Report

Notch Inhibition of RAS Signaling Through MAP Kinase Phosphatase LIP-1 During C. elegans Vulval Development

See allHide authors and affiliations

Science  09 Feb 2001:
Vol. 291, Issue 5506, pp. 1055-1058
DOI: 10.1126/science.1055642

Abstract

During Caenorhabditis elegans vulval development, a signal from the anchor cell stimulates the RTK/RAS/MAPK (receptor tyrosine kinase/RAS/mitogen-activated protein kinase) signaling pathway in the closest vulval precursor cell P6.p to induce the primary fate. A lateral signal from P6.p then activates the Notch signaling pathway in the neighboring cells P5.p and P7.p to prevent them from adopting the primary fate and to specify the secondary fate. The MAP kinase phosphatase LIP-1 mediates this lateral inhibition of the primary fate. LIN-12/NOTCH up-regulateslip-1 transcription in P5.p and P7.p where LIP-1 inactivates the MAP kinase to inhibit primary fate specification. LIP-1 thus links the two signaling pathways to generate a pattern.

MAP kinase phosphatases (MKPs) belong to the family of dual-specificity phosphatases that inactivate different types of MAP kinases by dephosphorylating the critical phosphotyrosine and phosphothreonine residues of the kinases (1). The transcription of MKPs is rapidly induced by various stimuli such as growth factors and cellular stresses that activate MAP kinases, suggesting that MKPs may participate in an autoinhibitory feedback loop.

To study the role of MKPs in RTK/RAS/MAPK signaling during development, we searched the C. elegans genome sequence for homologs of vertebrate MKPs. Among the 185 predicted phosphatases, we identified a candidate, termed lip-1 (lateral signal induced phosphatase−1, open reading frame C05B10.1), that is similar to human MKPs (Fig. 1A). LIP-1 exhibits 32% overall sequence identity and 49% similarity to human MKP-3/PYST1.

Figure 1

(A) Structure of LIP-1 with the MAP kinase binding domain (dark box) in the NH2-terminal and the catalytic domain (hatched box) in the COOH-terminal part. A sequence alignment of the catalytic domains of LIP-1, human MKP-1/CL100, MKP-2/hVH-2, MKP-3/PYST1, and PAC-1 is shown below. Identical and similar residues are highlighted with black and gray shading, respectively. The arrow indicates the essential Asp236 residue. The dotted line shows the 3′ breakpoint of the zh15 deletion. (B) Intron-exon structure of the lip-1 gene. The dotted line underneath shows the extent of the zh15 deletion. The numbers indicate the positions of the breakpoints (in bp) relative to the ATG initiation codon. The arrows indicate the three primers used for the PCR analysis shown in the lower panel. (C) Northern blot analysis of RNA isolated from wild-type and lip-1(zh15) animals. As a loading control, the total RNA stained in the gel before blotting is shown below. (D) Relative MAP kinase activity was measured in vitro as described (26).let-60(n1046gf) animals were used as a positive control in one experiment. The black diamonds indicate the results obtained in four independent experiments, and the gray bars are the average of the four experiments. Abbreviations for the amino acid residues are as follows: A, Ala; C, Cys; D, Asp; E, Glu; F, Phe; G, Gly; H, His; I, Ile; K, Lys; L, Leu; M, Met; N, Asn; P, Pro; Q, Gln; R, Arg; S, Ser; T, Thr; V, Val; W, Trp; and Y, Tyr.

Using a polymerase chain reaction (PCR)– based assay to screen a library of ethylmethane sulfonate–mutagenized worms (2), we isolated a deletion in lip-1 (Fig. 1B). The zh15 deletion removes 1416 base pairs (bp) that encode the putative MAP kinase binding domain and part of the catalytic domain of LIP-1 including the conserved Asp236 residue that is required for the proton transfer during the phosphatase reaction (Fig. 1, A and B). Moreover, no lip-1 transcripts could be detected in lip-1(zh15) mutants by Northern blot analysis (Fig. 1C). Thus, the homozygous viable lip-1(zh15) mutation probably represents a loss-of-function (lf) allele. lip-1(lf) animals contained higher levels of MAP kinase enzymatic activity, whereas animals carrying a transgene with a hyperactive form of lip-1([lip-1::nls::gfp], see below) exhibited lower kinase activity when compared with wild type (Fig. 1D). Thus, LIP-1 may negatively regulate MAP kinase activity.

