WO2013169749A1 - Phosphatidylinositol monophosphate levels as disease prognosticators - Google Patents

Abstract

A method of screening a test subject for a disparity in PTEN activity, indicative of the severity of a proliferative disease, or the propensity for developing a proliferative disease, diabetes, obesity, a developmental disorder or autoimmune disease, a neurological disorder or other PTEN-related disease, disorder or condition by determining the relative amount of PI(3)P or optionally PI(3)P and one or more additional phosphatidylinositol phosphate species selected from PI, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃. Also provided are uses of such assays for monitoring a proliferative disease or other health disparity in a subject undergoing therapy for a proliferative disease; and methods of assessing the effect of a prospective therapeutic agent on the tumor suppressor activity of PTEN by determining or inferring the effect of the prospective therapeutic agent on the relative amount of PI(3)P and optionally one or more phosphatidylinositol phosphate species in cells from a normal subject.

Description

PHOSPHATIDYLINOSITOL MONOPHOSPHATE LEVELS AS DISEASE

PROGNOSTICATORS

Government Funding

This work was supported in part by grant No. W81XWH-09-1-0557 from the Department of the Army, and grant No. 1R01CA137050-01A2 from the National Institutes of Health. The government has certain rights in the invention.

Technical Field

The invention relates to the field of diagnostics for determining the severity of disease or the future risk for cancer, diabetes, obesity, developmental disorders and autoimmune diseases, neurological disorders and other related conditions, wherein the clinical benefits of the prediction of propensity for development of a disease, disorder or condition is merited. The invention further relates to the prognosis and monitoring of such diseases, disorders and conditions.

More specifically, in one embodiment the invention relates to assay methods to determine PTEN activity through the detection and localization of PTEN and the quantification of phosphatidylinositol phosphate species in cellular or tissue-derived lipid samples for use in screening individual subjects and populations to assess the risk of development of a proliferative disease, diabetes, obesity, developmental disorder, autoimmune disease, neurological disorder or a related disease, disorder or condition.

In another embodiment, the invention relates to assay methods to determine the ratio of the total amount of the phosphatidylinositol monophosphate species including PI(3)P, PI(4)P and PI(5)P to the unphosphorylated phosphatidylinositol in the lipid sample; comparing the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in the lipid sample with the normal range of the ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in lipid samples from subjects not known to have prostate proliferative disease, and thereby assessing the severity of the proliferative disease of the prostate in the subject .

Finally, the invention relates to use of the specified assay methods for the discovery and/or improvement of therapeutic interventions for the treatment of the above-listed diseases, disorders and conditions. Background

The PTEN (phosphatase and tensin homolog, deleted on chromosome 10) a tumor suppressor, identified 15 years ago (Li et al., 1997: PTEN, a putative protein tyrosine phosphatase gene mutated in human brain, breast, and prostate cancer: Science 275, 1943-1947; and Steck et al., 1997: Identification of a candidate tumour suppressor gene, MMAC1, at chromosome 10q23.3 that is mutated in multiple advanced cancers. Nat Genet 15, 356-362), is among the most frequently altered genes in cancer. Realizing that this phosphatase controls phospholipid second messengers (Maehama and Dixon, 1998 The tumor suppressor, PTEN/MMAC1, dephosphorylates the lipid second messenger, phosphatidylinositol 3,4,5- trisphosphate. J Biol Chem 273, 13375-13378) has profoundly changed our understanding of signaling pathways in disease. PTEN antagonizes the class I PI 3-Kinases (PI 3-Ks), which produce the lipid second messenger PI(3,4,5)P₃ (PIP₃) in response to activation of receptor tyrosine kinases (RTKs), G-protein coupled receptors or membrane bound oncogenes (reviewed in Engelman J. A. et al., 2006, The evolution of phosphatidyl-inositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet 7, 606). PIP₃ then serves as a membrane platform for the coordinate activation of downstream signaling kinases, adaptor proteins and phosphatases that bind PIP₃ via their PH domain. Central among those kinases is the AKT proto-oncogene, which controls downstream targets to promote cell growth, proliferation, survival, metabolism and motility pathways (reviewed in Fayard, E. et al., 2010, Protein kinase B (PKB/Akt), a key mediator of the PI3K signaling pathway. Curr Top Microbiol Immunol 346: 31).

PTEN is the major negative regulator of PI 3-Kinase/AKT signaling in cells. While the gene is frequently deleted and mutated in sporadic and familial cancer, research in model organisms has demonstrated that partial loss of function suffices to initiate disease (Alimonti et al., 2010: Subtle variations in Pten dose determine cancer susceptibility. Nature Genetics 42, 454-458; Trotman et al., 2003: Pten dose dictates cancer progression in the prostate. PLoS Biol 1, E59). In contrast, complete PTEN loss triggers growth arrest (Chen et al., 2005 Crucial role of p53-dependent cellular senescence in suppression of Pten-deficient tumorigenesis, Nature 436, 725-730; Kim et al., 2007: Activation of p53-Dependent Growth Suppression in Human Cells by Mutations in PTEN or PIK3CA. Mol Cell Biol 27(2):662-77). Hence, PTEN has emerged as the prototypic haploinsufficient tumor suppressor (Berger et al., 2011 A continuum model for tumour suppression. Nature 476, 163-169). Based on these findings, the regulation of PTEN localization and activity have been under intense investigation. However, a major paradox has remained unsolved: while PTEN function is clearly ascribed to the cell's plasma membrane, only scant evidence for such localization has been found in cells or tissues.

Summary

The present findings provide new approaches for assessment of the severity of proliferative diseases and conditions in a subject and susceptibility or risk of a subject for developing such proliferative diseases and conditions based on particular interactions of PTEN with different phosphatidylinositol phosphate species. The inventors have found that the ratio in a cell of the amount of the sum of the three monophosphorylated phosphatidyl inositols to the amount of the unphosphorylated phosphatidyl inositol is a predictor of the severity of the disease of a cancerous cell or of the propensity of a cell to develop a proliferative disease or condition.

The inventors have determined that in normal cells, the ratio of the amount of the sum of the three monophosphorylated phosphatidyl inositols to the amount of the unphosphorylated phosphatidyl inositol is a ratio of about 1 part by weight of monophosphorylated phosphatidyl inositols to about five parts by weight of unphosphorylated phosphatidyl inositol. By contrast, in cells having a propensity for developing a proliferative disease, the ratio of the amount of the sum of the three monophosphorylated phosphatidyl inositols to the amount of the unphosphorylated phosphatidyl inositol is between about 1: 15 to 1:50, i.e., is about 1 part by weight of monophosphorylated phosphatidyl inositols to a range from about fifteen to fifty parts by weight of unphosphorylated phosphatidyl inositol.

The present invention provides a method of assessing the severity of a proliferative disease in a test subject suffering from a proliferative disease, the method comprising: providing a lipid sample from a biopsy comprising one or more cells of the test subject; determining the ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in the lipid sample; comparing the ratio of the amount of the phosphatidylinositol monophosphate species to the amount of unphosphorylated phosphatidylinositol in the lipid sample with the normal range of the ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in lipid samples from subjects not known to have prostate proliferative disease, and thereby assessing the severity of the proliferative disease of the prostate in the subject.

In one embodiment test cells are compared with control cells from the same organ or tissue of the test subject, wherein the control cells are not suspected of having a propensity for developing a proliferative disease or condition. For example, the test cells may be from an adenoma from a tissue or organ of the test subject and the control cells may be from a normal portion of the tissue or organ of the test subject, for example, as indicated by histology or by the absence of tumor markers. Alternatively, the test cells can be compared with control cells from a different tissue from the same subject. Alternatively, the test cells can be compared with control cells from a different subject.

In one embodiment of the above-described method, the proliferative disease is a proliferative disease of an exocrine or endocrine gland, including an adenoma, a benign hyperplasia or a cancer. For example, in one embodiment, the proliferative disease is a proliferative disease of the prostate, for example, benign prostatic hyperplasia or prostate adenoma, or prostate cancer, and test cells may be obtained from tissue within the prostate, or more specifically from a prostate biopsy conducted on a test subject.

In one embodiment of the above-described method, the phosphatidylinositol monophosphate species comprises PI(3)P.

In one embodiment of the above-described method, the phosphatidylinositol monophosphate species comprises PI(3)P, PI(4)P and PI(5)P, and a lower ratio of PI(3)P, PI(4)P and PI(5)P to unphosphorylated phosphatidylinositol in the lipid from the test subject than the normal range of PI(3)P, PI(4)P and PI(5)P to unphosphorylated phosphatidylinositol in the lipid is indicative of a more severe proliferative disease in the test subject. The phosphatidylinositol monophosphate species as a group can be resolved from other phosphatidylinositol species by any suitable method, such as for instance, by mass spectroscopy, or by thin layer chromatography (TLC), ion-exchange chromatography, high performance liquid chromatography (HPLC), gas chromatography, gel chromatography, or chromatography in a microfluidic system, or by an antibody-based separation system. In one embodiment, the present invention provides a method of screening a test subject for the propensity for developing a proliferative disease or condition, the method includes the following steps:

providing a lipid sample from one or more cells of the test subject, wherein the cells are suspected of having or having a propensity for a proliferative disease or condition; determining the amount of the phosphatidylinositol monophosphate species: i.e. the sum of the three PI monophosphate species: PI(3)P + PI(4)P + PI(5)P in the lipid sample relative to the amount of phosphatidylinositol (PI); comparing the relative amount of the phosphatidylinositol monophosphate species to the phosphatidylinositol (PI) in the lipid sample from the test subject with the relative amount of the phosphatidylinositol monophosphate species to the phosphatidylinositol (PI) in a lipid sample from one or more normal subjects not known to have a propensity for developing a proliferative disease or condition, and thereby assessing the propensity of the test subject for developing a proliferative disease or condition.

In one embodiment test cells are compared with control cells from the same organ or tissue of the test subject, wherein the control cells are not suspected of having a propensity for developing a proliferative disease or condition. For example, the test cells may be from an adenoma from a tissue or organ of the test subject and the control cells may be from a normal portion of the tissue or organ of the test subject, for example, as indicated by histology. Alternatively, the test cells can be compared with control cells from a different normal tissue or organ from the same subject, for example, as indicated by histology or the absence of tumor markers. Alternatively, the test cells can be compared with control cells from a different subject, where such control cells are not suspected of having a propensity for developing a proliferative disease or condition.

In one embodiment of the above-described method, the proliferative disease is a proliferative disease of an exocrine or endocrine gland, including an adenoma, a benign hyperplasia or a cancer. For example, in one embodiment, the proliferative disease is a proliferative disease of the prostate, for example, benign prostatic hyperplasia or prostate neoplasia, or prostate cancer, and test cells may be obtained from tissue within the prostate, or more specifically from a prostate biopsy conducted on a test subject. In one embodiment of the above-descibed method, the relative amount of the phosphatidylinositol monophosphate species to the phosphatidylinositol (PI) in the lipid sample from the test subject is compared with the range of phosphatidylinositol monophosphate species to the phosphatidylinositol (PI) in a normal population.

In another embodiment, the invention provides a method of monitoring a test subject having a proliferative disease or condition, the method includes the following steps: obtaining a first lipid sample from one or more cells of the test subject at a first time point;determining the amount of the phosphatidylinositol monophosphate species: i.e. the sum of the three PI monophosphate species: PI(3)P + PI(4)P + PI(5)P in the lipid sample relative to the amount of phosphatidylinositol (PI) in the first lipid sample; obtaining a lipid sample from one or more cells of the test subject at a second time point subsequent to the first time point;determining the amount of the phosphatidylinositol monophosphate species: i.e. the sum of the three PI monophosphate species: PI(3)P + PI(4)P + PI(5)P in the lipid sample relative to the amount of phosphatidylinositol (PI) in the second lipid sample; comparing the relative amount of the phosphatidylinositol monophosphate species to the phosphatidylinositol (PI) in the first lipid sample with the relative amount of the phosphatidylinositol monophosphate species to the phosphatidylinositol (PI) in the second lipid sample, and thereby monitoring the proliferative disease or condition in the test subject.

In one embodiment of the above-described method, the proliferative disease is a proliferative disease of an exocrine or endocrine gland, including an adenoma, a benign hyperplasia or a cancer. For example, in one embodiment, the proliferative disease is a proliferative disease of the prostate, for example, benign prostatic hyperplasia or prostate neoplasia, or prostate cancer, and test cells may be obtained from tissue within the prostate, or more specifically from a prostate biopsy conducted on a test subject.

In one embodiment of the above monitoring method, the test subject is undergoing therapy for the proliferative disease or condition. In an alternative embodiment, the test subject is not undergoing a therapeutic regiment, but the proliferative disease or condition may be monitored in order that the clinicial may decide on an appropriate therapeutic regimen. In still another alternative, the test subject may elect to have the the proliferative disease or condition monitored before any therapeutic intervention.

