logic is not homology of genes, a distinction Chapter 10 will generalise.

9.3.2 The quiescent centre and the radial root #

The root apical meristem is organised around the quiescent centre, a small group of slowly dividing cells. The initials around it are the stem cells: they divide to produce, in Arabidopsis, distinct files for columella, lateral root cap, epidermis, cortex and endodermis, and the stele. The quiescent centre keeps the contacting initials undifferentiated. Kill those few central cells and the abutting initials differentiate. The ablation result is the operational definition of this niche, the same logic as an organiser graft, done by subtraction instead of addition.

WOX5 is expressed in the quiescent centre and maintains the distal columella stem cells, repressing their differentiation. It is the root counterpart of WUS. Expressed in each other’s domain, WUS and WOX5 can largely substitute, which is evidence that shoot and root organising centres are variations on one gene family, swapped in address, not two unrelated inventions. The root cap’s columella is also the gravity-sensing tissue. How statoliths are read is Plant Physiology. That the columella exists because WOX5 kept its stem cells dividing is developmental.

Radial pattern is a separate, beautifully short pathway. SHORTROOT (SHR) is transcribed in the stele. SHR protein moves outward exactly one cell layer into the cortex/endodermis initial. There SCARECROW (SCR) is expressed, and SCR protein, with other factors, traps SHR in the nucleus. The pair specifies endodermis and triggers the periclinal division that separates an endodermal file from a cortical file. A shr mutant has a single ground-tissue layer and no endodermis. A scr mutant fails the asymmetric division and the remaining layer has mixed character. One moving transcription factor, one cell of range, two tissues. It is the plant equivalent of an inductive signal with a range of one cell diameter, and it does not require anyone to migrate.

An auxin maximum sits at the quiescent centre because PIN proteins reflux auxin back toward it. PLETHORA transcription factors are induced by auxin and form a graded distribution, high at the tip. High PLETHORA maintains stem cells, intermediate levels permit transit amplification, and low levels permit differentiation. The gradient is dose-sensitive in the French-flag sense, with the caveats that the cells are walled, that PLETHORA also feeds back to maintain the PIN pattern, and that this is not a syncytial transcription factor diffusing past naked nuclei. It is the right comparison to draw, and the right comparison not to overdraw.

9.3.3 Phyllotaxis without a receptor lecture #

Leaves and flowers are not placed at random on the flank of the meristem. A new primordium forms at an auxin maximum in the peripheral zone. PIN1 in the epidermis polarises toward the incipient primordium, which drains auxin from the nearby flank, so the next maximum cannot form next door. It forms where the remaining auxin is highest, at a characteristic distance. In spiral phyllotaxis that geometry often yields a divergence angle of about 137.5°, the golden angle, as an outcome of packing, not as a gene that encodes the number. Distichous shoots, with leaves opposite in two ranks, and decussate shoots, with successive pairs at right angles, are the same auxin-spacing mechanism on a different meristem shape. The TIR1 receptor and the transcriptional auxin response are Plant Physiology. The patterning statement — maxima, lateral inhibition by depletion, a repeating angle — is the developmental one, and it is juxtacrine-plus-transport rather than Notch, aimed at a hormone instead of a membrane ligand.

9.4 Floral organ identity #

9.4.1 The ABCDE model #

A flower is a short shoot whose primordia have been given floral identities. Four whorls, in Arabidopsis and in the classical model, are sepals, petals, stamens, and carpels, from outside in. Floral meristem identity comes first: LEAFY and APETALA1 commit an axillary meristem to being a flower rather than a leafy shoot. Organ identity then comes from combinations of transcription factors, almost all of them MADS-box proteins. Coen and Meyerowitz stated the ABC model from homeotic mutants in Arabidopsis and Antirrhinum. The modern form adds D and E.

A and C mutually repress. B does not repress A or C; it modifies them. E is required in every floral whorl because the proteins work as quartets, tetramers bound at pairs of DNA sites, and an E-class protein is a partner in each quartet. Without E, the A, B, and C proteins do not specify floral organs, and the whorls become leaf-like. That is the sepallata quadruple mutant, in which SEP1, SEP2, SEP3, and SEP4 are all gone. E is the reason there is no such thing as “B alone”.

The genes: A function is APETALA1 and APETALA2. AP2 is not a MADS protein; it is repressed in the inner whorls by the microRNA miR172, and it represses the C gene AGAMOUS in the outer whorls. B function is APETALA3 and PISTILLATA, which must heterodimerise. C function is AGAMOUS. D function, ovule identity, is SEEDSTICK together with the SHATTERPROOF genes. E function is the four SEPALLATA genes. AGAMOUS has a second job that ties this chapter to the meristem: it represses WUS in the floral centre, so the flower stops after four whorls. Lose AG and C identity is lost, A activity spreads inward, and WUS stays on. The flower makes sepals, petals, petals, and then another flower’s worth of organs, indefinitely.

A function is the combination later work revised most. Sepals are close to a ground state of floral organs when B and C are off, and AP1 is as much a meristem-identity gene as an organ-identity gene. Strong ap2 mutants do still convert outer whorls toward carpels and stamens, so the classical table is the right one to predict from. It is not a claim that A is as clean a molecular class as B or C. D function should not be drawn as an extra whorl outside the carpels. Ovules are organs inside whorl 4. Lose D function and ovules become carpel-like or leaf-like structures still sitting inside that whorl.

Florigen, the FT protein that moves from leaf to meristem and promotes flowering, and the photoperiod and gibberellin pathways that regulate it, are Plant Physiology. So is the detailed auxin transcription machinery that phyllotaxis uses. The developmental result is the combination table. A flower mutant with sepals, sepals, carpels, carpels has lost B. A flower that never stops, and has no stamens or carpels, has lost C. Those two predictions are the point of the model.

9.5 Key terms, summary, and self-check #

Key terms. Meristem; suspensor; hypophysis; PIN protein; WUSCHEL; CLAVATA3; quiescent centre; WOX5; SHORTROOT; SCARECROW; PLETHORA; ABCDE model; MADS-box; floral quartet.

Summary. Plant cells do not migrate; division plane and oriented expansion make the shape, and meristems add organs all through life. The Arabidopsis seedling is a polarised axis — shoot meristem, cotyledons, hypocotyl, root — oriented by PIN-dependent auxin flow, and gnom and monopteros break that axis. WUSCHEL maintains shoot stem cells and is repressed by the CLAVATA3 peptide they secrete. WOX5 in the root quiescent centre maintains columella stem cells, and SHORTROOT moving one layer outward specifies endodermis with SCARECROW. Floral organs are combinations: A+E sepals, A+B+E petals, B+C+E stamens, C+E carpels, C+D+E ovules, with E in every quartet and AGAMOUS also turning the floral meristem off.

Self-check.

1. Why can a plant not build a face by neural-crest migration, and what two cellular behaviours replace migration?

2. What does a gnom seedling look like, and which proteins are no longer correctly placed?

3. If CLAVATA3 is lost, what happens to WUSCHEL transcription and to the size of the meristem?

4. A flower is sepals, sepals, carpels, carpels, and it stops. Which class is missing, and why did it still stop?

5. Why is E required in every whorl, and why are ovules not a fifth whorl on the surface?

—

—

—