WO2008009869A1

WO2008009869A1 - Prediction of disease using lipopolysaccharide assay

Info

Publication number: WO2008009869A1
Application number: PCT/GB2006/002714
Authority: WO
Inventors: Philip Mcternan; Sudhesh Kumar
Original assignee: University Of Warwick
Priority date: 2006-07-20
Filing date: 2006-07-20
Publication date: 2008-01-24

Abstract

The progression of metabolic and cardiovascular diseases in predisposed individuals is monitored by means of measuring levels of lipopolysaccharide (LPS) in tissue samples. A blood test for lipopolysaccharide provides a means of monitoring the progression, or of diagnosing the onset of, insulin resistance, obesity, non-alcoholic fatty liver disease, atherosclerosis, coronary heart disease and type 2 diabetes, particularly in obese individuals. The screening of active agents against metabolic and cardiovascular diseases is achieved by an assay involving expression of toll-like receptors and the monitoring of the innate immune response in adipose tissues and cells.

Description

Prediction of disease using lipopolysaccharide assay

Field of the Invention

The invention relates to the field of metabolic and cardiovascular diseases, in particular the so-called metabolic syndrome, hyperinsulinaemia, insulin resistance, obesity, non-alcoholic fatty liver disease (NAFLD), atherosclerosis, coronary heart disease and type 2 diabetes (T2DM). The invention also concerns methods of predicting the degree of risk in onset and development of diabetes in susceptible patient groups, in particular obese patients, and also methods of treating such individuals. The invention also concerns methods of screening active agents to treat such metabolic and cardiovascular disease.

Background of the Invention

Diabetes mellitus is an increasing problem among Western societies and those adopting a Western lifestyle. Type 2 diabetes, in particular, is associated with poor diet, lack of exercise and predisposing genetic background. By 2003 the United States had 13.8 million diagnosed diabetics, with an estimated 5 million undiagnosed and a further 41 million pre-diabetic individuals with evidence of increasing insulin resistance (Bloomgarten, 2006, Diabetes Care 29:161-167). Type 2 diabetes accounts for roughly 90% of total prevalence.

'Metabolic syndrome' is a term used to describe the combination of risk factors for type 2 diabetes and cardiovascular disease (Reaven, 1988, Diabetes 37: 1595). It includes obesity (especially abdominal obesity - also known as central or visceral obesity) insulin resistance, impaired glucose metabolism, dyslipidaemia of the high triglyceride/low HDL cholesterol type and hypertension. It is estimated that as many as a quarter of the world's adult population suffer from metabolic syndrome, which carries with it three-fold increased risk of stroke or heart attack and a five-fold increased risk of developing type 2 diabetes (Dunstan etal, 2002, Diabetes Care 27: 2676; lsomaa etal, 2001, Diabetes Care 24: 683). Metabolic syndrome is linked to genetic factors, ageing, physical inactivity, dietary factors and is strongly linked to obesity and insulin resistance.

Obesity is known to represent one of the most important risk factors for the increased risk of type 2 diabetes and cardiovascular disease (Field et al, 2001 , Arch Intern Med 161:1581). In addition, an increase in central (visceral) adiposity confers higher metabolic risk. This increased metabolic risk is associated with sub-clinical inflammation, with several studies demonstrating increased proinflammatory adipocytokines in patients with obesity and type 2 diabetes (Pickup et al, 1997, Diabetologia 40: 1286; Pickup et al, 2000, Life Sci 67: 291). It is also known that adipose tissue secreted products such as plasminogen activator inhibitor type 1 (PAI-1), IL-6 and TNF-α are mediators of sub-clinical inflammation and potential cardiovascular risk through activation of NFKB, a key transcriptional factor in the inflammatory cascade (Pickup et al , 1997, Diabetologia 40: 1286; Festa etal, 2002, Diabetes 51:1131; Kern etal, 2002, Am J Physiol Endocrinol Metab 280: E745-51; Vozarova et al, 2001, Obes Res 9:414; Senn etal, 2002, Diabetes 51: 3391; Ahmad et al , 1997, J Cell Biochem 64:117; Uysal etal , 1997, Nature 389: 610; Collart ef a/ , 1990, MoI Cell Biol 10:1498; Baltimore etal , 1990, MoI Cell Biol 10:2327).

In contrast to the pro-inflammatory adipoeytokines, adipocytes also secrete adiponectin, which has been shown to possess anti-inflammatory properties through its action on NFKB, and is inversely correlated with obesity and diabetes (Tomas etal, 2002, Proc Natl Acad Sci USA 99:16309; Yamauchi et al, 2002, Nat Med 8: 1288; Ouchi et al , 2001 , Circulation 103: 1057; Ouchi et al, 2000, Circulation 102:1296; Yamauchi et al, 2002, J Biol Chem 278:2461).

A potential pathway for the induction of sub-clinical inflammation in obesity and type 2 diabetes involves activation of the innate immune pathway, which represents the first site of defence against infection (Medzhitov & Janeway, 1998, Curr Opin Immunol 10:12). Unlike the adaptive immune response, which depends on antigen-specific responses mediated by lymphocytes, the innate pathway has the capacity to recognise a wide range of pathogens based on the binding of common bacterial and viral structural motifs or pathogen-associated molecular patterns (PAMPs). Central to this system is the Toll-like receptor (TLR) family of receptors expressed on a variety of cells. TLRs contribute to multi-subunit receptors recognising antigens including LPS, viral CpG sequences and zymosan, which allow rapid reaction to infection (Kaisho et al, 2002, Biochim Biophys Acta 1589:1).

In bacteraemia from Gram-negative gut organisms such as Escherichia coli, Klebsiella, Proteus or Pseudomonas the immediate innate immune response is largely to lipopolysaccharide (LPS) and othercomponents derived from bacterial cell walls. Circulating LPS and, in particular, its constituent lipid A, provokes a wide range of systemic reactions. It is probably contact with Kupffer cells in the liver that first leads to IL-1 release and the onset of pyrexia. Activation of circulating monocytes and macrophages leads to release of cytokines such as IL- 6, IL-12, IL-15, IL-18, TNF-α, macrophage migration inhibitory factor (MlF), and cytokine-like molecules such as high mobility group B1 (HMGB1), which, in turn activate neutrophils, lymphocytes and vascular endothelium, up-regulate cell adhesion molecules, and induce prostaglandins, nitric oxide synthase and acute- phase proteins. Release of platelet activating factor (PAF), prostaglandins, leukotrienes and thromboxane activates vascular endothelium, regulates vascular tone and activates the extrinsic coagulation cascade. Dysregulation of these responses results in the complications of sepsis and septic shock in terms of peripheral vasodilation leading to hypotension, and abnormal clotting and fibrinolysis producing thrombosis and intravascular coagulation (Cohen, 2002, Nature 420: 885-891).

The best characterised high-affinity LPS receptor on myeloid cells comprises the LPS receptor CD 14, together with TLR4 and the leucine-repeat rich molecule

MD-2 (Palsson-McDermott & O'Neill, 2004, Immunology, 113: 153; Triantafilou et a/, 2004, Biochem J 381: 208). However, another TLR, TLR2, also appears to be involved in a functional LPS receptor on, for instance, pancreatic islet β cells (Vives-Pi et al, 2003, Clin Exp Immunol 133: 208). Activation of TLRs leads to translocation of NFKB to the nucleus and transcription of IL-6, IL-1 and TNF-α to initiate an acute phase response ( Muzio et al , 2000, Biochem Soc Trans

28:563). To date TLR 2 and 4 remain the most studied of these receptors, and have been found to be expressed in 3T3-L1 adipocytes, with LPS being shown to stimulate TLR 2 expression and induce IL-6 and TNF-α from these cells (Lin et al, 2000, J Biol Chem 275: 24255).

CD14 is a glycosyl phosphatidylinositol (GPI) -anchored membrane protein capable of binding LPS complexed with the acute phase serum glycoprotein, LPS-binding protein (LBP). However, like many GPI-anchored proteins, a soluble form of CD14 exists, which appears to be able to contribute to functional LPS receptors on cells that do not, themselves, express CD14 (Ulevitch &

Tobias, 1995, Annu Rev Immunol 13: 437). The presence of at least sCD14 has been shown to be required for human TLR2-mediated responsiveness to LPS (Kirschning et al, 1998, J Exp Med 188: 2091). Diagnostic assays of CD14 and its function have been proposed as a measure of sepsis or systemic inflammatory response syndrome (SIRS). German patent application DE

0027308 discloses a method of measuring cell-surface CD14 receptor clusters comprising the integrin subunit CD11b for the diagnosis of systemic inflammation including arteriosclerosis. European patent application EP 1571160 and Japanese application JP 2005/106694 propose determining the concentration of soluble CD14 isoforms by use of specific antibodies as an indicator of sepsis.

Similarly, it has been proposed that assays of various LPS-binding proteins be used as indicators of systemic infection, sepsis or SIRS. WO 94/21280 describes the use of an assay for neutrophil bactericidal/ permeability increasing protein (BPI), a protein with 44% identity with LBP and which also binds the lipid A portion of LPS. US 5,618,675 describes a method blocking LPS-meditated activation of the inflammatory response by use of CAP18 protein (obtained from mammalian granulocytes) or peptides derived therefrom. CAP18 is capable of binding LPS and so compete with functional LPS receptors.