The therapeutic intervention may be any suitable therapeutic intervention, such as for instance, therapy with a PI 3-Kinase inhibitor, i.e. a drug that replace PTEN enzyme function, or PTEN agonists, e.g. a lipid or other mimicking substances capable of activating PTEN enzyme. Alternatively, the therapeutic intervention may be administration of a kinase inhibitor that blocks the reverse of the PTEN-catalyzed reaction, shifting the balance of PIP-3 to PIP-2 towards PIP-2, which is the product of the normal functioning of the PTEN enzyme.

In one embodiment, the therapeutic intervention can be administration of monophosphoryl-phosphatidyl inositols (monophosphorylated PIPs) that are capable of entering the affected cells and activating PTEN.

The tumor suppressive action of PTEN is believed to occur at the plasma membrane. However, evidence for its localization to plasma membranes has remained surprisingly scarce both in vitro and in vivo. Here the inventors show that PTEN functions on early endosomes to which it is recruited by the class III PI 3-kinase, VPS34, the enzyme required for PI(3)P synthesis. PTEN is shown to localize in punctae along microtubules as it binds to PI(3)P, the signature lipid of early endosomes. Inhibition of VPS34 blocks generation of PI(3)P, reducing PTEN protein levels and obstructing its function. These results reveal a surprising constitutive upstream activation mechanism of PTEN.

VPS 34 and its product PI(3)P translate nutrient (amino acids and glucose) uptake into cell growth through activation of mTOR Complex 1. Hence, drugs which summarily target VPS34 and the class I PI 3-Kinases, their downstream effectors AKT kinase and mTOR kinase are currently being tested and developed as anti-cancer agents. The PTEN activating recruitment to early endosomes by PI(3)P is entirely unexpected. The inventors now show that surprisingly and paradoxically, the lowering of VPS34 and therefore also of PI(3)P levels leads to mislocalization of PTEN and a consequent lowering of PTEN activity, which is expected to lead to a reduction in tumor suppressor activity in the cell. The inventors now also show how negative effects of PI 3-Kinase inhibitors on PTEN can be tested in cancer therapy. The methods of the present invention permit screening for PTEN cytoplasmic mislocalization and also for reduced cellular PTEN protein levels and for reduced PI(3)P levels, each of which indicate an increased risk of developing a proliferative disease, a developmental disorder or autoimmune disease, a neurological disorder or other related disease, disorder or condition. Conversely, a finding of higher cellular PTEN protein levels and/or PI(3)P levels indicate an increased risk of developing obesity, diabetes or a diabetic condition. Such proliferative diseases, diabetes and diabetic conditions, obesity, developmental disorders, autoimmune diseases, neurological disorders and other related diseases, disorders and conditions are interchangeably individually referred to herein as "a health disparity" or collectively as "health disparities."

In one embodiment, the present invention provides a method of screening a test subject for the propensity for developing a health disparity. The method includes the steps of: providing a lipid sample from one or more cells of the test subject; determining the relative amount of one or more phosphatidylinositol phosphate species in the lipid sample, wherein the one or more of the phosphatidylinositol phosphate species are selected from the group consisting of PI (phosphatidyl inositol), PI(3)P (phosphatidyl inositol 3-phosphate), PI(4)P (phosphatidyl inositol 4-phosphate), PI(5)P (phosphatidyl inositol 5-phosphate), PI(3,4)P₂, (phosphatidyl inositol 3,4- phosphate) PI(3,5)P₂ (phosphatidyl inositol 3, 5 -phosphate), PI(4,5)P₂ (phosphatidyl inositol 4,5- phosphate) and PI(3,4,5)P₃ (phosphatidyl inositol 3,4,5-phosphate); comparing the relative amount of the one or more of the phosphatidylinositol phosphate species in the lipid sample from the test subject with the relative amount of the one or more phosphatidylinositol phosphate species in a lipid sample from one or more normal subjects not known to have a propensity for developing a health disparity, and thereby assessing the propensity of the test subject for developing a proliferative disease, diabetes, obesity, a developmental disease, autoimmune disease, neurological disorder or other related disease, disorder or condition. Alternatively, the relative amount of the one or more of the phosphatidylinositol phosphate species in the lipid sample from the test subject can be compared with one or more lipid samples derived from cells or tissues from parts of the body of the test subject other than that of the original lipid test sample. In another embodiment, the invention provides a method of determining the cellular localization of PTEN in a sample of cells: The method includes the steps of: providing a sample of one or more cells of the subject; contacting the cells with a PTEN-binding agent under suitable conditions for binding of the PTEN-binding agent to PTEN in the cells to produce PTEN-binding agent:PTEN complexes; detecting the PTEN-binding agent:PTEN complexes to display the distribution of PTEN in the one or more cells; and determining the distribution of PTEN in the one or more cells as being punctuate or diffuse, wherein a punctuate distribution is indicative of a PI(3)P:PTEN association at a vesicle, and a diffuse distribution is indicative of a non-lipid-bound cytoplasmic localization of PTEN. Similarly, in the nucleus, a punctuate distribution is indicative of a PI(3)P:PTEN association, and a diffuse nuclear distribution is indicative of non-PI(3)P bound PTEN. Without wishing to be bound by theory, it is believed that attenuation of the interaction between PTEN and PI(3)P leads to less PTEN enzyme activity and hence a reduction in its regulatory effects, such as for instance, tumor suppression.

In another embodiment, the invention provides a method of monitoring progression a proliferative disease, diabetes, a diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or a related disease, disorder or condition in a subject undergoing a therapy for the disease, disorder or condition. The method includes the steps of: providing two or more lipid samples from cells of the subject taken at different times during the therapy; determining the relative amount of one or more phosphatidylinositol phosphate species selected from the group consisting of PI, PI(3)P, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃ in the lipid samples from the subject; comparing the relative amount of the one or more phosphatidylinositol phosphate species of the lipid sample from the subject at the earliest time before or during therapy with the relative amount of the one or more phosphatidylinositol phosphate species in each of the lipid samples taken from the subject at one or more subsequent times during or after the therapy; and thereby monitoring the progression of or remission from the disease, disorder or condition, or the progress of the therapy.

In another embodiment, the invention provides a method of assessing the effect of a prospective therapeutic agent on PTEN tumor suppressor activity. The method includes the steps of: providing a first lipid sample from one or more cells from a normal subject or a tissue culture cell line and a second lipid sample from one or more cells of the normal subject or tissue culture cell line after exposure to the prospective therapeutic agent; comparing the relative amounts of one or more phosphatidylinositol phosphate species selected from the group consisting of PI, PI(3)P, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃ in the lipid sample from the one or more cells treated with the prospective therapeutic agent with the relative amount of the one or more phosphatidylinositol phosphate species in the lipid sample from the one or more otherwise identical untreated cells; and thereby assessing the effect of the prospective therapeutics agent on PTEN catalytic activity. The effect on PTEN catalytic activity forewarns of an increased likelihood of the prospective therapeutic agent having a downstream effect on PTEN, such as for instance, an effect on tumor suppressor activity.

In another embodiment, the invention provides a method of assessing the capacity of a prospective therapeutic agent to support PTEN tumor suppressor activity using cell lines or tissue culture cells to screen for PTEN promoting agents, such as for instance inhibitors of the PI(3)P phosphatase MTMR2, that act by raising the level of PI(3)P species and/ or PTEN vesicle localization. Such prospective therapeutic agents are candidate therapies for the treatment of proliferative diseases, developmental disorders, autoimmune diseases, neurological disorders or related diseases, disorders and conditions.

In another embodiment, the invention provides a method of assessing the effect of a prospective therapeutic agent on PTEN tumor suppressor activity using cell lines or tissue culture cells to screen for indirect PTEN inhibitors that act by lowering the PI(3)P species and/ or PTEN vesicle localization. Such prospective therapeutic agents are candidate therapies for the treatment of obesity, diabetes and diabetic conditions.

Brief Description of the Figures

Fig. 1. Shows PTEN organized along microtubules. (A) Immunofluorescence demonstrates that PTEN distribution is punctate, in both nuclear and cytoplasmic compartments. Punctate distribution is conserved in both mouse (primary wild type mouse embryonic fibroblasts (MEF), top panels) and human (HEK293, middle panels) cell lines. PTEN null cells (human prostate cancer-derived LNCaP) display no staining, confirming antibody specificity (bottom panels). Images in extended focus. Scale bars, 10 μιη. (B) Cytoplasmic Pten distribution is indistinguishable from that of transferrin (Tf). Pten and Tf punctate stains show a similar close association with microtubules (MTs). Images of MEFs, Z=l, scale bar, 10 μιη. (C and D) PTEN is organized along MTs, as demonstrated by super-resolution light microscopy (OMX). The distance of Pten signal intensity from MTs is statistically significantly lower than randomly selected locations in the cell cytoplasm (p=3.5x10-36, χ2). Scale bar, 2 μιη. (E) The punctate distribution of Pten is retained subsequent to nocodazole induced MT depolymerization. Images of MEFs, Z=l, scale bar, 10 μιη. (F) Overexpression of PTEN in PTEN-null human prostate cancer-derived PC3 cells. High levels of exogenous PTEN expression (high) provide a blanket PTEN distribution (left and center panels). Intermediate expression of PTEN recapitulates endogenous distribution (left and right panels). Z=l, scale bar, 10 μιη.

Fig. 2. Validation and quantification of PTEN localization. (A) The punctate distribution of PTEN is observed across multiple cell lines and was retained using three independent fixation and permeabilization methodologies (phemo, methanol and paraformaldehyde, see Materials and Methods). MEF and NIH3T3 shown, HEK293 not shown. PTEN-null LNCaP cells demonstrate no PTEN signal when processed under identical conditions to PTEN-positive cells and using identical microscopy settings, confirming signal specificity. All images in extended focus, scale bar, 10 μιη. (B) Additional example of Pten/MT dual labeled immunofluorescence, visualized using super-resolution microscopy. WT MEFs, Z=l, scale bar, 2 μιη. (C) Demonstration of quantification of the distance of Pten signal intensity relative to MT position showing distance markers placed (left panel) and comparative randomly selected cytoplasmic positions (right panel; see Materials and Methods). The thick intermittent white line represents the cell boundary. WT MEFs, Z=l, scale bar, 2 μιη. (D) PTEN overexpression in PC3 cells visualized subsequent to fixation and antibody staining reveals that only low or very low (1 and vl respectively) levels of expression recapitulate endogenous distributions. High levels (h) of expression demonstrate blanket, non-punctate distributions (top panel). PC3 cells un-transfected demonstrate no PTEN signal subsequent to similar fixation, permeabilization and visualization (bottom panel). Z=l, scale bar, 3 μιη.

Fig. 3. PTEN binds directly to the signature early endosomal lipid phosphatidylinositol 3 phosphate, PI(3)P. (A) Recombinant and catalytically active PTEN protein directly binds PI(3)P. Incubation of GST-PTEN and untagged PTEN with immobilized lipids demonstrates similar binding affinity to phopshatidylinositol monophosphates with PI(3)P being most avidly bound. (B) Quantitative assessment of GST-PTEN binding to immobilized phospholipids using a serial dilution PIP- strip and quantification of the results. (C) The cytoplasmic distribution of Pten mimics that of the PI(3)P-binding early endosomal antigen 1 (Eeal) and Tf. Pten, Tf and Eeal display a punctate distribution in the cell periphery (asterisks, left panels) with limited observable colocalization (arrows). In the juxtanuclear region, colocalization of Pten, Tf and Eeal is widely observed (arrows) in larger, putative multi-vesicualar structures. WT MEFs, Z=l, scale bars, 10 μιη. (D) Sedimentation of Pten and EEA1 from NIH3T3 cells using sucrose gradient fractionation identifies Pten and EEA1 in the same fractions. Pten/EEAl sediments partially in fractions containing golgi bodies (Ndfipl) and independently of ribosomal units (S6).

