Roles of Gß in membrane recruitment and activation of p110/p101 phosphoino
1 Institut für Physiologische Chemie II, Klinikum der Heinrich-Heine-Universität, 40225 Düsseldorf, Germanyw'(y\1, 百拇医药
2 Institut für Pharmakologie, Chemie, Pharmazie, Freie Universität Berlin, 14195 Berlin, Germanyw'(y\1, 百拇医药
3 Institut für Klinische Pharmakologie und Toxikologie, Chemie, Pharmazie, Freie Universität Berlin, 14195 Berlin, Germanyw'(y\1, 百拇医药
4 Fachbereich Biologie, Chemie, Pharmazie, Freie Universität Berlin, 14195 Berlin, Germanyw'(y\1, 百拇医药
Abstractw'(y\1, 百拇医药
Receptor-regulated class I phosphoinositide 3-kinases (PI3K) phosphorylate the membrane lipid phosphatidylinositol (PtdIns)-4,5-P2 to PtdIns-3,4,5-P3. This, in turn, recruits and activates cytosolic effectors with PtdIns-3,4,5-P3–binding pleckstrin homology (PH) domains, thereby controlling important cellular functions such as proliferation, survival, or chemotaxis. The class IB p110/p101 PI3K is activated by Gß on stimulation of G protein–coupled receptors. It is currently unknown whether in living cells Gß acts as a membrane anchor or an allosteric activator of PI3K, and which role its noncatalytic p101 subunit plays in its activation by Gß. Using GFP-tagged PI3K subunits expressed in HEK cells, we show that Gß recruits the enzyme from the cytosol to the membrane by interaction with its p101 subunit. Accordingly, p101 was found to be required for G protein–mediated activation of PI3K in living cells, as assessed by use of GFP-tagged PtdIns-3,4,5-P3–binding PH domains. Furthermore, membrane-targeted p110 displayed basal enzymatic activity, but was further stimulated by Gß, even in the absence of p101. Therefore, we conclude that in vivo, Gß activates PI3K by a mechanism assigning specific roles for both PI3K subunits, i.e., membrane recruitment is mediated via the noncatalytic p101 subunit, and direct stimulation of Gß with p110 contributes to activation of PI3K.
Key Words: G protein; fluorescence imaging; phosphoinositide 3-kinase; FRET; PH domaina&8n, 百拇医药
Introductiona&8n, 百拇医药
Phosphoinositide 3-kinases (PI3Ks) phosphorylate the D3 position of the inositol ring of phosphoinositides, thereby generating intracellular signaling molecules . These phospholipids, i.e., phosphatidylinositol (PtdIns)-3-P, PtdIns-3,4-P2, and PtdIns-3,4,5-P3, have been implicated in cellular functions including chemotaxis, differentiation, glucose homeostasis, proliferation, survival, and trafficking . They transmit signals by recruiting effector molecules that possess specific lipid-binding domains. FYVE domains and PX domains bind to PtdIns-3-P with high affinity, whereas a subgroup of pleckstrin homology (PH) domains, containing a highly basic motif, preferentially binds PtdIns-3,4-P2 and PtdIns-3,4,5-P3 (a&8n, 百拇医药
PtdIns-3,4,5-P3 represents the major product of class I PI3Ks in vivo. These lipid kinases are under tight control of cell surface receptors, including receptor tyrosine kinases (RTK) or G protein–coupled receptors (GPCR; ). All class I members are heterodimers consisting of a p110 catalytic and a p85- or p101-type noncatalytic subunit. The class IA p110 isoforms , ß, and form a complex with p85 adaptor subunits, whereas the only class IB member (p110) is associated with a p101 noncatalytic subunit. The p85 subunit binds to tyrosine-phosphorylated RTKs. This interaction has two consequences resulting in activation of the PI3K. First, the cytosolic enzyme translocates to the inner leaflet of the plasma membrane, giving p110 access to its lipid substrate . Second, the interaction with the tyrosine-phosphorylated receptor induces a conformational change of p85 that results in disinhibition of p110 enzymatic activity . Because the p85 regulatory subunit inhibits the p110 catalytic subunit, it is feasible that constitutively membrane-associated class IA p110 mutants trigger downstream responses characteristic of growth factor action .
Similar to class IA PI3Ks, unstimulated PI3K is predominantly localized in the cytosol, whereas GPCRs induced an increase of PI3K in the membrane fraction . GPCRs activate PI3K through direct interaction with Gß, whereas a stimulation by G is quantitatively less important. However, little is known about the activating mechanism, e.g., it remains elusive whether Gß functions as a membrane anchor or an allosteric regulator of PI3K. Although the p101 subunit has been proposed to act as an indispensable adaptor linking Gß with p110 , in vitro studies have suggested that p101 is not mandatory for Gß-induced stimulation of PI3K . Moreover, a direct interaction of Gß with NH2- and COOH-terminal regions of p110 has been demonstrated . Further support for a direct interaction of Gß with the p110 catalytic subunit came from the observation that the p110ß isoform can also be stimulated by Gß, regardless of whether p85 is present or not . Nevertheless, the noncatalytic p101 subunit of PI3K binds tightly to Gß, suggesting specific roles in G protein–induced regulation of this lipid kinase. For instance, in vitro studies showed that Gß-stimulated p110 potently produced PtdIns-3-P, whereas the Gß-activated heterodimeric p110p101 counterpart potently catalyzed the formation of PtdIns-3,4,5-P3 . This suggests that p101 may affect the interaction of Gß-stimulated p110 with the lipid interface.
Therefore, the exact roles of Gß and PI3K subunits in G protein–induced activation of PI3K in living cells remain unclear. Hence, we asked whether Gß functions as a membrane anchor and/or allosteric activator, and whether p101 is required for proper function of p110 in vivo. To tackle these questions, we examined cellular events leading to activation of PI3K using fluorescent fusion proteins in living cells.h&ahdq, 百拇医药
Resultsh&ahdq, 百拇医药
Characterization of fluorescent PI3K fusion proteinsh&ahdq, 百拇医药
To visualize PI3K subunits in vivo, we generated constructs encoding p110 and p101 fused to YFP or CFP . Their expression in human embryonic kidney (HEK) 293 cells was verified by immunoblot analysis using an antibody against GFP . The catalytic activity and Gß sensitivity of PI3K fusion proteins was confirmed by in vitro lipid kinase assays after immunoprecipitation with an mAb recognizing only intact p110. Precipitates from vector-transfected control cells neither exhibited p110 immunoreactivity nor lipid kinase activity. These results demonstrate that YFP-fused p110 or p101 retain essential functions of the wild-type PI3K.
fig.ommtteddo, 百拇医药
Construction and characterization of fluorescent PI3Kfusion proteins. (A) Schematic representation of the wild-type, fluorescent, and membrane-targeted PI3K subunits. A YFP (or CFP) tag was fused to either the NH2- or the COOH terminus of p110 and p101. The p110-CAAX fusion protein contains an isoprenylation motif (bars for lipid modifications). (B) Differently fluorescence-tagged PI3K subunits were coexpressed in HEK cells, and cell lysates were subjected to SDS-PAGE followed by immunoblotting with a GFP-specific antibody. (C) Catalytic activity of fluorescent PI3K fusion proteins. HEK cells were transfected with the indicated plasmids. Lysates were subjected to immunoprecipitation (IP) with (+) or without (-) an anti-p110 antibody. immunoprecipitation was controlled by immunoblotting (IB) with another anti-p110 antibody. Shown are chemiluminescence images (top panels). Immunoprecipitates were assayed for in vitro PI3K activity in the absence or presence of 120 nM Gß using PtdIns-4,5-P2– containing lipid vesicles and [32P]ATP as substrates. Depicted are autoradiographs of the generated 32P-PtdIns-3,4,5-P3 (bottom panels).
Subcellular distribution of PI3K subunits and instability of monomeric p101m, 百拇医药
Next, we examined the subcellular distribution of the YFP-tagged PI3K subunits in living cells by confocal laser scanning microscopy. Although YFP was uniformly distributed inside HEK cells (unpublished data), fluorescence signals from YFP fused to the NH2 or COOH termini of p110 were only detected in the cytosolic compartment (top). However, a small fraction of the YFP-fused p110 was also detected in the nucleus when coexpressed with wild-type p101 (top). The same distribution was seen for YFP-tagged p101 coexpressed with wild-type p110 ( B, bottom). Thus, fluorescent PI3K subunits form heterodimers (see next section), present in the cytosol and nucleus of HEK cells, whereas distribution of the p110 subunit alone seems restricted to the cytosol. Interestingly, YFP-fused p101 alone showed a predominant nuclear localization with a significantly lower overall fluorescence intensity ( bottom).
fig.ommttedm9-{, 百拇医药
Expression and subcellular distribution of fluorescent p110 and p101 in HEK cells. (A) Cells were transfected with plasmids for YFP-tagged single PI3K subunits as indicated. Images were taken by confocal laser scanning microscopy. Images of typical cells are shown. White bars indicate a 10-µm scale. (B) Cells were cotransfected with both p110 and p101. Only one subunit was fused to YFP as indicated. (C) Cells were transfected with plasmids encoding fluorescent p110 and p101 at different ratios. The total amount of transfected cDNA was kept constant by the addition of pcDNA3. Equal amounts of whole-cell lysates were subjected to SDS-PAGE followed by immunoblotting with an anti-GFP antibody.m9-{, 百拇医药
To explore the possibility that expression levels of p101 could depend on the coexpression of p110, we transfected HEK cells with different ratios of plasmids encoding fluorescent p110 or p101 and analyzed total cell lysates by immunoblotting using an anti-GFP antibody . The expression level of p110 correlated with the amount of the p110 (but not the p101) plasmid. In contrast, the expression level of p101 was also dependent on the amount of the p110 plasmid. Hence, the stability of p101 depends on the coexpression of p110 but not vice versa. The poor cytosolic fluorescence of YFP-tagged p101 in the absence of p110 may thus be explained by cytosolic degradation of monomeric p101.
In vivo dimerization of PI3K subunits(, 百拇医药
To directly demonstrate heterodimerization of PI3K subunits in living cells, we used fluorescence resonance energy transfer (FRET). FRET between two fluorophores, e.g., CFP and YFP, is restricted to distances of <100 Å, and therefore, provides direct evidence for a protein–protein interaction . We coexpressed CFP- and YFP-tagged PI3K subunits and determined FRET by following the donor (CFP) recovery during acceptor (YFP) bleach . All combinations showed significant FRET, although quantitative differences were evident.(, 百拇医药
fig.ommtted(, 百拇医药
Dimerization of p110 and p101. (A) HEK cells were cotransfected with plasmids encoding NH2- or COOH-terminally YFP-tagged p110 and NH2- or COOH-terminally CFP-tagged p101 in different combinations. FRET was measured in vivo. An increase in CFP (donor) fluorescence during YFP (acceptor) bleach indicates FRET between fluorescent PI3K subunits. The depicted data represent means ± SEM of at least 18 single cells in three independent transfection experiments. (B) HEK cells were transfected with the indicated plasmids. Cytosols were prepared and subjected to gel filtration. The elution profiles were analyzed by immunoblotting with an anti-GFP antibody (L, load diluted 1:5; V, void volume; 20–26, fraction numbers).
Because FRET efficiency is a function of the distance between two fluorophores, quantitative differences may represent different distances of the fluorescent tags fused to the NH2 or COOH termini of p101 and p110 in a particular combination. However, the different FRET efficiencies may also reflect different degrees of heterodimerization. To exclude the latter, we analyzed heterodimerization of different p101/p110 combinations by size-exclusion chromatography The elution profile of YFP-p110 ( bottom) was completely shifted to earlier fractions on coexpression with CFP-p101 ( top), indicating a high degree of heterodimerization. A similar elution profile was found for p110-YFP + CFP-p101 ( middle), albeit a lower FRET efficiency was detected for this combination. Hence, the different FRET efficiencies observed are not due to different degrees of dimerization, but likely reflect different distances of the fluorescent tags in the p101/p110 heterodimer. Thus, their two NH2 termini as well as their two COOH termini are in close proximity, whereas the NH2 terminus of each subunit is relatively far from the COOH terminus of the other.
Gß{gamma} dimers recruit PI3K to membranes via the p101 subunit6og{!, 百拇医药
Current concepts of PI3K activation are based on the recruitment of the cytosolic lipid kinase to the plasma membrane . In the case of PI3K, the major stimulus is assumed to be Gß, which is membrane-bound and has been shown to directly bind to both kinase subunits under in vitro conditions. Therefore, we asked whether overexpression of free Gß directs YFP-fused PI3K subunits to the cell membrane in vivo. Surprisingly, p110 was not membrane-localized in the presence of coexpressed Gß in HEK cells ( A, top). In contrast, Gß recruited p101 to the membrane, resulting in accumulation of NH2- or COOH-terminally YFP-tagged p101 at both plasma- and endomembranes ( A, bottom, and B). This corresponds to the subcellular distribution of overexpressed Gß dimers (see below), suggesting that p101 interacted with membrane-bound Gß.6og{!, 百拇医药
fig.ommtted
Membrane recruitment of PI3K by Gß. The effect of coexpression of Gß on the subcellular distribution of p110, p101, and heterodimeric PI3K in HEK cells was analyzed by confocal laser scanning microscopy. Images of typical cells are shown (white bars, 10 µm). (A) Coexpression of monomeric PI3K subunits with Gß. (B) Controls; coexpression of YFP-p101 with Gß1, G2, Gß12, or Gß12 and the Gß- scavenging Gi2. (C) Coexpression of heterodimeric PI3K with Gß. Note the round shape of the cells. Controls; coexpression of Gß with YFP or with kinase-deficient YFP-p110-K833R and p101; effect of 100 nM wortmannin. (D) Subcellular localization of Gß. Cells were transfected with a plasmid encoding CFP-tagged Gß1 alone or together with the G2 plasmid. (E) Immunoblot analysis of membrane fractions. HEK cells were transfected with the plasmids encoding PI3K, Gß, or both together. To avoid rounding and detachment of the cells, the kinase-deficient mutant YFP-p110-K833R was used. Membrane fractions were prepared and analyzed by immunoblotting with anti-p110 and anti-Gß antibodies.