Although the main function of TLRs is to form the basis of the innate immune response to infection, there is increasing evidence of their involvement in inappropriate inflammatory reactions contributing to a number of disease states. For instance, CD14 and TLR4 are implicated in the development of atherosclerosis (Arroyo-Espliguero et al, 2006, Heart 90: 983) and inflammatory bowel disease. Blocking of TLR4 has a beneficial effect in murine models of inflammatory bowel disease (Fort et al, 2005, J Immunol 174: 6416). However, attempts to establish a link between TLR4 polymorphisms and diabetes or metabolic syndrome have been unsuccessful (lllig et al, 2003, Diabetes 52: 2861).

Studies have shown that diabetic individuals and animal models may be more sensitive to the effects of LPS (Song et al, 2003, Cardiovasc Toxicol 3: 363; Plesner et al, 2002, Scand J Immunol 56: 522), although in these cases type 1 diabetes was the model. It has been suggested that high levels of glucose itself might have pro-inflammatory effects (Das, 2002, Critical Care 6: 389). It is also known that LPS is able to stimulate IL-6 production by adipocytes and this may be blocked by thizolidinediones such as pioglitazone (Yamaguchi et al, 2005, J Dent Res 84: 240). It has been suggested that this may be of relevance in the observed correlation between periodontal disease (which provides a means for LPS to enter the circulation) and coronary heart disease. Periodontal disease is also associated with diabetes and it has been suggested that thiazolidinedione might suppress this by an action on diabetic macrophages, which over-produce TNF-α and have up-regulated expression of activation markers such as CD14 and CD18 (Salvi et al, 1997, J Clin Periodontol 24: 8; Fogelstrand et al, 2004, Diabetologia 47: 1948). The inter-relationship between chronic exposure to bacterial products such as LPS and disease states such as atherosclerosis is poorly understood and complex.

It is known that the prevalence of metabolic syndrome, type 2 diabetes and cardiovascular disease is disproportionately high in some ethnic groups. In particular, individuals of South Asian or Indian ethnic origin carry a greatly increased risk of atherosclerosis and type 2 diabetes, with British Asians having a 40% increased mortality as compared with the European white population (Cappuccio et al, 1997, Heart 78:555).

It is also known that the majority of patients with type 2 diabetes mellitus also have fatty liver (steatosis), and it has recently been appreciated that steatosis may progress through necroinflammation - so called non-alcoholic steatohepatitis (NASH) - to fat with fibrosis and ultimately to cirrhosis, liver failure and even hepatocellular carcinoma in a significant proportion of patients (Day, 2002, Gut 5:585). Moreover in obese patients with steatosis the presence of type 2 diabetes has been consistently associated with a higher risk of advanced fatty liver disease and non-alcoholic fatty liver disease (NAFLD) is a significant cause of morbidity and mortality in patients with type 2 diabetes (Erbey et al , 2000, American Journal of Medicine 109: 588; De Marco et al, 1999, Diabetes Care 22:756).

In addition, both steatosis and type 2 diabetes are associated with insulin resistance/hyperinsulinaemia and chronic sub-clinical inflammation (Yki-Jarvinen and Westerbacka, 2005, Curr MoI Med 5:287). The role of the gastrointestinal tract as a source of endotoxin and sub-clinical inflammation in chronic diseases such as fatty liver disease and type 2 diabetes has received limited attention to date, with portal vein circulating endotoxin levels implicated as a consequence of disease rather than the potential mediator. Given the growing significance of type 2 diabetes and the long sub-clinical prediabetic period associated with it, attempts have been made to identify reliable diagnostic markers. One of the defining features of the pre-diabetic period is the development of impaired glucose tolerance and insulin resistance, that is, the failure of target tissues to respond to elevated levels of insulin by effectively lowering blood glucose levels. Current knowledge suggests that development of glucose intolerance or diabetes is initiated by insulin resistance and is worsened by the compensatory hyperinsulinaemia. However, insulin resistance is difficult to measure directly (by the complex and invasive 'hyperinsulinaemic euglycaemic clamp' method) and is usually inferred from measuring fasting glucose levels and glucose tolerance testing. A one of a number of variations on this approach is disclosed by WO 2005/017532, which describes oral administration of a controlled quantity of glycaemic carbohydrate of known composition.

A number of markers for frank diabetes or pre-diabetic insulin resistance have been developed as an alternative. It is known that chronic hyperglycaemia leads to abnormal non-enzymatic glycosylation ('glycation') of serum proteins such as albumin. US2006121532 discloses the use of glycated insulin as a biomarker for diabetes. US 5,183,764 discloses an indirect test for insulin resistance based on a quantitative assay for cΛ/ro-inositol and reports an inverse relationship between serum levels of this and insulin resistance.

There are reported correlations between the development of type 2 diabetes and serum C-reactive protein and IL-6 (Pradham et al, 2001, JAMA 286: 2233); IL-18 (Thorand et al, 2005, Diabetes 54: 2932); and markers of endothelial dysfunction, such as intercellular adhesion molecule-1 (ICAM-1), vascular cell adhesion molecule-1 (VCAM-1) (Meigs et al, 2004, JAMA 291: 1978) and E-selectin (Wexler et al, 2005, Obesity Research 1_3: 1772). It is also known that adipose tissue secreted products such as plasminogen activator inhibitor type 1 (PAI-1) and TNF-α are mediators of sub-clinical inflammation and potential cardiovascular risk through activation of NFKB, a key transcriptional factor in the inflammatory cascade, as is IL-6 ((Pickup et al , 1997, Diabetologia 40: 1286; Festa et al, 2002, Diabetes 51; 1131; Kern et al, 2002, Am J Physiol Endocrinol Metab 280: E745-51 ; Vozarova et al, 2001 , Obes Res 9:414; Senn et al, 2002, Diabetes 51: 3391; Ahmad etal , 1997, J Cell Biochem 64:117; Uysal et al , 1997, Nature 389: 610; Collart et al , 1990, MoI Cell Biol 10:1498; Baltimore et al , 1990, MoI Cell Biol 10:2327). Many of these markers are, however, greatly influenced by transient minor infections.

There is a known inverse correlation between adiponectin levels and type 2 diabetes (Tshritter et al, 2003, Diabetes 52: 239), although this is complicated by the different signalling responses that appear to be triggered by trimers of adiponectin and higher-order multimers (Hug and Lodish, 2005, Curr Opin

Pharmacol 2005, 5:129). However, current assays do not reliably distinguish between these different forms

Serum levels of retinol-binding protein 4 (RBP4) has recently been shown to correlate with insulin resistance (Graham et al, 2006, N Eng J Med 354:2596).

WO 2005/059564 discloses methods of diagnosing insulin resistance by detecting modulation of RBP4 activity.

US 2005/0244892 discloses the monitoring of serum levels of resistin, a product of human monocytes and macrophages and, to a lesser extent, adipocytes.

Levels of resistin correlate with chronic inflammation and may be used as a biomarker for cardiovascular disease.

WO 2005/028509 discloses the use of a cross-reacting anti-TLR2 antibody to block TLR-2-mediated immune ceil interaction, particularly in the case of septic shock.

Assays for the measurement of LPS in biological samples such as serum or whole blood are well-known. Traditionally, the Limulus amoebocyte lysate (LAL) assay has been used to detect endotoxin, but alternative immunoassays and agglutination assays such as that described by Rylatt et al (1995, Prog Clin Biol Res 392: 273) and WO 92/01228 are within ordinary skill in the art.

WO 00/53165 discloses the blocking of the production and absorption of LPS from the gut in individuals suffering from cachexia and wasting syndromes. Many chronic diseases are linked to cachexia and weight loss linked to an inflammatory response and this application proposes blocking LPS from the gut as a source of this inflammation. Among the methods disclosed include administration of bile acids, but also blocking the action of LPS by anti-LPS antibodies, LPS binding protein, soluble CD14 and TLR 2- and TLR4-blocking agents. The method is suggested as suitable for those suffering from cachexia, including cachexia resulting from diabetes.

Thiozolidinediones are a class of insulin-sensitising compounds, commonly used as an adjunct treatment for the insulin resistance characteristic of type 2 diabetes. Thiozolidinediones act by binding to nuclear peroxisome proliferator- activated receptors (PPARs), primarily PPARy, which are normally activated by free fatty acids and eicosanoids. Methods of treatment of diabetes-related conditions with such compounds are well-known in the art (see, for instance, EP 1671637).

It has been proposed that LPS derived from commensal gut bacteria might contribute to sub-clinical inflammation and perhaps, in part, to the pathogenesis of type 2 diabetes, with elevated levels of LPS being demonstrated in type 2 diabetes patients (McTeman et al, 21 January 2006, Keystone Conference Symposia Proceedings. Diabetes Mellitus and the Control of Cellular Energy Metabolism. Abstract No. 235: Sub-clinical Inflammation in Type 2 Diabetes: Effect of Endotoxemia on Initiating the Inflammatory Cascade in Human Adipose Tissue. pp168). At present, it is still not possible for physicians to reliably identify even among obese individuals, those who may be most at risk of developing insulin resistance, type 2 diabetes, coronary heart disease, non-alcoholic fatty liver disease, or other metabolic and cardiovascular diseases. There is a need for a simple blood test to detect insulin resistance indicating the individual has a high risk of a metabolic or cardiovascular disorder. The current methods for testing are either unreliable or time-consuming and therefore at present obese patients are not routinely tested. There is a clear need for a quick and reliable test to provide an early indication of those patients at risk so that that the disease can be prevented or its effects greatly reduced. This would also reduce the burden on hospitals, saving millions of pounds a year in health care costs.

