WO2013076501A2 - Screening method - Google Patents

Abstract

The present invention provides a method for identifying agents useful in the treatment and/or prevention of a disease associated with insulin resistance and/or glucose intolerance which comprises the step of investigating the capacity of a test agent to inhibit the Vps34 signalling pathway and/or the ΡΙ3Κ-02β signalling pathway. The present invention also provides a transgenic non-human animal which comprises a mutation in the gene encoding Vps34 or ΡΙ3Κ-C2β such that the active site is inactivated.

Description

SCREENING METHOD

FIELD OF THE IN VENTION

The present invention relates to methods for identifying agents useful in the treatment of diseases associated with insulin resistance and/or glucose intolerance, such as type II diabetes, non-alcoholic fatty liver disease (NAFLD), and cardiovascular diseases. BACKGROUND TO THE INVENTION

Insulin controls glucose and lipid homeostasis by modulating the functions of multiple organs and tissues, including liver, muscle and fat. In muscle and fat, insulin stimulates glucose uptake resulting in glucose clearance from circulation, and induces lipid synthesis. In the liver, insulin blocks glucose production and stimulates fatty acid synthesis. Any impairment in the insulin signalling pathway plays a key role in the development of insulin resistance.Insulin resistance initially manifests itself as the constellation of symptoms called "insulin resistance syndrome", which include glucose intolerance, obesity, and hypertension. Insulin resistance promotes the development of diseases such as type II diabetes andnon-alcoholic fatty liver (hepatic steatosis). Insulin resistance syndrome is increasing in prevalence with alarmingrapidity, affectingmore than 25% of adults in the United States. More than 50% of obese children have insulin resistance syndrome.

Among the insulin resistance related diseases,Type II diabetes and non-alcoholic fatty liver are the most prevalent and are associated with obesity.

The Type II diabetes is characterised by insulin resistance associated with an insulin secretion defect.Pancreatic beta cells initially compensate for this insulin resistance by increasing their insulin output. Over time, these cells become unable to produce enough insulin to maintain normal glucose levels, indicating progression to Type II diabetes.

Strongly associated with obesity and insulin resistance,non-alcoholic fatty liverdisease (NAFLD) can occur in individuals whose alcohol consumption is insignificant. NAFLDis rapidly becoming the most common cause of liver disease in Western countries, where estimates of NAFLD prevalence vary between 20 and 30%, rising up to 90% in morbidly obese individuals.NAFLD has a strong association with type II diabetes, with steatosis presenting 70% of type II diabetics screened with ultrasound, and thus it is now recognized to represent the hepatic manifestation of the metabolic syndrome. NAFLD is characterized by the deposition of lipids in the liver leading to fatty liver (hepatic steatosis). Fatty liver is often self -limited but it can progress to steatohepatitis and liver cancer. Although NAFLD is strongly associated with obesity and insulin resistance, its pathogenesis remains poorly understood, and therapeutic options are limited.

There is currently no efficient cure to treat patients with insulin resistance metabolic diseases. Conventional treatments for insulin resistance are very limited, and focus on attempting to control blood glucose levels in order to minimize or delay complications. Current treatments target either insulin resistance (for example metformin which inhibits hepatic glucose production, and thiazolidinediones ("TZDs") which increase glucose uptake), or insulin release from the beta cell (for example sulphonylureas, exenatide). Sulphonylureas and other compounds that act by depolarizing the beta cell, have the side effect of hypoglycemia since they cause insulin secretion independent of circulating glucose levels. One approved drug, Byetta (exenatide), stimulates insulin secretion only in the presence of high glucose, but is not orally available and must be injected. Januvia (sitagliptin) is another recently approved drug that increases blood levels of incretin hormones, which can increase insulin secretion, reduce glucagon secretion and have other less well characterized effects. However, Januvia and other dipeptidyl peptidases IV inhibitors may also influence the tissue levels of other hormones and peptides, and the long-term consequences of this broader effect have not been fully investigated.

There is therefore a need for improved agents to treat diseases associated with insulin resistance and/or glucose intolerance.

DESCRIPTION OF THE FIGURES Figurel: Generation of Vps34 D761A/+ mice

Figure 2: Vps34 D761A/+ mice exhibit enhancedinsulin sensitivity and better glucose tolerance.

Figure 3: Vps34 D761A/+ mice remain insulin-sensitive under obese condition.

Figure 4: Reduced High Fat Diet-induced hepatic steatosis in Vps34 D761A/+- mice (A and

B) is correlated with enhanced blood adiponectin level (C).

Figure 5: Vps34 D761A/+ female mice also exhibit enhanced insulin sensitivity and better glucose tolerance.

Figure 6: Generation of PI3K-C2p^{Dl212A/Dl2l2A} mice Figure 7:PI3K-C2p^{DI2i2A/D1212A} mice exhibit enhanced insulin sensitivity and better glucose tolerance

Figure 8:PI3K-C2 ^{D1212A/D1212A} mice remain insulin sensitive under obese condition and there is a decrease in accumulation of triglycerides in the liver (F and G).

Figure 9:A and B.Confocal analysis of mouse hepatocytesC. Comparing levels of P-S6K (T389) and P-S6 in WT and PI3K-C2p^{D1 12A/D1212A} mice. D. No difference in Insulin receptor (IR) and increase in IKS levels in C2p^{D121 A/D1212A} hepatocytes verswswild-type.

Figure 10: Comparing leptin and adiponectin levels in pi3K-C2 ^{D,212A/D,212A} and wild type mice.

Figure 11: Table showing IC50 values (μΜ) for selected PI3K inhibitors against Lipid Kinases according to Knight et al. Cd\125, 733-747 (2006), Kong et al, EJC 46 1 1 11 -1121 (2010) and Miller, et al. Science, 327, 1638-1642 (2010).