We therefore investigated whether LIP-1 inhibits MAP kinase MPK-1 signaling during vulval induction (3, 4). In lip-1(lf) single mutants, vulval development appeared normal (Table 1, row 2). However, loss oflip-1(+) function suppressed the vulvaless (Vul) phenotype caused by mutations that partially reduce the activity of the RTK/RAS/MAPK signaling pathway, such as mutations in let-23 egfr(5), lin-7 pdz (6),sem-5 grb2 (7), or mpk-1 MAP kinase (Table 1, rows 3 to 10). Furthermore, animals expressingmpk-1(+) under control of the heat-shock promoter ([HS-mpk-1(+)]) exhibited wild-type levels of vulval induction at 20°C (8), whereaslip-1(lf); [HS-mpk-1(+)] double mutants displayed a strong multivulva (Muv) phenotype (Table 1, rows 11 and 12). Thus, loss of lip-1(+) function increases the sensitivity of the vulval precursor cells (VPCs) P3.p through P8.p toward the inductive signal. Consistent with the hypothesis that LIP-1 inhibits MPK-1 signaling, a mutation inlin-25 that acts downstream of or parallel tompk-1 (9) was only weakly suppressed bylip-1(lf) (Table 1, rows 13 and 14).

Table 1

Genetic interactions between lip-1 and genes controlling vulval induction.

View this table:

Overexpression of lip-1(+) under the control of the Pn.p cell-specific lin-31 promoter (10) caused a Vul phenotype (Table 1, row 15). Furthermore, whereas animals expressing a lip-1–green fluorescent protein (GFP) fusion ([lip-1::gfp]) under the control of the lip-1 promoter only rarely exhibited a Vul phenotype, the insertion of a nuclear localization signal ([lip-1::nls::gfp]) caused a penetrant Vul phenotype (Table 1, rows 16 and 17), suggesting that LIP-1 may act more efficiently when translocated into the nucleus. Using this activated (nuclear) form of LIP-1, we determined at what step LIP-1 inhibits the RTK/RAS/MAPK signaling pathway. The[lip-1::nls::gfp]transgene efficiently suppressed the Muv phenotype caused by alet-60 ras(gf) mutation (11) or by overexpression of mpk-1(+), but it failed to suppress thelin-1(rf) Muv phenotype (Table 1, rows 18 to 23). lin-1 encodes an ETS-domain transcription factor that inhibits vulval induction downstream of MPK-1 (12). Inlin-1(rf) mutants, the VPCs adopt vulval fates independently of MPK-1 activity. Thus, LIP-1 inhibits vulval induction upstream of LIN-1, most likely at the level of MPK-1.

To determine the LIP-1 expression pattern, we analyzed twolip-1::gfp reporter lines (Fig. 2A). LIP-1::GFP expression was observed in most somatic cells starting during embryogenesis and persisting throughout larval development and adulthood. In particular, LIP-1::GFP was uniformly expressed at a low level in all VPCs until the L2 stage. In early L3 larvae however, LIP-1::GFP expression was up- regulated in the secondary (2°) VPCs P5.p and P7.p in 71% of[lip-1::nls::gfp]animals (Fig. 2B) (n = 24). To examine if the increase in LIP-1 expression was regulated at the level of transcription, we inserted a 2280-bp genomic fragment from the lip-1promoter/enhancer region upstream of a minimal promoter (Fig. 2A). This transcriptional reporter displayed the up-regulation of LIP-1::GFP in P5.p and P7.p in all cases examined (Fig. 2C) (n = 30), confirming that the induction of LIP-1 in 2° VPCs is regulated at the transcriptional level. At later stages, LIP-1::GFP expression remained high in the descendants of P5.p and P7.p and low in the descen- dants of P6.p (Fig. 2, D and E). In the descendants of the uninduced VPCs (P3.p, P4.p, and P8.p), LIP-1::GFP expression increased after the cells had fused to the hypodermal syncytium hyp7 where LIP-1::GFP was strongly expressed (Fig. 2E).

Figure 2

(A) Structure of the rescuing translational LIP-1::GFP fusionzhEx17[lip-1::nls::gfp](27) and the transcriptional fusion (zhIs4& zhEx23). The black triangles shown in the transcriptional fusion represent the four CSL binding motifs at positions −1439, −1214, −1094, and −913. The minimal Δpes-10 promoter controlling nls::gfpwas fused to lacZ to achieve entirely nuclear GFP localization. (B) LIP-1::GFP expression (translational fusion) in the Pn.p cells of an early L3 larva. The position of the Pn.p cell nuclei is indicated by the arrows; the smaller arrowheads indicate ventral cord neurons that expressed LIP-1::GFP. The transcriptional fusion is shown in (C) through (I). (C) LIP-1::GFP expression in the Pn.p cells of an early L3 larva; (D) in the daughters of the VPCs; and (E) in the granddaughters of P5.p through P7.p and the daughters of P3.p, P4.p, and P8.p after they had fused with hyp7, and (F) in the Pn.p cells oflet-60(n1046gf), (G)lin-12(n137gf), and (H)lin-12(n137n720lf) early L3 larvae. In (G) and (H), the extrachromosomal array zhEx23 was used becausezhIs4 had integrated near the lin-12 locus. (I) Expression of LIP-1::GFP from a reporter (zhEx25) containing point mutations in the four CSL binding motifs. Bar (I), 20 μm.