In addition, obese individuals who have hyperlipidaemia, hypertension or diabetes are currently treated specifically for these conditions; there are no broader treatments that try and target the underlying cause. Although some medication for lowering blood pressure and some treatments for hyperlipaemia (statins) may have the incidental effects of reducing inflammation, there is a need for improved treatment strategies for the underlying causes of these conditions.

Summary of the Invention

Accordingly, the present invention provides a method of determining the risk of metabolic or cardiovascular condition or disease in an individual comprising the steps of: (a) obtaining a biological sample from the individual, and

(b) determining the concentration or amount of LPS in the sample wherein the concentration or amount of LPS in the sample is indicative of the risk of the individual developing said metabolic or cardiovascular condition or , disease. The invention also provides a method for monitoring an individual for the onset or stage of metabolic or cardiovascular condition or disease, comprising the steps of:

(a) obtaining a biological sample from the individual, and (b) determining the concentration or amount of LPS in the sample, wherein the concentration or amount of LPS in the sample is indicative of the onset or stage of metabolic or cardiovascular condition or disease in the individual.

An elevated concentration or amount of LPS may usually be indicative of the onset or progression of said metabolic or cardiovascular condition or disease, the concentration or amount of LPS being measured:

(a) by reference to a standard value, or

(b) by reference to at least one previous measurement from the same individual.

The standard value may be one which has previously been established through clinical studies. The value may be for patients in general regardless of sex, age, BMI or ethnic origin. For example, for non-diabetic individuals in a population, the mean serum level of LPS is about 3.1 +/- 1.7 EU ml^"1.

When assessing onset, progression or stage of disease in an individual on the basis of LPS measurements over time for that individual, a significant increase over a baseline average correlates with onset of a metabolic or cardiovascular disease. Similarly, a significant decrease in LPS measurements from a baseline average in patients with disease correlates with amelioration of the disease. Such information may be useful in tailoring the treatment regimes for patients with metabolic or cardiovascular disease. When assessing the risk of an individual for a particular disease, the level of LPS in the sample from the individual may be combined with other data taken from the individual. Such data may include one or more of gender, age, fat mass, body mass index (BMI) and/or ethnicity. The quantitative relationships between such additional patient parameters and the relative risk of developing the various metabolic or cardiovascular diseases will be familiar to one of skill in the art.

In terms of the ethnic origin of individuals, some are more susceptible to metabolic or cardiovascular disease than others.

It is also preferred that the individual selected for testing is obese or alternatively or additionally is a member of a defined ethnic origin, for example South Asian, Indian, Black African or Caucasian. Most preferably, the subject is of South Asian or Indian origin.

Where elevated concentrations of LPS are detected by the methods of the invention then onset of T2DM in any given individual is indicated by a mean serum level in the range of 3.1 to 5.5 EU ml^"1.

An individual who has developed T2DM is indicated by an LPS serum level of at least 5.5 +/- 1.6 EU mI^"1.

It is also preferred that the biological sample used for the method is blood or plasma, preferably serum.

The invention also provides a method of diagnosing a metabolic or cardiovascular condition or disease in an individual comprising the steps of

(a) obtaining a sample from the individual

(b) determining the concentration or amount of LPS in the sample, wherein the concentration or amount of LPS in the sample indicates that the individual has the metabolic or cardiovascular disease. The metabolic or cardiovascular condition or disease may be one selected from obesity, polycystic ovary syndrome (PCOS), non-alcoholic fatty liver disease (NAFLD), atherosclerosis, coronary artery disease, metabolic syndrome, hyperinsulinaemia, insulin resistance and type 2 diabetes mellitus (T2DM).

More preferably, the metabolic condition or disease is type 2 diabetes mellitus (T2DM), non-alcoholic fatty liver degeneration or insulin resistance.

An elevated concentration or amount of LPS may indicate that the individual has the metabolic or cardiovascular disease or condition. In preferred embodiments it is the elevated concentration or amount of LPS which is determined:

(a) by reference to a standard value, or

(b) by reference to at least one previous measurement from the same individual.

An LPS serum value of 4 to 20 EU ml^"1, and more preferably 5 to 15 EU ml^"1 may be indicative of the presence of type 2 diabetes mellitus (T2DM).

The subject of any of the methods of the invention is preferably a member of a high risk group, more preferably the subject is obese or of a defined ethnic origin, for example South Asian or Indian, Black African or Caucasian. More preferably they are of South Asian or Indian origin. When employing methods of the invention for assessment of risk, onset or stage of cardiovascular disease, such disease is indicated by an LPS serum level of about 5.5 EU ml^"1 and above.

When employing methods of the invention for assessment of risk, onset or stage of non-alcoholic fatty liver disease, this is indicated by an LPS serum level of about 5.5 EU ml^"1 and above. The presence of NAFLD in an individual is indicated by LPS serum levels in the range 5.5 - 12 EU ml^"1. The invention also provides the use of LPS as a marker in an assay for determining the risk of metabolic or cardiovascular condition or disease in an individual, wherein the assay is carried out on a biological sample from the individual as hereinbefore described.

Alternatively, LPS is used as a marker in an assay for monitoring an individual for the onset or stage of metabolic or cardiovascular condition or disease, wherein the assay is carried out on a biological sample from the individual as hereinbefore described.

The invention further provides a method of treating an individual at risk of developing a metabolic or cardiovascular disease or condition comprising

(a) obtaining a biological sample from the individual,

(b) determining the concentration or amount of LPS in the sample, wherein the concentration or amount of LPS in the sample is indicative of the risk of the individual developing said metabolic or cardiovascular condition or disease, and

(c) administering to the individual an anti-inflammatory agent.

The anti-inflammatory agent may be a thiazolidinedione or a pharmaceutically acceptable salt thereof, preferably rosiglitazone.

Alternatively, the anti-inflammatory agent may be an anti-LPS antibody, an anti- CD 14 antibody, an anti-TLR 2 antibody or a TLR 2 antagonist.

The invention further provides a method of identifying an agent for the prevention or treatment of metabolic or cardiovascular disease comprising:

(a) administering endotoxin (LPS) to an animal, adipose tissue or adipocyte, (b) administering a candidate agent to said test animal, adipose tissue or adipocyte, (c) determining the level of expression of at least one toll-like receptor in adipose tissue taken from the animal, the adipose tissue, or adipocytes, whereby a decreased expression of at least one Toll-like receptor (TLR) identifies a said agent.

The at least one Toll-like receptor may be Toll-like receptor 2 (TLR 2) and/or Toll- like receptor 4 (TLR 4).

The invention also provides a method of identifying an agent for the prevention or treatment of metabolic or cardiovascular disease comprising:

(a) administering endotoxin (LPS) to an animal, adipose tissue or adipocyte

(b) administering a candidate agent to said test animal, adipose tissue or adipocyte, (c) monitoring the innate immune pathway, whereby a decreased activation of the innate immune pathway identifies a said agent.

The monitoring of the innate immune pathway may comprise monitoring of the level of (a) pro-inflammatory cytokines and/or (b) adiponectin, or the expression of (a) pro-inflammatory cytokines and/or (b) adiponectin, whereby decreased levels or expression of pro-inflammatory cytokines and/or increased levels or expression of adiponectin identifies a said agent.

The pro-inflammatory cytokines are preferably adipocytokines. Alternatively, the pro-inflammatory cytokines or adipocytokines may be selected from one or more of lL-1. IL.-6, or TNFα.

The innate immune pathway may be monitored or additionally monitored by determining the level or expression of CD14, soluble CD14 and/or plasminogen activator inhibitor type 1 (PAI-1). Alternatively, the immune pathway may be monitored or additionally monitored by determining activation of N FκBand/or translocation of NFKB to the nucleus.

Preferably said animal is a rodent, preferably a rat or a mouse.

The adipose tissue may be subcutaneous abdominal adipose tissue, preferably human subcutaneous abdominal adipose tissue.

When employed, the adipocyte is preferably a cultured adipocyte, more preferably a mature human adipocyte or a human subcutaneous abdominal adipocyte.

The various methods of the invention provides various technical advantages over the state of the art. The invention permits a simple test for indicating the risk, onset or progression of a variety of metabolic and cardiovascular diseases, including insulin resistance, type 2 diabetes, atherosclerosis, coronary heart disease, obesity and non-alcoholic fatty liver disease. Hitherto it has been difficult to identify patients at risk before they develop clinical symptoms. The methods of the invention, also when combined with other routine examination and measurement of patients, e.g. height, weight, BMI, fat mass, gender, age and ethnicity, for example, permit earlier diagnoses of nascent medical conditions than has hitherto been possible.

The invention is predicated in part by a discovery and realisation of the inventors which is that the metabolic and cardiovascular diseases are linked by low-grade chronic inflammation related to circulating bacterial lipopolysaccharide (LPS, endotoxin) and that measuring the levels of LPS provides a quantitative biomarker for the risk of an individual developing such disease states, as well as the onset and progression (stage) of these diseases. In particular, the invention is useful as a simple means of indicating individuals developing asymptomatic but progressive insulin resistance that often precedes frank type 2 diabetes, allowing therapeutic intervention at an early stage to reduce the risk of developing diabetes or cardiovascular disease in this high risk group. The finding of raised serum LPS in an individual is closely linked to diseases related to the so-called 'metabolic syndrome'. A finding of raised serum LPS is therefore useful in itself and, when taken together with the medical history, risk profile and other relevant biochemical and physical measures well-known in the art, offers an early diagnosis of hyperinsulinaemia, insulin resistance, type 2 diabetes, cardiovascular disease including coronary heart disease and atherosclerosis, and inflammatory liver disease such as non-alcoholic fatty liver disease (NAFLD), including its progressive stages of steatosis, steatohepatitis (non-alcoholic steatohepatitis - NASH), fibrosis and cirrhosis.