Next, we examined the subcellular localization of p110/p101 heterodimers in the presence of free Gß{gamma} . When coexpressed with p101 and Gß, the p110 fluorescent signal was at, or close to, plasma and endosomal membranes of HEK cells ( C, top). The same picture emerged when the YFP-tag was fused to p101 (unpublished data). Correspondingly, CFP-tagged Gß1 coexpressed with G2 exhibited a similar subcellular distribution like fluorescent heterodimeric PI3K (compare C and 4 D), and immunoblotting of membrane-derived proteins confirmed an enrichment of PI3K{gamma} in the membrane compartment after overexpression of Gß . Together, these data imply that Gß directs cytosolic PI3K to the membrane via interaction with p101.fe-p)t, 百拇医药
Interestingly, coexpression of heterodimeric PI3K together with Gß produced a rounded morphology and detachment of the cells This morphological change was reversed by treatment with 100 nM of the PI3K inhibitor wortmannin in approximately half of the cells within 30 min, and not seen when a kinase-deficient YFP-p110-K833R mutant was used . Hence, the morphological change was related to the enzymatic activity of PI3K{gamma} stimulated by the coexpressed Gß.
p101 is required for G protein–mediated activation of p110 in vivo-$0}, 百拇医药
Because changes in morphology are not a very reliable measure of PI3K activity, we examined phosphorylation of endogenous protein kinase B (PKB or Akt) as an established PI3K-specific read-out system. To avoid detachment of the cells as a consequence of constitutive PI3K stimulation by overexpressed Gß, we transiently stimulated the heterologously expressed Gi-coupled formyl-methionyl-leucyl-phenylalanine (fMLP) receptor, which is known to activate PI3K . Stimulation with fMLP induced Akt phosphorylation in HEK cells in the same range as with FCS ( A). However, Akt phosphorylation was only seen when the receptor was coexpressed with both PI3K subunits, and not with p110 alone. These data imply that p101 is required for GPCR-induced activation of PI3K in vivo.-$0}, 百拇医药
fig.ommtted-$0}, 百拇医药
Activation of PI3K by a G protein–coupled receptor. (A) Akt phosphorylation. Whole-cell lysates were analyzed by immunoblotting using an antibody that specifically recognizes the phosphorylated form of Akt. Equal loading was shown by using an anti-ERK antibody. Left; FCS-induced Akt phosphorylation in untransfected HEK cells. Right; fMLP-induced Akt phosphorylation in cells expressing the fMLP receptor and only the catalytic (p110) or both PI3K subunits. (B) Membrane recruitment of the PtdIns-3,4,5-P3–binding PH domain of GRP1. HEK 293 cells were transfected with plasmids encoding the PH domain of GRP1 fused to GFP (GFP-GRP1PH), and the human fMLP receptor (fMLP-R), p110, and p101 in different combinations. The localization of the GFP-GRP1PH was monitored before and after the addition of 1 µM fMLP and 100 ng/ml EGF (added 8 min later) by confocal laser scanning microscopy. Pictures were taken 4 min after addition of either agonist. Images of typical experiments are shown (white bars, 10 µm). (C) Membrane recruitment of the PtdIns-3,4,5-P3–binding PH domain of Btk. Instead of GFP-GRP1PH, the PH domain of Btk fused to CFP (BtkPH-CFP) was used as a PtdIns-3,4,5-P3 sensor.
To measure PI3K activity by a complementary experimental approach, the effects of fMLP receptor stimulation on the subcellular distribution of GFP-GRP1PH fusion proteins were examined. Recruitment of the GFP-labeled PH domain of general receptor for phosphoinositides-1 (GRP1), a PI3K-activated exchange factor for ADP-ribosylation factors , has been applied to detect formation of PtdIns-3,4,5-P3 with high affinity and selectivity . fMLP stimulation had no effect on the subcellular distribution of GFP-GRP1PH in HEK cells expressing either the receptor or PI3K alone ( B, top two panels), whereas subsequent stimulation with EGF induced a rapid mmbrane recruitment of the PtdIns-3,4,5-P3 sensor, presumably through activation of an endogenously expressed RTK-sensitive class IA PI3K. In cells coexpressing the fMLP receptor and heterodimeric PI3K, fMLP stimulation induced a rapid membrane translocation of GFP-GRP1PH ( B, third panel), indicating that this effect indeed reflects fMLP receptor–mediated stimulation of PI3K. Accordingly, the fMLP-induced effect was sensitive to pretreatment with pertussis toxin (PTX; unpublished data). Interestingly, a slight basal membrane localization of GFP-GRP1PH was visible in the cells coexpressing heterodimeric PI3K and the fMLP receptor even in the absence of agonist ( B, third panel). Because this was not seen after PTX treatment (unpublished data) or in cells omitting the receptor ( B, second panel), it most likely reflected a constitutively active fMLP receptor.
Translocation of an isolated PH domain in single cells may be a more sensitive read-out for PtdIns-3,4,5-P3 production than the Akt phosphorylation assay. Hence, we used the GFP-GRP1PH translocation approach to reexamine the requirement of p101 for receptor-induced PI3K activation . Nevertheless, an fMLP-induced membrane recruitment of GFP-GRP1PH in cells coexpressing the receptor and p110 (but not p101) was not visible ( B, fourth panel). This finding confirms that the noncatalytic p101 subunit is required for GPCR-mediated stimulation of PI3Kin living cells.i3f&!9d, 百拇医药
To further strengthen the findings, we replaced GFP-GRP1PH by the GFP-fused PH domain of Bruton's tyrosine kinase (BtkPH), which has also been described to bind PtdIns-3,4,5-P3 with high specificity and affinity . As expected, fluorescent BtkPH was distributed equally between cytosol and nucleus in HEK cells . However, cells coexpressing the receptor and both PI3K subunits exhibited a slight basal membrane localization of BtkPH-CFP followed by a prominent translocation from the cytosol to the membrane on exposure to fMLP. Similar to GFP-GRP1PH, no fMLP-induced redistribution of BtkPH-CFP was seen in the absence of p101.
Activation of a membrane-targeted p110-CAAX in vivoth?, http://www.100md.com
The presented data suggest that in living cells, p101 is an indispensable adaptor for GPCR-induced translocation and activation of class IB PI3K, which is equivalent to the role of p85 in RTK-induced class IA PI3K activation. Vice versa, Gß behaves as a membrane anchor recruiting PI3K through association with p101. So far, these experiments did not clarify whether membrane recruitment itself is sufficient for activation of the enzyme or whether additional allosteric stimulation is required. To tackle this question, we generated p110 mutants containing a COOH-terminal isoprenylation signal, i.e., a CAAX-box motif, which constitutively localizes p110 to the plasma membrane. p110-CAAX accumulated at the plasma membrane of HEK cells, and was able to complex with p101 (, top). Coexpression of Gß{gamma} together with YFP-p110-CAAX increased not only fluorescent staining of endomembranes, but also produced a wortmannin-sensitive rounding of the cells (, bottom) even in the absence of p101 (not depicted), which was similar to the effect after coexpression of Gß with p110/p101 dimers . These observations imply that membrane-bound PI3K is not fully active, but can be further activated by Gß.
fig.ommtted7, 百拇医药
Characterization of a constitutively membrane-associated p110-CAAX. HEK cells were transfected with the indicated plasmids and analyzed by confocal laser scanning microscopy. Images of typical cells are shown. White bars indicate a 10-µm scale. Top panel; Subcellular localization of YFP-p110-CAAX (left) or YFP-p101 expressed together with p110-CAAX (right). Bottom panel; Coexpression with Gß.7, 百拇医药
To substantiate this assumption, we analyzed the subcellular redistribution of GFP-GRP1PH in HEK cells coexpressing membrane-targeted p110-CAAX, p101 and the fMLP receptor in various combinations . A slight basal membrane localization of the fluorescent PH domain was evident on coexpression with p110-CAAX ( A, top panel) indicating an elevated PtdIns-3,4,5-P3 level. The same picture emerged when p101 was present in the panel of transfected plasmids ( A, second panel). On coexpression of the receptor, fMLP induced a significant additional translocation of GFP-GRP1PH from the cytosol to the membrane ( A, third and fourth panel) indicating additional stimulation of the membrane-bound lipid kinase. Interestingly, the effect was similar irrespective of whether p101 was present or not, suggesting direct activation of monomeric p110. PTX sensitivity confirmed that fMLP-induced activation of p110-CAAX was mediated by Gi proteins (unpublished data). Because Gß can activate Ras, which in turn may activate p110, one may imagine that fMLP-induced, p101-independent stimulation of p110-CAAX was mediated by Ras. However, coexpression of dominant-negative Ras N17 had no effect on fMLP-induced p110-CAAX stimulation, whereas its inhibitory effect was evident when EGF-induced MAP kinase activation was assessed ( A, bottom panel). In addition, an indirect autocrine stimulatory effect was excluded because supernatants of fMLP-stimulated cells failed to induce GFP-GRP1PH translocation in cells omitting fMLP receptors (unpublished data).
fig.ommttedq^%[25, 百拇医药
Activation of a constitutively membrane-associated p110-CAAX. (A) GFP-GRP1PH translocation. HEK cells were transfected with plasmids for GFP-GRP1PH and p110-CAAX, p101, fMLP-R, and dominant-negative Ras N17 in different combinations. Translocation of the PtdIns-3,4,5-P3–binding GFP-GRP1PH in response to agonist stimulation was monitored as described above (see ). White bars indicate a 10-µm scale. As a positive control for H-Ras N17, cells were transfected with the H-Ras N17 plasmid (same amount as for the GFP-GRP1PH translocation experiment) or empty vector. Cells were stimulated with 10 ng/ml EGF, and whole-cell lysates were analyzed by immunoblotting with an antibody that specifically recognizes the phosphorylated form of ERK (p-ERK). Equal loading was shown using the anti-ERK-antibody. (B) BtkPH-CFP translocation. (C) Akt phosphorylation. The experiment shown in was repeated. In addition, cells were transfected with the plasmid for the membrane-targeted p110-CAAX instead of wild-type p110.q^%[25, 百拇医药
By using BtkPH-CFP instead of GFP-GRP1PH as the PtdIns-3,4,5-P3 sensor, we confirmed both the basal enzymatic activity and fMLP-induced stimulation of membrane-targeted p110-CAAX, regardless of coexpressed p101. Furthermore, the Akt phosphorylation assay corroborated that p101 was dispensable for GPCR-induced stimulation of membrane-targeted p110-CAAX .
Next, we sought to confirm our principal findings using nontransformed cells in primary culture with a higher degree of differentiation than HEK cells. Vascular smooth muscle (VSM) cells were injected with plasmids encoding the fMLP receptor, p101, p110 or p110-CAAX, and GFP-GRP1PH in different combinations, and redistribution of the fluorescent PtdIns-3,4,5-P3 sensor was monitored. Again, this series of experiments confirmed that p101 is required for GPCR-mediated activation of PI3K{gamma} in living cells, whereas membrane-targeted p110 can be stimulated even in the absence of p101.yg'}m, 百拇医药
fig.ommttedyg'}m, 百拇医药
PI3K-mediated membrane translocation of GFP-GRP1PH in vascular smooth muscle (VSM) cells. VSM cells were microinjected with plasmids for GFP-GRP1PH, fMLP-R, p101, and p110 or p110-CAAX in different combinations. The localization of GFP-GRP1PH was monitored before and after the addition of 1 µM fMLP (picture taken after 4 min) and 10 ng/ml PDGF-BB (added 12 min later, picture taken after 8 min). White bars indicate a 10-µm scale.
Discussionc:, http://www.100md.com
PI3Kis a key player in the regulation of leukocyte functions such as chemotaxis and superoxide production . To elucidate the mechanism of how Gß activates the enzyme in vivo, we studied the subcellular localization and translocation of GFP-tagged PI3K subunits and PH domains in living cells.c:, http://www.100md.com
A p110-GFP construct has been previously reported , but data describing a functional interaction of the fusion protein with G proteins are not available. Therefore, we systematically tagged each PI3K subunit and found heterodimerization, enzymatic activity, and Gß-sensitivity of the fusion proteins unaffected by the tags, whereas a GFP-tagged Gß complexed to G did not interact with PI3K (unpublished data). In resting cells, fluorescent PI3K was predominantly detected in the cytosol. This resembles the native situation in which the vast majority of the endogenous PI3K pool has been assigned to the cytosol . Thus, the GFP-tagged proteins are appropriate tools to study the cellular events leading to activation of PI3K.
Recent analysis of p110 crystals gave insights into the three-dimensional structure of the catalytic PI3K subunit . However, the structure of p101 and the molecular determinants for the interaction of both PI3K subunits are currently unknown. In vitro data derived from p101 and p110 deletion mutants suggest that large areas of p101 may interact with the NH2-terminal side of p110 . FRET data presented here indicate that NH2 and both COOH termini of p110 and p101 are in close proximity. The latter finding is of interest because the p110 COOH terminus harbors the catalytic domain. Previous data have implied that not only Ras , but also Gß directly interact with the COOH terminus of p110. Therefore, interaction of p101 with the catalytic domain of p11 would be in line with the functional data suggesting that p101 affects interaction of Gß-stimulated p110 with the lipid interface .8ix4s, http://www.100md.com
Gß is considered to be the principal, direct stimulus of GPCR-induced PI3K enzymatic activity. Convincing evidence for this assumption has come from reconstitution of purified proteins, demonstrating that in vitro, the enzymatic activity of heterodimeric and monomeric PI3K is significantly stimulated by Gß . However, when we challenged the concept in living cells, marked differences in the sensitivities between heterodimeric and monomeric PI3K emerged. The G protein–coupled fMLP receptor activated p110/p101, but not p110, as evident from the translocation of fluorescent PtdIns-3,4,5-P3 sensors or the stimulation of Akt phosphorylation. Likewise, coexpressed Gß recruited p110/p101, but not p110, to membranes of HEK cells. Interestingly, PI3Kß, another Gß-regulated PI3K, did not translocate to the membrane on GPCRs or Gß stimulation, whereas RTKs induced membrane recruitment in vivo (unpublished data).
