The human tumour suppressor PTEN regulates longevity and ... - Nature

Oncogene (2005) 24, 20–27

& 2005 Nature Publishing Group All rights reserved 0950-9232/05 $30.00 www.nature.com/onc

The human tumour suppressor PTEN regulates longevity and dauer formation in Caenorhabditis elegans Florence Solari1, Ange´lique Bourbon-Piffaut1, Ingrid Masse1, Bernard Payrastre2, Andrew M-L Chan3 and Marc Billaud*,1 1

Laboratoire Geńe´tique et cancer, FRE 2692 CNRS, Universite´ Claude Bernard Lyon 1, Domaine Rockefeller, 8, avenue Rockefeller, 69373 Cedex 08, Lyon, France; 2Inserm U563, De´partement d’Oncogene`se et signalisation dans les cellules he´matopoıë´tiques, IFR30, Hoˆpital Purpan, Toulouse 31059, France; 3The Derald H Ruttenber Cancer Center, Mount Sinai School of Medicine, New York, NY 10029, USA

The PTEN tumour suppressor is a phosphatase that dephosphorylates phosphatidylinositol 3, 4, 5 triphosphate (PIP3) and protein substrates. PTEN function is modulated by its carboxy-terminal region, which contains several clustered phosphorylation sites and a PDZ-binding motif (PDZbm). Although PTEN growth suppression effect is well demonstrated, its additional biological roles are less well understood. DAF-18, a Caenorhabditis elegans homologue PTEN, is a component of the insulin/IGF-I signalling pathway that controls entry to the dauer larval stage and adult longevity. To further explore the role of PTEN in the insulin signalling cascade and its possible involvement in the mechanisms of ageing, we undertook a study of PTEN function in C. elegans. We now report that human PTEN can substitute for DAF-18 and restores the dauer and longevity phenotypes in worms devoid of DAF-18. Furthermore, we provide genetic and biochemical evidence that dauer and lifespan control depends on PTEN-mediated regulation of PIP3 levels. Finally, we established that phosphorylation sites in the Cterminus of PTEN and its PDZbm are necessary for PTEN control of the insulin/IGF-I pathway. These results demonstrate that PTEN negatively regulates the insulin/ IGF pathway in a whole organism and raise the hypothesis that PTEN may be involved in mammalian ageing. Oncogene (2005) 24, 20–27. doi:10.1038/sj.onc.1207978 Keywords: PTEN; insulin pathway; longevity; C. elegans

Introduction The PTEN tumour suppressor gene is somatically mutated in a wide range of tumours, including glioblastoma, melanoma, prostate and endometrial neoplasia (Cantley and Neel, 1999; Bonneau and Longy, 2000; Simpson and Parsons, 2001). Furthermore, germline mutations of PTEN are the underlying genetic cause *Correspondence: M Billaud; E-mail: [email protected] Received 10 February 2004; revised 15 June 2004; accepted 16 June 2004

of three related multiple hamartoma disorders: Cowden disease characterized by an increased risk of breast and thyroid cancers; Bannayan–Zonana and Proteus syndromes (for a review see Eng, 2003). PTEN encodes a phosphatase, which dephosphorylates phosphatidylinositol 3, 4, 5 triphosphate (PIP3), and protein substrates (Myers et al., 1997; Gu et al., 1998; Maehama and Dixon, 1998; Tamura et al., 1998). PIP3 is a product of class I phosphoinositide 3-kinase (PI3K) and acts as a lipid second messenger to activate multiple downstream effectors such as the serine threonine kinase AKT/PKB. Several observations have demonstrated that the PTEN PIP3 3-phosphatase activity is essential to its tumour suppressor function (Li and Sun, 1998; Myers et al., 1998). The PTEN G129E mutation found in Cowden patients disrupts PTEN lipid phosphatase activity but preserves the protein phosphatase activity (Myers et al., 1998). Furthermore, loss of PTEN in tumour cell lines leads to an increased concentration of PIP3 and to AKT hyperactivation (Stambolic et al., 1998). Restoration of PTEN WT in mutated tumour cells results in the reduction of PIP3 levels and of AKT activation (HaasKogan et al., 1998; Li and Sun, 1998). The FOXO subfamily of Forkhead transcription factors (FKHR/ FOXO1, FKHRL1/FOXO3a and AFX/FOXO4) are negatively regulated by AKT and evidence have been provided that they are critical effectors of PTENmediated tumour suppression effect (Nakamura et al., 2000). In Caenorhabditis elegans, a conserved insulin-like signalling pathway regulates worm metabolism, development and lifespan. The insulin-like signal is conveyed through DAF-2, a member of the insulin/insulin-like growth factor I (IGF1) family of receptor tyrosine kinases (Paradis and Ruvkun, 1998). Daf-2 mutants arrest development at a diapausing larval stage known as dauer and their metabolism is shifted dramatically towards energy storage. Dauer formation is an adaptive response to unfavourable conditions such as overcrowding and starvation (Riddle, 1997). Genetic analyses have revealed that DAF-2 signals to AGE-1, which encodes a homologue of the p110 catalytic subunit of class I PI3Ks (Dorman et al., 1995; Morris et al., 1996). Two C. elegans AKT homologues AKT-1 and AKT-2