The inventors have shown a close relationship between circulating LPS and insulin levels; patients with increased insulin levels also show an increase in LPS levels. Furthermore they show that LPS works through the cell-surface receptor TLR-2 to initiate a pro-inflammatory response from adipose tissue.

Conversely, a negative or negligible level of circulating LPS may be regarded as a 'well-being' marker in an otherwise high-risk individual.

Obese individuals are particularly at risk from this spectrum of diseases and so constitute a important target group for testing. Obesity of the central, visceral or abdominal type, characterised by a raised body mass index (BMI) but a disproportionately increased waist measurement, is a notable risk factor. Assessment of obesity and significant waist measurement is dependent on an overall clinical judgement, but the American Heart Association guidelines suggest that a measurement of 40 inches (102 cm) for men, and 35 inches (88cm) for women as indicative of central obesity. The invention provides a means of further defining the risk for such individuals. It is equally applicable to obese children and provides a particularly useful early indication of disease onset in this group.

The risk of metabolic syndrome and consequent metabolic and cardiovascular disease is also increased in certain ethnic populations. Individuals of South Asian or Indian descent are at higher risk than many other populations and also form an especially suitable group for testing.

Without being bound any one particular theory or model, it seems that it is LPS ' stimulation of the receptor, TLR-2, that produces the pro-inflammatory milieu that leads to both diabetes and cardiovascular disease in patients with obesity. Thus, by blocking the action of endotoxin (LPS) on the receptor TLR-2 in various tissues with a new drug may reduce both the risk for progression to diabetes and also vascular damage. Alternative strategies include blocking the absorption of endotoxin (LPS) from the gut or to neutralise circulating LPS in serum with blocking antibodies.

Brief Description of the Drawings

The invention will be described in detail with reference to examples and the drawings in which:

Figure 1 shows the correlation between log fasting serum insulin levels (IU/ml) and serum LPS (EU/ml) in the overall non-diabetic population. The line of best fit is also shown (r=0.604, p<0.001).

Figure 2 shows TLR 2 expression by quantitative Western blot in (A) obese and non-obese abdominal subcutaneous adipose tissue; and (B) abdominal subcutaneous adipose tissue from non-diabetic and type 2 diabetic individuals. Figure 3 shows the increase in TLR 2 expression in response to LPS in abdominal subcutaneous adipose tissue (relative fold increase, quantitative Western blotting).

Figure 4 compares IL-6 and TNF-α production in response to LPS in cultured adipocytes.

Figure 5 shows endotoxin serum levels expressed in Case Controls (CC) and Fatty Liver Disease (FLD) patients: A) CC (n=34) versus FLD (n= 71, **[<0.0-\); B) FLD diabetics (DB, n=22) versus FLD non-diabetics (Non-DB, n=49).

Figure 6 shows that LPS levels are increased at all stages of non-alcoholic fatty liver disease (NAFLD) NASH = non-alcoholic steatohepatitis.

Figure 7 shows expression of (A) the circulating pro-inflammatory markers sCD14 (CC: n=33, FLD: n=80, *p<0.05) and sTNFRII (CC: n=20, FLD: n=72) and (B) Correlation between circulating levels of sCD14 and sTNFRII in FLD patients (n=72, p=0.020, calculated by Pearson correlation).

Figure 8 shows the elevation in sCD14 levels in NAFLD; highest in cirrhosis.

Figure 9 shows the correlation between sCD14 levels and severity of liver cirrhosis.

Figure 10 shows that sCD14 is higher in diabetes than non-diabetics with NAFLD.

Figure 11 shows sCD14 correlates with fasting blood glucose

Figure 12 shows distribution of LPS levels before (A) and after (B) by transformation. Figure 13 shows age-adjusted circulating levels of endotoxin (and 95%CI) by ethnic group in men (A) and women (B). there was no significant interaction by gender (p=0.65).

Figure 14A shows relationships between circulating plasma endotoxin levels and metabolic variables in men (White; African origin; South Asian).

Figure 14B shows relationships between circulating plasma endotoxin levels and metabolic variables in women (White; African origin; South Asian).

Figure 15 shows the correlation between TNF-α and log LPS levels in type 2 diabetes patients.

Figure 16 shows the correlation between circulating LPS levels and fasting insulin levels in a non-diabetic cohort (see Example 2).

Detailed Description of the Invention

Example 1 : LPS activates the innate immune response in human adipose tissue in obesity and type 2 diabetes

Research Methods & Materials

Cross-sectional studies of serum endotoxin.

Fasting serum samples were obtained from 25 non-diabetic subjects who had no significant medical illness and 25 sex, BMI and age matched patients with type 2 diabetes without known diabetic complications or chronic inflammatory conditions. These patients had a history of type 2 diabetes for at least one year. From this diabetic cohort 11 of the patients were diet-controlled, 10 treated with oral hypoglycemics and 4 treated with insulin and oral hypoglycaemics. All patients remained on therapy during sampling. The study was conducted with institutional ethical approval and written informed consent was obtained for subjects for all studies undertaken. Samples were analysed as further described below.

Study of effect of rosiglitazone.

Fasting baseline and 10 weeks post-treatment sera were also obtained from a cohort of newly diagnosed previously untreated T2DM patients treated with the insulin sensitizer, rosiglitazone (RSG) (Table 2). These samples were utilized to examine the effect of reducing circulating insulin levels on endotoxin concentrations. Samples were analysed for assessment of cytokines, insulin and glucose levels as further described below.

In vitro studies of whole adipose tissue studies.

Whole abdominal subcutaneous adipose tissue was obtained from lean non- diabetic subjects (M:F ratio 3:1; n=5: age: 49-8±5-4yrs; BMI: 22-9+0-9kg/m²) and obese, non-diabetic subjects (M:F ratio 2:1; n=6: age: 48±10-2yrs; BMI: 31 -4±2-9kg/m²) from redundant liposuction material obtained during elective abdominal surgery. Type-2 diabetic subjects (male n=5; age: 61±10-2yrs; BMI: 28-2±3-6kg/m²) provided adipose tissue by elective abdominal adipose tissue biopsy performed under local anaesthetic. Subjects providing fat samples were not on endocrine therapy, (e.g. steroids, HRT, thyroxine) or receiving any antihypertensive therapy and were not diabetic unless otherwise stated. All studies were performed with the approval of the local ethics committee and patients provided informed consent (Cardiff/Coventry & Warwickshire/South Birmingham). Analysis of serum samples.

The aforementioned serum samples were analyzed for the determination of insulin, (Linco Research Inc. Missouri, USA) leptin, (Unco Research Inc. Missouri, USA), IL-6, (Bender MedSystems, Vienna, Austria) TNF-α, (Bender MedSystems, Vienna, Austria) and sCD14 (R&D Systems, Abingdon, UK) protein concentrations. Insulin, leptin, sCD14, TNF-α and IL-6 levels were analyzed by a solid phase enzyme linked immunosorbent assay. Insulin CV intra-assay 5-96%, inter-assay 10-3±0-9%; leptin CV intra-assay 2-6%, inter-assay 3-8±3-4%; sCD14 intra-assay 5-4%, inter-assay 6-3%; TNF-α CV intra-assay 6-0%, inter-assay 9-3%; IL-6 CV intra-assay 6-2%, inter-assay 70%. For these studies anthropometric data was collected-detailed in tables: 1 and 2. Fasting glucose was analyzed using a glucose oxidase method (YSL 200 STAT plus). Insulin resistance (HOMA IR) was derived using the HOMA equation (Matthews etal, 1985, Diabetologia 28: 412). Serum endotoxin was assayed using a QCL-1000 LAL Endpoint Assay (Cambrex, New Jersey, USA). Intra-assay CV 3-9±0-46, inter-assay CV 9-6±0-75.

Studies on mature human adipocytes.

Subcutaneous abdominal adipose tissue taken from patients at elective liposuction surgery (BMI: 24.5±5 kg/m²; Age: 41.5+10.1yrs; n=10) was digested as previously described to isolate mature adipocytes (McTernan etal, 2000, lnt J Obes Relat Metab Disord 24:875). Following isolation of these cells, compacted aliquots of mature adipocytes (1mL; 500,000 cells) were cultured in phenol red- free DMEM.F12 medium (5mL; containing 15mmol/L glucose, penicillin (100units/mL) and streptomycin (100 μg/mL)). The cells were maintained in this medium for 14h either treated with 100ng/mL LPS (Sigma Aldrich, Gillingham, UK) or untreated to act as matched paired controls. Following treatment, the conditioned media and adipocytes were separated by centrifugation. Conditioned medium was removed and stored at -8O⁰C. Protein was extracted from the adipocytes using RIPA buffer (McTernan et al, 2000, lnt J Obes Relat Metab Disord 24:875) containing; 15OmM NaCI, 1 -0% IGEPAL® CA-630, 0-5% sodium deoxycholate, 0-1% SDS, and 5OmM Tris. These samples were subsequently flash-frozen in liquid nitrogen, thawed and centrifuged (80Og, 4⁰C). The resulting infranatant was extracted from under the lipid layer and stored immediately at - 8O⁰C. Viability of adipocytes was assessed using the trypan blue exclusion dye method as previously documented (McTernan et al, 2000, lnt J Obes Relat Metab Disord 24:875) (Sigma, UK).