Does membrane recruitment of PI3K by itself result in constitutive activation of p110? To answer this intriguing question, we used an isoprenylated mutant of p110, i.e., p110-CAAX, which is permanently attached to the membrane, and thereby, is in close proximity to its lipid substrates. Expression of p110-CAAX only slightly enhanced membrane association of fluorescent PtdIns-3,4,5-P3 sensors in HEK or VSM cells. Notably, it did not affect the morphology of HEK cells. In contrast, coexpression of Gß together with p110-CAAX induced morphological changes comparable to cells transfected with plasmids encoding Gß and p110/p101. Furthermore, stimulation of the cells with fMLP stimulated Akt phosphorylation and membrane recruitment of PH domains regardless of whether p110/p101 or p110-CAAX was the effector. This establishes a second role for Gß, i.e., the activation of the membrane-attached catalytic p110subunit even in the absence of p101 . Although Gß{gamma} interacts with either monomeric PI3K subunit through individual binding sites, it does not exclude the possibility that heterodimeric PI3K forms either a common or different binding site(s) for Gß. Attempts to determine the stoichiometry of the interaction between heterodimeric PI3K and one or more Gßwere inconclusive. One possible reason may be a weak affinity between p110 and Gß, whereas p101 binds at least with moderate affinity to Gß as determined by copurification studies and by a Biacore plasmon resonance approach (unpublished data).
fig.ommttedm!'d, 百拇医药
Hypothetical model for receptor- induced membrane recruitment and activation of heterodimeric PI3K. fMLP receptor, a prototypical heptahelical receptor coupled to Gi proteins. PI3K; the cytosolic enzyme consists of a noncatalytic p101 subunit, which is in a tight complex with p110, thereby stabilizing p101. The NH2 and COOH termini of p101 and p110 are oriented in close proximity, respectively. Contact sites involve the NH2 termini of both subunits. The modular domain structure of p110 (RBD, Ras binding domain; C2, C2 domain; hel, helical domain; N-cat, C-cat, NH2- and COOH-terminal lobes of the catalytic domain) is based on the crystal structure of an NH2-terminally truncated p110. (A) Membrane recruitment. The agonist-stimulated receptor induces the release of Gß from Gi proteins (dotted arrow). Gß recruits the PI3K heterodimer to the plasma membrane (dotted arrow) by binding to the noncatalytic p101 (hollow arrow). Accordingly, Gß and p101 function as a membrane anchor and an adaptor for PI3K, respectively. In addition, the C2 domain of p110 may facilitate membrane attachment through interaction with phospholipids. p101 may also affect the interaction of PI3K with the lipid interface. Membrane-attached PI3K exhibits basal enzymatic activity. (B) Allosteric activation. At the membrane, Gß activates PI3K by direct interaction with p110(hollow arrows). This stimulation does not require p101. However, p101 may participate in Gß-induced stimulation of membrane-attached p110. The stoichiometry of the Gß–PI3K interaction is unknown, i.e., Gß may interact with p101 and p110 through individual or common binding sites. Gß binds to the NH2- and COOH-terminal part of p110, the latter harboring the catalytic domain. Thus, Gß may allosterically activate PI3K through a conformational change in the catalytic domain. Accordingly, Gß significantly increases Vmax of PtdIns-3,4,5-P3 production. PtdIns-3,4,5-P3, in turn, recruits PH domain–containing effectors such as GRP1 or Btk to the plasma membrane (dotted arrow).
In our paper, we present evidence that p110 is stable as a monomer without p101, but not vice versa. This may explain why relevant amounts of native monomeric p101 do not occur in vivo (Wetzker, R., personal communication). p110 can be expressed as a stable protein independent of the presence of p101, as shown previously . Interestingly, the p110 subunit of class IA PI3Ks needs stabilization by forming a complex with the p85 adaptor Accordingly, class IA p110{alpha} ,ß,{delta} subunits are thought to occur only in their adaptor-bound form. Moreover, the p85 adaptor is more abundant than the p110 counterparts In contrast, p110 is also suggested to occur in the absence of p101 .1'tv, http://www.100md.com
Increasing evidences suggest that monomeric p110 may also function as a downstream regulator of GPCR-dependent signal transduction pathways in cells . However, under in vivo conditions the p101 subunit seems mandatory for G protein–mediated activation of PI3K by mediating membrane recruitment. Therefore, the question arises of how monomeric p110 translocates to the membrane. In this context, recent in vitro evidence points to the possibility that membrane attachment of p110 may involve binding to anionic phospholipids . In line with this finding, the majority of PI3K is already associated with lipid vesicles in the absence of Gß under in vitro conditions, and vesicle-bound PI3K can still be activated by Gß . Alternatively, p110 may be recruited by binding to membrane-anchored proteins other than Gß. In this respect, p110, like all other class I PI3Ks, is directly activated by GTP-bound Ras
In conclusion, we present in vivo evidence that Gß may stimulate PI3K in a dual and complementary way, i.e., by recruitment to the membrane, and by activation of the membrane-bound enzyme . In this scenario, the p101 subunit functions as an adaptor molecule necessary to recruit the catalytic subunit to the plasma membrane through high affinity interaction with Gß. In turn, direct interaction between Gß and the membrane-attached catalytic p110 subunit contributes to a final activation of the enzyme by a mechanism other than translocation.+zc67, http://www.100md.com
Materials and methods+zc67, http://www.100md.com
Construction of expression plasmids+zc67, http://www.100md.com
For generation of expression plasmids of PI3K subunits, restriction sites were introduced, and stop codons were removed by 8–12 cycles of PCR using the indicated primers, a PCR system (Expand HF; Roche), and a pcDNA3.1/V5-His-TOPO vector for first subcloning. Subcloned PCR fragments were confirmed by sequencing with fluorescent dye terminators (ABI-Prism 377; ). Expression plasmids encoding NH2- and COOH-terminally tagged fusion proteins were generated by subsequent subcloning into the vectors pEYFP-C1 and pECFP-C1 or into pcDNA3-YFP and pcDNA3-CFP , respectively. The constructs are designated indicating the color and the position of the tag In detail, wild-type p110; human p110 cDNA ( a corrected nucleotide sequence for p110 is available from GenBank/EMBL/DDBJ, accession no. ) was amplified with the primers 5'-GCC ACC ATG GAG CTG GAG AAC TAT AA-3' and 5'-GGA TCC AGC TTT CAC AAT GTC TAT TG-3', and subcloned into pcDNA3 via KpnI and XhoI. p110-YFP and -CFP; the stop codon was replaced by an XbaI site using the primer 5'-GTC TAG AGC TGA ATG TTT CTC TCC CTT GT-3' and the same forward primer as above. Subcloning into pcDNA3-YFP and -CFP was done via BamHI and XbaI. YFP- and CFP-p110; XhoI and BamHI sites were introduced using the primers 5'-CTC GAG GCA TGG AGC TGG AGA ACT A-3' and 5'-GGA TCC AGC TTT CAC AAT GTC TAT TG-3' for subcloning into pEYFP-C1 and pECFP-C1. p110-K833R; position 2498 was mutated from A to G using the QuikChange® Mutagenesis Kit and appropriate primers. p110-CAAX; two AfeI sites in p110 were removed, and the stop codon was replaced by a new AfeI site by silent mutations using the QuikChange® Mutagenesis Kit and appropriate primers. Adaptor oligonucleotides encoding the 18 COOH-terminal amino acids of H-Ras were inserted into the AfeI and BamHI sites. Wild-type p101; the cDNA for porcine p101 was subcloned in pcDNA3 via EcoRI and NotI. An optimized ribosomal docking sequence was introduced by PCR using the primers 5'-GCC ACC ATG CAG CCA GGG GCC ACG GA-3' and 5'-GGC CCG AGA CGA AGG AGG T-3' and subsequent exchange of the 5'-end via HindIII and BsmBI. YFP- and CFP-p101; the EcoRI/NotI fragment of p101 was subcloned into EcoRI/Bsp120I-digested pEYFP-C1 or pECFP-C1. To adapt the reading frames, the HindIII site was blunted with Klenow fragment and religated. p101-YFP and -CFP; PCR with the primers 5'-GTC CTC TCC TCA CAC GGT TCT T-3' and 5'-GTC TAG AGG CAG AGC TCC GCT GAA AGT-3' generated the 3'-end of p101 with an XbaI site instead of the stop codon. To restore the full-length p101 cDNA, the 5'-part was excised from wild-type p101 in pcDNA3 with HindIII and ClaI (partial digest) and ligated to the HindIII/ClaI-digested 3'-end. Subsequent subcloning in pcDNA3-YFP and pcDNA3-CFP was done via HindIII and XbaI.
The human fMLP receptor cDNA was amplified with the primers 5'-GCC ACC ATG GAG ACA AAT TCC TCT CTC-3' and 5'-TCA CTT TGC CTG TAA CTC CAC-3' and subcloned in pcDNA3 via HindIII and XhoI. The cDNAs of Gi2 , human Gß1 and bovine G2 were subcloned into pcDNA3. Plasmids for GFP-GRP1PH and Ras N17 are described elsewhere .19t]]i5, http://www.100md.com
For generation of expression plasmids encoding CFP-tagged Gß1 and BtkPH, restriction sites were introduced by PCR using the indicated primers, the AdvantageTM II PCR enzyme system and the pGEM®-T Easy Vector for first subcloning. The Gß1 cDNA was amplified using the primers 5'-TAC AAG TCC GGA CAA GCT TCC ATG AGT GAG CTT GAC CAG TTA CGG C-3' and 5'-CGG GAT CCG TCG ACC CAT GGT GGC GTT AGT TCC AGA TCT TGA GGA AGC-3', allowing for in-frame subcloning into the HindIII and BamHI sites of pECFP-C1. The cDNA encoding the PH domain of human Btk and adjacent 5' untranslated bases was amplified from cDNA of dibuturyl-cAMP–differentiated HL-60 cells using the primers 5'-CCA AGT CCT GGC ATC TCA ATG CAT CTG-3' and 5'-TGG AGA CTG GTG CTG CTG CTG GCT C-3'. A nested PCR was performed using the primers 5'-GGA AGA TCT CGA GCC ACC ATG GCC GCA GTG ATT CTG G-3' and 5'-GGG GAT CCC GGG CCC GAG GTT TTA AGC TTC CAT TCC TGT TCT CC-3', allowing for in-frame subcloning into the XhoI and BamHI sites of pECFP-N1 The cDNA inserts and flanking regions of the resulting CFP-Gß1 and BtkPH-CFP constructs were confirmed by sequencing.
Cell culture, transfection, and intranuclear microinjection1@ze, http://www.100md.com
HEK 293 cells were grown at 37°C with 5% CO2 in DME or in MEM with Earle's salts supplemented with 10% FCS, 100 µg/ml streptomycin, and 100 U/ml penicillin. All transfections were done with a FuGENETM 6 transfection reagent (Roche) following the manufacturer's recommendations. For fluorescence microscopy experiments, cells were seeded on glass coverslips. For confocal imaging of the subcellular localization of the PI3K subunits, cells were transfected with 0.1 µg of plasmid encoding YFP-tagged PI3K subunits, 0.2 µg of plasmid for the untagged complementary subunit, and 1 µg each of plasmid for Gß1 and G2 (except for the experiment shown in B; 0.1 µg YFP-p101 + 2 µg Gß1 or G2 alone, or 0.5 µg Gß1 + 0.5 µg G2 + 1 µg Gi2). For monitoring PH domain translocation, or Akt- or ERK phosphorylation, HEK cells were transfected with 0.2 µg plasmid encoding the fMLP receptor, and 0.4 µg each of the plasmids encoding the PI3K subunits and the fluorescent PH domain. The total amount of transfected cDNA was always kept constant (2.5 µg/well) by the addition of empty expression vector. For FRET analysis, cells were transfected with 1.5 µg of plasmid encoding the YFP-tagged p110 and 0.5 µg of plasmid for CFP-tagged p101. All experiments were performed one or two days after transfection.
VSM cells from neonatal rat aortas were cultured as described previously . Cells were seeded on glass coverslips and microinjected with the plasmids for GFP-GRP1PH (0.4 µg/ml), fMLP receptor (0.05 µg/ml), p110 (0.05 µg/ml), and p101 (0.05 µg/ml) using an Eppendorf micromanipulator and transjector. Confocal images were taken one day after microinjection.2j, 百拇医药
Immunoblot analysis of whole-cell lysates, gel filtration, and membrane fractions2j, 百拇医药
p110/p101 titration; transfected HEK cells were lysed in Laemmli buffer and whole-cell lysates including nuclei were subjected to a 10% SDS-PAGE. Proteins were blotted on nitrocellulose membranes, probed with anti-GFP and peroxidase-coupled anti–rabbit IgG antibodies and visualized by ECL . Akt and ERK phosphorylation; HEK 293 cells were transfected as indicated with the plasmids encoding the fMLP receptor (0.2 µg), PI3K subunits (each 0.4 µg), and H-Ras N17 (1.5 µg). Cells were serum-starved overnight, then stimulated as indicated, and lysed in Laemmli buffer. Whole-cell lysates were subjected to SDS-PAGE and immunoblotting, and probed with anti–Phospho-Akt or anti–Phospho-ERK1/2 and anti-ERK1/2 as described previously . Gel filtration; HEK cells were transfected with equal amounts of the plasmids for YFP-p110 (or p110-YFP) and CFP-p101 (or pcDNA3). Cells were washed in PBS and lysed in 20 mM Hepes, pH 7.5, 150 mM NaCl, 10 mM ß-mercaptoethanol, 4 mM EDTA, 0.2% polyoxyethylene-10-laurylether (C12E10), 200 µM Pefabloc® SC, 30 µg/ml TPCK, 30 µg/ml trypsin inhibitor, and 50 µg/ml benzamidine by repetitive aspiration through a 26-gauge needle. Cytosols were prepared (100,000 g for 30 min at 4°C), and subjected to gel filtration on a 24-ml Superdex 200 column and eluted with the same buffer on an ÄKTATM purifier system . Fractions were collected, and aliquots were subjected to SDS-PAGE and immunoblotting with anti-GFP. Preparation of membrane fractions; HEK cells were transfected as indicated with the plasmids for YFP-p110-K833R (0.1 µg), p101 (0.2 µg), Gß1 (1.0 µg), and G2 (1.0 µg). Cells were washed with PBS and lysed in PBS with protease inhibitors (see above) by repetitive aspiration through a 26-gauge needle. Membranes were prepared (16,000 g for 20 min at 4°C), redissolved in Laemmli buffer, subjected to SDS-PAGE and immunoblotting, and probed with anti-p110 (see next section) and anti-Gßcommon (AS398; .
In vitro PI3K assay'', 百拇医药
HEK cells were transfected with equal amounts of the plasmids encoding p110 and p101 (or empty pcDNA3 to keep the total amount of transfected cDNA constant). Cell lysates were subjected to immunoprecipitation using a monoclonal anti-p110 antibody . In brief, protein A Sepharose was preincubated with or without antibody, washed, incubated overnight with cleared lysates, and washed again. The in vitro PI3K assay was performed as described previously . Radioactive PtdIns-3,4,5-P3 was visualized with a PhosphorImager (Fuji Bas-Reader 1500; Ray Test).'', 百拇医药
Fluorescence imaging and detection of FRET'', 百拇医药
Glass coverslips were mounted on a custom-made chamber and covered with a Hepes-buffered solution containing 138 mM NaCl, 6 mM KCl, 1 mM MgCl2, 1 mM CaCl2, 5.5 mM glucose, 10 mM Hepes, pH 7.5, and 2 mg/ml BSA. For confocal imaging, an inverted confocal laser scanning microscope with a Plan-Apochromat 63x/1.4 objective (model LSM 510; Carl Zeiss MicroImaging, Inc.) was used. YFP or GFP were excited at 488 nm, and fluorescence emission was detected through a 505-nm long pass filter. CFP was excited 458 nm, and detected through a 475-nm long pass filter. Pinholes were adjusted to yield optical sections of 0.5–1.5 µm.