Fig. 4. In vitro and in vivo validation of PTEN-PI(3)P interaction (A) PIP strip controls validating PTEN-PI(3)P binding. GST-PTEN/PTEN lipid interactions are consistent across increasing protein concentrations (0.1 - 4 μg/ml) and increasing duration (1 - 3.5 hours) of protein incubation with lipid strips. Relative PTEN- lipid interaction increases with protein concentration used and incubation duration. Interactions assessed at low temperature (4°C) show reduced although evident binding of PTEN to PI(3)P (all, top panels). Use of two independent anti-PTEN antibodies confirms observed binding. (B) Use of recombinant protein with known lipid binding characteristics (GST tagged PI(3)P-grip, a known PI(3)P binding protein, GST tagged MultiPIP-grip, a know binding protein of several phospholipids with highest affinities for PI(3,4)P2 and PI(3,4,5)P3 lipids, bottom left panel; see Methods and Materials) validates the observed PTEN binding pattern as PTEN- specific. (C) Secondary antibody only controls (no primary antibody used) demonstrates that detection signal is specific for primary antibody binding with no secondary antibody-produced signal detectable even after long exposures (bottom right panel). PIP strip lipid positions are the same across all strips (shown in top left example). PTEN-PI(4,5)P2 interactions are detectable with both antibodies only subsequent to extended exposures (A right and C right, shown with two independent antibodies). (D) Super resolution light microscopy reveals that larger Eeal intensities evident in the juxtanuclear region comprise of multiple smaller Eeal intensities, with frequent evidence of these multi-signal intensities in the immediate vicinity of or enveloping PTEN intensities (right panel, arrows). Images of WT MEFs, Z=l, scale bar, 5 μιη. (E) Immunoblot analysis of lysates derived from PTEN-null (PC3) and PTEN-positive (A549) cells using two independent anti-PTEN antibodies demonstrates their monospecificity. (F) PTEN shows limited colocalization with golgi markers TGN46 (human, Hela cells) or Ndfipl (mouse, WT MEFs). Top panels, Z=l, scale bar, 10 μιη. Bottom panels, in extended focus, scale bar, 10 μιη.

Fig. 5. Amino acid withdrawal alters PTEN distribution and function. (A) Amino acid withdrawal (24 hours) results in a reduction in discernible PTEN and EEAl punctate signal in A549 (left) and WT MEF cells (middle and right panels). More diffuse PTEN stain is observed, ranging from total loss of PTEN punctate signal intensity (MEF 1, signal is only observable subsequent to enhancement) to mixed populations of punctate and diffuse PTEN stains (MEF 2). Z=l, scale bar, 5 μιη. (B) Amino acid starvation (24 hours) followed by 6 hours of starvation in low serum (0.1%) and epidermal growth factor (EGF) stimulation in A549 cells (top panel). Stimulation in the amino acid starved cells gives rise to increased levels of pAkt compared to unstarved cells. Conversely, in starved cells, pS6 and pMEK levels are significantly diminished comparable to unstarved controls. Similar effects are observed in HeLa cells (bottom panel). (C) Mass spectroscopy analysis of A549 cells subsequent to lipid extraction with chloroform (see Materials and Methods) highlights that a single lipid family (consisting of phosphatidylinositol lipids containing a total of 38 carbon atoms and 3 double bonds in both fatty acid chains: 38:3) is present in these cells. The unphosphorylated phosphatidylinositol (PI) lipid type is the most abundant (left hashed red box). The monophosphate phosphoinositides (PIPs; comprising PI(3)P, PI(4)P and PI(5)P) are less abundant at steady state (center red box). As are the phosphatidylinositol bisphosphates (comprising PI(3,4)P2, PI(4,5)P2 and PI(3,5)P2). Of note, phosphotidylinositol triphosphate (PI(3,4,5)P3 (predicted m/z 1127, gray hashed box, arrow) is not discernible above background readings. (D) Fragmentation of predicted PI and PIP species parent ions (887.6 and 967.5 m/z, see Fig. S3) shows characteristic and unique fragment ions corresponding to PI and PIP lipid species (left panels). Amino acid starvation and subsequent quantification of PI and PIP specific ions demonstrates a significant change in PIP:PI ion ratio subsequent to withdrawal (right panel). +AA: A549 cells with amino acids. -AA: A549 cells in the absence of amino acids.

Fig. 6. Validation of PI(3)P suppression by amino acid withdrawal. (A) Amino acid starvation of MEFs (24 hours) gives rise to a more diffuse, Eeal distribution (bottom panel). Conversely, transferrin distribution retains its characteristic punctate pattern (top panel). Z=l scale bar, 3 μιη. (B) shRNA knockdown of Vps34 (top left panel). Four independent shRNA constructs were virally infected into NIH3T3 cells and selected. Forty individual clones were grown-out and Vps34 transcript levels assessed using qPCR. Five clones exhibiting the highest level of knockdown are shown. Clone identifiers: 2J, 3C, 3D, 3E, 31. Control luciferase hairpins (LF and LJ) were also infected and selected. A single clone (3C) demonstrated prominent Vps34 transcript reduction; levels represented as a ratio of Vps34:HPRT crossing points (Cp), error bars SEM, confirmed by immunoblot. siRNA knockdown of Vps34 in NIH3T3 cells using four independent siRNAs (siRNA identifiers: 1, 2, 3 and 4- siRNA 1 not shown); pooling of all four generated the greatest observable reduction in Vps34 protein levels (bottom left panel). Pool siRNA transfection (all 4 simultaneously) into the stable anti-Vps34 shRNA line 3C generated -80-90% knockdown of Vps34 protein levels. Molecular analysis shows only a modest reduction in pS6^ser325/236, a known readout of Vps34 activity. In contrast, amino acid withdrawal (See Fig. 5B) gives rise to strong pS5^ser325/236 suppression. (C) Predicted mass to charge (m/z) ration of cellular lipid species (top panel, Table 1). C12 represents the carbon fatty acid and saturation level (i.e. 32: 1 depicts two fatty acid tails comprising a total of 32 carbon atoms and a single double bond). PI, PIP, PIP2 and PIP3 depict the phosphorylation status of the lipid head group. Wide scan spectra of WT MEFs (middle panel) and HeLa cells (bottom panel) displaying the lipid species detected in these cell lines. The most abundant types are annotated. Note that while A549 cells have one major type of phosphatidylinositol lipid, primary MEFs have two types and HeLa 4 (3 annotated) (D) Biological replicate of PIP:PI lipid quantification subsequent to amino acid withdrawal (24 hours) on A549 cells. Error bars represent SEM of three technical replicates. +AA: with amino acids, -AA: without amino acids. PIP and PI specific ions were measured and quantified under both conditions and used to generate relative ratios (see Methods and Materials).

Fig. 7. PI3K inhibitor treatment leads to mislocalization of PTEN. (A and B) Treatment of A549 and MEF cells with the pan-PI3K inhibitors wortmannin and NVP-Bez235 mislocalizes PTEN. Z=l, PTEN-green, EEAl-red, nuclei-blue (top panels, scale bars 5 μιη). Lower panels show magnifications of PTEN only in cell periphery (scale bars 1 μιη). (C) Mass spectroscopy analysis of HeLa cells subsequent to amino acid starvation, wortmannin and NVP-Bez235 treatment (40 hours). PIP:PI ratios for the four most abundant lipid types present in these cells (see Fig. 6) are diminished subsequent to inhibitor treatment and amino acid withdrawal. NVPBez-235 (red data points), the most effective PI3K inhibitor (see Fig. 8), diminished PIP:PI ratios most effectively in this cell type (all lipid types, p<0.001, two-tailed t-test vs DMSO). (D) Visualization of NVP-Bez235 effects on PTEN localization in HeLa cells. Z=l, scale bar, 5 μιη, top panels, 1 μιη, lower panels. (E) PI3K axis interaction in protozoans and metazoans. At the protozoan evolutionary level, nutrient availability suffices to trigger cell growth via the ancestral VPS34. In metazoans, the hormone/growth factor sensitive class I PI3K axis has evolved, relinquishing growth control to specialized endocrine cells. The constant abundance of nutrients is a prime achievement of multicellular organisms. By coupling nutrient activation to PTEN activity metazoans ensure that the barrier for transmitting hormone activation to molecular effectors is high.

Fig. 8. Validation of PI3K inhibitor effects on PTEN. (A) PI3K inhibitor treatment leads to mislocalization and loss of EEA1 stain. Z=l, scale bars 5 μιη. (B) Biological replicate of measurements of HeLa cell lipid ratios subsequent to amino acid withdrawal, wortmannin and NVP-Bez235 treatment (left panel, identical experiment presented in Fig. 4C). NVP-Bez-235 treated cells had a statistically significant different mean PIP:PI ration comparative to DMSO treated controls for all lipid types (all lipids, p<0.001, two-tailed t-test vs DMSO). Subsequent to treatment, HeLa cell number was assessed (right panel). Error bars, SEM. (C) PI3K inhibitor efficacy on pAKT and pS6 levels. PC3 cells were serum starved overnight and then treated with inhibitors for 30 minutes prior to serum stimulation. All inhibitors reduce pAKT and pS6 levels below those observed in DMSO treated controls, with NVP-Bez235 demonstrating the strongest effects on both pAKT and pS6. Wort.: wortmannin, 3-MA: 3 methyladenine, NVP-Bez: NVP- Bez-235. (D) The molecular determinants of cellular PI(3)P levels include VPS 34 and INPP4b (positive determinants) and MTM/MTMR (negative determinants). The novel implication of these proteins in PTEN localization, and therefore functionality, provides compelling evidence that they function in tumor suppressive or tumor supportive capacities respectively.

Fig. 9. PTEN enzymatic activation by phosphatidylinositol 3 phosphate (PI(3)P). (A) Wildtype PTEN in the absence of its canonical substrate phosphatidylinositol 3,4,5 triphosphoate (PIP3) and presence of PI(3)P (20 uM) shows no enzyme activity. Wildtype PTEN in the presence of PIP3 (20 uM) and the absence of PI(3)P demonstrates phosphatase activity. In the presence of PI(3)P and PIP3 (both 20 uM) wildtype PTEN is super-activated. (B) Deletion-mutant PTEN protein (absent the first 16 n-terminal residues) demonstrates no activity in the presence of substrate. Nor is it active on PI(3)P. In the presence of PI(3), this mutant PTEN protein reveals phosphatase activity, demonstrating full activation by PI(3)P.

Fig. 10. Identification of the PI(3)P-binding region of PTEN. (A) The domain structure of PTEN. PTEN consists of an N-terminal region, phosphatase (catalytic) domain, C2 domain, a C- terminal tail and a PDZ binding motif. The amino acid region 260-269 encodes a peripheral PIP- binding loop on the protein's surface. Mutation of this loop ("loop mutant") renders PTEN unable to bind phosphatidylinositol monophosphates as shown below (B): In vitro binding of PTEN to phosphatidyl monophosphates. Wildtype PTEN binds phosphatidylinositol

monophosphates (left panel). The C2 domain alone also binds monophosphates (middle panel). Mutation of the PIP-binding domain in an otherwise wildtype PTEN protein abolishes monophosphate binding (right panel). (C) Loop-mutant PTEN is active in vitro but but cannot be further stimulated in the presence of PI(3)P.

Fig.l 1. Incorporation of lipid-mediated activation of PTEN into an activity-screening method. (A) The identification of a novel PTEN mutant which only demonstrates tumor-suppressive lipid phosphatase activity in the presence of additional, non-substrate lipids, can be used to assess effects of test sample components (e.g. phosphoinositides or drugs etc.) on PTEN function. Treatment of test cells with exogenous agents (such as clinical PI3K inhibitors) or taking samples from cancer tissue and extraction of the cellular lipid content provides a test sample to assess its effects on PTEN activity. If exogenous cellular treatments enhance activating-lipid levels, the PTEN mutant will demonstrate measurable phosphatase activity. If exogenous treatment diminishes activating-lipid levels, minimal activity will be observed. Such activity can be assessed in the context of known mutant PTEN 'active' (i.e. + PI(3)P) and inactive ( - PI(3)P) settings. (B) PTEN activating compounds can be screened to identify agents that elevate

(advantageous as clinical intervention in proliferative pathologies) or diminish (advantageous as clinical intervention in e.g. post-stroke treatment) the catalytic activity of PTEN.

Fig.12. Mechanistic hypothesis of PTEN activation by non-substrate lipids.

Dimerisation/multimerisation of PTEN locks the enzyme into an inactive confirmation.

Wildtype protein is in equilibrium between monomer/multimer forms such that enzymatic activity, performed by monomers, is measurable. In the presence of non-substrate lipids (PIPs), WT PTEN undergoes conformational change, dramatically shifting the equilibrium towards the monomer state, and enhanced phosphatase activity. The mutant PTEN enzyme is in a state such that the equilibrium favours dimer/multimerisation, such that lipid phosphatase activity is negligible. Addition of activating lipids shifts this equilibrium towards the monomeric state, such that enzymatic activity is restored and is measurable.

Detailed Description

The invention provides methods of assessing the propensity of a subject for developing health disparity, such as and without limitation, a proliferative disease, diabetes, a diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or a related disease, disorder or condition by providing cell samples or fractions such as cell lipids from such cell samples from the subject to be tested and assaying these cells or cell fractions for the assays disclosed herein, including assays for determination of phosphatidylinositol phosphate content of cell samples or cell membrane fractions, such as total membrane lipid samples.

The subject to be tested according to the methods of the present invention can be a normal human subject undergoing a routine health assessment, or a human subject suspected of having a disorder, disease or condition due to family history of a proliferative disease, diabetes, a diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or a related disease, disorder or condition, or other genetic factors, or other risk factors. Alternatively, the subject can be an animal, such as a laboratory animal and the test is a test for the effect of a treatment with a therapeutic agent, a prospective therapeutic agent, or a nutritional supplement or other regimen that may have an effect on PTEN pattern and/or levels of one or more phosphatidylinositol phosphates.