Conditioned media (100μl) from untreated (control) and LPS treated abdominal subcutaneous adipocytes, were assayed for IL-6 and TNF-α (QuantiGlo ELISA, R&D Systems, Abingdon, UK) using a solid phase enzyme linked immunosorbant assay: IL-6 intra-assay CV 3-1%, inter-assay CV 2-7%; TNF-α intra-assay CV 6-7%, inter-assay CV 11 -0%.

Western blot analysis:

Total protein was extracted from homogenized abdominal subcutaneous tissue from the obese, lean and diabetic cohorts, as previously described, using the RIPA buffer technique. Homogenized whole human adipose tissue (cohort as discussed) was extracted using a 10% RIPA buffer method. Western blot analysis was performed and relative expression was standardized using densitometry quantification software (GeneTools, Geneflow, Fradley, UK). In brief, 10-40μg of protein was loaded onto a 10% polyacrylamide gel (Geneflow, Fradley, UK). A human monoclonal toll-like receptor 2 primary antibody (TLR-2; HyCuIt Biotechnology, Uden, Netherlands) and a human polyclonal anti-TLR 4 (Santa Cruz Biotechnology, Santa Cruz, California) were used to assess the presence of the TLR's. TLR-2 and TLR-4 were developed using mouse and rabbit horseradish-peroxidase (HRP) secondary antibodies respectively (2mg/mL, The Binding Site, Birmingham, UK). A chemiluminescent detection system ECL/ECL+ (GE Healthcare, Little Chalfont, UK) enabled visualization after exposure to X-ray film. Equal protein loading for these samples was confirmed by examining α- tubulin expression using Western blotting, as previously described (McTeman et al, 2000, lnt J Obes Relat Metab Disord 24:875). Protein concentrations were determined using the Bio-Rad DC (Detergent Compatible) protein assay kit.

Statistical analysis.

All variables were initially analyzed using bivariate Pearson correlation to address any relationship between the clinical parameters. Linear logged regression was used to assess the relationship between measurements of endotoxin and insulin. Residual plots were applied and all statistical outliers removed from the study. Protein expression data between control and treatment regimen were compared using an unpaired t-test. The threshold for significance was p<0 05. Data in the text and figures are presented as mean ± SD unless otherwise stated. Analyses were carried out using the SPSS version 12 (SPSS Inc. Chicago) software package.

Circulating Endotoxin (LPS) levels in non-diabetic and T2DM subjects Endotoxin levels were significantly higher in the BMI, sex and age matched T2DM group than in the ND subjects (p=<0.0001; Table 1).

W

25

Table 1

Furthermore, fasting insulin significantly correlated with serum endotoxin levels in the whole ND population (r=0-679, p<00001; Figure 1). This correlation remained when controlling for sex, age and BMI. (r=0-731, pO-0001). No correlation was observed in any of the subjects between glucose and endotoxin.

The data show that circulating endotoxin levels are significantly higher in type 2 diabetes and are positively correlated with the pro-inflammatory cytokine, TNF-α (see also Figure 15) and with the marker of endotoxin activity and presence, sCD14. Furthermore, circulating endotoxin correlated with serum insulin in non- diabetic subjects. Table 2

Effect of rosiglitazone on circulating endotoxin in treatment naive T2DM patients:

Serum endotoxin fell following treatment with rosiglitazone. Patients treated with RSG who demonstrated a significant reduction in fasting insulin levels also showed significantly reduced endotoxin levels. (Endotoxin, p=0O198; Insulin, p=0.0393; Table 2) whilst there was a positive correlation between change in insulin and change in endotoxin which was also highly significant (r=0-673, p=0 016). Pro-inflammatory cytokine profile in T2DM subjects

TNF-α was raised in T2DM subjects compared with ND (p=0-007; Table 1). Soluble CD14, a marker of macrophage activation and endotoxin clearance, was also significantly higher in T2DM than controls (p<0O001 ; Table 1) and correlated significantly with TNF-cc in the T2DM population (r=0-601, p<0 001). This correlation remained when controlling for age, sex and BMI (r=0-593, p=0003). There was also a strong correlation between HOMA-IR and endotoxin in controls, which is accounted for by the correlation with insulin as stated.

TLR protein expression is higher in obese and Type 2 diabetic subjects

The studies determined TLR-2 and TLR-4 expression in human abdominal subcutaneous adipose tissue. TLR-2 protein expression was significantly higher in adipose tissue from obese subjects compared with lean individuals (p<0 001, n=5; Figure 2). Furthermore, TLR-2 was also higher in diabetic abdominal subcutaneous adipose tissue compared with non-diabetic matched controls (p=0O13, n=5; Figure 2). TLR-4 showed no change in adipose tissue expression in the obese cohort (obese: TLR-44384± (SEM) 1098 OD units vs. lean TLR-4 5993± (SEM) 863 OD units, p=NS) or diabetic cohort (T2DM TLR-44355± (SEM)269 OD units vs. non-diabetic TLR-44805± (SEM)263, p=NS).

TLR-2 protein expression is upregulated by LPS in human isolated abdominal subcutaneous human adipocytes.

TLR-2 and 4 protein expression was shown in isolated subcutaneous adipocytes. Treatment of abdominal subcutaneous adipocytes with LPS increased the expression of TLR-2 two-fold compared with control (p<0 05; Figure 3). There was no change observed in TLR-4 expression with LPS treatment (Data not shown). Secretion oflL6 and TNF-α in response to LPS in cultured abdominal subcutaneous adipocytes.

Secretion of IL-6 and TNF-α were increased in the in vitro cultured adipocytes treated with LPS (IL-6: Control p<0-01; TNF-α: p<001; Figure 4). This corresponded with the increase in TLR-2 expression as previously discussed (Fig. 3).

Serum endotoxin is significantly reduced by the insulin sensitizer, rosiglitazone, and change in serum endotoxin correlates with the fall in insulin observed in the diabetic patients. In vitro studies with cultured isolated subcutaneous adipocytes treated with LPS confirmed a significant increase in the secretion of potentially diabetogenic pro-inflammatory cytokines. From these results it appears that adipose tissue secretes these pro-inflammatory cytokines in response to LPS, which also induces expression and/or upregulation of the key Toll-like receptor, TLR 2 (Lin et al, 2000, J Biol Chem 275:24255). Without wishing to be bound by any particular theory, the inventors have found an explanation for the induction of the sub-clinical inflammation associated with type 2 diabetes and obesity- associated insulin resistance and vascular disease.

The relationship between insulin and LPS was further explored by utilizing the insulin sensitizer, rosiglitazone (RSG) to ascertain whether alteration in insulin levels affected circulating LPS levels. RSG has no known direct effects on bacterial flora in the gut, but treatment with RSG results in reduction of circulating insulin. RSG-treated type 2 diabetes patients showed an absolute change in insulin levels that also correlated with the change in circulating LPS levels. RSG is known to have direct immunomodulatory properties on adipose cells and appear to reduce inflammation in both in vitro and in vivo model systems (Cuoco et al, 2002, Hepatogastroenterology 49:1582). The reduction in inflammation seen during RSG treatment may, at least in part, be due to the change in circulating LPS levels. Type 2 diabetes patients have significantly higher levels of serum endotoxin than non-diabetic subjects and the innate immune system is upregulated within obese and type 2 diabetes adipose tissue. Serum insulin and endotoxin and are closely related.

Example 2: chronic endotoxaemia in non-alcoholic fatty liver disease

Materials and Methods

Subjects.

Fasted human blood was collected from NAFLD and case control (CC) subjects (NAFLD: Age: 36.8 ± 11.5yrs (mean ± standard deviation, SD) years, BMI: 26.5 ± 4.4(mean ± SD) kg/m² n=80; CC: Age: 38.9 ± 12.4 years, BMI: 26.5 ± 4.4 kg/m², n=35) accordingly to Local Ethics Research Committee consent. The sub- categories of FLD were determined by liver biopsies and liver function tests by ballooning and/or fibrosis (Brunt etal, 1999). Diabetic status was also ascertained by glucose and insulin levels. The subjects were divided into four categories: fatty liver or steatosis (FL, n= 58), steatohepatosis (NASH, n= 54), fibrosis and cirrhosis (CRR, n= 12) and type 2 diabetes mellitus (DB, n= 31).

Analysis of circulating endotoxin levels.

Serum endotoxin was assayed using a commercially available QCL-1000 LAL Endpoint Assay (Cambrex, New Jersey, USA). Intra-assay CV 3.9±0.46, inter- assay CV 9.6±0.75. Assessment of inflammatory markers and adiponectin.

Serum was analysed by enzyme-linked immunosorbent assay (ELISA) for quantitative detection of the inflammatory markers: soluble CD14, soluble TNF-α Receptor Il (TNFRII), Resistin, TNF-α (R&D Systems, UK) and Leptin (Biogenesis Ltd, UK) and of the anti-inflammatory cytokine: Adiponectin (Biogenesis Ltd, UK); according to manufacturer's instructions. Insulin, leptin, CD14, TNF-α TNF-α Receptor II, resistin and IL-6 levels were analyzed by a solid phase enzyme amplified sensitivity immunoassays. Insulin CV intra-assay 5.96%, inter-assay 10.3±0.9%; leptin CV intra-assay 2.6%, inter-assay 3.8±3.4%; sCD14 intra-assay 5.4%, inter-assay 6.3%; TNF-αCV intra-assay 6.0%, inter-assay 9.3%; IL-6 CV intra-assay 6.2%, inter-assay 7.0%. Adiponectin was determined by an enzyme immunoassay (Linco Research, Inc. Missouri, USA) with a sensitivity of 0.78ng/mL and an intra- and interassay CV of 3.3% and 5.5% respectively.