Human PTEN regulates longevity in Caenorhabditis elegans F Solari et al

21 Table 1 Effects of wild-type DAF-18, PTEN and mutant proteins expression on dauer formation at 251C Genotypea

Percent dauer formation

n

ExROL-6 ExDAF-18 WT ExDAF-18 G174E

0 100 0

601 523 217

ExPTENWT ExPTEN C124S ExPTEN G129E ExPTEN 399STOP ExPTEN T382/T383A ExPTEN S380/T382/T383A

100 0 0 100 100 100

483 128 209 131 317 163

a

All transgenics are daf-2(e1370); daf-18(mg198) double mutant and carried an extrachromosomal array (Ex) containing the rol-6 coinjection marker. n, total number of animals scored

a

ExpROL-6

100

ExDAF-18 WT 80 % alive

act downstream of AGE-1 (Paradis and Ruvkun, 1998). DAF-16, a Forkhead transcription factor, is the major negatively regulated target of the DAF-2/AGE-1/AKT1/2 signalling cascade (Lin et al., 1997; Ogg et al., 1997; Paradis and Ruvkun, 1998). DAF-18 has been recently characterized as the closest homologue of human PTEN in C. elegans (Ogg and Ruvkun, 1998; Gil et al., 1999; Mihaylova et al., 1999; Rouault et al., 1999). DAF-18 negatively regulates the insulin-like pathway and, based on its homology with PTEN, it has been suggested that DAF-18 may antagonize the activity of AGE-1 by catalysing the dephosphorylation of PIP3. PTEN activity has been extensively studied as a tumour suppressor and very little is known about its role in normal cell function. To further explore the role of PTEN in the insulin signalling cascade and its possible involvement in the mechanisms of ageing, we undertook a study of PTEN function in C. elegans. Here, we report that human PTEN can substitute for DAF-18 to regulate dauer and lifespan. Furthermore, we show that DAF-18, like PTEN, functions as a PIP3 phosphatase and that PIP3 is a critical lipid second messenger for dauer and lifespan regulation. Finally, we found that mutation of PTEN regulatory modules affect its ability to antagonize the insulin pathway.

ExPTEN WT ExPTEN C124S

60

ExPTEN G129E

40 20 0 0

Results

5

10

15 20 age (days)

25

30

PTEN controls dauer entry and lifespan in C. elegans

b

ExpROL-6

100

ExPTEN WT 80 % alive

To investigate whether human PTEN could substitute for DAF-18 to regulate the insulin/IGF-I pathway in C. elegans, we constructed vectors driving either PTEN cDNA or daf-18 cDNA expression under the control of 1 kb of 50 regulatory sequences of daf-18. This sequence is believed to contain the complete daf-18 promoter since it is bounded by an upstream gene located within 1 kb of its start codon. These constructs were injected into double mutants daf-2(e1370); daf-18(mg198). daf2(e1370) single mutants constitutively form dauers at 251C and become long-lived adults when transferred at 251C from the L4 stage (Kenyon et al., 1993; Larsen et al., 1995; Lin et al., 1997). daf-18(mg198) null mutation suppresses daf-2(e1370) mutant phenotypes: daf-2(e1370); daf-18(mg198) double mutants do not form dauer at 251C and have a shorter lifespan compared to daf-2(e1370) mutants (data not shown and Table 1). We found that expression of PTEN WT in daf2(e1370); daf-18(mg198) mutants restored the daf2(e1370) constitutive dauer phenotype at 251C as efficiently as DAF-18 (Table 1). Furthermore, the extension of lifespan observed for transgenic animals expressing PTEN WT or DAF-18 were alike (15.9 and 16.2 days, respectively, compared to 11.4 days for double mutants, Figure 1a and Table 2). To determine whether the functioning of PTEN in dauer and lifespan regulation requires its phosphatase activity, we introduced a catalytically inactive form of PTEN (PTEN C124S; Maehama and Dixon, 1998; Myers et al., 1998)

ExPTEN T382/T383A ExPTEN S380/T382/T383A

60

ExPTEN 399STOP

40 20 0 0

5

10

15 20 age (days)

25

30

35

Figure 1 Effects of wild-type DAF-18, PTEN and mutant proteins on lifespan. The percentage of live animals was plotted as a function of time (days). The curves shown represent the sum of all animals examined. All transgenic animals carried the coinjection marker rol-6. (a) DAF-18 WT and PTEN WT expression in daf2(e1370); daf-18(mg198) extended lifespan. Catalytic inactive PTEN mutant (PTEN C124S) or lipid phosphatase inactive PTEN mutant (PTEN G129E) did not extend lifespan. (b) Lifespan regulation by PTEN mutant proteins, carrying mutations in the phosphorylation sites (PTEN T382/T383A and PTEN S380/T382/ 383A) or carrying a deletion of the PDZbm (PTEN399STOP)

in double mutants. We neither observed dauer formation (Table 1) nor lifespan extension (Figure 1a and Table 2), although PTEN C124S was expressed in worms (Figure 2a). These results show that PTEN can substitute for DAF-18 to regulate dauer and lifespan in worms and that PTEN protein and/or phospholipids Oncogene


22 Table 2 Effects of wild-type DAF-18, PTEN and mutant proteins expression on lifespan at 251C Genotype

Mean (s.e.m.)

