The Development of the Caenorhabditis elegans Vulva

 Mark Blaxter for Developmental Biology 3,
IEB, University of Edinburgh

---

a model problem of developmental biology involving cell differentiation, cell-cell signalling and morphogen gradients

---

Back to C. elegans Introduction | Research Techniques |

These are the lecture notes.
References

---

PDFs of Overheads from the lecture

The C. elegans vulva	The lineage of C. elegans with Pnp cells highlighted	The Pnp cell lineage
The cells that make up the vulva	Laser ablation of the Pnp cells	The position of the Pnp cells
HOX genes and the Pnp cells	Signalling from the anchor cell 1	Signalling from the anchor cell 2
Signalling from the anchor cell 3	Signalling from the anchor cell 4

---

The Normal Vulva

Wild type hermaphrodite nematode from Kornfeld, K. (1997). Vulval development in Caenorhabditis elegans. Trends in Genetics 13, 55-61. The vulva is in the centre of the body, where the eggs are stacked up in the uterus.

The vulva is a structure on the ventral surface of adult hermaphrodite females of C. elegans. It forms the connection between the gonad and the outside, and comprises

• a link to the gonad, mediated by the anchor cell and central cells of the vulva
• a "tube" formed by vulval cells
• an opening through the cuticle (the nematode exoskeleton)
• a set of vulva-specific muscles which mediate opening of the vulva

The vulva is usually held closed by the positive internal hydrostatic pressure of the nematode.

The nematode moves using longitudinal muscles acting against a hydroskeleton. The external cuticle is a radially inextensible tube and its integrity is essential to the function of the hydroskeleton. The formation of the vulval opening compromises the hydroskeleton, and is strengthened.

The vulva develops in the L3-L4 stages, but the final opening to the outside world is not made until the L4-adult moult.

Making a Normal Vulva

The anchor cell is derived from the same cell lineage as the somatic gonad primordium.

The early embryonic cell divisions of C. elegans give rise rapidly to 558 cells which then undergo differentiation and morphogenesis. The newly hatched L1 has a set of cells, blast cells, which undergo additional post-embryonic divisions.

The vulva is derived from a set of ventral epidermal blast cells called the Pn.p cells.

There are 12 Pn.p cells, (P1.p to P12.p) and each follows a particular developmental pathway during post-embryonic development.

The Pn.a cells divide to give neuronal daughter cells.

From anterior to posterior these cells give rise to:

• P1.p and P2.p fuse with the syncytial hypodermis hyp
• P3.p and P4.p divide once (in the L3 stage) and the two daughters join the syncytial hyp7. This is called the tertiary fate
• P5.p divides in the L3 stage to give rise to 7 progeny which form vulval structures (the vulA, vulB, vulC and vulD cells). Note that the division pattern is asymmetric (with the vulval-proximal daughter P5.pp only having 3 progeny and P5.pa having 4). This is called the secondary fate.
• P6.p divides in the L3 to give eight progeny (four each of vulE and vulF cells). These cells contact the anchor cell and make the tube from the uterus. The divisions of P6.pp and P6.pa are symmetrical. This is called the primary or 1deg. fate
• P7.p divides in the L3 stage to give rise to 7 progeny which form vulval structures (the vulA, vulB, vulC and vulD cells). Note that the division pattern is asymmetric (with the vulval-proximal daughter P7.pa only having 3 progeny and P7.pp having 4 - the opposite orientation to P5.p's daughters). This is also a secondary fate.
• P8.p divides once and the two daughters join the syncytial hyp7 (tertiary fate)
• P9.p, P10.p and P11.p fuse with the syncytial hypodermis hyp7
• P12.p gives rise to two daughters, one of which (P12.pp) undergoes apoptotic cell death. The other cell (P12.pa) forms anal hypodermis, called hyp12

---

How Is this Developmental Pattern Achieved?

Making a non-functional vulva is detrimental to organismal survival. Either the animal will burst at the weak spot of a malformed vulva or the fertilised eggs will not be laid and will hatch out inside the animal giving rise to "a bag of worms". Experiments have shown that the development of the vulva involves cell-cell signalling and instruction. Two methods have been used to follow this. One involves perturbing the normal worm. The other involves the isolation of strains which are genetically defective in making a vulva. These mutants have been analysed at the molecular level and reveal an intracellular signalling cascade.

Laser Ablation

Using a finely tuned laser, directed down the objective of a microscope it is possible to selectively kill cells (by frying their nuclei - called laser ablation) in the developing nematode, leaving neighbours untouched. The procedure is carried out on precursor cells as killing differentiated cells may leave signalling functions intact.