The proliferative disease, diabetes or a diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition, the propensity for which is assessed by the methods of the invention can be any proliferative disease, diabetic condition, neurological disorder or related disease, disorder or condition regulated by or responsive to the activity of PTEN. Proliferative diseases regulated by PTEN include, for instance and without limitation, thyroid cancers, melanomas, brain cancers (including glioblastomas and astrocytomas), neuroblastomas, liver cancer, prostate cancer, breast cancer, and uterine cancer, including endometrial cancer. Diabetic conditions in which PTEN regulation has an impact include type I and type II diabetes and metabolic syndrome. Neurological disorders in which PTEN regulation has been implicated include a wide range of neurological disorders, such as Parkinsons disease and Alzheimers disease, mental retardation, ataxia and seizures, among others. Other PTEN regulation-related conditions include autoimmune diseases, including Cowden syndrome, Proteus syndrome, Bannayan-Riley-Ruvacaba syndrome and other non-cancerous growths and hamartomas; and also in autism and diabetes (see Tamguney & Stokoe 2007 New insights into PTEN 120(23):4072-4079). Other developmental disorders in which PTEN is implicated include developmental delay, autism spectrum disorders and macroencephaly. In proliferative diseases, developmental disorders, autoimmune diseases, neurological disorders or related diseases, disorders or conditions the propensity for which can be assessed by the methods of the invention, a cellular mislocalization and/or reduction in level or activity of PTEN is an indication of an increased propensity for development of the disease, disorder or condition.

By contrast, the propensity for obesity, diabetes, and diabetic conditions are are at lower risk in subjects with cellular mislocalization and/or reduction in level or activity of PTEN.

The assays of the invention include assays for the determination of PI(3)P levels in a lipid sample for example by determining the ratio between the PI(3)P and the non-phosphorylated PI.

In one embodiment, the invention provides a method of screening a test subject for the propensity for developing a proliferative disease or condition, wherein the method includes: providing a lipid sample from one or more cells of the test subject, wherein the cells are suspected of having a propensity for a proliferative disease; determining the relative amount of one or more phosphatidylinositol monophosphate species in the lipid sample; comparing the relative amount of phosphatidylinositol monophosphate species in the lipid sample from the test subject with the relative amount of the phosphatidylinositol monophosphate species in a lipid sample from one or more normal subjects not known to have a propensity for developing a proliferative disease or condition, and thereby assessing the propensity of the test subject for developing a proliferative disease or condition. In one embodiment, the phosphatidylinositol monophosphate species includes PI(3)P. In another embodiment, the phosphatidylinositol monophosphate species consists essentially of PI(3)P, PI(4)P and PI(5)P.

In another embodiment, the invention provides a method of screening a test subject for the propensity for developing a proliferative disease or condition, wherein the method includes: providing a lipid sample from one or more cells of the test subject, wherein the cells are suspected of having a propensity for a proliferative disease; determining the relative amount of one or more phosphatidylinositol monophosphate species in the lipid sample; comparing the relative amount of phosphatidylinositol monophosphate species in the lipid sample from the test subject with the relative amount of phosphatidylinositol monophosphate species in a lipid sample from one or more normal subjects not known to have a propensity for developing a proliferative disease or condition, wherein a lower relative amount of PI(3)P, PI(4)P and PI(5)P in the lipid from the test subject than the relative amount of PI(3)P, PI(4)P and PI(5)P in the lipid from the one or more normal subjects is indicative of a propensity of the test subject for developing a proliferative disease or condition.

In one embodiment, the phosphatidylinositol monophosphate species are resolved by mass spectroscopy. Alternatively, the phosphatidylinositol monophosphate species are resolved by thin layer chromatography (TLC), ion-exchange chromatography, high performance liquid chromatography (HPLC), gas chromatography, gel chromatography, or chromatography in a microfluidic system, or by an antibody-based separation system.

In another embodiment, the invention provides a method of screening test cells from a subject for the propensity for developing a proliferative disease or condition, wherein the method includes: providing a lipid sample from the one or more test cells of the subject, wherein the cells are suspected of having a propensity for a proliferative disease; determining the relative amount of one or more phosphatidylinositol monophosphate species in the lipid sample; comparing the relative amount of the phosphatidylinositol monophosphate species in the lipid sample from the test cells if the subject with the relative amount of the phosphatidylinositol monophosphate species in a lipid sample from one or more normal cells not suspected to have a propensity for developing a proliferative disease or condition from the same subject, and thereby assessing the propensity of the test cells for developing a proliferative disease or condition. Also provided are assays of PI(3)P with one or more optional additional phosphatidylinositol phosphate (PIP) species including one or more of PI, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃; by determining the ratio between PI(3)P plus the optional additional PIP species and the non-phosphorylated PL

In another embodiment, the assays of the invention include assays for the determination of the levels of PI(3)P plus PI(4)P and/ or PI(5)P, by determining the ratio between these monophosphate(s) and the total phosphatidyl inositols (including both phosphorylated PI and non-phosphorylated PI).

In another embodiment, the assays of the invention include assays for the determination of the levels of PI(3)P and one or more additional PI and/or PIP species, wherein the ratio of PI(3)P and one or more additional PI and/or PIP species including one or more of PI, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃, is determined as the ratio of the PI(3)P plus the optional additional PI and/or PIP species over the non-phosphorylated PI, or alternatively, over total phosphatidyl inositols (including both phosphorylated PI and non- phosphorylated PI), or over any suitable subset combination of PI and/or PIPs.

The assays of the invention for the determination of PI(3)P levels, and also the assays for the determination of levels of PI(3)P plus one or more PIP species, including one or more of PI, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃; can be determined as a ratio over the level of any suitable PI and/or PIP species, such as for instance, and without limitation, non- phosphorylated PI; phosphorylated PI plus non-phosphorylated PI; or non-phosphorylated PI plus any subset of PIP species; or even over a subset of PIP species alone, such as for instance monophosphorylated PIPs: i.e. the sum of (PI(3)P + PI(4)P + PI(5)P); or over diphosphorylated PIPs: i.e. the sum of (PI(3,4)P + PI(3,5)P + PI(4,5)P); or over the triphosphorylated PIP: i.e. PI(3,4,5)P; or over any suitable subset combination of PI and/or PIPs. .

The assays of the invention include assays for the determination of PTEN distribution and localization, including assessment of whether PTEN is distributed in distinct punctae characteristic of an active vesicle membrane association or active nuclear PI(3)P association or in a diffuse distribution characteristic of an inactive cytoplasmic or inactive nuclear location. Also provided are combination assays, including the above disclosed assays in combination with assays for total cellular PTEN content, which can be used as a measure of PTEN activity or surrogate for PTEN activity.

Any cells from a subject to be tested can be used in the methods of the invention, suitable cells include cells obtained by biopsy, including tumor cells, cells from tissues suspected of harboring cancerous or precancerous, and cells from apparently normal tissues, biopsy or other cellular or tissue sample. Cell samples useful for testing in the methods of the invention include white blood cells, such as peripheral blood mononuclear cells, including lymphocytes, monocytes and macrophages from blood samples.

In one embodiment the invention provides method of screening a test subject for the propensity for developing a proliferative disease, diabetes or diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition, wherein the method includes: providing a lipid sample from one or more cells of the test subject; determining the relative amount of PI(3)P or total phosphatidylinositol monophosphate; comparing the relative amount of PI(3)P or total phosphatidylinositol monophosphate in the lipid sample from the test subject with the relative amount of PI(3)P or total phosphatidylinositol monophosphate in a lipid sample from one or more normal subjects not known to have a propensity for developing a proliferative disease or condition, and thereby assessing the propensity of the test subject for developing a proliferative disease or condition, wherein a lower relative amount of PI(3)P in the lipid from the test subject than the relative amount of PI(3)P in the lipid sample from the one or more normal subjects is indicative of a propensity of the test subject for developing a proliferative disease, developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition.

Alternatively, a determination of no significant difference in the relative amount of PI(3)P in the lipid sample from the test subject than the relative amount of PI(3)P in the lipid sample from the one or more normal subjects is indicative of a normal risk of developing a health disparity such as a proliferative disease, obesity, diabetes, a diabetic condition, a developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition. It should be noted that the proliferative diseases, developmental disorders, autoimmune diseases, neurological disorders or related diseases, disorders or conditions the propensity for which can be assessed by the methods of the invention, wherein a lower relative amount of PI(3)P in the lipid from the test subject than the relative amount of PI(3)P in the lipid sample from the one or more normal subjects is an indication of an increased propensity for development of the disease, disorder or condition; whereas by contrast, a lower relative amount of PI(3)P in the lipid from the test subject than the relative amount of PI(3)P in the lipid sample from the one or more normal subjects is an indication of a decreased risk for developing obesity, diabetes, or a diabetic conditions, or amelioration of the diabetes, obesity or diabetic condition if already present in the test subject.

In another embodiment the invention provides a method of screening a test subject for the propensity for developing a proliferative disease, diabetes or diabetic condition, developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition. The method includes the steps of: providing a lipid sample from one or more cells of the test subject; determining the relative amount of PI(3)P or total phosphatidylinositol monophosphate and/or optionally one or more additional phosphatidylinositol phosphate species in the lipid sample, wherein the one or more additional phosphatidylinositol phosphate species are selected from the group consisting of PI, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃; comparing the relative amount of PI(3)P, or total phosphatidylinositol monophosphate and/or optional one or more additional phosphatidylinositol phosphate species in the lipid sample from the test subject with the relative amount of the PI(3)P and the optional one or more additional phosphatidylinositol phosphate species in a lipid sample from one or more normal subjects not known to have a propensity for developing a proliferative disease, diabetes or diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition, and thereby assessing the propensity of the test subject for developing a proliferative disease, diabetes or diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition. In one embodiment the PI(3)P or total phosphatidylinositol monophosphate and the optional one or more additional phosphatidylinositol phosphate species can be resolved by mass spectroscopy or immunofluorescence. Alternatively, or in addition, the PI(3)P total phosphatidylinositol monophosphate and the one or more additional phosphatidylinositol phosphate species can be resolved by thin layer chromatography (TLC), ion-exchange chromatography, high performance liquid chromatography (HPLC), gel chromatography, gas chromatography, or chromatography in a microfluidic system or an antibody-based separation system, such as an immunoaffinity column, immunobeads or magnetic immunobeads.

In another embodiment the invention provides a method of determining the cellular localization of PTEN in a sample of cells: The method includes the steps of: providing a sample of one or more cells of the subject; contacting the cells with a PTEN-binding agent under suitable conditions for binding of the PTEN-binding agent to PTEN in the cells to produce PTEN-binding agent:PTEN complexes; detecting the PTEN-binding agent:PTEN complexes to display the distribution of PTEN in the one or more cells; and determining the distribution of the PTEN in the one or more cells as being punctuate or diffuse, wherein a punctuate distribution is indicative of an active vesicle membrane association or active nuclear PI(3)P association of PTEN, and a diffuse distribution is indicative of an inactive cytoplasmic or an inactive nuclear localization of PTEN; wherein a determination of an inactive cytoplasmic or an inactive nuclear localization of PTEN is indicative of a high propensity for developing a proliferative disease, developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition and a low propensity for developing obesity, diabetes or a diabetic condition. Alternatively, a determination of an active vesicle membrane association or active nuclear PI(3)P association of PTEN is indicative of a low propensity for developing such a health disparity, but is indicative of a higher propensity for developing obesity, diabetes, or a diabetic condition.

The PTEN-binding agent can be any PTEN-binding agent, such as for instance a PTEN- specific antibody, a single chain antibody such as an Fv single chain antibody, or an antibody fragment such as an Fab fragment, an F(ab)₂ fragment, an Fv variable region fragment or a disulfide stabilized Fv antibody fragment, a phage display antibody or any other recombinant PTEN-binding agent. In one embodiment, the PTEN-antibody:PTEN complexes can be detected with a second antibody, optionally the second antibody can be labeled with a detectable label. The detectable label can be any detectable label, such as for instance, an enzyme, a fluorescent label, a phosphorescent label, a chemiluminescent label, a chromophore, a green fluorescent protein label, radioisotope label, a cytotoxic chemical label, a ligand such as biotin or any other suitable detectable label. In one embodiment, PTEN is detected by immunofluorescence. In another embodiment the invention provides a method of monitoring progress of a proliferative disease, diabetes or diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition in a subject undergoing a therapy for the disease, disorder or condition. The method includes the steps of: providing two or more lipid samples from cells of the subject taken at different times during the therapy; determining the relative amount of PI(3)P or total phosphoinositol monphosphate and optionally one or more optional additional phosphatidylinositol phosphate species selected from the group consisting of PI, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃ in the lipid samples from the subject; comparing the relative amount of PI(3)P and the one or more optional additional phosphatidylinositol phosphate species of the lipid sample from the subject at the earliest time before or during therapy with the relative amount of PI(3)P or total phosphoinositol monphosphate and the one or more optional additional phosphatidylinositol phosphate species in each of the lipid samples taken from the subject at one or more subsequent times during or after the therapy; and thereby monitoring the progress of the therapy. The total phosphatidylinositol monophosphate can be assessed, and the PI(3)P and one or more optional additional phosphatidylinositol phosphate species can be resolved by mass spectroscopy. Alternatively, the total phosphatidylinositol monphosphate can be assessed, and the PI(3)P and the one or more optional phosphatidylinositol phosphate species can be resolved by thin layer chromatography (TLC), ion-exchange chromatography, high performance liquid chromatography (HPLC) or gas chromatography or any other suitable separation method.