Figure 12: Structure of PI3K inhibitors

Figure 13:Vps34 inactivation protects against high-fat diet induced hepatic steatosis.

A. Histologic evaluation of the liver collected from the High-Fat-Diet subjected mice and those fed normal chow. H&E staining is the classic staining used for basic cellular cytology.

The "H" turns acidic structures (such as DNA) blue and the "E" turns the proteins red. B. Oil red O staining of liver sections of normal chow and High-Fat-Diet subjected mice.

Figure 14: Effect of long term High Fat Diet on WT and Vps34^D761A/+ body composition. The ratio of lean/fat tissue of wild type and Vps34^D761A+ mice which had been subjected to

45% High Fat Diet for 7 months

Figure 15: Echocardiographic Measurements.

Echocardiography was used to measure LV dimensionswith M-mode and to trace endocardial area with B-mode at the level of the papillary muscles with the average of 3 cardiac transmitral flow measured by Doppler traces. D and s: diastole and systole, respectively; IVS: intraventricularseptal wall thickness; LVD: left ventricular dimension; LVPW: posterior wall thickness; LV: left ventricle volume; EF: ejection fraction; FS: fractional shortening. Values are means ± SEM, n=6/6.

Figure 16: Metabolic parameters of high-fat-diet subjected mice.

A. Plasma Free Fatty Acid (FFA) levels were measured in Vps34^D761A/+ mice and control animals under normal chow (NC) or High-Fat_Diet (F1FD). (n=10/9)

B. Plasma triglycerides (TG) levels were measured in Vps34^D76lA/+mice andcompared to control animals under High-Fat_Diet (HFD). (n=10/9)

C. Plasma Insulin levels were measured in Vps34 mice and compared to control animals under High-FatJDiet (HFD). (n=10/9)

Figure 17: PI3P levels are reduced in ΡΓΚ3ΰ2β primary hepatocytes

PI3P levelsin primary hepatocytes from PIK3C2$ knock in and normal mice. Figure 18: Akt signalling is enhanced in PIK3C2p. insulin responsive tissues and primary hepatocytes and not in the spleen.

Western blot of cell lysates isolated from indicated tissues after insulin administration (0.75U/kg by i.p.) or from primary hepatocytes (lower panel) stimulated with lOOnM insulin at indicated time.

Figure 19: Enhanced Akt signalling is maintained in PfK3C2mice subjected to high fat diet.

A. Western blot of cell lysates isolated from indicated tissues after insulin administration (0.75U/kg by i.p.).

B. Liver Triglycerides content in the ΡΙΚ302βΚΙ livers compared to wild type.

SUMMARY OF ASPECTS OF THE INVENTION

PI3 -C2p and Vps34 are class II and III isoforms, respectively, of the phosphoinositide 3- kinase (PI3K) family, which are involved in endosomal trafficking and autophagy. The present inventors have created Vps34 kinase-dead (KI) mice, in which the active site of Vps34 has been inactivatedgiving rise to Vps34 D761A. The same approach has been employed to generate PI3K-C2p KI mice orPI3K-C2 ^{D1212A/D12,2A}mice.

It was found that heterozygous vps34 KI mice (where 50% of vps34 activity is inactivated), and homozygous and heterozygous ΡΙ3Κ-02β KI mice, display improved glucose tolerance and enhanced insulin sensitivity. Also, heterozygous male Vps34 KI mice and homozygous PI3K-C2p KI miceare protected from High Fat Diet-induced fatty liver (also known as hepatic steatosis). Thus, Vps34 and/or PI3K-C2 inhibitors will be useful to increase insulin sensitivity and/or improve glucose tolerance.

In a first aspect the present invention provides a method for identifying agents useful in the treatment and/or prevention of a disease associated with insulin resistance and/or glucose intolerance which comprises the step of investigating the capacity of a test agent to inhibit the Vps34 signalling pathway and/or the PI3K-C2 signalling pathway.

The method may comprise the step of investigating whether a test agent inhibits the kinase activity of Vps34 or PI3K-C2P or both.

The method may be conducted in vitro or in a cell in culture. The method may be used to screen for compounds capable of increasing insulin sensitivity in a subject.

The method may be used to screen for compounds capable of improving glucose tolerance in a subject.

Adiponectin may be used as a biomarker for agents capable of inhibiting the Vps34 signalling pathway and/or the PI3K-C2p signalling pathway. The disease associated with insulin resistance and/or glucose intolerance may, for example, be selected from the following group: type II diabetes, hepatic steatosis and non-alcoholic fatty liver disease (NAFLD).

In a second aspect, the present invention provides a transgenic non-human animal which comprises a mutation in Vps34 or PI3K-C2psuch that the active site is inactivated.

The transgenic non-human animal may comprise a mutation in the DFG motif of the ATP- binding site. The mutation may cause the motif to have the sequence AFG. Where the transgenic non-human animalcomprises a mutation in Vps34 it may comprise the mutation D761A.

Where the transgenic non-human animalcomprises a mutation in PI3K-C2P it may comprise the mutation D1212A.

DETAILED DESCRIPTION PI3K Phosphoinositide 3-kinases (PI 3-kinases or PI3Ks) are a family of enzymes involved in cellular functions such as cell growth, proliferation, differentiation, motility, survival and intracellular trafficking.

PI3Ks are a family of related intracellular signal transducer enzymes capable of phosphorylating the 3 position hydroxyl group of the inositol ring of phosphatidylinositol (Ptdlns). The PI3 family is divided into three different classes: Class I, Class II, and Class III. The classifications are based on primary structure, regulation, and in vitro lipid substrate specificity. Class II comprises three catalytic isoforms (PI3K-C2a, PI3K-C2p, and PI3K-C2y) which, unlike Classes I and III, arenot constitutively associated with a regulatory subunit. Class II PI3Ks catalyse the production of PI(3)P and PI(3,4)P₂ from the PI lipid. PI3K-C2a and PI3K- C2p are expressed throughout the body, whereas expression of PI3K-C2y is limited to hepatocytes. The distinct feature of Class II PI3Ks is the C-terminal C2 domain. This domain lacks critical Asp residues to coordinate binding of Ca²⁺, which suggests class II PBKs bind could bind lipids in a Ca²⁺-independent manner.