To investigate if LIP-1 was up-regulated in ectopic 2° cells, we examined LIP-1::GFP expression inlin-15(rf) (13) andlet-60(gf) mutants in which all VPCs frequently adopt primary (1°) or 2° vulval fates. In 72% oflin-15(rf) (n = 18) and 68% oflet-60(gf) animals (n = 34), LIP-1::GFP was up-regulated in P3.p, P4.p, or P8.p in addition to P5.p and P7.p (Fig. 2F). Moreover, in lin-12 notch(gf) mutants in which all VPCs adopt the 2° fate (14), up-regulation of LIP-1::GFP always occurred in all VPCs (Fig. 2G) (n = 20). On the other hand, in lin-12 notch(lf) mutants, the induction of LIP-1::GFP in P5.p or P7.p was not observed, and LIP-1::GFP was expressed at low levels in all VPCs (Fig. 2H) (n = 23) [note that the strong LIP-1::GFP expression in the hyp7 nuclei P2.p and P9.p was unchanged in lin-12(lf) mutants]. Downstream of LIN-12, the lateral signal is transduced by a CSL (CBF-1/Suppressor of Hairless/LAG-1) transcription factor that binds to RTGGGAA motifs in the regulatory regions of the target genes (15). Likewise, the lip-1 promoter/enhancer region contains a cluster of four potential CSL binding sites (Fig. 2A). Animals carrying a LIP-1::GFP reporter in which all four RTGGGAA motifs had been mutated to RAGGGAA did not show induction of LIP::GFP in P5.p or P7.p (Fig. 2I) (n = 26), similar to the pattern observed inlin-12(lf) mutants. Thus, lip-1may be a direct target of the Notch signaling pathway in the VPCs.

LIN-12/NOTCH performs two functions during vulval induction that are separated by the phase of the VPC cell cycle (16). Before completion of the S phase, LIN-12 inhibits the specification of the 1° fate and maintains the VPCs in an uncommitted state. After completion of the S phase, LIN-12 promotes the specification of the 2° fate. To determine in which phase of the cell cycle the up-regulation of LIP-1 takes place, we arrested the VPCs in the S phase by transferring mid-L2 larvae to hydroxyurea-containing medium. In all hydroxyurea-arrested animals carrying the transcriptional reporter, LIP-1::GFP expression was up-regulated in P5.p and P7.p (n = 30). Therefore, lip-1 may be induced before the end of the S phase to prevent P5.p and P7.p from adopting the 1° fate (17).

To test this hypothesis, we asked if loss of lip-1(+)function allows neighboring VPCs to adopt the 1° fate. For this purpose, we compared the expression pattern of the 1° fate marker EGL-17::GFP (18) in the presence and absence oflip-1(+) activity. In lip-1(lf) single mutants, EGL-17::GFP was expressed exclusively in P6.p (Table 2, rows 1 and 2). However, when the activity of the RTK/RAS/MAPK signaling pathway was increased, loss of lip-1(+) function frequently caused two or more adjacent VPCs to express the 1° fate marker EGL-17::GFP (Table 2, rows 3 to 8). In most cases, EGL-17::GFP expression could be observed in P5.p or P7.p in addition to P6.p, indicating that P6.p failed to inhibit the 1° fate in its neighbors when lip-1(+) was absent.

Table 2

Loss of lip-1(+) function allows neighboring VPCs to adopt the 1° fate.

View this table:

The data presented here suggest the following model. LIP-1 is initially expressed at a low level in all VPCs to repress basal MPK-1 activity. The anchor cell signal overcomes this constitutive inhibition in P6.p to induce the 1° fate. The lateral signal from P6.p up-regulates the transcription of lip-1 in P5.p and P7.p to reduce MPK-1 activity and thereby mediate the lateral inhibition of the 1° fate in these cells. Low MPK-1 activity in P5.p and P7.p, combined with the lateral LIN-12 signal, specifies the 2° fate (19, 20). Because loss oflip-1(+) function is not sufficient to disrupt lateral inhibition (Table 2, row 2), additional inhibitory mechanisms may exist. For example, the expression of the receptor for the inductive signal, LET-23 epidermal growth factor receptor (EGFR), increases in P6.p and decreases in P5.p and P7.p toward the end of the L2 stage. LET-23 EGFR in P6.p appears to sequester most of the inductive signal LIN-3 EGF (21), whereas the down-regulation of LET-23 EGFR in P5.p and P7.p may desensitize these cells toward the inductive signal.

Lateral inhibition of the RTK/RAS/MAPK signaling pathway by Notch has been observed in various cell types in different organisms (22–25). The model presented here may help explain how adjacent cells can translate the small differences in the amount of an extrinsic signal they receive into a binary decision, resulting in a defined pattern of cell fates.

  • * To whom correspondence should be addressed. E-mail: ahajnal{at}pathol.unizh.ch

REFERENCES AND NOTES

View Abstract

Navigate This Article