Ablating the Anchor Cell

The anchor cell is closest to P6.p in the developing larva. Killing the AC results in all the cells (P3.p-P8.p) assuming 3deg. fates, and no vulva is made. This phenotype is called Vul (for vulvaless). From this it is deduced that the AC send a vulva-inducing signal to the Pn.p cells.

a Vul nematode from Kornfeld, K. (1997). Vulval development in Caenorhabditis elegans. Trends in Genetics 13, 55-61. The eggs are not laid and the larvae hatch out inside the mother, leading to a "bag of worms" phenotype.

Ablating P6.p

With P6.p dead, one or other of the cells on either side (P5.p and P7.p) is observed to take up the 1deg. fate. In turn, either P4.p or P8.p is recruited to the 2deg. fate and a normal vulva is made. P5.p and P7.p are thus competent to play a primary role as well as a secondary one...

Ablating more than one Pn.p Cell

By a series of single and multiple ablations, it has been shown that all of the cells from P3.p through to P8.p form a vulval equivalence group: each of the cells is competent to play any of the roles in vulval development.

This implies that in normal vulvae there may be a lateral signal, where (usually) P6.p tells its neighbours that it is a primary cell and they should be secondary, and the other cells assume a tertiary default state.

Partial Transformations of Pn.p Cells

Usually when a cell is ablated, its neighbours move to take up the space it leaves. In the case of P6.p ablation, this means that P5.p or P7.p will move to take up a position close to the AC.

Sometimes when a single Pn.p cell (particularly P3.p or P8.p) is left, it fails to migrate fully to a position under the AC. When this occurs, it can take on a 2deg. or even a 3deg. fate instead of a 1deg. one. This suggests that the inductive signal from the AC is graded This signal fits the model of a graded morphogen.

---

Mutations in Genes Controlling Vulval Formation Reveal Inter- and Intra-Cellular Signalling

Defects in genes which control vulval development might be expected to have two phenotypes. Defects in signalling might mimic AC ablation and result in a Vul phenotype. Defects in responses might result in a constitutive "on" signal and result in multiple cells following the 1deg. fate.: a Muv or multivulva phenotype.

To isolate mutants defective in vulval differentiation, nematodes were mutagenised and strains isolated which

• generated the Pn.p cells normally (obviously deletion of the Pn.p cells would delete the vulva, but in an uninteresting way)
• had a Vul (0 vulvae) or Muv (>1 vulvae) phenotype.

The Vul nematodes can be propagated because there are a few eggs which hatch out inside the hermaphrodite, eat their mother and escape when she dies. Obviously there are problems doing genetics with Vul nematodes, because they cannot be mated. The same is sometimes also true of Muv nematodes, as the multiple vulvae may not be functional.

A Homeobox Gene Defines the Vulval Equivalence Group

Mutations at the lin-39 (lin standing for lineage defect) locus affect many cell lineages in the midbody region of C. elegans, including the formation of the Pn.p cells of the vulval equivalence group. lin-39 is one member of a cluster of homeobox genes which is thought to be homologous to the bithorax and Hom-C clusters of Drosophila and mammals. The cluster is arranged such that genes at one end affect anterior structures, while those at the other affect posterior ones, just as in flies and mammals.

A Novel Phenotype: An Inhibitory Signal from hyp7?

aMuv nematode from Kornfeld, K. (1997). Vulval development in Caenorhabditis elegans. Trends in Genetics 13, 55-61.

In the absence of an inductive signal from the AC all the cells have 3deg. fates. Mutations in the locus lin-15 cause Muv phenotypes, where more than one vulval precursor cell takes on the role of vulval formation. Genetic evidence suggests that an inhibitory signal is sent from the hyp7 syncytium surrounding the Pn.p cells to maintain them in the ground state, and that this signal is controlled by the complex lin-15 locus. lin-5 mutants can be divided into two classes which map to the same locus but affect differentially two alternate transcripts.

The inhibitory signal genes can in fact in general be divided into two classes, A and B. A mutation in both classes is required to generate a Muv phenotype. The original mutation in lin-15 which was isolated eliminates both A and B functions of this gene, which are genetically seperable. Class A contains lin-15A, lin-8 and lin-38. Class B includes lin-15B, lin-9, lin-35, lin-36, and lin-37 amongst others. Only lin-15 has been cloned to date, and the protein product has no similarity to other proteins.

The Anchor Cell -P6.p Inductive Signal

Mutants at a locus called lin-3 are Vul. lin-3 has been cloned and encodes a secreted epidermal growth factor-like product. the lin-3 gene is active in the AC, and lin-3 is thus thought to encode the AC signal.

Some mutants at the let-23 (lethal - most alleles are embryonic lethals) locus are also Vul. let-23 encodes a transmembrane receptor tyrosine kinase which is believed to be the receptor for the AC signal. LET23 is expressed on P6.p as well as P5.p and P7.p

Loss of function mutations at these two loci are thus the genetic equivalent of performing AC ablations.