In another embodiment the invention provides a method of qualifying a subject for treatment with indirect AKT pathway inhibitors such as PI 3-kinase inhibitors, or the rotenoids, Deguelin and Itraconazole or the like, in a subject suffering from a proliferative disease. The method includes the steps of: providing a lipid sample from one or more cells of the subject; determining the relative amount of PI(3)P or total phosphatidylinositol monphosphate and optionally one or more optional additional phosphatidylinositol phosphate species selected from the group consisting of PI, PI(4)P, PI(5)P, PI(3,4)P2, PI(3,5)P2, PI(4,5)P2 and PI(3,4,5)P3 in the lipid sample from the subject; comparing the relative amount of PI(3)P and the one or more optional additional phosphatidylinositol phosphate species of the lipid sample from the subject with the relative amount of PI(3)P or total phosphatidylinositol monphosphate and the one or more optional additional phosphatidylinositol phosphate species in each of the lipid samples taken from a normal subject or a pool of normal subjects; and thereby assessing qualifying or disqualifying the subject for treatment with AKT pathway inhibitors such as PI 3-kinase inhibitors, or the rotenoids, Deguelin and Itraconazole or the like, wherein a lower than normal level of PI(3)P and the one or more optional additional phosphatidylinositol phosphate species is a positive qualifying factor for the therapy. A normal PTEN level (e.g. from a PTEN immunoflourescence assay, indicating no PTEN deletion) and normal cellular localization of PTEN is another positive qualifying factor for the therapy with AKT pathway inhibitor.

In another embodiment of the above-recited method of monitoring progress of a proliferative disease, diabetes or diabetic condition, obesity, a developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition in a subject undergoing a therapy according to the present invention, an increase in the relative amount of PI(3)P in one or more of lipid samples taken from the subject at times during or after the therapy is indicative of a beneficial therapeutic effect of the therapy on the proliferative disease, developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition in the subject. Whereas, an observation of diminishing levels of the relative amount of PI(3)P in the lipid sample from therapeutic agent-treated cells and hence an expectation of a reduction in activity of PTEN can be a useful factor in making a clinical decision to cease the therapy responsible for the undesirable change in PTEN. The reverse would be true for treatment of obesity, diabetes or a diabetic condition, i.e. in monitoring these health disparities in a subject undergoing a therapy according to the present invention, an decrease in the relative amount of PI(3)P in one or more of lipid samples taken from the subject at times during or after the therapy is indicative of a beneficial therapeutic effect of the therapy on the obesity, diabetes or a diabetic condition; and similarly, an observation of increased levels of the relative amount of PI(3)P in the lipid sample from therapeutic agent-treated cells and hence an expectation of an increase in activity of PTEN can be a useful factor in making a clinical decision to cease the therapy responsible for the undesirable change in PTEN.

The methods of the present invention can be used to specifically guide clinicians and care-givers to use "PI(3)P-friendly" therapeutics (i.e. therapeutics which do not block the VPS34 kinase which is responsible for the synthesis of PI(3)P ) as an AKT pathway-targeted therapy in cancer patients. The inventors have found that in cell culture experiments, AKT pathway inhibitors such as the rotenoid Deguelin, or the triazole compound Itraconazole, fulfill the requirement of blocking the AKT pathway while showing no adverse effects on PTEN localization as measured by punctate vesicle staining. Furthermore, that such treatment leads to little or no reduction in PI(3)P levels as can be inferred from mass spectrometry data.

In another embodiment the invention provides a method of assessing the effect of a prospective therapeutic agent on PTEN tumor suppressor activity. The method includes the steps of: providing a first lipid sample from one or more cells of a normal subject and a second lipid sample from one or more cells of the normal subject after exposure to the prospective therapeutic agent; comparing the relative amounts of PI(3)P and optional one or more additional phosphatidylinositol phosphate species selected from the group consisting of PI, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃ in the lipid sample from the one or more cells treated with the prospective therapeutic agent with the relative amount of PI(3)P and the one or more optional additional phosphatidylinositol phosphate species in the lipid sample from the one or more untreated cells; and thereby assessing the effect of the prospective therapeutics agent on PTEN tumor suppressor activity.

Alternatively, the cells used in this embodiment can be from cell lines and tissue culture cells. The method can be used with laboratory animals as the test subject to optimize pharmacokinetic and pharmacodynamic parameters during development of novel therapeutic agents, and also in later clinical trials with human subjects as part of the safety and efficacy determinations.

In another alternative, the method of assessing the effect of a prospective therapeutic agent to support PTEN tumor suppressor activity can be used with cell lines and tissue culture cells to screen for PTEN promoting agents such as inhibitors of the PI(3)P phosphatase MTMR2 that act by raising the PI(3)P species and/ or PTEN vesicle localization as therapeutic agent candidates against proliferative disease, developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition.

In another alternative, the method of assessing the effect of a prospective therapeutic agent on PTEN tumor suppressor activity can be used with cell lines and tissue culture cells to screen for indirect PTEN inhibitors that act by lowering the PI(3)P species and/ or PTEN vesicle localization as therapeutic agent candidates for treatment of obesity, diabetes and diabetic conditions.

In another embodiment of the above-recited method, a finding that the relative amount of PI(3)P in the lipid sample from the prospective therapeutic agent-treated one or more normal cells is less than the relative amount of PI(3)P in the lipid sample from the one or more untreated normal cells is indicative of the prospective therapeutic agent having a propensity for increasing the likelihood of developing a proliferative disease, developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition in a subject treated with the therapeutic agent. A finding that exposure of cells to the prospective therapeutic agent brings an increased risk of developing a proliferative disease or condition, a developmental disorder or autoimmune disease, a neurological disorder or other related disease, disorder or condition is an important factor in determining the risk/benefit of therapy with the prospective therapeutic agent. However, as noted above, a reduction in the relative amount of PI(3)P in the lipid sample from the prospective therapeutic agent-treated one or more normal cells can be considered beneficial in terms of reducing the likelihood of developing obesity, diabetes or a diabetic condition. The relative amount of PI(3)P in the lipid sample can be for instance the amount of PI(3)P relative to the total cell lipid, or relative to the total phosphatidylinositol species (i.e. the sum of PI, PI(3)P, PI(4)P, PI(5)P, PI(3,4)P₂, PI(3,5)P₂, PI(4,5)P₂ and PI(3,4,5)P₃), or relative to the most abundant species, unphosphorylated phosphatidylinositol (PI). The phosphatidylinositol species may be resolved by chromatographic techniques, such as for instance high performance liquid chromatography (HPLC), or thin layer chromatography (TLC), or analyzed by mass spectroscopy and resolution of fragments of the different phosphatidylinositol species.

Alternatively, in another embodiment of the above-recited method, a finding that the relative amount of PI(3)P in the lipid sample from the prospective therapeutic agent-treated one or more normal cells is more than the relative amount of PI(3)P in the lipid sample from the one or more untreated normal cells is indicative of the prospective therapeutic agent having a propensity for reducing the likelihood of developing a proliferative disease, developmental disorder, autoimmune disease, neurological disorder or related disease, disorder or condition in a subject treated with the therapeutic agent. MATERIALS AND METHODS

PTEN activity assay method.

For activity assays, 1 ug of protein was used per well, all time points were performed in duplicate. Reaction mixtures comprised 20 uM of each individual lipid (Echelon) and PTEN protein, made up to 50 ul total volume in 100 mM Tris-HCl, pH7.5. Reaction mixtures were incubated at 37 degrees C for the duration of the assay. Reactions were quenched by the addition of 100 ul, room temperature, malachite green solution (Echelon). Thirty minutes post- quenching, absorbance at 620 nm was recorded and compared to known free-phosphate standards.

Protein production.

Recombinant proteins were produced as described previously (Berger I., Fitzgerald D.J., Richmond T.J., 2004; Baculovirus expression system for heterologous multiprotein complexes. Nat biotechnol. 322(12): 1583-7.)

Immunofluorescence microscopy

Cells were plated onto coverslips of 0.13-0.17 mm thickness (Electron Microscopy Sciences, PA, USA) and incubated in DMEM (Mediatech inc., VA, USA), 10% FBS, penicillin 50 units/ml, (Sigma, MO, USA) streptomycin 100 μg/ml (Sigma) at 37°C, 5% C0₂, 100% humidity overnight. Growth medium was then removed and coverslips were washed once with phosphate buffered saline (PBS). Cells were then fixed and permeabilized by one of three methods: methanol fixation and permeabilization by addition of ice-cold methanol for 5 minutes. Paraformaldehyde (PFA) fixation by addition of PFA for 5 minutes, then a subsequent PBS wash, followed by addition of 25 mM NH₄C1 for five minutes as a fixative quencher. Another PBS wash preceded permeabilization with 0.5% Triton X-100 in PBS for 5 minutes. Phemo fixation and permeablization by addition of a solution of 3.7% formaldehyde, 0.05% glutaraldehyde, 0.5% Triton X-100 in phemo buffer (0.068 M PIPES, 0.025 M HEPES, 0.015 M EGTA^»Na₂, 0.003 M MgCl₂'H₂0, 10% DMSO, pH 6.8) for 5 minutes. Following fixation and permeabilization, coverslips were washed three times with PBS and non-specific antibody binding blocked by addition of 10% goat serum in PBS for 30 minutes. Coverslips were then incubated at 4°C for 60 minutes in primary antibody diluted in 10% goat serum (see Table 2 for antibody concentration working dilutions). Coverslips were then washed three times in PBS and incubated with secondary, fluorescent-conjugated antibodies (Table 2) for 15 minutes at room temperature. Primary and secondary incubations were performed in a light proof, humidified chamber.

Subsequently, coverslips were washed 3 times in PBS and mounted using Prolong Gold anti-fade reagent containing DAPI (Invitrogen, OR, USA) and allowed to cure overnight. Coverslips were finally sealed using clear nail polish. Slides were imaged using a Perkin Elmer Ultra VIEW VoX spinning disk confocal microscope (Perkin Elmer, MA, USA) using Volocity v.6.0.1 software (Perkin Elmer). For visualization of transferrin-containing vesicles, cells were incubated for 60 minutes prior to fixation with fluorescent-conjugated transferrin (Invitrogen), 5 μg/ml in complete growth medium.

Structured Illumination for Super-Resolution Imaging (OMX microscopy)

Cell fixation and preparation was as described above. High resolution images were acquired using an OMX 3D Structured Illumination Microscope (Applied Precision, WA, USA). Solid state lasers were used for excitation at 405, 488 or 593 nm. Excitation light was coupled to a fiber optic cabling, scrambled and passed through a diffraction grating prior to sample illumination.

The diffraction grating was mounted on a piezoelectric and a rotational stage to allow for control of lateral shift and angular orientation (+/- 60 degree: required for 3D structured illumination imaging). Subsequent to beam diffraction, the innermost 3 beam orders (orders 0 and +/- 1) were used for image reconstruction with additional orders obstructed by a beam blocker and not utilized. The interference pattern of the beams in the focal plane of the objective (UPlanS Apochromat lOOx 1.4NA: Olympus, PA, USA) was used to generated 3D sinusoidal pattern. The fluorescent light emitted by the sample was gathered by the same objective, passed through different dichroic mirrors and filters and measured by Cascade II EMCCD back- illuminated cameras (Photometries, AZ, USA). Exposure times of each frame were typically between 100 and 200 ms, and the power of each laser was adjusted to achieve optimal intensities in the raw image of a 16-bit dynamic range. Multi-channel images were achieved through sequential acquisition. The original z-stacks were saved and processed using SoftWoRx 4.5.0 (Applied Precision) to reconstruct the high resolution information. The dataset was further processed to achieve a 3D reconstruction or maximum intensity projections using the same software. PTEN/microtubule association quantification

See Figure 2. Quantification of Pten position relative to microtubules was achieved by assessment of Pten staining distances from microtubule stain. For 3 individual OMX images of wildtype MEFs, the Pten (red) channel was visualized in the absence of the a-tubulin (microtubule: green) channel. Thirty-three Pten intensities were chosen at random per image. The distribution of randomly selected intensities was throughout the image. A scale marker, as determined by the digitally assigned marker added by the SoftWoRx software, was placed over the centre pixel of the Pten intensity. Subsequently, the green (a-tubulin) channel was overlaid onto the red channel/scale marker image. Each Pten intensity was then assigned a relative distance to the closest green pixel (microtubule stain). Intensities were assigned as either 0 μΜ from the closest microtubule (i.e. red and green channels demonstrated direct colocalization, with pixels both red and green in the corresponding channels), 0.5 μιη, 1 μιη or 2 μιη. Scoring was Outward', such that any intensity greater than 1 μΜ but less than 2 μΜ from the closest microtubule was assigned as '2 μιη'. As a control comparison, random pixels in the same images were selected and assessed in a similar manner. In order to select random pixels, a random number generator (www.randomizer.org) was used to provide co-ordinates of pixels within each image by providing X and Y values. Random pixels were only used if they were within the cell periphery in the image (co-ordinates corresponding to regions of the image outside of the cell were discarded). Thirty-three random pixels were assessed in each image. The relative distances were then compared using the non-parametric chi-squared (χ ) test, as not all observed data demonstrated a Gaussian distribution. Pten/random pixels were categorized as either overlapping (0 μιη from microtubules) or other (> 0 μιη) for statistical testing.