FRET analysis was performed using an inverted microscope with a Plan-Apochromat 63x/1.4 objective (Axiovert 100; Carl Zeiss MicroImaging, Inc.). CFP and YFP were alternately excited at 440 and 500 nm with a monochromator (Polychrome II; TILL Photonics) in combination with a dual reflectivity dichroic mirror (<460 nm and 500–520 nm; ). Emitted light was filtered through 475–505-nm (CFP) and 535–565-nm (YFP) band pass filters, changed by a motorized filter wheel (Lambda 10/2; and detected by a cooled CCD camera (Imago; TILL Photonics). FRET was assessed as recovery of CFP (donor) fluorescence during YFP (acceptor) bleach. First, CFP and YFP fluorescences without acceptor bleach were recorded during 30 cycles with a 10–20-ms exposure. Then, 60 cycles were recorded with an additional 2-s illumination per cycle at 510 nm to bleach YFP.o-n\cpc, http://www.100md.com
Acknowledgmentso-n\cpc, http://www.100md.com
Thanks to Claudia Plum, A. Schulz and J. Malkewitz for technical assistance. We are grateful to Drs. P. Gierschik, A. Gray and P. Downes, M.I. Simon, L. Stephens and P. Hawkins, and R. Wetzker for providing experimental tools. Thanks to Celsa Antao-Paetsch for assistance in preparation of the manuscript. We appreciate valuable discussions with Drs. Len Stephens, Phil Hawkins, Lewis Cantley, Reinhard Wetzker, and Roland Piekorz.
This work was supported by the Deutsche Forschungsgemeinschaft and Fonds der Chemischen Industrie.j)r, 百拇医药
Submitted: 21 October 2002j)r, 百拇医药
Revised: 2 December 2002j)r, 百拇医药
Accepted: 2 December 2002j)r, 百拇医药
Referencesj)r, 百拇医药
Al-Aoukaty, A., B. Rolstad, and A.A. Maghazachi. 1999. Recruitment of pleckstrin and phosphoinositide 3-kinase into the cell membranes, and their association with Gß after activation of NK cells with chemokines. J. Immunol. 162:3249–3255.j)r, 百拇医药
Balla, T., T. Bondeva, and P. Várnai. 2000. How accurately can we image inositol lipids in living cells? Trends Pharmacol. Sci. 21:238–241.j)r, 百拇医药
Baier, R., T. Bondeva, R. Klinger, A. Bondev, and R. Wetzker. 1999. Retinoic acid induces selective expression of phosphoinositide 3-kinase in myelomonocytic U937 cells. Cell Growth Differ. 10:447–456.j)r, 百拇医药
Bondev, A., T. Bondeva, and R. Wetzker. 2001. Regulation of phosphoinositide 3-kinase {gamma} protein kinase activity in vitro and in COS-7 cells. Signal Transduction. 1:79–85.
Bondeva, T., L. Pirola, G. Bulgarelli-Leva, I. Rubio, R. Wetzker, and M.P. Wymann. 1998. Bifurcation of lipid and protein kinase signals of PI3K{gamma} to the protein kinases PKB and MAPK. Science. 282:293–296..fg%&, 百拇医药
Boulay, F., M. Tardif, L. Brouchon, and P. Vignais. 1990. The human N-formylpeptide receptor. Characterization of two cDNA isolates and evidence for a new subfamily of G-protein-coupled receptors. Biochemistry. 29:11123–11133..fg%&, 百拇医药
Cantley, L.C. 2002. The phosphoinositide 3-kinase pathway. Science. 296:1655–1657..fg%&, 百拇医药
Codina, J., D. Stengel, S.L. Woo, and L. Birnbaumer. 1986. ß-subunits of the human liver Gs/Gi signal-transducing proteins and those of bovine retinal rod cell transducin are identical. FEBS Lett. 207:187–192..fg%&, 百拇医药
Conklin, B.R., Z. Farfel, K.D. Lustig, D. Julius, and H.R. Bourne. 1993. Substitution of 3 amino acids switches receptor specificity of Gq to that of Gi. Nature. 363:274–276..fg%&, 百拇医药
Fruman, D.A., R.E. Meyers, and L.C. Cantley. 1998. Phosphoinositide kinases. Annu. Rev. Biochem. 67:481–507.
Gautam, N., M. Baetscher, R. Aebersold, and M.I. Simon. 1989. A G protein {gamma} subunit shares homology with ras proteins. Science. 244:971–974.w\sz}6., 百拇医药
Gillham, H., M.C. Golding, R. Pepperkok, and W.J. Gullick. 1999. Intracellular movement of green fluorescent protein-tagged phosphatidylinositol 3-kinase in response to growth factor receptor signaling. J. Cell Biol. 146:869–880.w\sz}6., 百拇医药
Gray, A., J. Van Der Kaay, and C.P. Downes. 1999. The pleckstrin homology domains of protein kinase B and GRP1 (general receptor for phosphoinositides-1) are sensitive and selective probes for the cellular detection of phosphatidylinositol 3,4-bisphosphate and/or phosphatidylinositol 3,4,5-trisphosphate in vivo. Biochem. J. 344:929–936.w\sz}6., 百拇医药
Hurley, J.H., and S. Misra. 2000. Signaling and subcellular targeting by membrane-binding domains. Annu. Rev. Biophys. Biomol. Struct. 29:49–79.w\sz}6., 百拇医药
Katso, R., K. Okkenhaug, K. Ahmadi, S. White, J. Timms, and M.D. Waterfield. 2001. Cellular function of phosphoinositide 3-kinases: implications for development, homeostasis, and cancer. Annu. Rev. Cell Dev. Biol. 17:615–675.
Kirsch, C., R. Wetzker, and R. Klinger. 2001. Anionic phospholipids are involved in membrane targeting of PI3-kinase . Biochem. Biophys. Res. Commun. 282:691–696.oud4, http://www.100md.com
Klippel, A., C. Reinhard, W.M. Kavanaugh, G. Apell, M.A. Escobedo, and L.T. Williams. 1996. Membrane localization of phosphatidylinositol 3-kinase is sufficient to activate multiple signal-transducing kinase pathways. Mol. Cell. Biol. 16:4117–4127.oud4, http://www.100md.com
Krugmann, S., P.T. Hawkins, N. Pryer, and S. Braselmann. 1999. Characterizing the interactions between the two subunits of the p101/p110 phosphoinositide 3-kinase and their role in the activation of this enzyme by Gßsubunits. J. Biol. Chem. 274:17152–17158.oud4, http://www.100md.com
Krugmann, S., M.A. Cooper, D.H. Williams, P.T. Hawkins, and L.R. Stephens. 2002. Mechanism of the regulation of type IB phosphoinositide 3OH-kinase by G-protein ß{gamma} subunits. Biochem. J. 362:725–731.oud4, http://www.100md.com
Kurosu, H., T. Maehama, T. Okada, T. Yamamoto, S. Hoshino, Y. Fukui, M. Ui, O. Hazeki, and T. Katada. 1997. Heterodimeric phosphoinositide 3-kinase consisting of p85 and p110ß is synergistically activated by the ß subunits of G proteins and phosphotyrosyl peptide. J. Biol. Chem. 272:24252–24256.
Leopoldt, D., C. Harteneck, and B. Nürnberg. 1997. G proteins endogenously expressed in Sf 9 cells: interactions with mammalian histamine receptors Naunyn Schmiedebergs Arch. Pharmacol. 356:216–224.2lp, 百拇医药
Leopoldt, D., T. Hanck, T. Exner, U. Maier, R. Wetzker, and B. Nürnberg. 1998. Gß{gamma} stimulates phosphoinositide 3-kinase-by direct interaction with two domains of the catalytic p110 subunit. J. Biol. Chem. 273:7024–7029.2lp, 百拇医药
Lietzke, S.E., S. Bose, T. Cronin, J. Klarlund, A. Chawla, M.P. Czech, and D.G. Lambright. 2000. Structural basis of 3-phosphoinositide recognition by pleckstrin homology domains. Mol. Cell. 6:385–394.2lp, 百拇医药
Lopez-Ilasaca, M., P. Crespo, P.G. Pellici, J.S. Gutkind, and R. Wetzker. 1997. Linkage of G protein-coupled receptors to the MAPK signaling pathway through PI3-kinase . Science. 275:394–397.2lp, 百拇医药
Lopez-Ilasaca, M., J.S. Gutkind, and R. Wetzker. 1998. Phosphoinositide 3-kinase is a mediator of Gß-dependent Jun kinase activation. J. Biol. Chem. 273:2505–2508.2lp, 百拇医药
Maier, U., A. Babich, and B. Nürnberg. 1999. Roles of non-catalytic subunits in Gß{gamma} -induced activation of class I phosphoinositide 3-kinase isoforms ß and . J. Biol. Chem. 274:29311–29317.
Maier, U., A. Babich, N. Macrez, D. Leopoldt, P. Gierschik, D. Illenberger, and B. Nürnberg. 2000. Gß52 is a highly selective activator of phospholipid-dependent enzymes. J. Biol. Chem. 275:13746–13754.l{#xk*, 百拇医药
Metjian, A., R.L. Roll, A.D. Ma, and C.S. Abrams. 1999. Agonists cause nuclear translocation of phosphatidylinositol 3-kinase . A Gß-dependent pathway that requires the p110 amino terminus. J. Biol. Chem. 274:27943–27947.l{#xk*, 百拇医药
Murga, C., L. Laguinge, R. Wetzker, A. Cuadrado, and J.S. Gutkind. 1998. Activation of Akt/protein kinase B by G protein-coupled receptors. A role for and ß subunits of heterotrimeric G proteins acting through phosphatidylinositol-3-OH kinase . J. Biol. Chem. 273:19080–19085.l{#xk*, 百拇医药
Naccache, P.H., S. Levasseur, G. Lachance, S. Chakravarti, S.G. Bourgoin, and S.R. McColl. 2000. Stimulation of human neutrophils by chemotactic factors is associated with the activation of phosphatidylinositol 3-kinase . J. Biol. Chem. 275:23636–23641.l{#xk*, 百拇医药
Pacold, M.E., S. Suire, O. Perisic, S. Lara-Gonzalez, C.T. Davis, E.H. Walker, P.T. Hawkins, L. Stephens, J.F. Eccleston, and R.L. Williams. 2000. Crystal structure and functional analysis of Ras binding to its effector phosphoinositide 3-kinase . Cell. 103:931–943.
Reusch, H.P., M. Schaefer, C. Plum, G. Schultz, and M. Paul. 2001. Gß mediate differentiation of vascular smooth muscle cells. J. Biol. Chem. 276:19540–19547./utf[, 百拇医药
Ridley, A.J., H.F. Paterson, C.L. Johnston, D. Diekmann, and A. Hall. 1992. The small GTP-binding protein rac regulates growth factor-induced membrane ruffling. Cell. 7:401–410./utf[, 百拇医药
Schaefer, M., N. Albrecht, T. Hofmann, T. Gudermann, and G. Schultz. 2001. Diffusion-limited translocation mechanism of protein kinase C isotypes. FASEB J. 15:1634–1636./utf[, 百拇医药
Simonsen, A., A.E. Wurmser, S.D. Emr, and H. Stenmark. 2001. The role of phosphoinositides in membrane transport. Curr. Opin. Cell Biol. 13:485–492./utf[, 百拇医药
Stephens, L.R., T.R. Jackson, and P.T. Hawkins. 1993. Agonist-stimulated synthesis of phosphatidylinositol(3,4,5)-trisphosphate: a new intracellular signalling system? Biochim. Biophys. Acta. 1179:27–75./utf[, 百拇医药
Stephens, L., A. Smrcka, F.T. Cooke, T.R. Jackson, P.C. Sternweis, and P.T. Hawkins. 1994. A novel phosphoinositide 3 kinase activity in myeloid-derived cells is activated by G protein ß subunits. Cell. 77:83–93.
Stephens, L.R., A. Eguinoa, H. Erdjument-Bromage, M. Lui, F. Cooke, J. Coadwell, A.S. Smrcka, M. Thelen, K. Cadwallader, P. Tempst, and P.T. Hawkins. 1997. The Gß sensitivity of a PI3K is dependent upon a tightly associated adaptor, p101. Cell. 89:105–114.r5, 百拇医药
Stoyanov, B., S. Volinia, T. Hanck, I. Rubio, M. Loubtchenkov, D. Malek, S. Stoyanova, B. Vanhaesebroeck, R. Dhand, B. Nürnberg, et al. 1995. Cloning and characterization of a G protein-activated human phosphoinositide-3 kinase. Science. 269:690–693.r5, 百拇医药
Tang, X., and C.P. Downes. 1997. Purification and characterization of Gß-responsive phosphoinositide 3-kinases from pig platelet cytosol. J. Biol. Chem. 272:14193–14199.r5, 百拇医药
Teruel, M.N., and T. Meyer. 2000. Translocation and reversible localization of signaling proteins: a dynamic future for signal transduction. Cell. 103:181–184.r5, 百拇医药
Ueki, K., D.A. Fruman, S.M. Brachmann, Y.H. Tseng, L.C. Cantley, and C.R. Kahn. 2002. Molecular balance between the regulatory and catalytic subunits of phosphoinositide 3-kinase regulates cell signaling and survival. Mol. Cell. Biol. 22:965–977.