There is only one known class III PI 3 -kinase, Vps34, which is also the only PI3 expressed in all eukaryotic cells. In humans it is encoded by the PIK3C3 gene. In human cells Vps34 associates with a regulatory subunit, pl50.

The class III kinase produces PI(3)P from PI.

VPS34 AND PI3K-C2p SIGNALLING PATHWAYS

In the method of the first aspect of the invention, the capacity of a test agent to inhibit the Vps34 and/or PI3K-C2 signalling pathway is investigated.

The agent may directly inhibit Vps34 and/or PI3K-C2P, for example by inhibiting the kinase activity of Vps34 and/or PI3K-C2 . The agent may downregulate Vps34 and/or PI3K-C2 expression, or downregulate one or more binding partner(s) of Vps34 and/or PI3K-C2 .

SCREENING METHOD In the first aspect, the present invention relates to a method for identifying agents useful in the treatment and/or prevention of a disease.

Agents are identified on the basis of their capacity to inhibit the Vps34 signalling pathway or the PI3K-C2 signalling pathway.

An in vitro screen may be conducted for agents such as for small molecule inhibitors that inhibit the in vitro kinase activity of vps34 or PI3K- C2p or both. Agents, such as RNAi, capable ofdownregulating binding partners of Vps34, may interfere with Vps34 stability, and thus indirectly with its function. The method may be a cell-based assay, looking for agentswhichinterfere with the biology and signaling of these PD s. For Vps34, the assay may use autophagy, a biological phenomenon in which this kinase is involved in cells. Such assays, which are based on monitoring intracellular vesicular traffic, are known in the art. Alternative cell-based assays include those based on forced over/inducible expression of Vps34 in cells, which may lead to cell death. Agentswhich inhibit the Vps34 signalling pathway would then rescue this cell death.

For PI3K-C2 , cell-based assays may be based on potentiating insulin signaling in cells.

In vivo studies may be conducted, for example in rodents to investigate the time and dose- dependent impact of putative Vps34 inhibitors on blood glucose levels, for example under fasting conditions in mice. Glucose tolerance tests (GTTs) and insulin tolerance tests (ITTs) and metabolic analysis may also be conducted to assess improvements in insulin sensitivity.

The tissue levels of the Vps34 lipid product PtdIns3P may be assayed in primary tissues in vivo. This provides a direct readout of target inhibition in vivo.

In vivo Vps34- or PI3K-C2p-dependent phosphorylation response/biomarkers may be used to investigate the effect of test agents on Vps34-dependent orPI3K-C2p-dependent cellular phosphorylation.

In vivo metabolic assays may also be conducted.

The discriminating characteristics of PI3K-C2p KI Vps34 I mice are shown in Table 1.

PI3K-C2P KI Vps34 KI

metabolic phenotype - gender in males only in both males and females phenotypes in heterozygous mice very mild strong

Phosphorylation of Akt kinase upon ††T O to†

insulin stimulation

circulating leptin †(in High fat diet) No effect circulating adiponectin no effect † (in High Fat Diet)

Impact on cellular PI3P lipid Reduced in MEFs Not affected in MEFs upon upon insulin insulin stimulation stimulation

vesicular traffic defects endosomal defect autophagy flux defect

Characteristics associated with ΡΙ3Κ-02β or Vps34 kinase-dead mice may be used as markers for ΡΙ3Κ-02β or Vps34 inhibition. For example, adiponectin, an adipokine known to reduce hepatic and serum triglyceride levels and to protect from nonalcoholic hepatic steatosis, may be used as a biomarker for agents capable of inhibiting Vps34. Leptin, on the other hand, may be used as a biomarker for agents capable of inhibiting PI3K-C2p.

TEST AGENT The test agent may be based on one of the known class H/III PI3K inhibitors. Vps34 and PI3K-C2p activity can be inhibited by pan-PI3K inhibitors such as wortmannin and 2-(4- mo holinyl)-8-phen lchromone (LY294002). PI3K-C2p is also inhibited by these compounds, but at higher doses. In previous studies, 3-methyladenine (3 -MA) has often been used and at very high concentration (e.g., 10 mM for 3 -MA) to inhibit Vps34. For instance, 3-MA has been used to test whether protein degradation processes are autophagy-dependent in cells. However, it was later demonstrated that other general PI3K inliibitors, such as wortmannin also inliibit autophagy suggesting that these agents have additional off-target effects (see table 1) that may confound interpretations in certain contexts.

Several inhibitors targeting class I PI3Ks or mTOR have been identified and show some inhibitory activity towards class II and III PI3Ks (Figure 1 1 and Figure 12 for structure). For example, the IC50s of ZSTK474, GDC-0941, NVP-BEZ235 and LY294002 for PBK-C2p were 0.176, 0.590, 0.044 and 10.4 μΜ, respectively (Figure 1 1).

The test agent may be based on one of the PI3K inhibitors shown in Figure 1 1.

The term "based on" is used to indicate that the test agent may have a similar chemical structure to one of the PI3K inhibitors shown in Figure 1 1 , with one or more minor variation(s), such as to side chain or ring substiruents to increase the selectivity of the molecule for class II or III PDKs. The test agent may alternatively have previously unreported chemical structure.

Where the test agent is capable of inhibiting the expression of Vps34 or ΡΙ3Κ-€2β, a binding partner, or another component involved in the Vps34 or ΡΙ3Κ-02β signalling pathways, it may be a nucleic acid-based molecule, such as an antisense sequence or ansiRNA.