P-value

Maximum

n

daf-2(e1370); daf-18(mg198) ExROL-6 ExDAF-18 WT ExDAF-18 G174E

10.7 11.4 16.2 11.3

(70.2) (70.3) (70.4) (70.1)

— — 109a 109b

16 16 30 15

137 90 158 171

ExPTEN ExPTEN ExPTEN ExPTEN ExPTEN ExPTEN

15.9 11.0 10.4 13.0 17.1 13.7

(70.4) (70.2) (70.3) (70.3) (70.3) (70.1)

109a 109c 109c 107c 0.02c 105c

26 15 15 20 32 21

112 92 99 110 212 95

WT C124S G129E 399STOP T382/383A S380/T382/T383A

Experiments were performed at least three times with independent transgenic lines and gave similar results. s.e.m., standard error of the mean. Pvalue corresponds to comparisons of lifespan using Student’s t-test: abetween DAF-18 WT or PTEN WT and ROL-6; bbetween DAF-18 WT and DAF-18 G174E; and cbetween PTEN mutants and PTEN WT. n, total number of animals scored

Figure 2 Expression of WT and PTEN mutant proteins in double mutants daf-2(e1370); daf-18(mg198). Samples for Western blot analysis were prepared as described in Material and methods. PTEN proteins were revealed with a monoclonal anti-human PTEN antibody directed against the PTEN C-terminus. Protein loading was controlled with an anti-b-tubulin antibody. PTEN is revealed as a 60 kDa band in all lines expressing PTEN WT, PTEN C124S, PTEN T382/T383A and PTEN G129E except for the control line that contains only the rol-6 marker used to generate all transgenic lines (a). PTEN is barely detected in PTEN S380/T382/ T383A expressing transgenic lines (b)

phosphatase activity is essential for its function in C. elegans. DAF-18 regulates PIP3 levels in worms PTEN specifically dephosphorylates the D3 position of PIP3 (Maehama and Dixon, 1998; Lee et al., 1999). Several reports have convincingly shown that tumour cell lines with an inactive mutant form of PTEN have Oncogene

Figure 3 Increased PtdIns(3,4,5)P3 level in daf-18(mg198) mutants. HPLC analysis profiles showing the separation and the relative level of PtdIns(4,5)P2 and PtdIns(3,4,5)P3 in C. elegans. Data shown are representative of two independent experiments. A table indicating the [32P]PtdIns(3,4,5)P3/[32P]PtdIns(4,5)P2 ratio measured in the two independent experiments (Exp. 1–3) is included

elevated levels of PIP3 (Furnari et al., 1998). Although genetic evidences place DAF-18 downstream of AGE-1/ PI3K and suggests that DAF-18 functions as a PIP3 3phosphatase, no biochemical data supporting this model have been reported to date. According to this notion, daf-18 mutation should result in an increase of PIP3 level in worms. To test this hypothesis, we metabolically labelled phospholipids from wild-type and daf18(mg198) animals and analysed the PIP3/PIP2 ratio by high-performance liquid chromatography (HPLC, see Materials and methods). We reproducibly observed a two- to threefold increase of PIP3 in daf-18(mg198)


23

mutants compared to wild type (Figure 3), thus indicating that DAF-18, like PTEN, displays PIP3 3phosphatase activity. The lipid phosphatase activity of PTEN and DAF-18 is essential for dauer entry and lifespan regulation While our results provide biochemical evidences that DAF-18 regulates PIP3 levels in worms, they do not show that the lipid phosphatase activity of PTEN and DAF-18 is required for dauer and lifespan regulation. To address this question, we tested the biological activity of a PTEN mutant (G129E), which has lost the ability to recognize PIP3 but which remains active towards protein substrates (Myers et al., 1998; Lee et al., 1999). This glycine is found in all PTEN orthologues at the same position within the catalytic core domain and corresponds to Gly174 in DAF-18 (Maehama et al., 2001). The introduction of transgenes encoding either PTEN G129E or DAF-18 G174E into daf-2(e1370); daf-18(mg198) mutants resulted in transgenic animals, which did not enter the dauer larval stage when grown at 251C (Table 1), although Western blot analysis confirmed that the PTEN G129E protein was efficiently expressed in worms (Figure 2a). Interestingly, the lifespan of these transgenic worms was comparable to the daf-2(e1370); daf-18(mg198) mutant lifespan (Figure 1a and Table 2). These results strongly indicate that the protein phosphatase activity of PTEN and DAF-18 is not sufficient to regulate development and longevity in C. elegans, and that the lipid phosphatase activity is essential for these physiological processes. Mutations of PTEN phosphorylation sites enhance its biological function in C. elegans Although the tumour suppressor role of PTEN has been firmly established, the molecular mechanisms involved in PTEN regulation remain poorly understood. Phosphorylation has been proposed as a possible mode of PTEN regulation (Georgescu et al., 1999; Vazquez et al., 2000; Torres and Pulido, 2001; Vazquez et al., 2001; Birle et al., 2002). Mutation of each of three putative phosphorylation sites (Ser380, Thr382 and Thr383) to alanine residues increases PTEN biological activity but reduces PTEN protein stability (Vazquez et al., 2000; Torres and Pulido, 2001). Based on these results, we first asked whether PTEN was phosphorylated in C. elegans. For this purpose, we used a polyclonal serum that reacts with PTEN when phosphorylated on Ser380/Thr382/ Thr383 and found that this antibody recognized PTEN WT when expressed in worms. Conversely, PTEN mutated on both Thr382 and Thr383 (PTEN T382/ T383A) was barely detected with this antibody (Figure 4a), whereas it was revealed with a monoclonal antibody recognizing a phospho-independent epitope (Figure 2a). These results indicate that PTEN is phosphorylated on Ser380, Thr382 and Thr383 in C. elegans and that the protein kinase(s) responsible for phosphorylation at these residues is (are) present in C. elegans.