Statistical analysis.

Statistical analysis was carried out using SPSS 12.0 for Windows software (SPSS UK, Woking, UK). Differences were considered statistically significant at p<0.05.

Clinical and biochemical characteristics of NASH patients and control subjects

The circulating Endotoxin levels in the stages of NAFLD

Endotoxin levels were significantly higher in patients with FLD compared with CC

(FLD: 11.06 ± 0.96EU/mL (mean±SEM); CC: 5.27 ± 0.45EU/mL, p<0.01; Figure 5A). FLD alone produced comparable endotoxin levels to T2DM (FLD:T2DM:10.82 ± 1.15EU/mL; Non-Diabetic: 11.16 ± 0.90EU/mL; Figure 5B). Analysis of the different stages of NAFLD demonstrated that endotoxin levels were raised at all stages of NAFLD when compared to CC (FL: 12.95±1.37 EU/mL, NASH: 10.19±0.77 EU/mL and CRR: 8.24±2.10 EU/mL, Figure 6).

Circulating endotoxin levels and relationship with fasting insulin.

In the non-diabetic case control cohort there was a significant positive association between fasting insulin and circulating endotoxin levels (Data not shown).

The effect of a soluble CD14 and TNF RII in FLD

The serum receptor for LPS clearance, sCD14 and a marker for TNF-α activation, TNFRII, were also analysed, sCD14 circulating levels were raised in FLD compared to CC (FLD: 1645.87±46.12 ng/ml (mean±SEM) and CC: 1450.57±69.17ng/mL, p<0.05, Figure 7A). A significant positive association was also observed between sCD14 and sTNFRII in FLD patients (r=0.327, p=0.020, Figure 7B) which was not observed in the case control subjects (data not shown). In addition sCD14 was positively associated with the severity of fibrosis through the different stages of NAFLD (r=0.288, p=0.006, Figure 8 and 9). The levels of sCD14 in FLD were higher in FLD with diabetes (FLD:T2DM:1903±0.12ng/ml; Non-Diabetic: 1558±0.05ng/ml; Figure 10) and correlated with fasting blood glucose levels (Figure 11).

LPS levels were significantly higher in patients with FLD compared with case control subjects. In addition, FLD alone produced similar LPS levels to type 2 diabetes subjects with FLD. Type 2 diabetes patients are known to have gut dysmotility (Zietz et al, 2000) which may affect circulating LPS levels, however increased LPS was also identified in the non-diabetic FLD patients. Other factors appear to alter LPS levels prior to the onset of type 2 diabetes. The data here appears to confirm that whilst diabetic status may raise LPS levels compared with case control subjects, the impact of liver disease in a pre-diabetic state also represents a significant contributor to raised LPS levels.

Further analysis of the different stages of NAFLD demonstrated that circulating LPS levels were raised at all stages of NAFLD when compared with case control subjects. Further assessment of the stages of NAFLD also demonstrated that LPS was highest during the necro-inflammation stages with later stages showing a significant but small reduction in circulating LPS.

The role of sub-clinical inflammation associated with obesity in the pathogenesis of insulin resistance and vascular disease is also partially explained by these in vivo studies. These data demonstrate a significant correlation between LPS and insulin in healthy non-diabetic subjects. Without wishing to be bound by any particular hypothesis, the inventors observe that hyperinsulinaemia in the pre- type 2 diabetic state can lead to increased LPS secondary to its action on absorption in the gastrointestinal tract.

In relation to increased LPS levels the present findings also noted that the serum receptor for LPS clearance, sCD14 was raised in FLD compared with case control subjects with a clear positive association observed between sCD14 and severity of fibrosis through the different stages of NAFLD.

Example 3: Circulating LPS levels as a marker for cardiovascular and metabolic risk in ethnic minority groups

Methods

Subjects.

A nested study was carried out in individuals from the Wandsworth Heart & Stroke Study, a population-based cross-sectional study of randomly selected men and women, aged 40-59 yrs, from three different ethnic groups (Cappuccio et al, 1997, Heart 78:555; Cappuccio et al, 1998, Nutr Metab Cardiovasc Dis 8:371). Subjects on hypertension or lipid lowering medications, those taking the oral contraceptive pill or hormone replacement therapy and those with previous medical history of ischemic heart disease or stroke were excluded to rule out the possibility of a direct effect of disease status on the level of endotoxin. A randomly selected a sub-group of participants matched for age and gender from each ethnic group were randomly selected for endotoxin determination. The sample size was calculated to have a power of >90% to detect a significant difference (p<0.05) in circulating levels of endotoxin between groups (n=30) of 3.6 EU/mL [SD 1.9]. Of the 193 individuals studied, 62 were white (30 women), 68 were of African origin (33 women) and 63 were of South Asian origin (33 women). The Local Ethics Committee approved the study. All participants gave their informed consent to participate.

Procedures

Subjects, who had fasted overnight and had refrained from smoking or taking vigorous exercise, were seen between 08:00 am and 12:00 noon the following day. A detailed questionnaire was completed, height and weight were measured and BMi calculated and expressed as kg/m². Waist and hip circumference (in cm) were taken by standardised methods and waist-hip ratio calculated. Blood pressure (BP) was taken with standard methods and an automated recorder. Blood was taken from seated subjects without stasis. Biochemical measurements were performed as described previously Cappuccio eta\, 1997, Heart 78:555; Cappuccio etal, 1998, Nutr Metab Cardiovasc Dis 8:371).

Analysis of circulating endotoxin levels.

Endotoxin measurements were performed, using a commercially available QCL- 1000 Chromogenic Limulus Amoebocyte Lysate (LAL) Endpoint Assay (Cambrex, New Jersey, USA), on heparinised plasma, which had been stored at -40⁰C and defrosted at room temperature prior to analysis. In brief, the LAL Test is a quantitative test for gram-negative bacterial endotoxin. A sample is mixed with the LAL supplied in the test kit and incubated at 37°C (±1°C) for 10 minutes. The reaction is stopped after a further minute incubation (37°C [±1⁰C]) with the substrate solution. The amount of endotoxin present in the sample is proportional to the yellow colour that develops, which is determined spectrophotometrically at 405-410 nm and can be calculated by comparison with a standard curve. All the samples were diluted 1 in 7 with LAL water using a standard curve in the range of 0 to 5 EU/mL. All samples were run in duplicate within the same 96-well plate. To prevent contamination all items used in the assay were sterilised (by autoclave or treatment with microsol). Intra-assay coefficient of variation was 3-9 ± 0-46 and inter-assay CV 9-6 ± 0-75.

Statistical analysis

Plasma levels of endotoxin were not normally distributed. Therefore analyses were performed on log-transformed data (Figure 12). Differences between groups (with described adjustments) were tested using analysis of co-variance. Results are presented as geometric means and 95% confidence intervals (Cl). General Linear Models were used to estimate the relationship between endotoxin levels and cardiovascular risk factors adjusting for confounders. Interaction test was used to compare slopes between groups. Given the degree of multiple testing, only a p<0.005 (Bonferroni's correction) was considered as statistically significant.

The characteristics of the groups are shown in Table 3. They are consistent with those previously published. In particular, serum triglycerides are highest in South Asians and lowest in people of African origin, whilst HDL-cholesterol follows the opposite pattern. Serum insulin is lowest in whites than blacks or South Asians. The plasma levels of endotoxin were not in a Gaussian distribution and as such appeared skewed (Figure 12A): hence subsequent analyses were performed on log-transformed data (Figure 12B).

Plasma endotoxin levels varied by ethnic groups and by gender (Table 3). In men, they were 11.1 EU/mL (95% Cl 9.1 to 12.4) in people of African origin, 12.5 EU/mL (10.6 to 14.6) in whites and 14.2 EU/mL (12.0 to 16.8) in South Asians (p<0.001); in women they were 9.1 EU/mL (8.1 to 10.1), 9.5 EU/mL (8.1 to 11.2) and 12.6 EU/mL (10.7 to 14.7), respectively (p<0.001). In the total group, age- adjusted endotoxin levels were significantly higher in men than in women (p<0.001). Following adjustment for age and sex, endotoxin levels still varied significantly according to ethnicity: African origin 10.1 EU/mL (9.1 to 11.1), whites 10.9 EU/mL (9.5 to 12.5) and South Asians 13.2 EU/mL (11.9 to 14.7), p<0.001. This graded ethnic difference in endotoxin levels was seen in men (p<0.008) and in women (p=0.071; interaction p=0.65; Figure 13).

The prevalence of current smoking was 29% in whites, 12% in people of African origin and 14% in South Asians. However, endotoxin levels were not associated with smoking (p=0.157). Furthermore partial correlation analysis (adjusted for age, sex and ethnicity) showed that, in the total group, endotoxin levels were not associated with blood pressure (systolic r=0.027, p=0.071; diastolic r=0.109, p=0.13) and that there were no interactions with gender (SBP: p=0.16; DBP: p=0.47) or ethnicity (SBP: p=0.76; DBP: p=0.86). There was, however, a significant gender difference in the association between BMI and endotoxin (interaction p<0.001) in that there was a significant association in men (r=0.350, p=0.001) but not in women (r=-0.109, p=0.29).