Vanhaesebroeck, B., S.J. Leevers, K. Ahmadi, J. Timms, R. Katso, P.C. Driscoll, R. Woscholski, P.J. Parker, and M.D. Waterfield. 2001. Synthesis and function of 3-phosphorylated inositol lipids. Annu. Rev. Biochem. 70:535–602.(u, 百拇医药
Várnai, P., K.I. Rother, and T. Balla. 1999. Phosphatidylinositol 3-kinase-dependent membrane association of the Bruton's tyrosine kinase pleckstrin homology domain visualized in single living cells. J. Biol. Chem. 274:10983–10989.(u, 百拇医药
Walker, E.H., O. Perisic, C. Ried, L. Stephens, and R.L. Williams. 1999. Structural insights into phosphoinositide 3-kinase catalysis and signalling. Nature. 402:313–320.(u, 百拇医药
Wishart, M.J., G.S. Taylor, and J.E. Dixon. 2001. Phoxy lipids: Revealing PX domains as phosphoinositide binding modules. Cell. 105:817–820.(u, 百拇医药
Yu, J., Y. Zhang, J. McIlroy, T. Rordorf-Nikoloic, G.A. Orr, and J.M. Backer. 1998. Regulation of the p85/p110 phosphatidylinositol 3'-kinase: stabilization and inhibition of the p110 catalytic subunit by the p85 regulatory subunit. Mol. Cell. Biol. 18:1379–1387.(Carsten Brock Michael Schaefer H. Peter Reusch Cor)
2 Institut für Pharmakologie, Chemie, Pharmazie, Freie Universität Berlin, 14195 Berlin, Germanyw'(y\1, 百拇医药
3 Institut für Klinische Pharmakologie und Toxikologie, Chemie, Pharmazie, Freie Universität Berlin, 14195 Berlin, Germanyw'(y\1, 百拇医药
4 Fachbereich Biologie, Chemie, Pharmazie, Freie Universität Berlin, 14195 Berlin, Germanyw'(y\1, 百拇医药
Abstractw'(y\1, 百拇医药
Receptor-regulated class I phosphoinositide 3-kinases (PI3K) phosphorylate the membrane lipid phosphatidylinositol (PtdIns)-4,5-P2 to PtdIns-3,4,5-P3. This, in turn, recruits and activates cytosolic effectors with PtdIns-3,4,5-P3–binding pleckstrin homology (PH) domains, thereby controlling important cellular functions such as proliferation, survival, or chemotaxis. The class IB p110/p101 PI3K is activated by Gß on stimulation of G protein–coupled receptors. It is currently unknown whether in living cells Gß acts as a membrane anchor or an allosteric activator of PI3K, and which role its noncatalytic p101 subunit plays in its activation by Gß. Using GFP-tagged PI3K subunits expressed in HEK cells, we show that Gß recruits the enzyme from the cytosol to the membrane by interaction with its p101 subunit. Accordingly, p101 was found to be required for G protein–mediated activation of PI3K in living cells, as assessed by use of GFP-tagged PtdIns-3,4,5-P3–binding PH domains. Furthermore, membrane-targeted p110 displayed basal enzymatic activity, but was further stimulated by Gß, even in the absence of p101. Therefore, we conclude that in vivo, Gß activates PI3K by a mechanism assigning specific roles for both PI3K subunits, i.e., membrane recruitment is mediated via the noncatalytic p101 subunit, and direct stimulation of Gß with p110 contributes to activation of PI3K.
Key Words: G protein; fluorescence imaging; phosphoinositide 3-kinase; FRET; PH domaina&8n, 百拇医药
Introductiona&8n, 百拇医药
Phosphoinositide 3-kinases (PI3Ks) phosphorylate the D3 position of the inositol ring of phosphoinositides, thereby generating intracellular signaling molecules . These phospholipids, i.e., phosphatidylinositol (PtdIns)-3-P, PtdIns-3,4-P2, and PtdIns-3,4,5-P3, have been implicated in cellular functions including chemotaxis, differentiation, glucose homeostasis, proliferation, survival, and trafficking . They transmit signals by recruiting effector molecules that possess specific lipid-binding domains. FYVE domains and PX domains bind to PtdIns-3-P with high affinity, whereas a subgroup of pleckstrin homology (PH) domains, containing a highly basic motif, preferentially binds PtdIns-3,4-P2 and PtdIns-3,4,5-P3 (a&8n, 百拇医药
PtdIns-3,4,5-P3 represents the major product of class I PI3Ks in vivo. These lipid kinases are under tight control of cell surface receptors, including receptor tyrosine kinases (RTK) or G protein–coupled receptors (GPCR; ). All class I members are heterodimers consisting of a p110 catalytic and a p85- or p101-type noncatalytic subunit. The class IA p110 isoforms , ß, and form a complex with p85 adaptor subunits, whereas the only class IB member (p110) is associated with a p101 noncatalytic subunit. The p85 subunit binds to tyrosine-phosphorylated RTKs. This interaction has two consequences resulting in activation of the PI3K. First, the cytosolic enzyme translocates to the inner leaflet of the plasma membrane, giving p110 access to its lipid substrate . Second, the interaction with the tyrosine-phosphorylated receptor induces a conformational change of p85 that results in disinhibition of p110 enzymatic activity . Because the p85 regulatory subunit inhibits the p110 catalytic subunit, it is feasible that constitutively membrane-associated class IA p110 mutants trigger downstream responses characteristic of growth factor action .
Similar to class IA PI3Ks, unstimulated PI3K is predominantly localized in the cytosol, whereas GPCRs induced an increase of PI3K in the membrane fraction . GPCRs activate PI3K through direct interaction with Gß, whereas a stimulation by G is quantitatively less important. However, little is known about the activating mechanism, e.g., it remains elusive whether Gß functions as a membrane anchor or an allosteric regulator of PI3K. Although the p101 subunit has been proposed to act as an indispensable adaptor linking Gß with p110 , in vitro studies have suggested that p101 is not mandatory for Gß-induced stimulation of PI3K . Moreover, a direct interaction of Gß with NH2- and COOH-terminal regions of p110 has been demonstrated . Further support for a direct interaction of Gß with the p110 catalytic subunit came from the observation that the p110ß isoform can also be stimulated by Gß, regardless of whether p85 is present or not . Nevertheless, the noncatalytic p101 subunit of PI3K binds tightly to Gß, suggesting specific roles in G protein–induced regulation of this lipid kinase. For instance, in vitro studies showed that Gß-stimulated p110 potently produced PtdIns-3-P, whereas the Gß-activated heterodimeric p110p101 counterpart potently catalyzed the formation of PtdIns-3,4,5-P3 . This suggests that p101 may affect the interaction of Gß-stimulated p110 with the lipid interface.
Therefore, the exact roles of Gß and PI3K subunits in G protein–induced activation of PI3K in living cells remain unclear. Hence, we asked whether Gß functions as a membrane anchor and/or allosteric activator, and whether p101 is required for proper function of p110 in vivo. To tackle these questions, we examined cellular events leading to activation of PI3K using fluorescent fusion proteins in living cells.h&ahdq, 百拇医药
Resultsh&ahdq, 百拇医药
Characterization of fluorescent PI3K fusion proteinsh&ahdq, 百拇医药
To visualize PI3K subunits in vivo, we generated constructs encoding p110 and p101 fused to YFP or CFP . Their expression in human embryonic kidney (HEK) 293 cells was verified by immunoblot analysis using an antibody against GFP . The catalytic activity and Gß sensitivity of PI3K fusion proteins was confirmed by in vitro lipid kinase assays after immunoprecipitation with an mAb recognizing only intact p110. Precipitates from vector-transfected control cells neither exhibited p110 immunoreactivity nor lipid kinase activity. These results demonstrate that YFP-fused p110 or p101 retain essential functions of the wild-type PI3K.
fig.ommtteddo, 百拇医药
Construction and characterization of fluorescent PI3Kfusion proteins. (A) Schematic representation of the wild-type, fluorescent, and membrane-targeted PI3K subunits. A YFP (or CFP) tag was fused to either the NH2- or the COOH terminus of p110 and p101. The p110-CAAX fusion protein contains an isoprenylation motif (bars for lipid modifications). (B) Differently fluorescence-tagged PI3K subunits were coexpressed in HEK cells, and cell lysates were subjected to SDS-PAGE followed by immunoblotting with a GFP-specific antibody. (C) Catalytic activity of fluorescent PI3K fusion proteins. HEK cells were transfected with the indicated plasmids. Lysates were subjected to immunoprecipitation (IP) with (+) or without (-) an anti-p110 antibody. immunoprecipitation was controlled by immunoblotting (IB) with another anti-p110 antibody. Shown are chemiluminescence images (top panels). Immunoprecipitates were assayed for in vitro PI3K activity in the absence or presence of 120 nM Gß using PtdIns-4,5-P2– containing lipid vesicles and [32P]ATP as substrates. Depicted are autoradiographs of the generated 32P-PtdIns-3,4,5-P3 (bottom panels).
Subcellular distribution of PI3K subunits and instability of monomeric p101m, 百拇医药
Next, we examined the subcellular distribution of the YFP-tagged PI3K subunits in living cells by confocal laser scanning microscopy. Although YFP was uniformly distributed inside HEK cells (unpublished data), fluorescence signals from YFP fused to the NH2 or COOH termini of p110 were only detected in the cytosolic compartment (top). However, a small fraction of the YFP-fused p110 was also detected in the nucleus when coexpressed with wild-type p101 (top). The same distribution was seen for YFP-tagged p101 coexpressed with wild-type p110 ( B, bottom). Thus, fluorescent PI3K subunits form heterodimers (see next section), present in the cytosol and nucleus of HEK cells, whereas distribution of the p110 subunit alone seems restricted to the cytosol. Interestingly, YFP-fused p101 alone showed a predominant nuclear localization with a significantly lower overall fluorescence intensity ( bottom).
fig.ommttedm9-{, 百拇医药
Expression and subcellular distribution of fluorescent p110 and p101 in HEK cells. (A) Cells were transfected with plasmids for YFP-tagged single PI3K subunits as indicated. Images were taken by confocal laser scanning microscopy. Images of typical cells are shown. White bars indicate a 10-µm scale. (B) Cells were cotransfected with both p110 and p101. Only one subunit was fused to YFP as indicated. (C) Cells were transfected with plasmids encoding fluorescent p110 and p101 at different ratios. The total amount of transfected cDNA was kept constant by the addition of pcDNA3. Equal amounts of whole-cell lysates were subjected to SDS-PAGE followed by immunoblotting with an anti-GFP antibody.m9-{, 百拇医药
To explore the possibility that expression levels of p101 could depend on the coexpression of p110, we transfected HEK cells with different ratios of plasmids encoding fluorescent p110 or p101 and analyzed total cell lysates by immunoblotting using an anti-GFP antibody . The expression level of p110 correlated with the amount of the p110 (but not the p101) plasmid. In contrast, the expression level of p101 was also dependent on the amount of the p110 plasmid. Hence, the stability of p101 depends on the coexpression of p110 but not vice versa. The poor cytosolic fluorescence of YFP-tagged p101 in the absence of p110 may thus be explained by cytosolic degradation of monomeric p101.
In vivo dimerization of PI3K subunits(, 百拇医药
To directly demonstrate heterodimerization of PI3K subunits in living cells, we used fluorescence resonance energy transfer (FRET). FRET between two fluorophores, e.g., CFP and YFP, is restricted to distances of <100 Å, and therefore, provides direct evidence for a protein–protein interaction . We coexpressed CFP- and YFP-tagged PI3K subunits and determined FRET by following the donor (CFP) recovery during acceptor (YFP) bleach . All combinations showed significant FRET, although quantitative differences were evident.(, 百拇医药
fig.ommtted(, 百拇医药
Dimerization of p110 and p101. (A) HEK cells were cotransfected with plasmids encoding NH2- or COOH-terminally YFP-tagged p110 and NH2- or COOH-terminally CFP-tagged p101 in different combinations. FRET was measured in vivo. An increase in CFP (donor) fluorescence during YFP (acceptor) bleach indicates FRET between fluorescent PI3K subunits. The depicted data represent means ± SEM of at least 18 single cells in three independent transfection experiments. (B) HEK cells were transfected with the indicated plasmids. Cytosols were prepared and subjected to gel filtration. The elution profiles were analyzed by immunoblotting with an anti-GFP antibody (L, load diluted 1:5; V, void volume; 20–26, fraction numbers).
Because FRET efficiency is a function of the distance between two fluorophores, quantitative differences may represent different distances of the fluorescent tags fused to the NH2 or COOH termini of p101 and p110 in a particular combination. However, the different FRET efficiencies may also reflect different degrees of heterodimerization. To exclude the latter, we analyzed heterodimerization of different p101/p110 combinations by size-exclusion chromatography The elution profile of YFP-p110 ( bottom) was completely shifted to earlier fractions on coexpression with CFP-p101 ( top), indicating a high degree of heterodimerization. A similar elution profile was found for p110-YFP + CFP-p101 ( middle), albeit a lower FRET efficiency was detected for this combination. Hence, the different FRET efficiencies observed are not due to different degrees of dimerization, but likely reflect different distances of the fluorescent tags in the p101/p110 heterodimer. Thus, their two NH2 termini as well as their two COOH termini are in close proximity, whereas the NH2 terminus of each subunit is relatively far from the COOH terminus of the other.
Gß{gamma} dimers recruit PI3K to membranes via the p101 subunit6og{!, 百拇医药
Current concepts of PI3K activation are based on the recruitment of the cytosolic lipid kinase to the plasma membrane . In the case of PI3K, the major stimulus is assumed to be Gß, which is membrane-bound and has been shown to directly bind to both kinase subunits under in vitro conditions. Therefore, we asked whether overexpression of free Gß directs YFP-fused PI3K subunits to the cell membrane in vivo. Surprisingly, p110 was not membrane-localized in the presence of coexpressed Gß in HEK cells ( A, top). In contrast, Gß recruited p101 to the membrane, resulting in accumulation of NH2- or COOH-terminally YFP-tagged p101 at both plasma- and endomembranes ( A, bottom, and B). This corresponds to the subcellular distribution of overexpressed Gß dimers (see below), suggesting that p101 interacted with membrane-bound Gß.6og{!, 百拇医药
fig.ommtted
Membrane recruitment of PI3K by Gß. The effect of coexpression of Gß on the subcellular distribution of p110, p101, and heterodimeric PI3K in HEK cells was analyzed by confocal laser scanning microscopy. Images of typical cells are shown (white bars, 10 µm). (A) Coexpression of monomeric PI3K subunits with Gß. (B) Controls; coexpression of YFP-p101 with Gß1, G2, Gß12, or Gß12 and the Gß- scavenging Gi2. (C) Coexpression of heterodimeric PI3K with Gß. Note the round shape of the cells. Controls; coexpression of Gß with YFP or with kinase-deficient YFP-p110-K833R and p101; effect of 100 nM wortmannin. (D) Subcellular localization of Gß. Cells were transfected with a plasmid encoding CFP-tagged Gß1 alone or together with the G2 plasmid. (E) Immunoblot analysis of membrane fractions. HEK cells were transfected with the plasmids encoding PI3K, Gß, or both together. To avoid rounding and detachment of the cells, the kinase-deficient mutant YFP-p110-K833R was used. Membrane fractions were prepared and analyzed by immunoblotting with anti-p110 and anti-Gß antibodies.