Figure 4 PTEN is phosphorylated in double mutants daf2(e1370); daf-18(mg198). Expression of the phosphorylated form of PTEN was analysed with a phosphospecific antibody that recognizes PTEN phosphorylated on Ser380, Thr382 and Thr 383. (a) Proteins extracted from worms expressing either PTEN wildtype, the ROL-6 marker or PTEN mutated on both Thr 382 and Thr 383 (substitution of threonine for an alanine) were blotted with the anti-phosphospecific PTEN. Mutation of Thr382 and Thr383 nearly abrogates the reactivity of the antibody with the phosphorylated epitope. Protein loading was controlled with an anti-btubulin antibody. (b) Proteins extracted from worms expressing either ROL- 6, PTEN WT, PTEN 399STOP, PTEN C124S and PTEN G129E were blotted with the anti-phospho-PTEN. Protein loading was controlled with an anti-actin serum. Note that the reduction of the molecular weight of PTEN 399STOP due to the deletion of the five amino acids at the C-terminus is detectable on the Western blot

Although the current model proposes that unphosphorylated PTEN is not stable, the level of PTEN T382/T383A protein observed by Western blot analysis was not diminished compared to PTEN WT (Figure 2a). We thus asked whether alanine substitution of the three clustered phosphorylation sites Ser380, Thr382 and Thr383 would affect PTEN protein stability and/or protein expression level. PTEN mutated on Ser380, Thr382, Thr383 was barely detectable by Western blot analysis with anti-PTEN monoclonal antibody (Figure 2b), thus indicating that phosphorylation of these three residues may regulate PTEN protein stability and that phosphorylation of Ser 380 may be critical for this effect. The prevailing view is that phosphorylation keeps PTEN stable but less active, whereas dephosphorylation activates PTEN. PTEN T382/T383A expression in daf2(e1370); daf-18(mg198) restored the daf-2(e1370) dauer phenotype at 251C as efficiently as PTEN WT (Table 1). Based on the mammalian model, we expected an enhancement of PTEN T382/T383A biological activity. However, this effect could not be examined in our assays since PTEN WT already induced 100% of worms to enter the dauer stage at 251C. We then took advantage of the temperature sensitivity of the daf-2(e1370) mutation to ask whether a lower temperature would reveal increased PTEN activity by the Oncogene


24 Table 3 Effects of wild-type and mutant PTEN proteins expression on growth arrest at 211C Genotype daf-2(e1370); daf-18(mg198) daf-2(e1370) ExPTEN WT ExPTEN 399STOP ExPTEN T382/T383A ExPTEN S380/T382/T383A

Percent growth arrest (s.e.m.) 8.7 10.7 3 79.5 19.9

0 (70.3) (71.9) (72.8) (71.1) (71.1)

P-value

0.22a Po105b Po105b Po105b

n 210 613 484 525 677 131

L4 animals were transferred from 15 to 211C and allowed to lay eggs for 24 h. Animals that arrest at the L2 stage were scored after 72 h. s.e.m., standard error of the mean. P-values correspond to comparisons of growth arrest using Khi2 test: abetween PTEN WT and daf2(e1370); and bbetween the corresponding strain and PTEN WT. n, total number of animals scored

unphosphorylated mutant. At the semipermissive temperature of 211C, daf-2(e1370) mutants do not enter the dauer stage but show a transient larval arrest at the L2 stage (Table 3). As expected, daf-2(e1370); daf18(mg198) worms expressing PTEN WT show a similar phenotype with a penetrance comparable to daf2(e1370) mutants, of 10.7 and 8.7%, respectively (Table 3). Consistent with the prediction that mutation of these phosphoacceptor sites increases PTEN biological activity, 79.5% of animals expressing PTEN T382/ T383A underwent larval arrest when raised at 211C (Table 3). Furthermore, despite the very low level of PTEN S380/T382/T383A protein, all animals expressing this PTEN mutant formed dauer at 251C (Table 1) and 19.9% underwent larval arrest when grown at 211C (Table 3). These results show that mutation of PTEN at sites of phosphorylation enhances its function for dauer regulation. Lifespan of worms expressing PTEN T382/ T383A was significantly increased compared to worm expressing PTEN WT (Figure 1b and Table 2). On the other hand, worms expressing PTEN S380/T382/T383A displayed a reduced longevity (Figure 1b and Table 2) compared to PTEN WT. Thus, although PTEN S380/ T382/T383A expression rescues the dauer phenotype of double mutants daf-2(e170); daf-18(mg198), this PTEN mutant does not recapitulate PTEN WT function for lifespan control. These results suggest that modulation of PTEN phosphorylation and stability do not affect PTEN function in the same way for dauer and lifespan regulation. The PTEN PDZ-binding motif (PDZbm) is required for lifespan and dauer regulation The last four amino acid of PTEN display a PDZPDZbm and several groups have reported association of PTEN with PDZ domain-containing proteins (Adey et al., 2000; Wu et al., 2000a, b). To explore the biological relevance of the PDZbm of PTEN, a transgene encoding a PDZbm-deleted form of PTEN (PTEN 399STOP) was introduced into daf-2(e1370); daf-18(mg198) mutants. To ensure that PTEN 399STOP was expressed in transgenic animals, we used Oncogene