A Gradient of Morphogen?

C. elegans can be transformed with exogenous DNA to make transgenic worms. The lin-3 gene was transformed into C. elegans under the control of a heat-shock inducible promoter derived from a heat shock protein gene. This allowed researchers to induce expression of the lin-3 EGF-like protein and observe the effects of no, a little or a lot of LIN-3 by simply giving no, short or longer heat shocks.

Transgenic nematodes were laser operated to leave only one Pn.p vulval precursor cell (VPC). The endogenous source of LIN-3 was removed by ablating the AC. In unshocked animals, the VPC assumed the 3deg. fate as expected. A long heat shock induced LIN3 expression and induced differentiation of the VPC: the 1deg. fate was followed. A short heat shock only induced 2deg. fates in VPCs.

This data suggests that the VPCs receive and interpret a graded morphogenic signal of LIN-3 from the AC. The P6.p cell will receive the strongest signal (as it is closest to the AC), and the P5.p and P7.p cells less signal.

These experiments are the transgenic equivalent of observations of isolated VPCs failing to migrate to central AC-proximal positions.

However, mosaic analysis of let-23 (receptor) function revealed that let-23 was not required by P5.p or P7.p in order to express 2deg. fates. P5.p cells lacking LET-23 and thus unable to percieve the LIN-3 signal could still adopt the 2deg. fate. The researchers suggest that perhaps the behaviour of isolated cells in ablationn experiments does not reflect the situation in the intact animal.

Genetic Mosaics Reveal Cross Talk Between Vulval Precursor Cells.

It is possible to make C. elegans genetic mosaics. A wild-type allele of a gene is isolated on an unstable chromosomal element (which is also tagged with a cell-level visible marker - a nucleolar size mutation) and maintained in a genetic background which carries a recessive mutation in the gene. This element can be lost at each cell division, resulting in animals with a mosaic of cells which either have wild type or lack wild type function of the gene.

These sorts of analyses were performed on let-23 mutant mosaic animals. It turns out that even if they lack let-23 function, P5.p and P7.p will still take on the 2deg. fate. From this is was deduced that there must be a lateral inductive signal from the P6.p primary cell working in a parallel but redundant pathway to the lin-3 graded morphogen signal.

The receptor for this lateral signal is encoded by the gene lin-12. LIN12 is a large transmembrane protein which has similarity to both Notch of Drosophila and other species, and to glp-1, another C. elegans cell-cell signalling gene.

---

Intracellular Signalling Cascades in Vulval Induction

From the large number of Vul and Muv mutations isolated it has been possible to construct a genetic pathway for vulval induction and differentiation. The lin-3 to let-23 signal is transduced inside the cells by a series of gene products which are homologues of mammalian and fly signal transduction pathway molecules. The following figure is taken from the K. Kornfeld review (Trends in Genetics, 13, 55-61) and summarises what is known (Feb 1997) about the genes involved in the vulval induction pathway.

The growth factor ligand, lin-3, expressed on the anchor cell interacts with the let-23 receptor tyrosine kinase. This activated kinase then interacts with a transducr molecule (an SH2-SH3 molecule; SH stands for src homology) encoded by sem-5. sem-5 induces activity of let-60 (a ras homologue), which in turn induces lin-45 (a raf homologue). lin-45 then activates a Mitogen Activated Protein Kinase signal transduction chain which inhibits the transcription factor encoded by the gene lin-1. The lin-31 gene also encodes a transcription factor of the HNF3/Forkhead family but its integration into the signalling cascade is less clear.

Other genes also interact with this pathway. lin-2 (which encodes a MAGUK [membrane-associated guanylate kinase] homologue) and lin-7 (a PDZ [PSD-95/discs-large/ZO-1] domain containing protein which can physically interact with LET-23) positively regulate let-23 activity. These two genes appear to act in concert with lin-10 (which is not similar to any other proteins). Two genes have been identified which weakly negatively regulate vulval cell fates, unc-101 (a component of clathrin-associated complexes of the trans-golgi) and sli-1 (a homologue of the mammalian oncoprotein CBL).

Three genes appear to act downstream of let-60 ras but are not in the main pathway. ksr-1 is a proitein kinase, and lin-25 and sur-2 encode proteins with no homologues.

Determination of Secondary Fates

The gene lin-12 encodes a receptor of the Notch family, and is believed to recieve an as yet uncharacterised signal from the 1deg. cells. There are a number of genes which act in the hypodermal syncytium, hyp7, and send an inhibitory signal to the signalling cascade. The point(s) in the cascade at which these inhibitory signals act have not been determined.

Mutant Strains (this is from 2001-2002!!!)

References