Next, we examined the subcellular localization of p110/p101 heterodimers in the presence of free Gß{gamma} . When coexpressed with p101 and Gß, the p110 fluorescent signal was at, or close to, plasma and endosomal membranes of HEK cells ( C, top). The same picture emerged when the YFP-tag was fused to p101 (unpublished data). Correspondingly, CFP-tagged Gß1 coexpressed with G2 exhibited a similar subcellular distribution like fluorescent heterodimeric PI3K (compare C and 4 D), and immunoblotting of membrane-derived proteins confirmed an enrichment of PI3K{gamma} in the membrane compartment after overexpression of Gß . Together, these data imply that Gß directs cytosolic PI3K to the membrane via interaction with p101.fe-p)t, 百拇医药
Interestingly, coexpression of heterodimeric PI3K together with Gß produced a rounded morphology and detachment of the cells This morphological change was reversed by treatment with 100 nM of the PI3K inhibitor wortmannin in approximately half of the cells within 30 min, and not seen when a kinase-deficient YFP-p110-K833R mutant was used . Hence, the morphological change was related to the enzymatic activity of PI3K{gamma} stimulated by the coexpressed Gß.
p101 is required for G protein–mediated activation of p110 in vivo-$0}, 百拇医药
Because changes in morphology are not a very reliable measure of PI3K activity, we examined phosphorylation of endogenous protein kinase B (PKB or Akt) as an established PI3K-specific read-out system. To avoid detachment of the cells as a consequence of constitutive PI3K stimulation by overexpressed Gß, we transiently stimulated the heterologously expressed Gi-coupled formyl-methionyl-leucyl-phenylalanine (fMLP) receptor, which is known to activate PI3K . Stimulation with fMLP induced Akt phosphorylation in HEK cells in the same range as with FCS ( A). However, Akt phosphorylation was only seen when the receptor was coexpressed with both PI3K subunits, and not with p110 alone. These data imply that p101 is required for GPCR-induced activation of PI3K in vivo.-$0}, 百拇医药
fig.ommtted-$0}, 百拇医药
Activation of PI3K by a G protein–coupled receptor. (A) Akt phosphorylation. Whole-cell lysates were analyzed by immunoblotting using an antibody that specifically recognizes the phosphorylated form of Akt. Equal loading was shown by using an anti-ERK antibody. Left; FCS-induced Akt phosphorylation in untransfected HEK cells. Right; fMLP-induced Akt phosphorylation in cells expressing the fMLP receptor and only the catalytic (p110) or both PI3K subunits. (B) Membrane recruitment of the PtdIns-3,4,5-P3–binding PH domain of GRP1. HEK 293 cells were transfected with plasmids encoding the PH domain of GRP1 fused to GFP (GFP-GRP1PH), and the human fMLP receptor (fMLP-R), p110, and p101 in different combinations. The localization of the GFP-GRP1PH was monitored before and after the addition of 1 µM fMLP and 100 ng/ml EGF (added 8 min later) by confocal laser scanning microscopy. Pictures were taken 4 min after addition of either agonist. Images of typical experiments are shown (white bars, 10 µm). (C) Membrane recruitment of the PtdIns-3,4,5-P3–binding PH domain of Btk. Instead of GFP-GRP1PH, the PH domain of Btk fused to CFP (BtkPH-CFP) was used as a PtdIns-3,4,5-P3 sensor.
To measure PI3K activity by a complementary experimental approach, the effects of fMLP receptor stimulation on the subcellular distribution of GFP-GRP1PH fusion proteins were examined. Recruitment of the GFP-labeled PH domain of general receptor for phosphoinositides-1 (GRP1), a PI3K-activated exchange factor for ADP-ribosylation factors , has been applied to detect formation of PtdIns-3,4,5-P3 with high affinity and selectivity . fMLP stimulation had no effect on the subcellular distribution of GFP-GRP1PH in HEK cells expressing either the receptor or PI3K alone ( B, top two panels), whereas subsequent stimulation with EGF induced a rapid mmbrane recruitment of the PtdIns-3,4,5-P3 sensor, presumably through activation of an endogenously expressed RTK-sensitive class IA PI3K. In cells coexpressing the fMLP receptor and heterodimeric PI3K, fMLP stimulation induced a rapid membrane translocation of GFP-GRP1PH ( B, third panel), indicating that this effect indeed reflects fMLP receptor–mediated stimulation of PI3K. Accordingly, the fMLP-induced effect was sensitive to pretreatment with pertussis toxin (PTX; unpublished data). Interestingly, a slight basal membrane localization of GFP-GRP1PH was visible in the cells coexpressing heterodimeric PI3K and the fMLP receptor even in the absence of agonist ( B, third panel). Because this was not seen after PTX treatment (unpublished data) or in cells omitting the receptor ( B, second panel), it most likely reflected a constitutively active fMLP receptor.
Translocation of an isolated PH domain in single cells may be a more sensitive read-out for PtdIns-3,4,5-P3 production than the Akt phosphorylation assay. Hence, we used the GFP-GRP1PH translocation approach to reexamine the requirement of p101 for receptor-induced PI3K activation . Nevertheless, an fMLP-induced membrane recruitment of GFP-GRP1PH in cells coexpressing the receptor and p110 (but not p101) was not visible ( B, fourth panel). This finding confirms that the noncatalytic p101 subunit is required for GPCR-mediated stimulation of PI3Kin living cells.i3f&!9d, 百拇医药
To further strengthen the findings, we replaced GFP-GRP1PH by the GFP-fused PH domain of Bruton's tyrosine kinase (BtkPH), which has also been described to bind PtdIns-3,4,5-P3 with high specificity and affinity . As expected, fluorescent BtkPH was distributed equally between cytosol and nucleus in HEK cells . However, cells coexpressing the receptor and both PI3K subunits exhibited a slight basal membrane localization of BtkPH-CFP followed by a prominent translocation from the cytosol to the membrane on exposure to fMLP. Similar to GFP-GRP1PH, no fMLP-induced redistribution of BtkPH-CFP was seen in the absence of p101.
Activation of a membrane-targeted p110-CAAX in vivoth?, http://www.100md.com
The presented data suggest that in living cells, p101 is an indispensable adaptor for GPCR-induced translocation and activation of class IB PI3K, which is equivalent to the role of p85 in RTK-induced class IA PI3K activation. Vice versa, Gß behaves as a membrane anchor recruiting PI3K through association with p101. So far, these experiments did not clarify whether membrane recruitment itself is sufficient for activation of the enzyme or whether additional allosteric stimulation is required. To tackle this question, we generated p110 mutants containing a COOH-terminal isoprenylation signal, i.e., a CAAX-box motif, which constitutively localizes p110 to the plasma membrane. p110-CAAX accumulated at the plasma membrane of HEK cells, and was able to complex with p101 (, top). Coexpression of Gß{gamma} together with YFP-p110-CAAX increased not only fluorescent staining of endomembranes, but also produced a wortmannin-sensitive rounding of the cells (, bottom) even in the absence of p101 (not depicted), which was similar to the effect after coexpression of Gß with p110/p101 dimers . These observations imply that membrane-bound PI3K is not fully active, but can be further activated by Gß.
fig.ommtted7, 百拇医药
Characterization of a constitutively membrane-associated p110-CAAX. HEK cells were transfected with the indicated plasmids and analyzed by confocal laser scanning microscopy. Images of typical cells are shown. White bars indicate a 10-µm scale. Top panel; Subcellular localization of YFP-p110-CAAX (left) or YFP-p101 expressed together with p110-CAAX (right). Bottom panel; Coexpression with Gß.7, 百拇医药
To substantiate this assumption, we analyzed the subcellular redistribution of GFP-GRP1PH in HEK cells coexpressing membrane-targeted p110-CAAX, p101 and the fMLP receptor in various combinations . A slight basal membrane localization of the fluorescent PH domain was evident on coexpression with p110-CAAX ( A, top panel) indicating an elevated PtdIns-3,4,5-P3 level. The same picture emerged when p101 was present in the panel of transfected plasmids ( A, second panel). On coexpression of the receptor, fMLP induced a significant additional translocation of GFP-GRP1PH from the cytosol to the membrane ( A, third and fourth panel) indicating additional stimulation of the membrane-bound lipid kinase. Interestingly, the effect was similar irrespective of whether p101 was present or not, suggesting direct activation of monomeric p110. PTX sensitivity confirmed that fMLP-induced activation of p110-CAAX was mediated by Gi proteins (unpublished data). Because Gß can activate Ras, which in turn may activate p110, one may imagine that fMLP-induced, p101-independent stimulation of p110-CAAX was mediated by Ras. However, coexpression of dominant-negative Ras N17 had no effect on fMLP-induced p110-CAAX stimulation, whereas its inhibitory effect was evident when EGF-induced MAP kinase activation was assessed ( A, bottom panel). In addition, an indirect autocrine stimulatory effect was excluded because supernatants of fMLP-stimulated cells failed to induce GFP-GRP1PH translocation in cells omitting fMLP receptors (unpublished data).
fig.ommttedq^%[25, 百拇医药
Activation of a constitutively membrane-associated p110-CAAX. (A) GFP-GRP1PH translocation. HEK cells were transfected with plasmids for GFP-GRP1PH and p110-CAAX, p101, fMLP-R, and dominant-negative Ras N17 in different combinations. Translocation of the PtdIns-3,4,5-P3–binding GFP-GRP1PH in response to agonist stimulation was monitored as described above (see ). White bars indicate a 10-µm scale. As a positive control for H-Ras N17, cells were transfected with the H-Ras N17 plasmid (same amount as for the GFP-GRP1PH translocation experiment) or empty vector. Cells were stimulated with 10 ng/ml EGF, and whole-cell lysates were analyzed by immunoblotting with an antibody that specifically recognizes the phosphorylated form of ERK (p-ERK). Equal loading was shown using the anti-ERK-antibody. (B) BtkPH-CFP translocation. (C) Akt phosphorylation. The experiment shown in was repeated. In addition, cells were transfected with the plasmid for the membrane-targeted p110-CAAX instead of wild-type p110.q^%[25, 百拇医药
By using BtkPH-CFP instead of GFP-GRP1PH as the PtdIns-3,4,5-P3 sensor, we confirmed both the basal enzymatic activity and fMLP-induced stimulation of membrane-targeted p110-CAAX, regardless of coexpressed p101. Furthermore, the Akt phosphorylation assay corroborated that p101 was dispensable for GPCR-induced stimulation of membrane-targeted p110-CAAX .
Next, we sought to confirm our principal findings using nontransformed cells in primary culture with a higher degree of differentiation than HEK cells. Vascular smooth muscle (VSM) cells were injected with plasmids encoding the fMLP receptor, p101, p110 or p110-CAAX, and GFP-GRP1PH in different combinations, and redistribution of the fluorescent PtdIns-3,4,5-P3 sensor was monitored. Again, this series of experiments confirmed that p101 is required for GPCR-mediated activation of PI3K{gamma} in living cells, whereas membrane-targeted p110 can be stimulated even in the absence of p101.yg'}m, 百拇医药
fig.ommttedyg'}m, 百拇医药
PI3K-mediated membrane translocation of GFP-GRP1PH in vascular smooth muscle (VSM) cells. VSM cells were microinjected with plasmids for GFP-GRP1PH, fMLP-R, p101, and p110 or p110-CAAX in different combinations. The localization of GFP-GRP1PH was monitored before and after the addition of 1 µM fMLP (picture taken after 4 min) and 10 ng/ml PDGF-BB (added 12 min later, picture taken after 8 min). White bars indicate a 10-µm scale.
Discussionc:, http://www.100md.com
PI3Kis a key player in the regulation of leukocyte functions such as chemotaxis and superoxide production . To elucidate the mechanism of how Gß activates the enzyme in vivo, we studied the subcellular localization and translocation of GFP-tagged PI3K subunits and PH domains in living cells.c:, http://www.100md.com
A p110-GFP construct has been previously reported , but data describing a functional interaction of the fusion protein with G proteins are not available. Therefore, we systematically tagged each PI3K subunit and found heterodimerization, enzymatic activity, and Gß-sensitivity of the fusion proteins unaffected by the tags, whereas a GFP-tagged Gß complexed to G did not interact with PI3K (unpublished data). In resting cells, fluorescent PI3K was predominantly detected in the cytosol. This resembles the native situation in which the vast majority of the endogenous PI3K pool has been assigned to the cytosol . Thus, the GFP-tagged proteins are appropriate tools to study the cellular events leading to activation of PI3K.
Recent analysis of p110 crystals gave insights into the three-dimensional structure of the catalytic PI3K subunit . However, the structure of p101 and the molecular determinants for the interaction of both PI3K subunits are currently unknown. In vitro data derived from p101 and p110 deletion mutants suggest that large areas of p101 may interact with the NH2-terminal side of p110 . FRET data presented here indicate that NH2 and both COOH termini of p110 and p101 are in close proximity. The latter finding is of interest because the p110 COOH terminus harbors the catalytic domain. Previous data have implied that not only Ras , but also Gß directly interact with the COOH terminus of p110. Therefore, interaction of p101 with the catalytic domain of p11 would be in line with the functional data suggesting that p101 affects interaction of Gß-stimulated p110 with the lipid interface .8ix4s, http://www.100md.com
Gß is considered to be the principal, direct stimulus of GPCR-induced PI3K enzymatic activity. Convincing evidence for this assumption has come from reconstitution of purified proteins, demonstrating that in vitro, the enzymatic activity of heterodimeric and monomeric PI3K is significantly stimulated by Gß . However, when we challenged the concept in living cells, marked differences in the sensitivities between heterodimeric and monomeric PI3K emerged. The G protein–coupled fMLP receptor activated p110/p101, but not p110, as evident from the translocation of fluorescent PtdIns-3,4,5-P3 sensors or the stimulation of Akt phosphorylation. Likewise, coexpressed Gß recruited p110/p101, but not p110, to membranes of HEK cells. Interestingly, PI3Kß, another Gß-regulated PI3K, did not translocate to the membrane on GPCRs or Gß stimulation, whereas RTKs induced membrane recruitment in vivo (unpublished data).