a polyclonal serum that reacts with PTEN phosphorylated on Ser380, Thr382 and Thr383, since the monoclonal anti-PTEN used in this study was raised against the PTEN C-terminus. Western blot analysis comparing the level of expression of PTEN WT and PTEN 399STOP did not reveal any major difference in protein level (Figure 4b), similarly to results obtained in mammalian cells (Leslie et al., 2001). Transgenic animals expressing PTEN 399STOP constitutively arrested at the dauer stage at 251C (Table 1). However, at 211C, only 3% of them show transient larvae arrest (Table 3). Furthermore, their lifespan was significantly reduced (13 days) compared to worms expressing PTEN WT (15.9 days, see Figure 1b and Table 2). Collectively, these results indicate that the deletion of the PDZbm significantly impairs PTEN biological activity in C. elegans and strongly suggest that PDZ-containing proteins play a role in dauer and lifespan regulation by PTEN.

Discussion PTEN can replace DAF-18 in C. elegans Previous genetic studies have shown that DAF-18, the C. elegans homologue of PTEN, antagonizes the action of both the receptor DAF-2 and the PI3K homologue AGE-1. Although the high conservation of the core enzymatic domain between DAF-18 and PTEN supports the notion that these enzymes recognize similar substrates, DAF-18 and PTEN differ in their C-terminal region (Ogg and Ruvkun, 1998; Gil et al., 1999; Mihaylova et al., 1999; Rouault et al., 1999), thus raising the possibility that their mode of regulation is not alike. In this study, we have demonstrated that expression of either human PTEN or DAF-18 under the control of daf-18 regulatory sequences complements dauer and lifespan phenotype of daf-18 mutants to the same extent. These data establish that PTEN and DAF18 are genuine orthologues and further suggest that mechanisms regulating PTEN function are conserved between human and C. elegans. PIP3 is a critical second messenger for dauer and lifespan regulation Biochemical determination of the phosphoinositide content in daf-18 mutant worms allowed us to show that the PIP3 level was significantly increased compared with wild-type animals, thus suggesting that this class of phosphoinositides may be a critical second messenger for the regulation of dauer and longevity in C. elegans. However, the insulin/IGF-I pathway delivers different biological outcomes that are both PI3K dependent and PI3K independent (Saltiel and Kahn, 2001). Therefore, the variation of PIP3 levels could be a secondary event in dauer and lifespan regulation by DAF-18. Using a previously characterized lipid phosphatase mutant (PTEN G129E), we firmly established that PTEN catalytic activity directed towards PIP3 is essential for


25

dauer and lifespan regulation. Similarly, mutation of the conserved glycine in DAF-18 (DAF-18 G174E) invalidates its capacity to rescue dauer and lifespan phenotype of daf-18 mutants. These results indicate that the PIP3 3phosphatase activity of PTEN and DAF-18 is essential for dauer and lifespan regulation and further suggest that PIP3 is a second messenger of the insulin signalling pathway that is critical for both physiological processes. Distinct PTEN regulatory modules are necessary for the control of insulin signalling In addition to the catalytic and C2 domains, PTEN displays several motifs that serve to regulate its activity and biological function. The 50 amino acids of the PTEN carboxy-terminal tail contain several putative phosphorylation sites including Ser380, Thr382 and Thr383. Using a phosphospecific antibody, we found that PTEN is phosphorylated on Ser380, Thr382 and Thr383 in C. elegans. In addition, these results show that PTEN phosphorylation is independent of its enzymatic activity, since phosphorylated PTEN is also detected in transgenic worms expressing a catalytic null mutant (Figure 4b). Moreover, mutation of each of the three phosphorylation sites (Ser380, Thr382 and Thr383) has been shown to enhance the capacity of PTEN to inhibit downstream targets and to result in a loss of PTEN protein stability (Tolkacheva et al., 2001; Torres and Pulido, 2001; Vazquez et al., 2001). In agreement with these findings, we found that phosphorylation of these residues are critical for PTEN stability and for the ability of PTEN to modulate the insulin/IGF1 signalling in worms. Furthermore, PTEN protein mutated at these three phosphorylation sites was barely detected by Western blot analysis. However, from these experiments, we cannot definitively exclude that mechanisms other than the loss of protein stability could account for the decreased level of mutant PTEN. The protein kinase CK2 was identified as one of the physiologically relevant kinase that phosphorylates PTEN C-terminus and inhibits its activity (Torres and Pulido, 2001; Vazquez et al., 2001). Following this line of investigation, it will be relevant to study whether the C. elegans orthologue of CK2 may regulate the insulin pathway in the nematode via its action on DAF-18/ PTEN. Another feature of the PTEN protein is the presence of a PDZbm in the last four C-terminus amino acids. PDZ domains are often found in scaffold proteins that organize multimolecular complexes at the plasma membrane. We show here that deletion of the PDZbm of PTEN (PTEN 399STOP) impairs its capacity to regulate dauer and lifespan. However, PTEN 399STOP does not behave like a catalytic null mutant and retains some biological activity. The PDZbm of PTEN can bind to PDZ-containing proteins such as membrane-associated guanlylate kinase hDlg and MAGI-1, MAGI-2, MAGI-3, the kinase hMAST205 and the multi-PDZ protein MUPP1 (Adey et al., 2000; Wu et al., 2000a, b). It has been suggested that these PDZ-containing proteins contribute to the recruitment of PTEN at the

