Essential roles of PI(3)K-p110beta in cell growth, metabolism and tumorigenesis

Abstract

On activation by receptors, the ubiquitously expressed class IA isoforms (p110alpha and p110beta) of phosphatidylinositol-3-OH kinase (PI(3)K) generate lipid second messengers, which initiate multiple signal transduction cascades. Recent studies have demonstrated specific functions for p110alpha in growth factor and insulin signalling. To probe for distinct functions of p110beta, we constructed conditional knockout mice. Here we show that ablation of p110beta in the livers of the resulting mice leads to impaired insulin sensitivity and glucose homeostasis, while having little effect on phosphorylation of Akt, suggesting the involvement of a kinase-independent role of p110beta in insulin metabolic action. Using established mouse embryonic fibroblasts, we found that removal of p110beta also had little effect on Akt phosphorylation in response to stimulation by insulin and epidermal growth factor, but resulted in retarded cell proliferation. Reconstitution of p110beta-null cells with a wild-type or kinase-dead allele of p110beta demonstrated that p110beta possesses kinase-independent functions in regulating cell proliferation and trafficking. However, the kinase activity of p110beta was required for G-protein-coupled receptor signalling triggered by lysophosphatidic acid and had a function in oncogenic transformation. Most strikingly, in an animal model of prostate tumour formation induced by Pten loss, ablation of p110beta (also known as Pik3cb), but not that of p110alpha (also known as Pik3ca), impeded tumorigenesis with a concomitant diminution of Akt phosphorylation. Taken together, our findings demonstrate both kinase-dependent and kinase-independent functions for p110beta, and strongly indicate the kinase-dependent functions of p110beta as a promising target in cancer therapy.

Figure 1. Mice with liver-specific deletion of p110β exhibit insulin resistance and glucose intolerance

Eight to ten week old mice were injected with adenoviruses expressing LacZ or Cre-recombinase. Two weeks post-injection, metabolism was analyzed as follows: a. Fasted serum insulin levels (n=6~7). b. Glucose-tolerance test (GTT) (n=6~11). c. Insulin tolerance test (ITT) (n=6~9). Results represent blood glucose concentrations as a percentage of starting value at time zero. d. Pyruvate challenge (n=6~7). Data represent the mean ± SEM (standard error of the mean). *p<0.05, ** p<0.01, ***p<0.001 (t test). “n” represents the number of mice used in each experiment.

Figure 2. Analyses of the effects of p110β deletion on cell growth and signaling

a. Loss of p110β retards cell growth of primary MEFs. Cells were stained with crystal violet followed by plate reading at 590 nm. Absorbance (Abs) at 590 nm was normalized to Day 0 to show relative growth. Data represent the mean from triplicate with SD (standard deviation). b. Deletion of endogenous p110β protein in immortalized MEFs. Cell lysates from immortalized MEFs were immunoprecipitated using antibody against p110β and blotted with an anti p85 antibody. In parallel the same lysates were immunoblotted with antibodies against p110α, p85, PTEN and tubulin. c. Loss of p110β has no negative effect on insulin signaling. MEFs were starved and then stimulated with insulin at different concentrations for 10 minutes. Phospho-Akt was used as the signaling readouts. d. Loss of p110β impairs LPA-induced signaling. MEFs were starved and then stimulated with LPA (10 mM). Phospho-Akt and phospho-S6 ribosomal protein were used as readouts. e. Comparing the responses of αKO, βKO and WT MEFs to EGF, IGF or LPA stimulation. MEFs were starved and then stimulated with EGF (10 ng/ml), IGF1 (5 ng/ml) or LPA (10 mM) for 10 minutes. Phospho-Akt was used as the readout. Tubulin served as the loading control. f. BrdU incorporation assay. The percentage of cells in S-phase was measured by incorporation of BrdU into newly synthesized DNA via FACS analysis in various MEF lines as indicated. g. Transferrin uptake in various MEF lines is shown as indicated. The red signal arises from Alexa Fluro 555-conjugated transferrin while blue color arises from DAPI staining of nuclear DNA.

Figure 3. Kinase activity of p110β contributes to transformation both in vitro and in vivo

a. Focus formation assay was carried our in various KO and reconstituted MEF lines as indicated. The assay was carried out as described in Methods and the means (±SEM) for 4 independent experiments are shown (*, p<0.05, **P<0.01, t test). b. Effects of genetic ablation of p110β or p110α on tumorigenesis caused by PTEN loss in the anterior prostate. Paraffin sections of anterior prostates of indicated strain aged 12 weeks were stained with haematoxylin and eosin (H&E) and phospho-Akt. c. Quantification of the weight (±SEM) of anterior prostate tissues of the indicated strain (n=10 per group; *, p < 0.001, t test). d. A model for the elevation of basal PIP3 signals derived from p110β catalytic activity induced by PTEN loss.