Does membrane recruitment of PI3K by itself result in constitutive activation of p110? To answer this intriguing question, we used an isoprenylated mutant of p110, i.e., p110-CAAX, which is permanently attached to the membrane, and thereby, is in close proximity to its lipid substrates. Expression of p110-CAAX only slightly enhanced membrane association of fluorescent PtdIns-3,4,5-P3 sensors in HEK or VSM cells. Notably, it did not affect the morphology of HEK cells. In contrast, coexpression of Gß together with p110-CAAX induced morphological changes comparable to cells transfected with plasmids encoding Gß and p110/p101. Furthermore, stimulation of the cells with fMLP stimulated Akt phosphorylation and membrane recruitment of PH domains regardless of whether p110/p101 or p110-CAAX was the effector. This establishes a second role for Gß, i.e., the activation of the membrane-attached catalytic p110subunit even in the absence of p101 . Although Gß{gamma} interacts with either monomeric PI3K subunit through individual binding sites, it does not exclude the possibility that heterodimeric PI3K forms either a common or different binding site(s) for Gß. Attempts to determine the stoichiometry of the interaction between heterodimeric PI3K and one or more Gßwere inconclusive. One possible reason may be a weak affinity between p110 and Gß, whereas p101 binds at least with moderate affinity to Gß as determined by copurification studies and by a Biacore plasmon resonance approach (unpublished data).
fig.ommttedm!'d, 百拇医药
Hypothetical model for receptor- induced membrane recruitment and activation of heterodimeric PI3K. fMLP receptor, a prototypical heptahelical receptor coupled to Gi proteins. PI3K; the cytosolic enzyme consists of a noncatalytic p101 subunit, which is in a tight complex with p110, thereby stabilizing p101. The NH2 and COOH termini of p101 and p110 are oriented in close proximity, respectively. Contact sites involve the NH2 termini of both subunits. The modular domain structure of p110 (RBD, Ras binding domain; C2, C2 domain; hel, helical domain; N-cat, C-cat, NH2- and COOH-terminal lobes of the catalytic domain) is based on the crystal structure of an NH2-terminally truncated p110. (A) Membrane recruitment. The agonist-stimulated receptor induces the release of Gß from Gi proteins (dotted arrow). Gß recruits the PI3K heterodimer to the plasma membrane (dotted arrow) by binding to the noncatalytic p101 (hollow arrow). Accordingly, Gß and p101 function as a membrane anchor and an adaptor for PI3K, respectively. In addition, the C2 domain of p110 may facilitate membrane attachment through interaction with phospholipids. p101 may also affect the interaction of PI3K with the lipid interface. Membrane-attached PI3K exhibits basal enzymatic activity. (B) Allosteric activation. At the membrane, Gß activates PI3K by direct interaction with p110(hollow arrows). This stimulation does not require p101. However, p101 may participate in Gß-induced stimulation of membrane-attached p110. The stoichiometry of the Gß–PI3K interaction is unknown, i.e., Gß may interact with p101 and p110 through individual or common binding sites. Gß binds to the NH2- and COOH-terminal part of p110, the latter harboring the catalytic domain. Thus, Gß may allosterically activate PI3K through a conformational change in the catalytic domain. Accordingly, Gß significantly increases Vmax of PtdIns-3,4,5-P3 production. PtdIns-3,4,5-P3, in turn, recruits PH domain–containing effectors such as GRP1 or Btk to the plasma membrane (dotted arrow).
In our paper, we present evidence that p110 is stable as a monomer without p101, but not vice versa. This may explain why relevant amounts of native monomeric p101 do not occur in vivo (Wetzker, R., personal communication). p110 can be expressed as a stable protein independent of the presence of p101, as shown previously . Interestingly, the p110 subunit of class IA PI3Ks needs stabilization by forming a complex with the p85 adaptor Accordingly, class IA p110{alpha} ,ß,{delta} subunits are thought to occur only in their adaptor-bound form. Moreover, the p85 adaptor is more abundant than the p110 counterparts In contrast, p110 is also suggested to occur in the absence of p101 .1'tv, http://www.100md.com
Increasing evidences suggest that monomeric p110 may also function as a downstream regulator of GPCR-dependent signal transduction pathways in cells . However, under in vivo conditions the p101 subunit seems mandatory for G protein–mediated activation of PI3K by mediating membrane recruitment. Therefore, the question arises of how monomeric p110 translocates to the membrane. In this context, recent in vitro evidence points to the possibility that membrane attachment of p110 may involve binding to anionic phospholipids . In line with this finding, the majority of PI3K is already associated with lipid vesicles in the absence of Gß under in vitro conditions, and vesicle-bound PI3K can still be activated by Gß . Alternatively, p110 may be recruited by binding to membrane-anchored proteins other than Gß. In this respect, p110, like all other class I PI3Ks, is directly activated by GTP-bound Ras
In conclusion, we present in vivo evidence that Gß may stimulate PI3K in a dual and complementary way, i.e., by recruitment to the membrane, and by activation of the membrane-bound enzyme . In this scenario, the p101 subunit functions as an adaptor molecule necessary to recruit the catalytic subunit to the plasma membrane through high affinity interaction with Gß. In turn, direct interaction between Gß and the membrane-attached catalytic p110 subunit contributes to a final activation of the enzyme by a mechanism other than translocation.+zc67, http://www.100md.com
Materials and methods+zc67, http://www.100md.com
Construction of expression plasmids+zc67, http://www.100md.com
For generation of expression plasmids of PI3K subunits, restriction sites were introduced, and stop codons were removed by 8–12 cycles of PCR using the indicated primers, a PCR system (Expand HF; Roche), and a pcDNA3.1/V5-His-TOPO vector for first subcloning. Subcloned PCR fragments were confirmed by sequencing with fluorescent dye terminators (ABI-Prism 377; ). Expression plasmids encoding NH2- and COOH-terminally tagged fusion proteins were generated by subsequent subcloning into the vectors pEYFP-C1 and pECFP-C1 or into pcDNA3-YFP and pcDNA3-CFP , respectively. The constructs are designated indicating the color and the position of the tag In detail, wild-type p110; human p110 cDNA ( a corrected nucleotide sequence for p110 is available from GenBank/EMBL/DDBJ, accession no. ) was amplified with the primers 5'-GCC ACC ATG GAG CTG GAG AAC TAT AA-3' and 5'-GGA TCC AGC TTT CAC AAT GTC TAT TG-3', and subcloned into pcDNA3 via KpnI and XhoI. p110-YFP and -CFP; the stop codon was replaced by an XbaI site using the primer 5'-GTC TAG AGC TGA ATG TTT CTC TCC CTT GT-3' and the same forward primer as above. Subcloning into pcDNA3-YFP and -CFP was done via BamHI and XbaI. YFP- and CFP-p110; XhoI and BamHI sites were introduced using the primers 5'-CTC GAG GCA TGG AGC TGG AGA ACT A-3' and 5'-GGA TCC AGC TTT CAC AAT GTC TAT TG-3' for subcloning into pEYFP-C1 and pECFP-C1. p110-K833R; position 2498 was mutated from A to G using the QuikChange® Mutagenesis Kit and appropriate primers. p110-CAAX; two AfeI sites in p110 were removed, and the stop codon was replaced by a new AfeI site by silent mutations using the QuikChange® Mutagenesis Kit and appropriate primers. Adaptor oligonucleotides encoding the 18 COOH-terminal amino acids of H-Ras were inserted into the AfeI and BamHI sites. Wild-type p101; the cDNA for porcine p101 was subcloned in pcDNA3 via EcoRI and NotI. An optimized ribosomal docking sequence was introduced by PCR using the primers 5'-GCC ACC ATG CAG CCA GGG GCC ACG GA-3' and 5'-GGC CCG AGA CGA AGG AGG T-3' and subsequent exchange of the 5'-end via HindIII and BsmBI. YFP- and CFP-p101; the EcoRI/NotI fragment of p101 was subcloned into EcoRI/Bsp120I-digested pEYFP-C1 or pECFP-C1. To adapt the reading frames, the HindIII site was blunted with Klenow fragment and religated. p101-YFP and -CFP; PCR with the primers 5'-GTC CTC TCC TCA CAC GGT TCT T-3' and 5'-GTC TAG AGG CAG AGC TCC GCT GAA AGT-3' generated the 3'-end of p101 with an XbaI site instead of the stop codon. To restore the full-length p101 cDNA, the 5'-part was excised from wild-type p101 in pcDNA3 with HindIII and ClaI (partial digest) and ligated to the HindIII/ClaI-digested 3'-end. Subsequent subcloning in pcDNA3-YFP and pcDNA3-CFP was done via HindIII and XbaI.
The human fMLP receptor cDNA was amplified with the primers 5'-GCC ACC ATG GAG ACA AAT TCC TCT CTC-3' and 5'-TCA CTT TGC CTG TAA CTC CAC-3' and subcloned in pcDNA3 via HindIII and XhoI. The cDNAs of Gi2 , human Gß1 and bovine G2 were subcloned into pcDNA3. Plasmids for GFP-GRP1PH and Ras N17 are described elsewhere .19t]]i5, http://www.100md.com
For generation of expression plasmids encoding CFP-tagged Gß1 and BtkPH, restriction sites were introduced by PCR using the indicated primers, the AdvantageTM II PCR enzyme system and the pGEM®-T Easy Vector for first subcloning. The Gß1 cDNA was amplified using the primers 5'-TAC AAG TCC GGA CAA GCT TCC ATG AGT GAG CTT GAC CAG TTA CGG C-3' and 5'-CGG GAT CCG TCG ACC CAT GGT GGC GTT AGT TCC AGA TCT TGA GGA AGC-3', allowing for in-frame subcloning into the HindIII and BamHI sites of pECFP-C1. The cDNA encoding the PH domain of human Btk and adjacent 5' untranslated bases was amplified from cDNA of dibuturyl-cAMP–differentiated HL-60 cells using the primers 5'-CCA AGT CCT GGC ATC TCA ATG CAT CTG-3' and 5'-TGG AGA CTG GTG CTG CTG CTG GCT C-3'. A nested PCR was performed using the primers 5'-GGA AGA TCT CGA GCC ACC ATG GCC GCA GTG ATT CTG G-3' and 5'-GGG GAT CCC GGG CCC GAG GTT TTA AGC TTC CAT TCC TGT TCT CC-3', allowing for in-frame subcloning into the XhoI and BamHI sites of pECFP-N1 The cDNA inserts and flanking regions of the resulting CFP-Gß1 and BtkPH-CFP constructs were confirmed by sequencing.
Cell culture, transfection, and intranuclear microinjection1@ze, http://www.100md.com
HEK 293 cells were grown at 37°C with 5% CO2 in DME or in MEM with Earle's salts supplemented with 10% FCS, 100 µg/ml streptomycin, and 100 U/ml penicillin. All transfections were done with a FuGENETM 6 transfection reagent (Roche) following the manufacturer's recommendations. For fluorescence microscopy experiments, cells were seeded on glass coverslips. For confocal imaging of the subcellular localization of the PI3K subunits, cells were transfected with 0.1 µg of plasmid encoding YFP-tagged PI3K subunits, 0.2 µg of plasmid for the untagged complementary subunit, and 1 µg each of plasmid for Gß1 and G2 (except for the experiment shown in B; 0.1 µg YFP-p101 + 2 µg Gß1 or G2 alone, or 0.5 µg Gß1 + 0.5 µg G2 + 1 µg Gi2). For monitoring PH domain translocation, or Akt- or ERK phosphorylation, HEK cells were transfected with 0.2 µg plasmid encoding the fMLP receptor, and 0.4 µg each of the plasmids encoding the PI3K subunits and the fluorescent PH domain. The total amount of transfected cDNA was always kept constant (2.5 µg/well) by the addition of empty expression vector. For FRET analysis, cells were transfected with 1.5 µg of plasmid encoding the YFP-tagged p110 and 0.5 µg of plasmid for CFP-tagged p101. All experiments were performed one or two days after transfection.
VSM cells from neonatal rat aortas were cultured as described previously . Cells were seeded on glass coverslips and microinjected with the plasmids for GFP-GRP1PH (0.4 µg/ml), fMLP receptor (0.05 µg/ml), p110 (0.05 µg/ml), and p101 (0.05 µg/ml) using an Eppendorf micromanipulator and transjector. Confocal images were taken one day after microinjection.2j, 百拇医药
Immunoblot analysis of whole-cell lysates, gel filtration, and membrane fractions2j, 百拇医药
p110/p101 titration; transfected HEK cells were lysed in Laemmli buffer and whole-cell lysates including nuclei were subjected to a 10% SDS-PAGE. Proteins were blotted on nitrocellulose membranes, probed with anti-GFP and peroxidase-coupled anti–rabbit IgG antibodies and visualized by ECL . Akt and ERK phosphorylation; HEK 293 cells were transfected as indicated with the plasmids encoding the fMLP receptor (0.2 µg), PI3K subunits (each 0.4 µg), and H-Ras N17 (1.5 µg). Cells were serum-starved overnight, then stimulated as indicated, and lysed in Laemmli buffer. Whole-cell lysates were subjected to SDS-PAGE and immunoblotting, and probed with anti–Phospho-Akt or anti–Phospho-ERK1/2 and anti-ERK1/2 as described previously . Gel filtration; HEK cells were transfected with equal amounts of the plasmids for YFP-p110 (or p110-YFP) and CFP-p101 (or pcDNA3). Cells were washed in PBS and lysed in 20 mM Hepes, pH 7.5, 150 mM NaCl, 10 mM ß-mercaptoethanol, 4 mM EDTA, 0.2% polyoxyethylene-10-laurylether (C12E10), 200 µM Pefabloc® SC, 30 µg/ml TPCK, 30 µg/ml trypsin inhibitor, and 50 µg/ml benzamidine by repetitive aspiration through a 26-gauge needle. Cytosols were prepared (100,000 g for 30 min at 4°C), and subjected to gel filtration on a 24-ml Superdex 200 column and eluted with the same buffer on an ÄKTATM purifier system . Fractions were collected, and aliquots were subjected to SDS-PAGE and immunoblotting with anti-GFP. Preparation of membrane fractions; HEK cells were transfected as indicated with the plasmids for YFP-p110-K833R (0.1 µg), p101 (0.2 µg), Gß1 (1.0 µg), and G2 (1.0 µg). Cells were washed with PBS and lysed in PBS with protease inhibitors (see above) by repetitive aspiration through a 26-gauge needle. Membranes were prepared (16,000 g for 20 min at 4°C), redissolved in Laemmli buffer, subjected to SDS-PAGE and immunoblotting, and probed with anti-p110 (see next section) and anti-Gßcommon (AS398; .
In vitro PI3K assay'', 百拇医药
HEK cells were transfected with equal amounts of the plasmids encoding p110 and p101 (or empty pcDNA3 to keep the total amount of transfected cDNA constant). Cell lysates were subjected to immunoprecipitation using a monoclonal anti-p110 antibody . In brief, protein A Sepharose was preincubated with or without antibody, washed, incubated overnight with cleared lysates, and washed again. The in vitro PI3K assay was performed as described previously . Radioactive PtdIns-3,4,5-P3 was visualized with a PhosphorImager (Fuji Bas-Reader 1500; Ray Test).'', 百拇医药
Fluorescence imaging and detection of FRET'', 百拇医药
Glass coverslips were mounted on a custom-made chamber and covered with a Hepes-buffered solution containing 138 mM NaCl, 6 mM KCl, 1 mM MgCl2, 1 mM CaCl2, 5.5 mM glucose, 10 mM Hepes, pH 7.5, and 2 mg/ml BSA. For confocal imaging, an inverted confocal laser scanning microscope with a Plan-Apochromat 63x/1.4 objective (model LSM 510; Carl Zeiss MicroImaging, Inc.) was used. YFP or GFP were excited at 488 nm, and fluorescence emission was detected through a 505-nm long pass filter. CFP was excited 458 nm, and detected through a 475-nm long pass filter. Pinholes were adjusted to yield optical sections of 0.5–1.5 µm.