plasma membrane and that interaction with MAGI proteins enhances the ability of PTEN to regulate negatively AKT, presumably by locating PTEN within close proximity of its substrate PIP3 (Wu et al., 2000a, b). Furthermore, a truncated form of PTEN that lacks solely the PDZbm is less efficient at inhibiting platelet-derived growth factor-induced cytoskeletal changes (Leslie et al., 2000, 2001). However, several reports have concluded that the PDZbm is dispensable for PTEN tumour suppressor activity (Georgescu et al., 1999; Morimoto et al., 1999; Leslie et al., 2000; Tolkacheva and Chan, 2000). Our data suggest that PTEN may interact with PDZ-containing proteins in C. elegans and consistent with this idea, homologues of hDlg and MAGI have been identified in the nematode (Bossinger et al., 2001; F Solari, unpublished data). The functional and physiological significance of the regulatory modules of PTEN are still a matter of controversy. Our results support the notion that the PTEN PDZbm and that phosphorylation sites Ser 380, Thr382 and Thr383 are critical for the downregulation of the insulin pathway in a whole organism, a finding that would deserve to be tested in mammals. Insulin signalling and regulation of lifespan: a conserved role of PTEN throughout evolution? Holzenberger et al. (2003) have recently shown that Igf1r heterozygous knockout mice live up to 33% longer that their wild-type littermates . Moreover, disruption of the gene encoding the insulin receptor in adipocytes increases the mean lifespan of mice of 18% (Blu¨her et al., 2003). These results combined with previous reports on both DAF-2 (Kenyon et al., 1993) and on the role of the insulin receptor, in Drosophila lifespan (Clancy et al., 2001; Tatar et al., 2001), strengthen the emerging view that the insulin/IGF-I system controls ageing in all organisms with similar molecular mechanisms. We have observed that over expression of DAF-18 is sufficient to increase substantially wild-type worms longevity (data not shown). With regard to our results, it is conceivable that PTEN might be one of the components of the insulin cascade implicated in mammalian lifespan regulation and overexpression of PTEN in mice with a transgenic approach should allow this question to be addressed. Overall, our data strongly suggest that proteins that regulate PTEN function are conserved in worms. Mechanisms responsible for PTEN regulation are still poorly understood. A genetic screen to identify genes interfering with PTEN lifespan rescue should elucidate this question and improve our understanding of PTEN tumour suppressor role, given the intertwined relationships between tumour suppression and longevity.

Materials and methods Srains Strains used were as follows: wild-type N2 bristol, daf2(e1370) III, daf-18(mg198) IV and daf-2(e1370); dafOncogene


26 18(mg198). The daf-18(mg198) null mutation was isolated in a EMS screen for suppressors of the dauer phenotype of age1(mg44) (Paradis and Ruvkun, 1998; SP, SK, GR, unpublished results). daf-18(mg198) mutants behave similarly to the previously described daf-18(nr2037) null allele (Gil et al., 1999; Mihaylova et al., 1999): daf-18(mg198) are dauer defective and short lived (F Solari, unpublished results). Constructs and transgenic lines To generate pPD95.75-daf-18p, 1 kb upstream sequence of daf-18 ATG was amplified by PCR from N2 genomic DNA using the primers XbaI-daf-18pF: 50 -GCTCTAGAGC GGAAACTCATTTCTGAAATGTC-30 ; and BamHI-daf18pR: 50 -CGGGATCCCG TGGGGGTTAGTAGATGTACC-30 , double digested with XbaI and BamHI and inserted into XbaI–BamHI double-digested pPD95.75 (provided by Andy Fire, Carnegie Institute of Washington). Daf-18p-daf18cDNA and daf-18p-PTENcDNA constructs were obtained by inserting, respectively, daf-18 cDNA (amplified by PCR with primers XhoI-daf-18cDNAF 50 -GCCCTGAAAACTCGAGAACAAATGG-30 ; and SalI/SmaI-daf-18cDNAR 50 GAATTTTGATCAAGCTATTTATTTGTAACCCGGGAG TCGACGATC-30 from yk400b8 cDNA provided by Yuji Kohara) or the PTEN full-length cDNA into pPD95.75 daf18p double digested with SmaI and EcoRI upstream unc-54 30 UTR. DAF-18-G174E, PTEN-C124S and PTEN-G129E mutants were generated by standard PCR-based site-directed mutagenesis using, respectively, primers 50 -CACTGTAAAGCTG GAAAAGAACGTACCGGAGTGATGATATG-30 ; 50 -GTTG CAGCAATTCACAGTAAGGCTGGAAAGGGAC-30 ; and 50 -GTAAAGCTGGAAAGGAGCGAACTGGTGTAATGA TATG-30 . The authenticity of all constructs was confirmed by nucleotide sequencing analysis. PTEN T382/T383A and PTEN S380/T382/T383A mutants were obtained as described previously (Tolkacheva et al., 2001). PTEN mutant cDNAs were recovered by BamHI–EcoRI digestion and inserted into pPD95.75 daf-18p digested with BamHI and EcoRI. PTEN399STOP was recovered by EcoRI digestion from pEGF-C2 PTEN-399STOP (kindly provided by NR Leslie) and cloned into pPD95.75-daf-18p after digestion with EcoRI and SmaI. Constructs were injected at 20 ng/ml together with pRF4 (ROL-6) at 100 ng/ml into daf-2(e1370); daf-18(mg198) mutants. For each construct, two to four independent lines have been generated and analysed. Lipid extraction and analysis Worms were washed in phosphate-free RPMI at 211C and labelled with 0.7 mCi/ml of [32P]orthophosphate (Amersham) during 12 h in phosphate-free RPMI at 211C. 32P-labelled worms were then washed once in phosphate-free RPMI at 211C and lipids were immediately extracted following the