DISEASE The disease may be any disease or medical condition associated with insulin resistance and/or glucose intolerance.

The disease may, for example, be type II diabetes ornon-alcoholic fatty liver disease (NAFLD).

Diabetes mellitus type II, also known as non-insulin-dependent diabetes mellitus (NIDDM) or adult-onset diabetes, is a metabolic disorder that is characterized by high blood glucose in the context of insulin resistance and relative insulin deficiency. Diabetes is often initially managed by increasing exercise and dietary modification. The classic symptoms of diabetes are polyuria (frequent urination), polydipsia (increased thirst), polyphagia (increased hunger), and fatigue.

Hepatic steatosis is an abnormal fat accumulation in the liver that can have detrimental consequences on liver functions and in some cases can lead to hepatocarcinomas.

Non-alcoholic fatty liver disease (NAFLD) is one cause of a fatty liver, occurring when fat is deposited (steatosis) in the liver not due to excessive alcohol use. It is related to insulin resistance and the metabolic syndrome and may respond to treatments originally developed for other insulin-resistant states (e.g. diabetes mellitus type Π) such as weight loss, metformin and thiazolidinediones.

TRANSGENIC ANIMALS

The second aspect of the invention relates to a transgenic non-human animal in which the active site of Vps34 or PI3K-C2 is inactivated. The non-human animal may be a mammal. The non-human animal may be, for example, a rodent, such as a mouse or rat.

The transgenic animal may comprise an inactivating mutation in Vps34 or PI3K-C2p. The mutation may be in the active site of the enzyme. For example, the mutation may be in the DFG motif of the ATP-binding site of the enzyme. The mutation may mean that the encoded amino acid sequence has an AFG sequence instead of a DFG motif.

For the Vps34 murine gene the mutation may be D761A. For other non-human animals, the mutation may be in the position equivalent to D761 in the mouse Vps34 sequence.

For the PBK-C2p murine gene the mutation may be D1212A. For other non-human animals, the mutation may be in the position equivalent to D1212A in the mouse PI3K-C2p sequence.

The invention will now be further described by way of Examples, which are meant to serve to assist one of ordinary skill in the art in carrying out the invention and are not intended in any way to limit the scope of the invention.

EXAMPLES

Example 1 - Generation of vps34 knockin mice

Knock-In (KI) vps34 mice were generated by introducing a germline inactivating mutation (D761A) in the ATP-binding site (DFG motif) of the PIK3C3 gene that encodesvps34 (Fig. 1A). The resulting mutant allele (hereafter designated as vps34^D7S1A), encodes vps34 protein in a constitutively inactive form (also called kinase dead). The Vps34^D7fiIA allele was subsequently maintained into the C57BL/6J mouse genetic background through backcrossing. The heterozygous vps34^{D761A +}(+ indicates the wild-type (WT) vps34 allele) mice are fertile, phenotypically comparable to the vps34^+/+controls and can live at least up to 54 months with no apparent behavioural defects during this period. However, no viable homozygous (vps34^{D761A/D7 1A}) mice were obtained from any heterozygous intercrosses suggesting that homozygosity for vps34 mutation leads to embryonic lethality. Further characterization iinnddiiccaatteecd that vps34^{D7 1A/D761A} mutant embryos died before day 6.5 of embryogenesis (data not shown).

We then examined the expression pattern of the vps34 protein in the vps34^{D761A +} mice. Vps34 was ubiquitously expressed in murine adult tissues with variable expression levels across the different organs (the highest being in the brain). Importantly, vps34 expression was unchanged in the vps34^D761A/+ mice (and Mouse Embryonic Fibroblasts (MEFs)) and was similar to WT levels (Fig. IB). As expected, tissues isolated from vps34 hetero2ygous mice display a 50% reduction of vps34 kinase activity (Fig. 1C) confirming that heterozygous mice carrying the D761A mutation have a 50% loss of function of vps34 activity.

Example 2 - Enhanced insulin sensitivity and better glucose tolerance in vps34 heterozygous mice

It was investigated whether vps34 inactivation in mice has an impact on insulin sensitivity. First, no differences in blood glucose levels were found in chow-fed mice whereas a mild hypoglycaemia was detectable in the vps34^D761A/+ mice compared to WTmice in starved conditions (-15% decrease) (Fig. 2A). This effect was also apparent in females (Fig. 5). We next performed insulin and glucose tolerance tests (ITTs and GTTs). Compared to WT mice, vps34^D761A/+ mice showed a marked reduction in glucose after insulin injection as well as an improvement in GTT (Fig. 2B). This suggests that vps34^D761A mice displayedimproved insulin sensitivity and a better glucose tolerance. The insulin levels in chow-fed and starved mice (males and females) werefound to be reduced in heterozygous mice compared to WT mice, correlating with the enhanced insulin sensitivity observed in mutant mice (Fig. 2C). Altogether, these data show that 50% reduction of vps34 activity strongly affects glucose homeostasis and insulin sensitivity in mice.

To test whether this effect could be due to an insulin signalling defect, signaling pathways were monitored in response of insulin in the different insulin-responsive tissues. Among all the tissues tested, only the liver of vps34^D761A/+ mice showed a mild increase in phosphorylation of PKB (P-Ser473) upon insulin stimulation compared to WT mice (Fig. 2D and data not shown), suggesting that the insulin sensitivity described above could be due to a specific role of vps34 in hepatic insulin signalling.

Example 3 - Vps34 D761/+ mice remain insulin sensitive under obese condition

Because inactivating 50% of vps34 activity strongly enhanced insulin sensitivity and glucose tolerance as described above, it was hypothesized that inactivating vps34 under conditions that mimic metabolic syndrome such as high-fat diet (HFD) could have a beneficial consequence.