FRET analysis was performed using an inverted microscope with a Plan-Apochromat 63x/1.4 objective (Axiovert 100; Carl Zeiss MicroImaging, Inc.). CFP and YFP were alternately excited at 440 and 500 nm with a monochromator (Polychrome II; TILL Photonics) in combination with a dual reflectivity dichroic mirror (<460 nm and 500–520 nm; ). Emitted light was filtered through 475–505-nm (CFP) and 535–565-nm (YFP) band pass filters, changed by a motorized filter wheel (Lambda 10/2; and detected by a cooled CCD camera (Imago; TILL Photonics). FRET was assessed as recovery of CFP (donor) fluorescence during YFP (acceptor) bleach. First, CFP and YFP fluorescences without acceptor bleach were recorded during 30 cycles with a 10–20-ms exposure. Then, 60 cycles were recorded with an additional 2-s illumination per cycle at 510 nm to bleach YFP.o-n\cpc, http://www.100md.com
Acknowledgmentso-n\cpc, http://www.100md.com
Thanks to Claudia Plum, A. Schulz and J. Malkewitz for technical assistance. We are grateful to Drs. P. Gierschik, A. Gray and P. Downes, M.I. Simon, L. Stephens and P. Hawkins, and R. Wetzker for providing experimental tools. Thanks to Celsa Antao-Paetsch for assistance in preparation of the manuscript. We appreciate valuable discussions with Drs. Len Stephens, Phil Hawkins, Lewis Cantley, Reinhard Wetzker, and Roland Piekorz.
This work was supported by the Deutsche Forschungsgemeinschaft and Fonds der Chemischen Industrie.j)r, 百拇医药
Submitted: 21 October 2002j)r, 百拇医药
Revised: 2 December 2002j)r, 百拇医药
Accepted: 2 December 2002j)r, 百拇医药
Referencesj)r, 百拇医药
Al-Aoukaty, A., B. Rolstad, and A.A. Maghazachi. 1999. Recruitment of pleckstrin and phosphoinositide 3-kinase into the cell membranes, and their association with Gß after activation of NK cells with chemokines. J. Immunol. 162:3249–3255.j)r, 百拇医药
Balla, T., T. Bondeva, and P. Várnai. 2000. How accurately can we image inositol lipids in living cells? Trends Pharmacol. Sci. 21:238–241.j)r, 百拇医药
Baier, R., T. Bondeva, R. Klinger, A. Bondev, and R. Wetzker. 1999. Retinoic acid induces selective expression of phosphoinositide 3-kinase in myelomonocytic U937 cells. Cell Growth Differ. 10:447–456.j)r, 百拇医药
Bondev, A., T. Bondeva, and R. Wetzker. 2001. Regulation of phosphoinositide 3-kinase {gamma} protein kinase activity in vitro and in COS-7 cells. Signal Transduction. 1:79–85.
Bondeva, T., L. Pirola, G. Bulgarelli-Leva, I. Rubio, R. Wetzker, and M.P. Wymann. 1998. Bifurcation of lipid and protein kinase signals of PI3K{gamma} to the protein kinases PKB and MAPK. Science. 282:293–296..fg%&, 百拇医药
Boulay, F., M. Tardif, L. Brouchon, and P. Vignais. 1990. The human N-formylpeptide receptor. Characterization of two cDNA isolates and evidence for a new subfamily of G-protein-coupled receptors. Biochemistry. 29:11123–11133..fg%&, 百拇医药
Cantley, L.C. 2002. The phosphoinositide 3-kinase pathway. Science. 296:1655–1657..fg%&, 百拇医药
Codina, J., D. Stengel, S.L. Woo, and L. Birnbaumer. 1986. ß-subunits of the human liver Gs/Gi signal-transducing proteins and those of bovine retinal rod cell transducin are identical. FEBS Lett. 207:187–192..fg%&, 百拇医药
Conklin, B.R., Z. Farfel, K.D. Lustig, D. Julius, and H.R. Bourne. 1993. Substitution of 3 amino acids switches receptor specificity of Gq to that of Gi. Nature. 363:274–276..fg%&, 百拇医药
Fruman, D.A., R.E. Meyers, and L.C. Cantley. 1998. Phosphoinositide kinases. Annu. Rev. Biochem. 67:481–507.
Gautam, N., M. Baetscher, R. Aebersold, and M.I. Simon. 1989. A G protein {gamma} subunit shares homology with ras proteins. Science. 244:971–974.w\sz}6., 百拇医药
Gillham, H., M.C. Golding, R. Pepperkok, and W.J. Gullick. 1999. Intracellular movement of green fluorescent protein-tagged phosphatidylinositol 3-kinase in response to growth factor receptor signaling. J. Cell Biol. 146:869–880.w\sz}6., 百拇医药
Gray, A., J. Van Der Kaay, and C.P. Downes. 1999. The pleckstrin homology domains of protein kinase B and GRP1 (general receptor for phosphoinositides-1) are sensitive and selective probes for the cellular detection of phosphatidylinositol 3,4-bisphosphate and/or phosphatidylinositol 3,4,5-trisphosphate in vivo. Biochem. J. 344:929–936.w\sz}6., 百拇医药
Hurley, J.H., and S. Misra. 2000. Signaling and subcellular targeting by membrane-binding domains. Annu. Rev. Biophys. Biomol. Struct. 29:49–79.w\sz}6., 百拇医药
Katso, R., K. Okkenhaug, K. Ahmadi, S. White, J. Timms, and M.D. Waterfield. 2001. Cellular function of phosphoinositide 3-kinases: implications for development, homeostasis, and cancer. Annu. Rev. Cell Dev. Biol. 17:615–675.
Kirsch, C., R. Wetzker, and R. Klinger. 2001. Anionic phospholipids are involved in membrane targeting of PI3-kinase . Biochem. Biophys. Res. Commun. 282:691–696.oud4, http://www.100md.com
Klippel, A., C. Reinhard, W.M. Kavanaugh, G. Apell, M.A. Escobedo, and L.T. Williams. 1996. Membrane localization of phosphatidylinositol 3-kinase is sufficient to activate multiple signal-transducing kinase pathways. Mol. Cell. Biol. 16:4117–4127.oud4, http://www.100md.com
Krugmann, S., P.T. Hawkins, N. Pryer, and S. Braselmann. 1999. Characterizing the interactions between the two subunits of the p101/p110 phosphoinositide 3-kinase and their role in the activation of this enzyme by Gßsubunits. J. Biol. Chem. 274:17152–17158.oud4, http://www.100md.com
Krugmann, S., M.A. Cooper, D.H. Williams, P.T. Hawkins, and L.R. Stephens. 2002. Mechanism of the regulation of type IB phosphoinositide 3OH-kinase by G-protein ß{gamma} subunits. Biochem. J. 362:725–731.oud4, http://www.100md.com
Kurosu, H., T. Maehama, T. Okada, T. Yamamoto, S. Hoshino, Y. Fukui, M. Ui, O. Hazeki, and T. Katada. 1997. Heterodimeric phosphoinositide 3-kinase consisting of p85 and p110ß is synergistically activated by the ß subunits of G proteins and phosphotyrosyl peptide. J. Biol. Chem. 272:24252–24256.
Leopoldt, D., C. Harteneck, and B. Nürnberg. 1997. G proteins endogenously expressed in Sf 9 cells: interactions with mammalian histamine receptors Naunyn Schmiedebergs Arch. Pharmacol. 356:216–224.2lp, 百拇医药
Leopoldt, D., T. Hanck, T. Exner, U. Maier, R. Wetzker, and B. Nürnberg. 1998. Gß{gamma} stimulates phosphoinositide 3-kinase-by direct interaction with two domains of the catalytic p110 subunit. J. Biol. Chem. 273:7024–7029.2lp, 百拇医药
Lietzke, S.E., S. Bose, T. Cronin, J. Klarlund, A. Chawla, M.P. Czech, and D.G. Lambright. 2000. Structural basis of 3-phosphoinositide recognition by pleckstrin homology domains. Mol. Cell. 6:385–394.2lp, 百拇医药
Lopez-Ilasaca, M., P. Crespo, P.G. Pellici, J.S. Gutkind, and R. Wetzker. 1997. Linkage of G protein-coupled receptors to the MAPK signaling pathway through PI3-kinase . Science. 275:394–397.2lp, 百拇医药
Lopez-Ilasaca, M., J.S. Gutkind, and R. Wetzker. 1998. Phosphoinositide 3-kinase is a mediator of Gß-dependent Jun kinase activation. J. Biol. Chem. 273:2505–2508.2lp, 百拇医药
Maier, U., A. Babich, and B. Nürnberg. 1999. Roles of non-catalytic subunits in Gß{gamma} -induced activation of class I phosphoinositide 3-kinase isoforms ß and . J. Biol. Chem. 274:29311–29317.
Maier, U., A. Babich, N. Macrez, D. Leopoldt, P. Gierschik, D. Illenberger, and B. Nürnberg. 2000. Gß52 is a highly selective activator of phospholipid-dependent enzymes. J. Biol. Chem. 275:13746–13754.l{#xk*, 百拇医药
Metjian, A., R.L. Roll, A.D. Ma, and C.S. Abrams. 1999. Agonists cause nuclear translocation of phosphatidylinositol 3-kinase . A Gß-dependent pathway that requires the p110 amino terminus. J. Biol. Chem. 274:27943–27947.l{#xk*, 百拇医药
Murga, C., L. Laguinge, R. Wetzker, A. Cuadrado, and J.S. Gutkind. 1998. Activation of Akt/protein kinase B by G protein-coupled receptors. A role for and ß subunits of heterotrimeric G proteins acting through phosphatidylinositol-3-OH kinase . J. Biol. Chem. 273:19080–19085.l{#xk*, 百拇医药
Naccache, P.H., S. Levasseur, G. Lachance, S. Chakravarti, S.G. Bourgoin, and S.R. McColl. 2000. Stimulation of human neutrophils by chemotactic factors is associated with the activation of phosphatidylinositol 3-kinase . J. Biol. Chem. 275:23636–23641.l{#xk*, 百拇医药
Pacold, M.E., S. Suire, O. Perisic, S. Lara-Gonzalez, C.T. Davis, E.H. Walker, P.T. Hawkins, L. Stephens, J.F. Eccleston, and R.L. Williams. 2000. Crystal structure and functional analysis of Ras binding to its effector phosphoinositide 3-kinase . Cell. 103:931–943.
Reusch, H.P., M. Schaefer, C. Plum, G. Schultz, and M. Paul. 2001. Gß mediate differentiation of vascular smooth muscle cells. J. Biol. Chem. 276:19540–19547./utf[, 百拇医药
Ridley, A.J., H.F. Paterson, C.L. Johnston, D. Diekmann, and A. Hall. 1992. The small GTP-binding protein rac regulates growth factor-induced membrane ruffling. Cell. 7:401–410./utf[, 百拇医药
Schaefer, M., N. Albrecht, T. Hofmann, T. Gudermann, and G. Schultz. 2001. Diffusion-limited translocation mechanism of protein kinase C isotypes. FASEB J. 15:1634–1636./utf[, 百拇医药
Simonsen, A., A.E. Wurmser, S.D. Emr, and H. Stenmark. 2001. The role of phosphoinositides in membrane transport. Curr. Opin. Cell Biol. 13:485–492./utf[, 百拇医药
Stephens, L.R., T.R. Jackson, and P.T. Hawkins. 1993. Agonist-stimulated synthesis of phosphatidylinositol(3,4,5)-trisphosphate: a new intracellular signalling system? Biochim. Biophys. Acta. 1179:27–75./utf[, 百拇医药
Stephens, L., A. Smrcka, F.T. Cooke, T.R. Jackson, P.C. Sternweis, and P.T. Hawkins. 1994. A novel phosphoinositide 3 kinase activity in myeloid-derived cells is activated by G protein ß subunits. Cell. 77:83–93.
Stephens, L.R., A. Eguinoa, H. Erdjument-Bromage, M. Lui, F. Cooke, J. Coadwell, A.S. Smrcka, M. Thelen, K. Cadwallader, P. Tempst, and P.T. Hawkins. 1997. The Gß sensitivity of a PI3K is dependent upon a tightly associated adaptor, p101. Cell. 89:105–114.r5, 百拇医药
Stoyanov, B., S. Volinia, T. Hanck, I. Rubio, M. Loubtchenkov, D. Malek, S. Stoyanova, B. Vanhaesebroeck, R. Dhand, B. Nürnberg, et al. 1995. Cloning and characterization of a G protein-activated human phosphoinositide-3 kinase. Science. 269:690–693.r5, 百拇医药
Tang, X., and C.P. Downes. 1997. Purification and characterization of Gß-responsive phosphoinositide 3-kinases from pig platelet cytosol. J. Biol. Chem. 272:14193–14199.r5, 百拇医药
Teruel, M.N., and T. Meyer. 2000. Translocation and reversible localization of signaling proteins: a dynamic future for signal transduction. Cell. 103:181–184.r5, 百拇医药
Ueki, K., D.A. Fruman, S.M. Brachmann, Y.H. Tseng, L.C. Cantley, and C.R. Kahn. 2002. Molecular balance between the regulatory and catalytic subunits of phosphoinositide 3-kinase regulates cell signaling and survival. Mol. Cell. Biol. 22:965–977.
Vanhaesebroeck, B., S.J. Leevers, K. Ahmadi, J. Timms, R. Katso, P.C. Driscoll, R. Woscholski, P.J. Parker, and M.D. Waterfield. 2001. Synthesis and function of 3-phosphorylated inositol lipids. Annu. Rev. Biochem. 70:535–602.(u, 百拇医药
Várnai, P., K.I. Rother, and T. Balla. 1999. Phosphatidylinositol 3-kinase-dependent membrane association of the Bruton's tyrosine kinase pleckstrin homology domain visualized in single living cells. J. Biol. Chem. 274:10983–10989.(u, 百拇医药
Walker, E.H., O. Perisic, C. Ried, L. Stephens, and R.L. Williams. 1999. Structural insights into phosphoinositide 3-kinase catalysis and signalling. Nature. 402:313–320.(u, 百拇医药
Wishart, M.J., G.S. Taylor, and J.E. Dixon. 2001. Phoxy lipids: Revealing PX domains as phosphoinositide binding modules. Cell. 105:817–820.(u, 百拇医药
Yu, J., Y. Zhang, J. McIlroy, T. Rordorf-Nikoloic, G.A. Orr, and J.M. Backer. 1998. Regulation of the p85/p110 phosphatidylinositol 3'-kinase: stabilization and inhibition of the p110 catalytic subunit by the p85 regulatory subunit. Mol. Cell. Biol. 18:1379–1387.(Carsten Brock Michael Schaefer H. Peter Reusch Cor)