modified procedure of Bligh and Dyer (Bligh and Dyer, 1959). Lipids were then separated by thin-layer chromatography (Merck, Nogent-sur-Marne, France) using CHCl3/ CH3COCH3/CH3COOH /H2O (80/30/26/24/14, v/v). The spots corresponding to [32P]PtdInsP2 and [32P]PtdInsP3 were visualized by a PhosphorImager 445 SI (Molecular Dynamic, Inc.) using appropriate standards, scraped off, deacylated and analyszed by HPLC on a Whatman Partisphere 5 SAX column (Whatman International Ltd, UK) as described previously (Payrastre et al., 2001; Niebuhr et al., 2002). Antibodies and immunoblotting Anti-human PTEN (clone 6H2.1) antibodies from Cascade BioScience; anti-phospho-PTEN (Ser380/Thr382/383) polyclonal antibodies from Cell Signalling; anti-b-tubulin monoclonal antibodies from Boehringer Mannheim Biochemica were used at 1 : 1000 dilution and anti-actin monoclonal antibodies from Sigma were used at 1 : 600. Samples for Western blot analysis were prepared essentially as follows: approximately 100 adult worms were hand picked, washed twice in M9 buffer (Brenner, 1974), once in water, frozen in liquid nitrogen and resuspended in an equal volume of 2 Laemmli buffer. Samples were boiled for 10 min, centrifuged for 10 min at 13 000 r.p.m. Proteins were separated on 10% SDS–PAGE. Lifespan assays Lifespan analyses were performed essentially as described (Kenyon et al., 1993). Animals were raised at 151C until the L4 stage and then shifted to 251C. The day of the shift is counted as day 0 in the adult lifespan assay. Animals were considered dead when they ceased moving and responding to prodding. Dauer assays The dauer phenotype was scored essentially as described (Gottlieb and Ruvkun, 1994). L4 animals were transferred from 15 to 251C then removed after 24 h and the progeny were scored for dauer formation 48 h later. Dauers were distinguished by a radially constricted body, darkly pigmented intestine, dauer alae and a constricted pharynx. Acknowledgements We are grateful to Drs G Ruvkun and P Hu for kindly providing the mg198 strain. We thank Dr NR Leslie for PTEN 399STOP mutant and Dr Yuji Kohara for cDNA clones. This work was supported by grants from la Ligue Nationale Contre le Cancer (e´quipe labelliseé), Association pour la Recherche contre le Cancer Contract no. 4794 and ARECA-Toulouse. AMC is supported by grants from the NIH (CA95063) and DOD (17-03-1-0682).

References Adey NB, Huang L, Ormonde PA, Baumgard ML, Pero R, Byreddy DV, Tavtigian SV and Bartel PL. (2000). Cancer Res., 60, 35–37. Birle D, Bottini N, Williams S, Huynh H, deBelle I, Adamson E and Mustelin T. (2002). J. Immunol., 169, 286–291. Bligh EG and Dyer WJ. (1959). Can. J. Biochem. Physiol., 37, 911–918. Blu¨her M, Kahn BB and Khan CR. (2003). Science, 299, 572–574. Bonneau D and Longy M. (2000). Hum. Mutat., 16, 109–122. Oncogene

Bossinger O, Klebes A, Segbert C, Theres C and Knust E. (2001). Dev. Biol., 230, 29–42. Brenner S. (1974). Genetics, 77, 71–94. Cantley LC and Neel BG. (1999). Proc. Natl. Acad. Sci. USA, 96, 4240–4245. Clancy DJ, Gems D, Harshman LG, Oldham S, Stocker H, Hafen E, Leevers SJ and Partridge L. (2001). Science, 292, 41–43. Dorman JB, Albinder B, Shroyer T and Kenyon C. (1995). Genetics, 141, 1399–1406.