The present inventors therefore subjected WT and vps34^D761A/+ male mice to 16 weeks of HFD (45% fat) and monitored weekly their body weight. Comparison of HFD-fed WT and vps34^D761A/+ miceshowed that these animals gained similar body mass, with a mild tendency for the heterozygous mice to be leaner (area under the curve (AUC) p<0.0577) (Fig. 3A). No significant differences between vps34 'and WT mice regarding food/water intake (data not shown) and leptin levels (Fig. 4C) were found. The HFD-fed vps34^D761A/+ mice continued to develop a mild fasting hypoglycaemia whereas no significant differences were apparent in fed states (Fig. 3B). As for chow-fed conditions, HFD-fed vps34^D761A/+ mice remained insulin sensitive as assessed during an insulin tolerance test (Fig. 3C, left panel). However, no differences were observed during a glucose tolerance test when compared to HFD-fed WT mice, indicating that HFD-fed vps34^D761A/+ mice lost their ability to clear glucose better under HFD conditions (Fig. 3C, right panel).

Moreover and in contrast to normal chow-fed (NC) conditions, HFD-fed vps34 mice did not show any effect on hepatic insulin signalling (Fig. 3D).

Example 4 - Reduction of hepatic steatosis in Vps34^{D761A +} mice

One aspect of the metabolic syndrome is represented by the nonalcoholic fatty liver disease (also called NAFLD) characterized in the early stages by hepatic steatosis (also known as fatty liver). Hepatic steatosis of HFD-fed mice from both WT and vps34 heterozygous mice groups was compared. It was found that vps34^D7<51A/+ mice showed a markedly reduced liver lipid accumulation (correlated with steatosis) when compared to controls as assessed by Oil Red O staining of liver sections (Fig. 4B, lower panel). Consistent with this, a significant decreased in liver tissue weight was observed in the vps34^D761A/+ mice as compared with WT whereas no changes were detected in other tissues (Fig. 4A). This supports the previous observation that constitutive inactivation of vps34 has a critical impact on liver function.

Furthermore, the present inventors monitored the level of adiponectin, an adipokine known to reduce hepatic and serum triglyceride levels and protects from nonalcoholic hepatic steatosis(Polyzos et al, Diabetes, Obesity and Metabolism (2010)). As shown in Fig. 4C, blood levels of adiponectin were strongly increased in the HFD-fed vps34^D7ilA/+ mice. This finding correlates with the steatosis protection observed in the mutant mice as adiponectin is often inversely correlated with body fat levels. For instance, high adiponectin in the lean state is linked to elevated insulin sensitivity, whereas low adiponectin in the obese state is linked to insulin resistance and type II diabetes. It is notable that the levels of leptin, an adipokinecontroling energy balance and body weight, were not changed between both groups under HFD ruling out a possible role of vps34 in leptin signalling. Together, these data demonstrate that 50% inactivation of the vps34 gene preventshigh-fat- diet-induced hepatic steatosis. Histologic evaluation of the liver collected from the High-Fat-Diet subjected mice showed a significant reduction in hepatic steatosis in vps34^D761A/+ mice (Figure 13 A). Red O staining was also carried out to assess the amount of fat in the liver tissue. As shown in Figure 13B, the reduction in hepatic steatosis correlated with a reduction of the oil red O staining.

Example 5: Effect of long term High Fat Diet on WT and Vps34^D7MA/+ body composition

Wild typeandVps34^D761A/+ mice were subjected to 45% High Fat Diet, using the methodology described in Example 3, for 7 months. Body composition measurements were performed with a BrukerBioSpec 70/30 7 Tesla magnetic resonance imaging (MRI) scanner (BrukerBiospin, Ettlingen, Germany). The ratio of lean/fat tissue was calculated from the nuclear magnetic resonance (NMR) data and expressed as % (Figure 14). A long term High Fat Diet did not have a further deleterious impact on body composition of Vps34^D761A/+ compared to WT control animals.

Example 6: Echocardiography Measurementson WT and Vps34^P761A/+mice

Malewild type and Vps34^D761A/+ mice (n=6/6, 13-15 week-old, fed on normal chow diet) were used to detect cardiac structure alterations and cardiac function in vivo by using echocardiography (Vevo770, Visualsonics) by a blinded operator. LV dimensions were measured with M-mode and endocardial area was traced with B-mode at the level of the papillary muscles with the average of 3 cardiac transrnitral flow measured by Doppler traces.

The results are shown in Figure 15 and Table 2. Vps34^{D761A +} mice showed a normal cardiac function, wall thickness, mass (representing hypertrophy) and cardiac dimension (for dilatation).

2. Echocardiography measurements

D and s: diastole and systole, respectively; IVS: intraventricularseptal wall thickness; LVD: left ventricular dimension; LVPW: posterior wall thickness; LV: left ventricle volume; EF: ejection fraction; FS: fractional shorten/rig. Values are means ± SEM, n=6/6. Example 7: Metabolic parameters of high-fat-diet subjected mice

The Plasma Free Fatty Acid (FFA) levels, Plasma triglycerides (TG) levels and Plasma Insulin levels were compared between Vps34^{D?61A +} mice and control animals under normal chow (NC) or High-Fat_Diet (HFD). The results are shown in Figure 16.

No significant differences in plasma FFA levels were detected between the Vps34^D76IA/+ mice and control animals (Figure 16A). A significant reduction ofplasma TG levels was detected in the Vps34^D761A/ mice compared to control animals under High-FatJDiet (Figure 16B).