PTEN overexpression

Human PTEN cDNA was overexpressed in PC3 cells using Lipofectamine reagent (Invitrogen) according to the manufacturer's instructions. PTEN cDNA was cloned using Xbal/Xhol restriction sites into the pcDNA4/TO/myc-His plasmid vector (Invitrogen).

Nocodazole/PDK inhibitor treatment/amino acid starvation

Cells were treated with nocodazole (Sigma) at a final concentration of 10 μg/ml (33 μΜ) for 30 minutes in DMEM, 10% FBS, penicillin 50 units/ml, streptomycin 100 μg/ml at 37°C, 5% C0₂, 100% humidity prior to fixation for microscopy. Stock solution was prepared in DMSO (Mallinckrodt, NJ, USA). Cells were treated with Wortmannin, 3-methylalanine (both Sigma) or NVPBez-235 (Axon Medchem, Groningen, Netherlands) at the final concentrations described in DMEM, 10% FBS, penicillin 50 units/ml, streptomycin 100 μg/ml at 37°C, 5% C0₂, 100% humidity for the duration described. Stock solutions of each drug were made in DMSO which was heated at 75°C for 5 minutes prior to use to ensure complete dissolving of compounds. Comparison controls, treated with DMSO, were treated in an identical manner with DMSO at a final concentration of 0.1%. Amino acid/glutamine starvation was achieved using amino acid, glucose and glutamine-free RPMI 1640 medium (BiOLOG, CA, USA) supplemented with D- glucose (Mallinckrodt) 5% FBS, penicillin 50 units/ml, streptomycin 100 μg/ml at 37°C, 5% C0₂, 100% humidity for the duration described. Comparison controls were formulated in the same way and treated under identical conditions except that RPMI 1640 medium containing amino acids (BiOLOG) was used and supplemented to make a final concentration of 0.3 mM L-glutamine (Sigma).

In vitro EGF stimulation

Cells were plated and incubated in medium containing amino acids or the control comparison. After 24 hours of treatment, growth medium was removed and the cells washed 3 times with PBS. The medium was then changed and the serum content reduced from 5 to 0.1% and the cells incubated for a further 6 hours at 37°C, 5% C0₂, 100%. Stimulation was achieved by removal of growth media and addition of identical medium supplemented with 3 ng/ml of recombinant human epidermal growth factor (Invotrogen) for the described duration.

Phosphatidylinositol (PIP) strip binding assay/Phosphatidylinositol binding quantification

GST-tagged human recombinant PTEN protein (Echelon, UT, USA) was resuspended in nuclease free water (Qiagen, MD, USA) to a concentration of 50 μg/ml. PIP strips (membrane blotted with immobilized lipids-Echelon) were blocked in 3% non-fat milk, PBS*0.1% Tween-20 (PBS-T) for 60 minutes at room temperature. Recombinant human PTEN protein was then added to the strip at the concentrations described in 3% non-fat milk, PBS-T for 60 minutes at room temperature. Subsequently, three five minute washes of the strip in PBS-T were followed by incubation of primary anti-PTEN antibody. Strips were then washed for 5 minutes three times and incubated with corresponding secondary antibody (see antibody list) and visualized using an enhanced chemiluminescence (ECL) detection reagent (GE Healthcare, Bucks, UK). Quantification of lipid binding was performed in an identical manner using quantification strips (Echelon). Relative binding was measured by assessment of blot intensities using ImageJ software (NIH, Maryland, USA). Control, untagged PTEN protein was expressed in insect cells using baculovirus and purified from Hi5 cell lysate by ion exchange chromatography on SP sepharose. Recombinant, GST-tagged lipid binding protein domains p40PX (a PI(3)P binding domain) and LL5a (a domain capable of binding multiple lipid species) were purchased (Echelon) and used as comparative binding controls.

Sucrose gradient fractionation

Sucrose fractionation was performed on cell homogenate obtained from 3, 10 cm petri dishes approximately 75% confluent. Cells were grown in DMEM, 10% FBS, penicillin 50 units/ml, streptomycin 100 μg/ml at 37°C, 5% C0₂, 100% humidity until the desired confluence and subsequently washed once in PBS. Cells from each plate were scraped and collected in 500 μΐ, 4 mM HEPES, ImM MgCl₂, 0.25 M sucrose, pH 7.5, containing cOmplete Mini protease inhibitor (Roche, IN, USA) and Phostop phosphate inhibitor (Roche). Homogenization was achieved using Qiashredder spin columns (Qiagen). The supernatant generated from all 3 plates by the homogenization columns was then layered onto the top of 4 ml of a 0.3 M sucrose - 1.5 M sucrose (in 4 mM HEPES, 1.2 mM EDTA, pH 7.5) gradient. Gradients were generated using an SG5 gradient maker (Hoefer, MA, USA). Sucrose/cell homogenate was centrifuged for 22 hours, 4°C in a SW55 swinging bucket rotor (Beckman Coulter, CA, USA) at 30,000 RPM. Following centrifugation, 200 μΐ sequential fractions were obtained and analyzed by immunoblot (see Table 2 for working concentrations).

Mass spectroscopy

Cells were counted and pelleted prior to lipid extraction. The same number of cells were used across experiments when comparing treatments. Lipid extraction has been described previously (Milne S. B. et al., 2005, A targeted mass spectrometric analysis of phosphatidylinositol phosphate species. / Lipid Res 46: 1796). Extracted lipids were analyzed by a Vantage triple-stage quadrupole mass spectrometer (Thermo Scientific, NC, USA) in negative ion mode with the Heated Electrospray Ionization (HESTII) ion source. The samples were dissolved in 200 μΐ of negative ion buffer (methanol:chloroform:water at 1: 1:0.3 v/v, made up to 30mM piperidine; Sigma) and 5 μΐ was injected at 35 μΐ/min. Spray voltage was 2.5 kV, vaporizer pressure 50°C, Sheath gas pressure 5 psi and capillary temperature 240°C. The full spectrum from 700 to 1200 m/z was scanned at 500 amu/s to determine the intact mass of each individual lipid. In HeLa cells we identified multiple phosphatidylinositol (PI) lipid species (m/z = 835.57, 861.55, 863.56, 889.54) and phosphatidylinositol phosphate (PIP) lipid species (m/z = 915.46, 941.56, 943.50, 969.74). In A549 cell lines we found only one pair of IP (m/z =887.30) and PIP lipids (m/z = 967.33). We selected these ions as the parent ions of the corresponding lipids and fragmented them in selected reaction monitoring (SRM) mode, using argon gas for collision-induced dissociation (CID) at 1.2 Torr and 40 V. The most abundant and unique fragments (see Table 1) were monitored (50 ms dwell time), and the data collected with Xcalibur software (Thermo Scientific). The sum of fragment ion intensities was then used to calculate PIP/PI ratios.

Generation of shVPS34 stable cell lines

Three independent anti-Vps34 hairpins, and an anti-luciferase control hairpin in the GIPZ lentiviral vector (Thermo Scientific) were transfected into HEK393 (phoenix) cells using calcium phosphate transfection methodology. Concurrent transfection of psPAX2 and pM2D.G packaging vectors (Addgene plasmids 12260 and 11259) allowed efficient virus production. Growth medium (complete DMEM) containing virus was removed from cells after 8 hours and added to NIH3T3 cells for infection. After 36 hours, growth medium was removed and replaced with virus free medium supplemented with puromycin (Sigma) at a concentration of 2 μg/ml. Cells were kept in selection medium for 5 days prior to subsequent analysis.

Generation of shVPS34 stable cell lines

Three independent anti-Vps34 hairpins, and an anti-luciferase control hairpin in the GIPZ lentiviral vector (Thermo Scientific) were transfected into HEK393 (phoenix) cells using calcium phosphate transfection methodology. Concurrent transfection of psPAX2 and pM2D.G packaging vectors (Addgene plasmids 12260 and 11259) allowed efficient virus production. Growth medium (complete DMEM) containing virus was removed from cells after 8 hours and added to NIH3T3 cells for infection. After 36 hours, growth medium was removed and replaced with virus free medium supplemented with puromycin (Sigma) at a concentration of 2 μg/ml. Cells were kept in selection medium for 5 days prior to subsequent analysis. siRNA transfection

Four independent siRNAs targetting Vps34 were obtained (Qiagen) and a scramble control comparison and transfected into NIH3T3 cells using Dharmafectl reagent (Thermo Scientific) according to the manufacturer's instructions. After 72 hours cells were analyzed for VPS 34 knockdown effects. qPCR assessment of Vps34 transcript levels

RNA was obtained from cells using RNEasy extraction kits (Qiagen) according to the manufacturer's instructions. RNA was then reverse transcribed using Superscript Reverse Transcriptase kits (Invitrogen) according to the manufacturer's instructions. qPCR assessment of Vps34 was performed using primers Vi (5' -CCA GGC ACG ACG TA A CTT CT-3') and V₂ (5'-TGT CAG ATG AGG AGG CTG TG). HPRT transcript levels were used to determine relative sample RNA levels and assessed with primers Hi (AGT CCC AGC GTC GTG ATT AG) and H₂ (TTT CCA AAT CCT CGG CAT AAT GA).

Assessment if PI3K inhibitor efficacy

PC3 cells were plated in DMEM, 10% FBS, penicillin 50 units/ml, streptomycin 100 μg/ml at 37°C, 5% C0₂, 100% humidity and allowed to adhere overnight. Cells were then serum starved (DMEM, 0.1% FBS, penicillin 50 units/ml, streptomycin 100 μg/ml at 37°C, 5% C0₂, 100%) overnight and subsequently treated with PI3K inhibitors (all at 10 μΜ) or DMSO (at 0.1% final concentration). Thirty minutes after treatment, cells were serum stimulated with inhibitor-containing medium (DMEM, 10% FBS, penicillin 50 units/ml, streptomycin 100 μg/ml at 37°C, 5% C0₂, 100% containing PI3K inhibitor at 10 μΜ or DMSO). Five minutes subsequent to stimulation, cells were washed 3 times with PBS and lysed for subsequent immunoblot analysis.

EXAMPLES

EXAMPLE 1 : Localization of PTEN within Mouse Embryonic Fibroblast Cells

The prominent cytoplasmic and nuclear localization of PTEN was confirmed using light microscopy in primary Mouse Embryonic Fibroblasts (MEFs) and several tissue culture cell lines (Fig. 1A and Fig. 2A). Upon close inspection however, cytoplasmic PTEN was surprisingly found to be distributed in discrete punctae that suggested organization into distinct cytoplasmic entities. Specificity of staining was confirmed with P EN-deficient cell lines (Fig. 1A) and by testing various staining procedures, primary and secondary antibody specificity, fixation protocols and different cell types (see Fig. 2A and Materials and Methods).

Next, the hypothesis that the observed PTEN pattern could come from cytoplasmic vesicles was tested using fluorescent Transferrin, a marker for endocytic vesicles, as a reference. As shown in Fig. IB, highly similar staining patterns for Pten and Transferrin were obtained in primary MEFs. Since microtubules (MTs) guide most internal membrane transport MT filaments were also stained. PTEN staining was generally found on or near microtubules in a pattern that was very similar to the co-staining of Transferrin and microtubules (Fig. IB, Pten - MT and Tf - MT, respectively), suggesting a possible association.