27 Eng C. (2003). Hum. Mutat., 22, 183–198. Furnari FB, Huang HJ and Cavenee WK. (1998). Cancer Res., 58, 5002–5008. Georgescu MM, Kirsch KH, Akagi T, Shishido T and Hanafusa H. (1999). Proc. Natl. Acad. Sci. USA, 96, 10182–10187. Gil EB, Malone Link E, Liu LX, Johnson CD and Lees JA. (1999). Proc. Natl. Acad. Sci. USA, 96, 2925–2930. Gottlieb S and Ruvkun G. (1994). Genetics, 137, 107–120. Gu J, Tamura M and Yamada KM. (1998). J. Cell Biol., 143, 1375–1383. Haas-Kogan D, Shalev N, Wong M, Mills G, Yount G and Stokoe D. (1998). Curr. Biol., 8, 1195–1198. Holzenberger M, Dupont J, Ducos B, Leneuve P, Geloen A, Even PC, Cervera P and Le Bouc Y. (2003). Nature, 421, 182–187. Kenyon C, Chang J, Gensch E, Rudner A and Tabtiang R. (1993). Nature, 366, 461–464. Larsen PL, Albert PS and Riddle DL. (1995). Genetics, 139, 1567–1583. Lee JO, Yang H, Georgescu MM, Di Cristofano A, Maehama T, Shi Y, Dixon JE, Pandolfi P and Pavletich NP. (1999). Cell, 99, 323–334. Leslie NR, Bennett D, Gray A, Pass I, Hoang-Xuan K and Downes CP. (2001). Biochem. J., 357, 427–435. Leslie NR, Gray A, Pass I, Orchiston EA and Downes CP. (2000). Biochem. J., 346 (Part 3), 827–833. Li DM and Sun H. (1998). Proc. Natl. Acad. Sci. USA, 95, 15406–15411. Lin K, Dorman JB, Rodan A and Kenyon C. (1997). Science, 278, 1319–1322. Maehama T and Dixon JE. (1998). J. Biol. Chem., 273, 13375–13378. Maehama T, Taylor GS and Dixon JE. (2001). Annu. Rev. Biochem., 70, 247–279. Mihaylova VT, Borland CZ, Manjarrez L, Stern MJ and Sun H. (1999). Proc. Natl. Acad. Sci. USA, 96, 7427–7432. Morimoto AM, Berson AE, Fujii GH, Teng DH, Tavtigian SV, Bookstein R, Steck PA and Bolen JB. (1999). Oncogene, 18, 1261–1266. Morris JZ, Tissenbaum HA and Ruvkun G. (1996). Nature, 382, 536–539. Myers MP, Pass I, Batty IH, Van der Kaay J, Stolarov JP, Hemmings BA, Wigler MH, Downes CP and Tonks NK. (1998). Proc. Natl. Acad. Sci. USA, 95, 13513–13518.

Myers MP, Stolarov JP, Eng C, Li J, Wang SI, Wigler MH, Parsons R and Tonks NK. (1997). Proc. Natl. Acad. Sci. USA, 94, 9052–9057. Nakamura N, Ramaswamy S, Vazquez F, Signoretti S, Loda M and Sellers WR. (2000). Mol. Cell. Biol., 20, 8969–8982. Niebuhr K, Giuriato S, Pedron T, Philpott DJ, Gaits F, Sable J, Sheetz MP, Parsot C, Sansonetti PJ and Payrastre B. (2002). EMBO J., 21, 5069–5078. Ogg S, Paradis S, Gottlieb S, Patterson GI, Lee L, Tissenbaum HA and Ruvkun G. (1997). Nature, 389, 994–999. Ogg S and Ruvkun G. (1998). Mol. Cell, 2, 887–893. Paradis S and Ruvkun G. (1998). Genes Dev., 12, 2488–2498. Payrastre B, Missy K, Giuriato S, Bodin S, Plantavid M and Gratacap M. (2001). Cell Signal., 13, 377–387. Riddle DL. (1997). C. elegans II Riddle DL, Blumenthal T, Meyer BJ and Priess JR (eds) Cold Spring Harbor Press: Cold Spring Harbor, NY, 739–768. Rouault JP, Kuwabara PE, Sinilnikova OM, Duret L, Thierry-Mieg D and Billaud M. (1999). Curr. Biol., 9, 329–332. Saltiel AR and Kahn CR. (2001). Nature, 414, 799–806. Simpson L and Parsons R. (2001). Exp. Cell Res., 264, 29–41. Stambolic V, Suzuki A, de la Pompa JL, Brothers GM, Mirtsos C, Sasaki T, Ruland J, Penninger JM, Siderovski DP and Mak TW. (1998). Cell, 95, 29–39. Tamura M, Gu J, Matsumoto K, Aota S, Parsons R and Yamada KM. (1998). Science, 280, 1614–1617. Tatar M, Kopelman A, Epstein D, Tu MP, Yin CM and Garofalo RS. (2001). Science, 292, 107–110. Tolkacheva T, Boddapati M, Sanfiz A, Tsuchida K, Kimmelman AC and Chan AM. (2001). Cancer Res., 61, 4985–4989. Tolkacheva T and Chan AM. (2000). Oncogene, 19, 680–689. Torres J and Pulido R. (2001). J. Biol. Chem., 276, 993–998. Vazquez F, Grossman SR, Takahashi Y, Rokas MV, Nakamura N and Sellers WR. (2001). J. Biol. Chem., 276, 48627–48630. Vazquez F, Ramaswamy S, Nakamura N and Sellers WR. (2000). Mol. Cell. Biol., 20, 5010–5018. Wu X, Hepner K, Castelino-Prabhu S, Do D, Kaye MB, Yuan XJ, Wood J, Ross C, Sawyers CL and Whang YE. (2000a). Proc. Natl. Acad. Sci. USA, 97, 4233–4238. Wu Y, Dowbenko D, Spencer S, Laura R, Lee J, Gu Q and Lasky LA. (2000b). J. Biol. Chem., 275, 21477–21485.

Oncogene