A tendency toward a reduction ofplasma insulin levels was detected in the Vps34^D761A/+ mice compared to control animals under High-Fat_Diet (FTFD) indicating that the Vps34^D761A/+ mice are less insulin resistant compared to WT mice. (Figure 16C)

Example 8 - Generation and characterisation of ΡΙ3Κ-02β knock-in mice

In order to study the organismal role of ΡΙ3Κ-02β, we created a germline knockin mouse line in which the DNA encoding the ATP-binding DFG motif of PIK3C2B, the gene encoding PI3K-C2P, is mutated to encode the AFG sequence instead, resulting in a kinase-dead protein, further referred to as ΡΙ3Κ-€2β ^D1212A (Fig. 6A). Homozygous mice for the mutated PIK3CB allele (hereafter called ΡΙ3Κ-02β^ο1212Αιτηΰβ) were bom at the expected Mendelian ratios. ΡΒΚ-02β protein is highly expressed in some tissues including brain, white adipose tissue (WAT), spleen, pancreas, lung and prostate (Fig. 6B). Expression of the mutant PI3K-C2 protein and other PI3 isofonns was unaffected, both in Mouse embryonic fibroblasts (MEFs) derived from E13.5 embryos and in WAT from adult mice (Fig. 6C). PI3K activity in PI3K- 02βίπΐΓηυηορΓεα^'ρίΐ3ί65 was completely abolished from pi₃K-C2 ^{D1212A D12,2A} brain and liver and no compensatory alterations in the activities of ΡΓ3Κ-02β (Fig. 6D). Despite considerable effort, the main physiological lipid products produced by ΡΙ3Κ-02βθΓ class II PI3Ks in general are still unknown. The present invention provides a relevant physiological model to address this question.

Phosphoinositide levels were measured in wt and ΡΙ3Κ-02β mutant MEFs in unstimulated conditions or after 10 minutes insulin stimulation. MEFs were labelled with ³ P- orthophosphate and the relative levels of each phosphoinositide species PIP₃, PI3,4P₂, PI3,5P_2, PI4P and PI3P were quantified. A small but significantincrease in the levels of PDP and PI4P production were observed after 10 min insulin stimulation (Fig6D). In contrast, the levels of PEP3,5P2, PI3,4P2 and PIP3in WT MEFs were not significantly changed upon insulin stimulation. The increased in PI3P production is clearly abrogated in the PI3K- _C2pDi2i2A/Dm2A_{MEFs whereas} p_I4p _{levels are not affected by} PB_K-C2p reactivation (Fig

6D). Moreover, it seems that P13K-C2p knockdown slightly affects PI3,4P2 and PIP3 production but not in a significant manner (Fig 6D). This result shows for the first time a clear evidence of the role ofPI3K-C2p lipid kinase in the PI3P production in vivo.

Example 9 - Increasedglucose tolerance and insulin sensitivity in PI3K-C2 ^{D1 12A/D1212A}mice PI3K-C2p^{D1212A/D1212A} mice are viable and fertile, with no apparent defects. No body weight difference was observed either in postnatal or at the embryonic stage in the mutant mice compared to WT mice (Fig.7A and data not shown). The levels of blood glucose in overnight- fasted or randomly-fed states are unaffected in 12 week old pi3 -C2p^{D,2I AD, 212A} mice (Fig. 7B). However the plasma insulin level in a randomly-fed state or after glucose stimulation was strongly reduced i the pi3 -C2p^{D1 12A/D1212A} compared to WT mice (Fig.7C, Fig. 7D). No difference in insulin levels was observed afteran overnight starvation where the insulin secretion is minimal (Fig.7C). Surprisingly, despite their hypoinsulinemic state, the PI3K- C2p^{D1212A/D1212A} mice are more glucose tolerant as assessed by a glucose tolerance test (Fig. 7E). Glucose clearance is clearly increased upon injection of glucose in the ΡΙ3Κ-02β D12I 2A/D1212A _mj_ce (pjg ₇£) These results show that despite their hypoinsulinemic state, the mutant mice are more glucose tolerant suggesting an increase in insulin sensitivity. To address this hypothesis, an insulin tolerance test was performed. A higher insulin hypoglycemic response was observed in the mutant compared to WT mice (Fig.7G). This increase in insulin sensitivity is correlated to an increase in insulin signalling as observed by the increase of Akt phosphorylation on the serine 473 site in the liver pBK-C2p^{D121 A/D1212A} mice after insulin stimulation (Fig. 7G).

Taken together, these data show that 12-week-old PI3K-C2p^{um2A U A} mice are more insulin sensitive and have an increase in insulin signalling. This metabolic phenotype has been also confirmed in older mice but was not found in females (data not shown and fig.10). These results suggest that the κ_£2β°^{1212Αωΐ212Α} mice could be more protected against age or high fat diet-induced metabolic disorders such as insulin resistance and steatosis.

Example 10 - PI3K-C2 "'^"^'^ mice are protected against HFD-induced liver steatosis To address this question, the pi3K_c2p^D12,2AD12I2A mice were challenged with a high fat diet for 16 weeks. Firstly, body weight was monitored over the HFD weeks. No difference in weight gain was observed between the WT and _PI3_K__C2p^DI212AyDm2A mice (Fig.8A). The levels of blood glucose and plasma insulin (Fig. 8B) are unaffected in the PI3K- C2 ^{D1212A/D1212A} mice, with both genotypes displaying increased in glycaemia. Surprisingly no difference in glucose tolerance (Fig. 8C) was observed between the wt and mutant mice after HFD. However the c2p^D1212AD,212A mice are less insulin resistant than WT mice (Fig. 8D). PBK-C2p^{D12l2A/D1212A} mice seem to be protected against the insulin resistant-detrimental effects such as steatosis. Indeed, the mutant livers are less "fatty" compared to the WT ones as observed on Hematoxylin eosin liver sections. This decrease of accumulation of triglycerides (TG) in pi3K-C2p^{D12UA/J 1 12A} livers has been confirmed by an Oil RedO staining that specifically stains neutral lipids (Fig. 8F). These results show that PI3K-C2p inactivation decreases HFD-induced steatosis and for some cases protects completely against steatosis. In parrallel the Triglycerides that are not accumulated in the liver appear to be partially accumulating in the WAT of the mutant mice. Indeed a slight increase in adipocyte size in the PI3K-C2p^{D1212A/D1212A} compared to WT mice was observed. This increase in WAT weight could have been detrimental for the general organism but it does not affect the body weight.. In light of these results it seems that PI3K-C2P inactivation protects against HFD-induced steatosis mainly due to the increase of insulin sensitivity in the liver.