To better define the relation between Pten and MTs super resolution microscopy, which surpasses the optical diffraction limit of light microscopes was employed. This approach revealed strong colocalization of Pten with MTs (Fig. 1C, and Fig. 2B). Furthermore, the Pten signal pattern revealed that the majority of cytoplasmic PTEN is arranged along tracks (Fig. ID top panel) that correspond to the local organization of the MT network (Fig. ID bottom panel). Quantification of these results showed that 63% of the Pten punctae were directly on MTs and 70% were within 0.5 μιη of an MT, an association that was significantly different from a the distance distribution of random locations (p=3.5E-36, Fig. ID graph, and Fig. 2C). Nocodazole treatment efficiently depolymerized MTs resulting in a fine grain pattern, yet this did not affect the Pten staining pattern or distribution (Fig. IE). This finding suggested that Pten did not directly associate with MTs. Finally, whether exogenously expressed PTEN also reveals this distinct organization was tested. In contrast to endogenous PTEN, tag-free exogenous PTEN was diffusely distributed in the cytoplasm of PTEN null cells (Fig. IF, 'high'). However, rare cells expressing low levels of exogenous PTEN revealed a pattern that was similar to endogenous PTEN (Fig. IF, 'low'), but absent from untransfected cells. In summary, the majority of endogenous PTEN was found to be organized along microtubules in a vesicle-type staining pattern.

EXAMPLE 2: Association of PTEN with Specific Vesicle Classes

Next, a test was run to determine if PTEN could be associated with any specific vesicle class. Organelle and vesicle identity of eukaryotic cells is generally encoded by the type of phosphoinositides (PIPs) displayed on their cytosolic surface (Di Paolo and De Camilli, 2006: Phosphoinositides in cell regulation and membrane dynamics. Nature 443, 651-657). To assess if PTEN binds to a specific PIP, binding assays to PIP- strip membranes onto which the various PIP isoforms are printed were performed (Fig. 3A). Recombinant GST-PTEN bound to PI(3)P, a monophosphorylated phosphoinositide. Similarly, recombinant untagged PTEN bound predominantly to PI(3)P. Next, the in vitro affinity for these lipids under non- saturating conditions was tested. As shown and quantified (Fig. 3B), PTEN bound most avidly to PI 3-P while several fold weaker binding to PI (5)P (at the non- saturating 25 pmol lipid amount) was also seen. Notably, PTEN did not bind to its substrate PI(3,4,5)P₃ under all conditions tested (Fig. 4A-C). These data indicate that PTEN preferentially binds to PI(3)P, the major membrane determinant for identity of early endosomes.

Localization of Pten in primary MEFs was compared with the localization the Early Endosomal Antigen 1 (EEA1), the prototypical PI(3)P binding protein, which binds PI(3)P by its FYVE domain (reviewed in Di Paolo and De Camilli, 2006 cited above). Eeal staining is found in two distinct patterns: small uniform punctae representing endocytic vesicles in the cell periphery (Fig. 3C, left panels, asterisks) and strongly staining larger vesicles of variable size (right panels). The latter likely represent fused early endosomes or late endosomes, which themselves often contain multiple intralumenal vesicles. Staining showed that in the cell periphery, a punctate Eeal pattern was most prominent and indistinguishable from the Pten and Transferrin staining pattern (Fig. 3C, left panels). Notably, these signals did not overlap in the cellular periphery as it is known that different PI(3)P binding domains (the FYVE and PX domains) are recruited to endocytic vesicles in a mutually exclusive fashion, which is likely due to their differential interaction with the second determinant of vesicle identity, the Rab proteins (Zerial and McBride, 2001: Rab proteins as membrane organizers. Nature reviews Molecular cell biology 2: 107-117). In the juxtaniclear region however, more complex Eeal staining was often found to include Pten and Transferin signals, as shown in (Fig. 3C right panels, arrows), suggesting their common inclusion into multivesicular bodies (MVBs). Super-resolution microscopy confirmed these results (Fig. 4D), revealing that the large perinuclear Eeal vesicles shown by conventional confocal light microscopy in Fig. 3C consisted of several smaller signals that often surrounded Pten signal, consistent with its localization into MVBs. To further probe the similarities between Eeal and Pten distribution in the cell, cytoplasmic cell lysates were fractionated by ultracentrifugation on sucrose gradients. As shown in Fig. 3D, the sedimentation properties of Pten and Eeal were essentially identical, both peaking in fractions 9 and 10. In contrast the transmembrane Golgi protein Ndfipl, for which we had previously shown an interaction with Pten only showed partial co- sedimentation while the ribosomal protein S6 migrated farthest in the gradient. Consistent with these results, the mono-specific PTEN antibodies detected only minor Pten colocalization with Golgi markers by immunofluorescence (IF) colocalization with Golgi markers (Fig. 4E-F). Since PI(5)P is a Golgi signature lipid, these results supported the in vitro binding preference of PTEN to PI(3)P. Collectively, these results suggest that cytoplasmic PTEN resides on endosomes by virtue of its binding to the endosomal signature lipid, PI(3)P.

If the localization of a substantial fraction of PTEN in cells were to be controlled by this phosphoinositide, then the major PI(3)P producing kinase, the class III PI 3-Kinase VPS34 (Backer, 2008: The regulation and function of Class III PDKs: novel roles for Vps34. The Biochemical journal 410, 1-17), should be a determinant of PTEN localization. To test this hypothesis, first depletion of amino acids, which are essential for VPS34 activity (Byfield et al., 2005 hVps34 is a nutrient-regulated lipid kinase required for activation of p70 S6 kinase. J Biol Chem 280, 33076-33082; Nobukuni et al., 2005: Amino acids mediate mTOR/raptor signaling through activation of class 3 phosphatidylinositol 30H-kinase. Proc Natl Acad Sci US 102, 14238-14243) was used. Mock treatment reproduced the punctate PTEN and EEA1 pattern in human cells (Fig. 5A, A549). In contrast, we found loss and strong reduction of both PTEN and EEA1 staining after 24 hours of amino acid depletion. This situation was exacerbated after 48 hours of amino acid withdrawal (not shown). In primary MEFs, two types of reactions to amino acid depletion were observed. First, a sharp drop in the overall Pten/ Eeal signal levels was observed (Fig. 5A, 'MEF Γ). Importantly, when enhanced, the entire cytoplasm revealed diffuse Pten and Eeal signal, with the exception of the perinuclear region, where some larger punctae remained. Since these punctae were positive for both Pten and Eeal, these punctae may represent MVBs that harbor the two proteins with a limiting membrane preventing their diffusion. To check if loss of endosomes had occurred upon amino acid depletion, Transferrin distribution was investigated. As shown (Fig. 6A), the transferrin signal persisted in strength and remained discretely punctate as expected for an intravesicular, receptor-bound molecule stain. Thus, we concluded that loss of Pten/ Eeal signal was due to the reduction in PI(3)P anchoring sites and not due to loss of endosomes per se. After shorter treatment (36 hours to 40 hours of amino acid depletion) cells without major Pten signal loss revealed diffuse clouds of Pten stain alongside the punctate stain (Fig. 5A, 'MEF ΙΓ). These results suggested that the degree of VPS34 inhibition may dictate the extent of PTEN perturbation. In this vein, combined siRNA and shRNA knockdown of Vps34 apparently did not yield enough suppression of VPS34 and pS6 (fig. S3B) and showed only minor effects on Pten localization (not shown). Interestingly, the A549 lung cancer cell line which responded strongly to amino acid depletion has been reported to have low VPS34 (PIK3C3 gene) mRNA expression (Oncomine database (D. R. Rhodes et ah, 2007; Oncomine 3.0: genes, pathways, and networks in a collection of 18,000 cancer gene expression profiles. Neoplasia 9: 166).

EXAMPLE 3: Effect of PTEN localization and levels on PTEN function

Two human cancer cell lines, which better endure amino acid withdrawal than primary MEFs were used. As shown (Fig. 5B immunoblots and quantification graphs), it was found that amino acid depletion stimulated receptor-mediated AKT activation between 2-fold to 14-fold on both activation sites. In contrast, amino acid withdrawal effectively blunted Ribosomal Protein S6 (RPS6)-activation, confirming efficient VPS34-inhibition through amino acid withdrawal. These findings are consistent with amino acid depletion mediating PTEN-suppression through VPS34 inhibition. A slight reduction in PTEN caused by the depletion was confirmed, consistent with the above immunofluorescence (IF) results. However, similar reduction in other signaling molecules, such as total AKT and ribosomal protein S6 were observed, yet their activation went in opposing directions. Interestingly, MEK activation was also blunted by amino acid withdrawal. This finding suggested that the observed signal activation was specific to loss of PTEN function, and not due to generally enhanced receptor tyrosine kinase (RTK) signaling. These data suggested that amino acids are needed for PTEN to block RTK signaling to AKT.

An assessment of whether amino acid depletion indeed altered PI(3)P levels in cells was then undertaken using mass spectrometry. As shown (Fig. 5C), A549 cells contain a typical pattern of phosphoinositide abundance: the major species is unphosphorylated (PI), followed by the mono- and di-phosphorylated forms, while the triphosphorylated PI(3,4,5)P3, represents only a minuscule amount of these lipids and is typically present below the quantification limit, an observation confirmed in other mass spectroscopy lipid analyses (Milne 2005, cited above). Fragmentation spectra were quantified for PI and PIP (Fig. 5D, left panels) and based on this the PIP:PI ratio was determined (Fig. 5D, right graph and see also lipid mass table and replicate in Fig. 6C-D). This technique revealed that amino acid withdrawal resulted in massive collective reduction of mono-phosphorylated PIPs (PI(3)P, PI(4)P, PI(5)P) phosphoinositides, consistent with the strong perturbation of PTEN in this cell line (Fig. 5A,B). Taken together, these results indicated that (a) PTEN acts on endosomes to block AKT activation and (b) VPS34 functions upstream of PTEN by controlling its localization to endosomes.

EXAMPLE 4: Pharmacological inhibition of VPS 34 and perturbation of PTEN

MEF cells were treated with different types of PI 3-Kinase inhibitors and studied PTEN localization. As shown (Fig. 7A), the fungal pan-specific PI 3-Kinase inhibitor wortmannin effectively mislocalized PTEN to yield reduced punctate stain and a more diffuse cytoplasmic background. Next, the pan PI 3-Kinase and mTOR dual specificity inhibitor NVP-BEZ235 (Maira, S.M. et ah, 2008; Identification and characterization of NVP-BEZ235, a new orally available dual phosphatidylinositol 3-kinase/mammalian target of rapamycin inhibitor with potent in vivo antitumor activity. Mol Cancer Ther 7: 1851) was tested. As shown in Fig. 7 A (right most panel) it also perturbed PTEN localization into a cloudy, diffuse cytoplasmic pattern. Importantly, we observed identical effects on the PI(3)P sensitive endosomal marker EEAl (see also Fig. 8A). Very similar results were obtained when testing the primary MEFs (Fig. 7B), except that loss of Pten and EEAl signal seemed to be the prominent result after wortmannin treatment, similar to amino acid depletion in these cells.

EXAMPLE 5: Mass spectrometry assessment of the effects of the inhibitors on PIPs

Only NVP-BEZ235 consistently lowered the PIP : PI ratio in HeLa cells, which have four major lipid types on their phosphoinositides (Fig. 7C and 8B left graphs). At the same time, this drug was also most efficient at blocking cell proliferation (Fig. 8B right graph) and at blocking both AKT and S6 activation (Fig. 8C). Consistent with these findings, NVP-BEZ235 was the treatment that most profoundly affected PTEN and EEAl punctae in HeLa cells (Fig. 7D). This observation indicates that the above-described PIP ratio analysis can yield a quantitative assessment or prediction of negative effects on PTEN. Collectively, these results showed that VPS34 blocking PI 3-Kinase inhibitors can mislocalize PTEN or reduce its levels.

EXAMPLE 6: Effects of PIPs on PTEN enzymatic activity. While the above results revealed that PTEN recruitment to vesicles affects its activity and is mediated by PI(3)P, we next studied if PI(3)P had any effects on PTEN in vitro enzymatic activity. As shown (Fig. 9A), addition of PI(3)P to a PTEN activity assay (conversion of PI(3,4,5)P₃ to PI(4,5)P₂) greatly enhanced PTEN activity. Testing the N-terminus deficient (delta N1-N16) mutant of PTEN (dN-PTEN or dN16-PTEN), [the N-terminus of PTEN has been proposed to mediate lipid interaction], we found that it had no activity on the PIP₃ substrate, as expected (Leslie N.R., Foti M., 2011; Non-genomic loss of PTEN function in cancer: not in my genes. Trends Pharmacol 5α.32(3): 131-40.). Strikingly and surprisingly, however, addition of PI(3)P to the reaction fully activated the mutant (Fig. 9B). These results demonstrated that (1) the N-terminus of PTEN is not needed for PI(3)P interaction, and (2) that the dN16-PTEN mutant had not lost its inherent enzymatic activity, but rather was in a state that needed PI(3)P for conversion back to functionality. Note that PTEN was unable to catalyze PI(3)P itself (Fig. 9A, no PIP₃, + PI(3)P).

EXAMPLE 7: The location and nature of the PIP-binding motif on PTEN.