There were significant positive associations between endotoxin and both waist and waist-to-hip ratio which were observed in men and women and were consistent across each ethnic group. Plasma endotoxin levels showed direct relationships with total cholesterol (r=0.502, p<0.001)) serum triglycerides (r=0.776, p<0.001) and an inverse relationship with HDL-cholesterol (r=-0.439, p<0.001). There were no interactions by gender or by ethnicity (data not shown). There was no association with fasting serum glucose levels (r=0.009, p=0.90); by contrast there was a significant association with fasting serum insulin (r=0.320, p<0.001) and there were no interactions with gender (p=0.26) or with ethnicity (p=0.21). The individual relationships are shown in Figure 14A for men and in Figure 14B for women, separately.

The results of this study clearly show that circulating endotoxin levels are associated with the expected differences in atherosclerotic vascular disease risk. That is endotoxin levels are higher in men than women and highest in South Asians and lowest in people of African origin. In addition, endotoxin levels are associated with cardiovascular risk factors such as waist-hip ratio and serum lipid levels.

The observed association between circulating endotoxin levels and various cardiovascular risk factors, including measures of central adiposity and in particular, serum lipids and insulin levels, suggests a consistent pattern with markers of the metabolic syndrome. Furthermore, the associations with serum lipids are consistent across both gender and ethnic groups.

Example 4 - LPS as a mediator of sub-clinical inflammation in obese children

Background

Obesity and its later progression to type 2 diabetes are both known to be associated with sub-clinical inflammation; although the underlying cause for this is unclear. However, studies suggest that bacterial endotoxin (LPS) derived from commensal bacteria in the human gastrointestinal tract may contribute directly to sub-clinical inflammation. Hence, with increasing adiposity the inflammatory response is exacerbated producing inflammatory adipocytokines, such as plasminogen activator inhibitor type-1 (PAI-1), IL-6 and TNF-α. This study therefore investigated the pathogenesis of the metabolic syndrome in obese children and its' association with inflammation, with local ethereal research committee approval. Therefore, for this study, the role of endotoxaemia in childhood obesity was examined through the application of multiplex cardiovascular disease (CVD) biomarker immunoassays to investigate the levels of a range of inflammatory and CVD risk markers.

Methods and materials

Fasted serum was obtained from children with varying degrees of obesity (age±SD:13.9±2.3yr; BMI±SD:35.1±5.2 Kg/m²; n=60). All children underwent an oral glucose tolerance test and insulin resistance was measured by homeostasis model assessment (HOMA-IR) and insulin secretion by Stumvoll index (ISI). The vascular injury markers for these obese children were measured using a multiplex assay system that analysed interleukin - (IL-) 1β, 6, 8 and 10, interferon- Y (IFN- Y), TNF-α, monocyte chemotactic protein-1 (MCP-1), n-terminal brain natriuretic peptide (NT-proBNP), vascular endothelial growth factor (VEGF), plasminogen activator inhibitor 1 (PAI-1), C-reactive protein (CRP), fibrinogen, serum amyloid A (SAA), serum amyloid P component (SAP), E-selectin, soluble intercellular adhesion molecule type 1 (slCAM-1), soluble vascular cell adhesion molecule type 1 (sVCAM-1), matrix metalloproteinase -9 (MMP9), myeloperoxidase, lipid peroxide modified protein (MPO) and adiponectin.

Results

BMI correlated strongly with many of the inflammatory markers. However, endotoxin levels demonstrated a significant and positive correlation with the majority of markers for vascular injury and atherogenesis (TNF-α, myeloperoxidase; p<0.05; PAI-1 , matrix metalloproteinase-9, monocyte chemotactic protein-1, soluble intercellular adhesion molecule type-1, vascular endothelial growth factor; p<0.01). These relationships remained significant following adjustment for sex, BMI and HOMA-IR.

The data highlights that sub-clinical inflammation in childhood obesity and type 2 diabetes appears to be mediated by circulating endotoxin levels. Furthermore, that children as young as 11 exhibit the same inflammatory profiles as identified in obese adults and may, as a result of long-term sub-clinical inflammation, have increased risk in the pathogenesis of type 2 diabetes and cardio vascular disease at a much earlier age.

Table 3. Age-adiusted characteristics of 193 men and women, aged 40-59 years, of different ethnic groups living in Wandsworth, South London, 1994-96.

African origin White South Asian

(n=68: Men=35; Women=33) (n=62: Men= 32;Women= 30) (n=63: Men = 30; Women=33)

Body mass index (kg/m²) Men 25.8 (24.7 to 27.0) 25.5 (24.4 to 26.7) 24.7 (23.5 to 26.0)

Women 29.7 (27.9 to 31.5) 26.4 (24.4 to 28.3) 27.5 (25.7 to 29.4)

Waist (cm) Men . 90.0 (87.0 to 93.0) 91.2 (88.0 to 94.2) 90.6 (87.4 to 93.8) -

Women 88.1 (84.2 to 92.0) 81.8 (77.6 to 85.9) 85.1 (81.1 to 89.0)

Waisfchip ratio Men 0.91 (0.89 to 0.93) 0.91 (0.89 to 0.93) 0.94 (0.92 to 0.96)

Women 0.82 (0.79 to 0.85) 0.79 (0.76 to 0.82) 0.82 (0.79 to 0.85)

Systolic Blood Pressure (mmHg) Men 130.4 (124.5 to 136.2) 128.6 (122.5 to 134.7) 131.7 (125.5 to 138.0)

Women 132.6 (125.6 to 139.7) 121.3 (113.8 to 128.7) 129.50 (122.4 to 136.7)

Diastolic Blood Pressure (mmHg) Men 84.2 (80.9 to 87.6) 84.5 (80.9 to 88.0) 86.7 (83.1 to 90.3)

Women 86.0 (82.4 to 89.5) 77.7 (73.9 to 81.4) 80.4 (76.8 to 84.0)

Total cholesterol (mrnol/L) Men 5.5 (5.2 to 5.9) 6.0 (5.6 to 6.4) 5.7 (5.4 to 6.1)

Women 5.5 (5.2 to 5.8) 6.0 (5.7 to 6.4) 5.6 (5.3 to 6.0)

Serum triglycerides (mmol/L)* Men 0.84 (0.73 to 0.97) 1.14 (0.98 to 1.33) 1.30 (1.11 to 1.53)

Women 0.68 (0.60 to 0.78) 0.93 (0.81 to 1.07) 1.19 (1.04 to 1.36)

HDL cholesterol (mmol/L) Men 1.3 (1.2 to 1.4) 1.3 (1.2 to 1.4) 1.2 (1.1 to 1.3)

Women 1.6 (1.5 to 1.8) 1.7 (1.6 to 1.9) 1.3 (1.2 to 1.5)

Serum glucose (mmol/L) Men 5.0 (4.8 to 5.2) 4.9 (4.8 to 5.1) 5.0 (4.8 to 5.2)

Women 4.8 (4.6 to 5.0) 5.0 (4.8 to 5.2) 5.2 (5.0 to 5.4)

Serum Insulin (mU/L)* Men 7.1 (6.0 to 8.5) 6.6 (5.6 to 7.9) 10.4 (8.6 to 12.5)

Women 8.6 (7.2 to 10.3) 6.7 (5.6 to 8.2) 11.2 (9.3 to 13.4)

Endotoxin EU/mL Men 11.1 (9.9 to 12.4) 12.5 (10.6 to 14.6) 14.2 (12.0 to 16.8)

Women 9.1 (8.1 to10.1) 9.5 (8.1 to11.2) 12.6 (10.7 to .14.7)

Results are means f*Geometric means) or percentages (95% CD . P values are for test of heterogeneity between ethnic groups by analysis of co-variance. There were no more than two values missing from any cell.

Example 5: Elevated Endotoxin levels as a mediator of continued Chronic sub-clinical inflammatory risk in coronary artery bypass graft (CABG) patients.

Chronic sub-clinical inflammation is associated with increased risk of type 2 diabetes and cardiovascular disease and is particularly observed in patients that have undergone CABG. The 'inflammatory risk profile' is altered in these patients. This example assessed whether circulating endotoxin may represent a potential trigger for immune activation in this cohort. Therefore the study:

(1) investigated endotoxin levels in BMI matched healthy lean case controls (CC) subjects (Age:43.0±SD 10.6yrs; BMI:29.6±SD 4.7kg/m²; n=27) compared with CABG subjects with (Age:62.9±SD 15.2yrs; BMI:29.1+SD 3.8kg/m²) using an Limulus Amoebocyte Lysate (LAL) gram negative endotoxin ELISA.

(2) assessed the levels of soluble CD14 (an immunological receptor involved in the presentation of LPS to the toll-like receptors) using a sCD14 ELISA.

(3) assessed the effect of CABG status on HsCRP, sCDI4, TNFRII and adiponectin levels.

The findings from the study demonstrated that endotoxin levels were significantly higher in CABG than controls (CABG: 6.7(mean±SD)2.3 IU/MI Vs CC:3.4±0.33 IU/mL, p<0.001).

HsCRP was significantly raised in CABG patients (HsCRP: CABG: 9.11(mean±SE) 0.92 vs CC: 1.27±0.55; p=0.001) however neither sCDI4 nor TNFRII were raised (sCD14: CABG: 1359.1±203.2ng/mL Vs CC:1487.5.1±429.7ng/mL; p=N.S). Adiponectin levels were however significantly reduced in CABG patients (CABG:6.22±6.1mg/mL Vs CC:29.01±6.09mg/mL p=0.013).