Example 11 - PI3K-C2p inactivation may affect endocytic trafficking

An immunofluorescence staining was perfonned using EEA1 as an endocytic marker or using the PI3P-binding protein GST-2X FYVE^mstransfected in hepatocytes. Confocal microscopic analysis revealed bigger and distorted endosomes in the pi3K-C2p^D1212A/m212A hepatocytes (Fig. 9A).Fig 9B shows a decrease of the POP levels in the ΡΙ3Κ-02β ^{DI212AyD1212A} hepatocytes after starvation. This endocytic phenotype and the decrease of the PI3P levels observed in the PI3K-C2P ^{D1212A D1212A} hepatocytes is a sign of a trafficking defect, suggesting that insulin receptor internalisation may be affected and delayed leading to a prolongation of the insulin signal.

Example 12: Investigating PI3P levels in ΡΙΚ3€2β primary hepatocytes

Mass assay on primary hepatocytes from PI3K-C2p^{D12,2A/D12,2A}and wild type mice was performed according to Chicanneei al., (Biochem J. 2012 Oct 1 ;447( 1): 17-23) to measure PI3P levels. A reduction of PI3P levels of approximately 60% was observed in PIK3C2 knock in primary hepatocytes (Figure 17).

Example 13: Investigating Akt signalling

A Western blot was conducted of cell lysates isolated from liver, epididymal (white) adipose tissue (WAT), spleen and muscle after insulin administration (0.75U/kg by i.p.) or from primary hepatocytes (lower panel) stimulated with ΙΟΟηΜ insulin over a time course. The results show that Akt signalling is enhanced in PIK3C2pinsulin responsive tissues and primary hepatocytes but not in the spleen (Figure 18).

A similar study was conducted in PIK3C2pmice subjected to high fat diet. The data shows an increase of Akt signalling after insulin stimulation in the ΡΓΚ302β knock in mice compared to wild type controls (Figure 19). The liver triglyceride content was significantly reduced in the PlK3C2 knock in livers compared to wild type. Thus enhanced Akt signalling is maintained in PIK3C2 mice subjected to high fat diet. Experimental Procedures for Examples 1-13

Mice

Mice were maintained at 22°C with a 12-hour dark, 12-hour light schedule with free access to water and housed in individually-ventilated cages and cared for according to United Kingdom Animals (Scientific Procedures) Act (1986). The mice used were on mixed C57BL/6- C57BL/6NT background. Animals were fed either a normal chow diet (20% protein, 75% carbohydrate, 5% fat) or high fat diet (D 12492; 20% protein, 20% carbohydrate and 60% fat) for 16 weeks. Metabolic and Physiological Methods

Serum levels of insulin, leptin, adiponectin and free fatty acids were measured using ELISA kits (CrystaChem Inc., for insulin and Millipore for the others). Glucose tolerance tests were performed by ip injection of 2g glucose/kg after an overnight fast and 4h fasting for the HFD fed mice. Insulin tolerance tests (ITTs) were performed by ip injection of 0.75U/kg of recombinant human insulin after an overnight fast and 4h fasting for the HFD fed mice. The pyruvate challenge (PTT) was performed by injecting 2 g/kg of pyruvate (Sigma-Aldrich) ip after an 18 h fast. Mice have been injected with insulin (lU/kg) by intraperitoneal injection.

Lipid extraction and analysis

MEFs were labeled with 0.6 mCi/ml [³²P]orthophosphate during 8h in a phosphate-free N-2- hydroxyethylpiperazine-N -2-ethanesulfonic acid-Tyrode buffer (pH 6.5) at 37°C. After insulin stimulation (ImM), reaction was stopped by addition of chloroform/methanol (vol/vol) containing 0.6 N HC1, and lipids were immediately extracted and analyzed by a combination of thin-layer chromatography and high-performance liquid chromatography (HPLC) as described previously (Gratacap MP J Biol Chem. 1998 Sep 18;273(38):24314-21.

Western blot analysis Tissues was lysed in 20 mMTris-HCl (pH 8,0), 5% glycerol, 138mMNaCl, 2.7 mMKCl, 1% NP-40, 20mMNaF, 5mM EDTA, I mM sodium orthovanadate, 20 μΜ leupeptin, 18 μΜ pepstatin, 4 μg/ml aprotinin, 1 mM DTT. To remove cell debris, homogenates were spun at 13000 rpm for 10 min at 4°C. Protein concentration was determined by colorimetric assay (Bradford assay, Biorad). Protein extract was resolved by SDS-PAGE before transfer onto PVDF membrane and incubation over-night at 4°C with the following antibodies : anti- phospho-ser 473 Akt antibody (Cell SignalingTechnology). PDK-C2a pl70 and ΡΙ3Κ-02β antibodies was from BD Biosciences (BD #61 1046 and 61 1342, respectively); anti ρΐ ΐ θβ and pl l06were from Santa Cruz (SC #602 and #7176, respectively); anti pi 1 Oct and vps34 antibody were from Cell Signalling Technology (CST #4249 and #9145, respectively). Aiiti- phospho-Thr389 p70S6K and anti phospho-ser240/244 S6 antibodies were from CST. Antigen-specific binding of antibodies was visualised using an ECL detection system.

Histology

For tissue sections, hematoxylin and eosin (H&E) staining was performed on 5 um paraffin sections of tissues fixed overnight in 4% phosphate-buffered paraformaldehyde at 4 °C.For staining of neutral lipids, cryosections were fixed with 4% PFA in PBS at room temperature for 15 min andwashed again with PBS and stained with Oil Red O (0.5% w/isopropanol, diluted 3 :2 in PBS) for 1 h at room temperature. Stained sections were rinsed in 60% isopropanol, followed by deionized water and mounted in Vectashield.