In view of the realization that the N-terminal 16 amino acids are dispensable: As shown (Fig. 10A), PTEN contains an unstructured N-terminal tail, a large catalytic phosphatase domain followed by the C2 domain and an unstructured C-terminal tail with a PDZ binding motif. By systematically deleting these domains, we found that the C2 domain alone retained full binding activity to PIPs, indistinguishable from the full length protein (Fig. 10B), demonstrating that the catalytic domain is dispensable for binding. Next, we systematically truncated the PTEN C2- domain residues between 186 and 300 and found that deletions beyond residue 270 abolished or greatly reduced PIP-binding, suggesting a critical feature in the first 100 amino acids of the C2 domain (not shown). Since a prominent feature of PTEN in this region is the CBR-loop, we replaced its features with alanine and glycine residues (see Fig. 10A). As shown (Fig. 10B), the CBR-mutant version of full length PTEN (loop-mutant) retained no PIP binding affinity demonstrating that the CBR-loop is the critical PIP ligand binding site on PTEN. We then tested if the CBR-loop is also essential for PTEN enzymatic activity. As shown (Fig. IOC), the loop mutant PTEN had solid activity that appeared similar to wt enzyme. However, we found that in contrast to wt enzyme, addition of PI(3)P had no stimulatory effect on mutant activity. Taken together, these results show that the CBR-loop of PTEN is essential for PTEN hyper- activation by PI(3)P but not for PTEN activity. To summarize, we have discovered that PI(3)P is a ligand of PTEN, which it binds via the CBR-loop in solution. Ligand binding hyper-activates enzymatic activity in the case of wt PTEN and switches it from inactive to active in the case of the dN16-PTEN mutant. Our findings suggest that PTEN can be activated in two fashions, first by recruitment to PI(3)P containing vesicles, which bring it to its PIP3 substrate. Secondly the binding of the PI(3)P ligand to PTEN on membranes will super-activate the enzyme. Thus, we expect PTEN to be most active on membranes that contain PIPs, such as endosomes, which are loaded with PI(3)P.

EXAMPLE 8: Assay for detection of PTEN activating lipids.

As shown (Fig. 11 A), the dN-PTEN mutant can be used as a highly responsive sensor for detection of PIPs (or other substances) that activate PTEN enzyme function (via the CBR-loop). In the absence of stimulating PIPs dN-PTEN activity will not be detected (or at most only minimally), whereas PIP containing solutions will trigger PTEN activity, which can be quantified relative to a PIP positive control standard curve. This assay thus measures the PTEN activating potential of a given test sample, taken for example from a patient's tumor tissue extract. Another application of our discovery involves combining the inactive dN16-PTEN mutant with compounds from a compound library to screen for activators of PTEN that work as agonists on the CBR-loop, and replace the need for PIPs. The CBR-mutant PTEN can be used as a control for ligand binding in this assay since it should not interact with a compound in analogy to PI(3)P. Such compounds could be able to boost PTEN function e.g. in the many cancer types that retain only little PTEN protein due to degradation, RNA-suppression or heterozygous gene deletion. Note that these two assays can also be performed using wt protein. However, the screening window (difference between no PIP-ligand and PIP-ligand control, Fig. 11 A-B) will be more narrow than when using the dN16- or similar PTEN mutant.

Since we found that the purified dN-PTEN mutant elutes on size exclusion at a molecular weight that is twice that of wt PTEN, we conclude that the mutant forms a homo-dimer of two PTEN units. Based on our activity assay results this dimer is catalytically inactive (Fig. 9B). However, it follows that addition of PI(3)P restores PTEN activity by converting it to the active monomer (see Fig. 9B). Since wt PTEN can typically be activated to some degree by PIPs, we conclude that a fraction of wt PTEN is present in the inactive homo-dimer form. Therefore, a PIP - PTEN activation assay requires any form of PTEN that is preferentially in the multimeric state so that addition of a PIP or other compound containing test solution can then measure activation that is the result of conversion to the enzymatically active monomeric state via binding to the CBR-loop (Fig. 12).

DISCUSSION

The invention provides quantitative assays for the determination of the levels of PTEN activity for the determination of severity of a proliferative disease, such as for instance prostate cancer in a prostate biopsy sample. The level of PTEN activity reflects the severity of the disease state and informs the clinician as to the most appropriate therapy for the patient providing the test sample. The inventors have shown that the PTEN activity level is reflected by the ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in the lipid sample; comparing the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in the lipid sample with the normal range of the ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in lipid samples from subjects not known to have prostate proliferative disease, and thereby assessing the severity of the proliferative disease of the prostate in the subject.

Determination of a normal ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in the lipid sample correlates with hyperplasia or slow progressing proliferative disease, which likely does not need to be addressed by surgical, radiological or chemotherapeutic intervention. Thus, the use of the present test on biopsy samples from prostate cancer patients would avoid needless surgeries and radiotherapies, benefiting the tested patients and sparing the healthcare system countless unnecessary and expensive procedures.

A lower than normal ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in the lipid sample correlates with aggressive disease in the patient with ongoing proliferative disease and a high likelihood of progression to more aggressive proliferative disease in patients with hyperplasia or early stage proliferative disease. The results of the above-described assays can be used in conjunction with the current standard Gleason score or any other measures of disease state to design appropriate therapies for cancer patients based on the availability of the new information provided by the present invention.

The data presented also define the membrane localization of PTEN. These data reconcile PTEN localization with activity. Surprisingly and intriguingly, a phosphoinositide is shown to control PTEN activity, which was previously thought to be constitutive. Thus, the PI(3)P containing endosomes emerge as a critical compartment for modulating signaling in cancer. PI(3)P levels are generally maintained by the class III PI 3-Kinase VPS34. The data presented herein suggest that VPS34 is a candidate tumor suppressor upstream of PTEN (see Fig. 8D). Consistent with this notion, frequent gene deletion of VPS34 (PIK3C3 gene) has been found in prostate cancer as part of the common chromosome 18q loss (Chen M., et al., 2011; Identification of PHLPP1 as a Tumor Suppressor Reveals the Role of Feedback Activation in PTEN-Mutant Prostate Cancer Progression. Cancer Cell 20: 173). Furthermore, the PI(3,4)P2 4- phosphatase INPP4b has previously been implicated in tumor suppression (Gewinner C, et al., 2009; Evidence that inositol polyphosphate 4-phosphatase type II is a tumor suppressor that inhibits PI3K signaling. Cancer Cell 16, 115; Fedele C. G., et al., 2010; Inositol polyphosphate 4-phosphatase II regulates PI3K/Akt signaling and is lost in human basal-like breast cancers. Proc. Natl. Acad. Sci. US 107:22231). Since its product is PI(3)P, the data presented herein immediately rationalize this finding mechanistically (See schematic in Fig. 8D).

These data strongly suggest that PI 3-Kinase inhibitors which adversely affect VPS34 can block PTEN and hence its tumor suppressor activity. This finding casts doubt on their intended use as a surrogate for PTEN. However, the PTEN-suppressing effects of any treatment can be measured using the immunofluorescence or mass spectrometry assays as demonstrated hereinabove. These assays as described can also be used to guide in vitro drug screens and to target therapy in patients by monitoring the PIP ratio or adverse effects on PTEN distribution in cells derived from patient blood or from biopsies. TABLE 1: Mass charge ratios of fragment

Subsequent to fragmentation of parent ions corresponding to intact lipid molecules, fragment ions were then measured and abundances used to determine PIP:PI ratios.

TABLE 2: Antibody manufacturer, uses and concentrations

Claims

Claims

1. A method of assessing the severity of a proliferative disease in a test subject, the method comprising:

providing a lipid sample from a biopsy comprising one or more cells of the test subject;

determining the ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in the lipid sample;

comparing the the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in the lipid sample with the normal range of the ratio of the total amount of the phosphatidylinositol monophosphate species to the unphosphorylated phosphatidylinositol in lipid samples from subjects not known to have a proliferative disease, and

thereby assessing the severity of the proliferative disease in the subject .

2. The method according to claim 1, wherein the proliferative disease is a proliferative disease of the prostate.

3. The method according to claim 1, wherein the phosphatidylinositol monophosphate species comprise PI(3)P.

4. The method according to claim 3, wherein the phosphatidylinositol monophosphate species comprises PI(3)P, PI(4)P and PI(5)P, and

a lower ratio of PI(3)P, PI(4)P and PI(5)P to unphosphorylated phosphatidylinositol in the lipid from the test subject than the normal range of PI(3)P, PI(4)P and PI(5)P to unphosphorylated phosphatidylinositol in the lipid is indicative of a more severe proliferative disease in the test subject.

5. The method according to claim 4, wherein the phosphatidylinositol monophosphate species as a group are resolved from other phosphatidylinositol species by mass spectroscopy.

6. The method according to claim 4, wherein the one or more phosphatidylinositol monophosphate species are resolved by thin layer chromatography (TLC), ion-exchange chromatography, high performance liquid chromatography (HPLC), gas chromatography, gel chromatography, or chromatography in a microfhiidic system, or by an antibody-based separation system.

7. A method of screening test cells from a subject for the propensity for developing a proliferative disease or condition, the method comprising:

providing a lipid sample from the one or more test cells of the subject, wherein the cells are suspected of having a propensity for a proliferative disease;

determining the relative amount of the phosphatidylinositol monophosphate species in the lipid sample;

comparing the relative amount of the phosphatidylinositol monophosphate species in the lipid sample from the test cells if the subject with the relative amount of the phosphatidylinositol monophosphate species in a lipid sample from one or more normal cells not suspected to have a propensity for developing a proliferative disease or condition from the same subject, and

thereby assessing the propensity of the test cells for developing a proliferative disease or condition.

8. The method according to claim 7, wherein the phosphatidylinositol monophosphate species comprises PI(3)P.

9. The method according to claim 8, wherein the phosphatidylinositol monophosphate species consists essentially of PI(3)P, PI(4)P and PI(5)P.

10. The method according to claim 9, wherein

a lower relative amount of PI(3)P, PI(4)P and PI(5)P in the lipid from the test cells than the relative amount of PI(3)P, PI(4)P and PI(5)P in the lipid from the one or more normal cells is indicative of a propensity of the test subject for developing a proliferative disease or condition.

11. The method according to claim 9, wherein the one or more phosphatidylinositol monophosphate species are resolved by mass spectroscopy.

12. The method according to claim 8, wherein the one or more phosphatidylinositol monophosphate species are resolved by thin layer chromatography (TLC), ion-exchange chromatography, high performance liquid chromatography (HPLC), gas chromatography, gel chromatography, or chromatography in a microfhiidic system, or by an antibody-based separation system.

13. A method of monitoring progress of a proliferative disease or condition in a subject undergoing a therapy for the disease or condition, the method comprising:

providing two or more lipid samples from cells of the subject taken at different times during the therapy;

determining the relative amount of one or more phosphatidylinositol monophosphate species in the lipid samples from the subject;

comparing the relative amount of the one or more phosphatidylinositol monophosphate species of the lipid sample from the subject at the earliest time before or during therapy with the relative amount of the one or more phosphatidylinositol monophosphate species in each of the lipid samples taken from the subject at one or more subsequent times during or after the therapy; and

thereby monitoring the progress of the therapy.

14. The method according to claim 13, wherein the one or more phosphatidylinositol monophosphate species comprises PI(3)P.

15. The method according to claim 14, wherein the one or more phosphatidylinositol monophosphate species are resolved by mass spectroscopy.

16. The method according to claim 14, wherein the one or more phosphatidylinositol monophosphate species are resolved by thin layer chromatography (TLC), ion-exchange chromatography, high performance liquid chromatography or gas chromatography.

17. The method according to claim 14, wherein an increase in the relative amount of PI(3)P is indicative of a therapeutic effect on the proliferative disease or condition in the subject.

18. A method of assessing the effect of a prospective therapeutic agent on PTEN tumor suppressor activity, the method comprising:

providing a first lipid sample from one or more cells of a normal subject or a tissue culture cell line and a second lipid sample from one or more cells of the normal subject or the tissue culture cell line after exposure to the prospective therapeutic agent; comparing the relative amounts of one or more phosphatidylinositol monophosphate species in the lipid sample from the one or more cells treated with the prospective therapeutic agent with the relative amount of the phosphatidylinositol monophosphate species in the lipid sample from the one or more otherwise identical untreated cells; and

thereby assessing the effect of the prospective therapeutics agent on PTEN tumor suppressor activity.

19. The method according to claim 18, wherein the one or more phosphatidylinositol monophosphate species comprises PI(3)P.

20. The method according to claim 19, wherein a finding that the relative amount of PI(3)P in the lipid sample from the prospective therapeutic agent-treated one or more normal cells is less than the relative amount of PI(3)P in the lipid sample from the one or more untreated normal cells is indicative of the prospective therapeutic agent having a propensity for increasing the likelihood of a proliferative disease or condition in a subject treated with the therapeutic agent.