Vulval Development: The Battle between Ras and Notch

Vulval development in the nematode Caenorhabditis elegans is a beautifully simple system for asking how cells choose between different possible fates [1]. Vulval fate patterning involves just seven cells: a signaling cell (the anchor cell), which initiates vulval fate induction, and six vulval precursor cells which can respond to that inductive signal by generating vulval descendants (Figure 1). Each vulval precursor cell is capable of adopting any of three different fates: 1°, central vulva; 2°, lateral vulva; or 3°, non-vulval. In wild-type animals, an invariant pattern of fates is seen: 3° 3° 2° 1° 2° 3°. The vulval precursor cell closest to the anchor cell, termed P6.p from its cell lineage position, adopts the 1° fate, the flanking vulval precursor cells, P5.p and P7.p, adopt the 2° fate, and vulval precursor cells farther away from the anchor cell adopt the 3° fate (Figure 1). The precise pattern of vulval precursor cell fates is controlled by a combination of Ras-mediated ‘inductive’ signaling and Notch-mediated ‘lateral’ signaling [1], so vulval development is also a good system for studying the molecular interactions between these two signaling pathways. The inductive signal from the anchor cell is an EGF-like growth factor, LIN-3, which stimulates a canonical receptor tyrosine kinase–Ras–MAP kinase cascade in nearby vulval precursor cells and promotes both the 1° and 2° vulval fates [1]. A lateral signal between neighboring vulval precursor cells stimulates the LIN12/Notch pathway which promotes the 2° fate in the signal-receiving cell [2]. The recent molecular identification of the lateral signal [3] and of various gene targets of lateral signaling [4] have clarified the role of lateral signaling and its relationship to the inductive Ras pathway. There has been a long-running debate about the relationship between inductive and lateral signaling, part of the larger debate about the roles of long-range and short-range signals in cell fate patterning [5]. The original ‘graded signal’ model proposed that the inductive signal is graded and acts like a morphogen: high levels induce the 1° fate in P6.p, and lower levels induce the 2° fate in P5.p and P7.p. [6,7]. But this model cannot be true in its simplest form, because mosaic analysis showed that LET-23, the receptor for the inductive signal, is required only in P6.p for normal vulval patterning [8]. These data led to the ‘sequential signal’ model, where the inductive signal induces the 1° fate in P6.p, which subsequently produces a lateral signal to induce the 2° fate in P5.p and P7.p. It is also possible that the combination of the lateral signal and a graded inductive signal cooperate to ensure ‘a perfect vulva every time’ [5]. Chen and Greenwald [3] have now identified the elusive lateral signal of C. elegans vulval patterning. Genetic evidence indicated that the receptor for the lateral signal is LIN-12/Notch [2,9], so the lateral signal seemed likely to be a member of the Delta/Serrate/LAG2 (DSL) family of Notch ligands. Chen and Greenwald [3] used a computational approach to identify new C. elegans DSL proteins that, unlike known Notch ligands, appear to be secreted proteins. They then used a combination of RNA interference (RNAi) [10] and mutants to systematically test the requirements for each DSL protein using a sensitized genetic background that could detect small defects in lateral signaling. Two known transmembrane proteins, LAG-2 and APX-1, and one newly identified secreted protein, DSL-1, showed phenotypes in this assay, suggesting that all three are components of the lateral signal. The sequential signaling model predicts that Rasmediated inductive signaling should stimulate production of the lateral signal in P6.p. Consistent with this prediction, Chen and Greenwald [3] found that transcriptional reporters for apx-1 and dsl-1 are turned on in P6.p at the time of inductive signaling; lag-2 is expressed in all six vulval precursor cells, but is strongly upregulated in P6.p. These increases in lateral signal gene expression depend on the inductive signal LIN-3 and on SUR-2, a component of the ‘mediator’ complex that is required for the activities of some Ras-dependent transcription factors [11]. Ras-mediated downregulation of LIN-12 in P6.p also appears to be required for production or activity of the lateral signal. Last year, Shaye and Greenwald [12] showed that this is a post-translational effect that requires a specific di-leucine motif in the cytoplasmic domain of LIN-12 and appears to involve increased LIN-12 endocytosis. This downregulation of LIN-12 seems to involve Ras-mediated changes in transcription, as it requires SUR-2. Downregulation of LIN-12 is important for lateral signaling, as expression of a persistent form of LIN-12 in P6.p interferes with 2° fate specification in P5.p and P7.p. Ras-mediated inductive signaling thus appears to do several important things in P6.p: it specifies the 1° fate, stimulates transcription of genes involved in LIN-12 endocytosis, and stimulates transcription of genes encoding the lateral signal (Figure 1). So how do the inductive and lateral signals affect P5.p and P7.p? Yoo et al. [4] recently reported the first Dispatch Current Biology, Vol. 14, R311–R313, April 20, 2004, ©2004 Elsevier Ltd. All rights reserved. DOI 10.1016/j.cub.2004.03.052