Morphometric analysis

To examine the size of the white adipocytes the area of each adipocyte was measured in at least 300 cells of representative sections per fat pad (epididymal WAT) per mouse (4 per genotype) was determined using Image J software (NIH). Then mean value was designated as in index of the cell size.

Lipid Kinase Assays

PI3K assays using Ptdlns as a lipid substrate were performed as previously described (Chaussade et al.Biochem. J. (2007) 404 (449^-58).

Statistical Analysis

All data unless otherwise indicated are shown as mean values ± SEM. data sets were compared for statistical significance using the two-tailed Students t test. All statistical analyses were generated using excel sofware. -values < 0.05 were considered to be statistically significant (designated by a single asterisk; double asterisk, P <0.01, triple asterisk, P <0.001). The number of animals in each group is indicated by n. Example 14 - In v/frobiochemical assays for Vps34 inhibitors

Vps34 activity is analysed/?? vitroby measuring the transfer of the gamma phosphate of ³²P- labelled ATP to the phosphatidyinositol (Ptdlns or PI) lipid followed by binding of the product to nitrocellulose membranes or thin layer chromatography plates. The capacity of test agents to modulate this activity is analysed/?? vitro.

High-throughput fluorescence polarization protocol is also used to test the effect of test agents on Vps34 kinase activity.

Example 15 - Cell-based assays for Vps34 inhibitors

Vps34 inactivation impacts on endocytosis and autophagy, which can be monitored by commercially cell-based imaging assays (for example from Biotek and Millipore). The capacity of test agents to modulate endocytosis and/or autophagy is analysed using such a cell- based assay.

An inducible vps34 cell-based system is also created, whereby expression of active vps34 is expected to induce cell death, under certain conditions. Putative Vps34 inhibitors may then be screened for their capacity to rescue cell lethality.

Stable cell lines that are homozygous for kinase-dead Vps34 are also created, in order to probe for cellular off-target effects of Vps34-selective inhibitors. Assays to probe cell-based functions of class I and III PI3 isoforms are known in the art. Example 16 - Rodent-based assays for Vps34 inhibitors.

The time and dose-dependent impact of the vps34 inhibitors is tested on blood glucose levels under fasting conditions in mice. This is complemented by glucose and insulin tolerance assays (GTT and ITT), followed by metabolic analysis (measurement of lipoproteins and adipokines - leptin, adiponectin, resistin- and analysis of the glycaemia control with a euglycaemic-hyperinsulinaemic clamps) to further assess improved insulin sensitivity and determine potential side-effects.

A mass assay is set up to determine the tissue levels of the Vps34 lipid product PtdIns3P in primary tissues in vivo. This provides a direct readout of target inhibition in vivo. P-proteomics technology allows label-free quantitation of phosphorylation in primary tissues. Primary hepatocytes and liver may be to develop robust in vivo vps34-dependent phosphorylation response/biomarkers. The effect of test agents on vps34-dependent cellular phosphorylation may then be assessed.

In vivo metabolic assays are also conducted in rodents. Vps34 inhibition is expected to reduce blood glucose levels under starvation and under high-fat diet conditions, and alter various disease markers. The impact of vps34 inhibition is also tested in more advanced models of insulin-resistance related disease in rodents, including (1) protection from high fat-diet- induced liver steatosis; (2) improvement of metabolic disease parameters in insulin-resistant mouse models, such as ob/ob or db/db mice, andlipodistrophy mouse models. Characterisation involves measurement of key hormones and blood parameters (such as lipoproteins and adipokines (leptin, adiponectin, resistin) and analysis of the glycaemia control with a euglycaemic-hyperinsulinaemic clamps.

All publications mentioned in the above specification are herein incorporated by reference. Various modifications and variations of the described methods and system of the invention will be apparent to those skilled in the art without departing from the scope and spirit of the invention. Although the invention has been described in connection with specific preferred embodiments, it should be understood that the invention as claimed should not be unduly limited to such specific embodiments. Indeed, various modifications of the described modes for carrying out the invention which are obvious to those skilled in molecular biology or related fields are intended to be within the scope of the following claims.

Claims

1. A method for identifying agents useful in the treatment and/or prevention of a disease associated with insulin resistance and/or glucose intolerance which comprises the step of investigating the capacity of a test agent to inhibit the Vps34 signalling pathway and/or the PI3K-C2P signalling pathway.

2. A method according to claim 1, which comprises the step of investigating whether a test agent inhibits the kinase activity of Vps34 and/or PI3K-C2 .

3. A method according to claim 1 or 2 which is conducted in vitro or in a cell in culture.

4. A method according to claim 1 or 2 for screening for compounds capable of increasing insulin sensitivity in a subject.

5. A method according to any preceding claim for screening for compounds capable of improving glucose tolerance in a subject.

6. A method according to any preceding claim, wherein adiponectin is used as a biomarker for agents capable of inhibiting the Vps34 signalling pathway and/or the ΡΙ3Κ-02β signalling pathway.

7. A transgenic non-human animal which comprises a mutation in the gene encoding Vps34 such that the active site of Vps34 is inactivated.

8. A transgenic non-human animal according to claim 7, which comprises a mutation in the DFG motif of the ATP-binding site.

9. A transgenic non-human animal according to claim 8, which comprises the mutation D761A.

10. A transgenic non-human animal which comprises a mutation in the gene encoding PI3K-C2P such that the active site of PI3K-C2p is inactivated.

11. A transgenic non-human animal according to claim 10, which comprises a mutation in the DFG motif of the ATP-binding site. A transgenic non-human animal according to claim 11, which comprises the mutation