In summary, endotoxin levels are significantly raised in CABG. In addition, the associated endotoxinaemia appears to mediate chronic inflammation observed by raised HsCRP levels. The unaltered sCD14 and TNFRII levels in disease indicate these factors correlate with the early developmental phases of this disease. The continued reduction in adiponectin levels indicates that the inflammatory cytokine profile arises through elevated endotoxin levels which may mediate the continued inflammatory state.

Example 6: South Asians with impaired glucose tolerance and subclinical inflammation have reduced endoxtin levels and adiponectin levels with increased weight loss using orlistat.

Sub-clinical inflammation is an important factor in the pathogenesis of type 2 diabetes and may be influenced by ethnicity and the degree of glucose intolerance. The study compared inflammatory markers and adipocytokines in South Asian individuals with impaired glucose tolerance (IGT-Group) to age-, sex- and BMI-matched individuals with normal glucose tolerance (NGT-Group). The impact of 12-months dietary treatment with and without the lipase inhibitor orlistat was also assessed.

Method

Levels of adiponectin, resistin, CRP, endotoxin and insulin were measured in the IGT- and NGT-Groups. Individuals in the IGT-Group were randomised to dietary treatment either alone or with orlistat and, after 12-months, levels were repeated. Results

Adiponectin was lower in the IGT-Group (n=33) compared to the NGT-Group

(n=33) (12.46+/-5.68μg/ml vs 17.88+/-9.20μg/ml, p<0.01). In the IGT Group randomised to treatment with orlistat, after 12-months there was both a favourable increase in adiponectin (6.73+/-6.50μg/ml, p<0.01) and a reduction in endotoxin (4.56+/-2.07IU/ml, p<0.01), however, in the Group treated with diet alone there was no significant change in adiponectin and a smaller reduction in endotoxin (1.91IU/ml, p<0.05). The endotoxin levels were higher than in comparative white Caucasian serum samples (3.5I+/-1.3U/mL).

Conclusion

In South Asians with IGT, treatment with orlistat results in a favourable increase in adiponectin and a reduction in endotoxin after 12-months.

Claims

1. A method of determining the risk of metabolic or cardiovascular condition or disease in an individual comprising the steps of:

(a) obtaining a biological sample from the individual, and

(b) determining the concentration or amount of LPS in the sample wherein the concentration or amount of LPS in the sample is indicative of the risk of the individual developing said metabolic or cardiovascular condition or disease.

2. A method of monitoring an individual for the onset or stage of metabolic or cardiovascular condition or disease, comprising the steps of:

(a) obtaining a biological sample from the individual, and (b) determining the concentration or amount of LPS in the sample; wherein the concentration or amount of LPS in the sample is indicative of the onset or stage of metabolic or cardiovascular condition or disease in the individual.

3. A method as claimed in claim 2, wherein an elevated concentration or amount of LPS is indicative, of the onset or progression of said metabolic or cardiovascular condition or disease, the concentration or amount of LPS being measured (a) by reference to a standard value, or (b) by reference to at least one previous measurement from the same individual.

4. A method as claimed in any of claims 1 to 3, wherein the metabolic or cardiovascular condition or disease is selected from obesity, non-alcoholic fatty liver disease (NAFLD), atherosclerosis, coronary artery disease, metabolic syndrome, hyperϊnsulinaemia, insulin resistance and type 2 diabetes mellitus (T2DM).

5. A method as claimed in claim 4, wherein the metabolic condition or disease is type 2 diabetes mellitus (T2DM).

6. A method as claimed in claim 3, wherein the metabolic disease or condition is non-alcoholic fatty liver degeneration.

7. A method as claimed in claim 3, wherein the metabolic disease or condition is insulin resistance.

8. A method as claimed in any preceding claim, wherein the individual is obese.

9. A method as claimed in any preceding claim, wherein the individual is selected from an individual of defined ethnic origin, e.g. South Asian, Black African or Caucasian.

10. A method as claimed in any of claims 3 to 9, wherein an elevated concentration of LPS is in the range 4 to 20 EU ml^"1

11. A method as claimed in claim 10, wherein an elevated concentration of LPS is in the range 5 to 15 EU ml^"1.

12. A method as claimed in any preceding claim wherein the biological sample is blood or serum

13. A method of diagnosing a metabolic or cardiovascular condition or disease in an individual comprising the steps of

(a) obtaining a sample from the individual

(b) determining the concentration or amount of LPS in the ample wherein the concentration or amount of LPS in the sample indicates that the individual has the metabolic or cardiovascular disease.

14. A method as claimed in claim 13, wherein the metabolic or cardiovascular disease or condition is selected from non-alcoholic fatty liver disease (NAFLD), atherosclerosis, coronary artery disease, metabolic syndrome, hyperinsulinaemia, insulin resistance and type 2 diabetes meliitus (T2DM).

15. A method as claimed in claim 13 or claim 14, wherein an elevated concentration or amount of LPS indicates that the individual has the metabolic or cardiovascular disease or condition.

16. A method as claimed in claim 15, wherein the elevated concentration or amount of LPS is determined (a) by reference to a standard value, or (b) by reference to at least one previous measurement from the same individual.

17. A method as claimed in any of claims 13 to 16, wherein a value of 4 to 20 EU ml^"1 is indicative of the presence of type 2 diabetes meliitus (T2DM).

18. A method as claimed in any of claims 13 to 17 wherein the individual is obese.

19. A method as claimed in any preceding claim, wherein the individual is selected from an individual of defined ethnic origin, e.g. South Asian, Black African or Caucasian

20. The use of LPS as a marker in an assay for determining the risk of metabolic or cardiovascular condition or disease in an individual, wherein the assay is carried out on a biological sample from the individual.

21. The use of LPS as a marker in an assay for monitoring an individual for the onset or stage of metabolic or cardiovascular condition or disease, wherein the assay is carried out on a biological sample from the individual.

22. The use of LPS as a marker in an assay for diagnosing a metabolic or cardiovascular condition or disease in an individual, wherein the assay is carried out on a biological sample from the individual.

23. The use as claimed in any of claims 20 to 22, wherein an elevated concentration or amount of LPS in the sample indicates the risk of metabolic or cardiovascular condition or disease; the onset or stage of metabolic or cardiovascular condition or disease; or the metabolic or cardiovascular condition or disease in an individual

24. The use as claimed in any of claims 20 to 22, wherein the concentration or amount of LPS is determined (a) by reference to a standard value, or (b) by reference to at least one previous measurement from the same individual.

25. The use as claimed in any of claims 20 to 24 wherein the individual is obese.

26. The use as claimed in claim 21 , wherein a value of more than 4 EU ml^"1, preferably a value in the range 4 to 20 EU ml^"1 is indicative of the presence of type 2 diabetes mellitus (T2DM).

27. A method of treating an individual at risk of developing a metabolic or cardiovascular disease or condition comprising:

(a) obtaining a biological sample from the individual, (b) determining the concentration or amount of LPS in the sample, wherein the concentration or amount of LPS in the sample is indicative of the risk of the individual developing said metabolic or cardiovascular condition or disease, and

(c) administering to the individual a an anti-inflammatory agent.

28. A method as claimed in claim 27, wherein the disease or condition is selected from: obesity, non-alcoholic fatty liver disease (NAFLD), atherosclerosis, coronary artery disease, metabolic syndrome, hyperinsulinaemia, insulin resistance and type 2 diabetes mellitus (T2DM).

29. A method as claimed in claim 28 wherein an elevated concentration or amount of LPS or of a soluble LPS receptor is detected, indicating the onset or progression of said metabolic or cardiovascular disease.

30. A method as claimed in claim 29, wherein the elevated serum level of LPS or soluble LPS receptor is measured by reference to a standard value.

31. A method as claimed in claim 29, wherein the elevated serum level of LPS or soluble LPS receptor is measured by reference to at least one previous measurement from the same individual.

32. A method as claimed in claim 31, wherein a value of 4 to 20 EU ml^"1 is indicative of the presence of said metabolic or cardiovascular disease.

33. A method as claimed in any of claims 27 to 32, wherein the anti-inflammatory agent is a thiazolidinedione or a pharmaceutically acceptable salt thereof.

34. A method as claimed in claim 33, wherein the thiazolidinedione is rosiglitazone.

35. A method as claimed in any of claims 27 to 32, wherein the anti-inflammatory agent is an anti-LPS antibody.

36. A method as claimed in any of claims 27 to 32, wherein the anti-inflammatory agent is an anti-TLR 2 antibody or a TLR 2 antagonist.

37. A method of identifying an agent for the prevention or treatment of metabolic or cardiovascular disease comprising:

(a) administering endotoxin (LPS) to an animal, adipose tissue or adipocyte,

(b) administering a candidate agent to said test animal, adipose tissue or adipocyte,

(c) determining the level of expression of at least one toll-like receptor in adipose tissue taken from the animal, the adipose tissue, or adipocytes, whereby a decreased expression of at least one toll-like receptor (TLR) identifies a said agent.

38. A method as claimed in claim 37, wherein the at least one toll-like receptor is toll-like receptor 2 (TLR 2) and/or toll-like receptor 4 (TLR 4).

39. A method of identifying an agent for the prevention or treatment of metabolic or cardiovascular disease comprising:

(a) administering endotoxin (LPS) to an animal, adipose tissue or adipocyte,

(c) monitoring the innate immune pathway, whereby a decreased activation of the innate immune pathway identifies a said agent.

40. A method as claimed in claim 39, wherein the monitoring of the innate immune pathway comprises monitoring of the level of (a) pro-inflammatory cytokines and/or (b) adiponectin, or the expression of (a) pro-inflammatory cytokines and/or (b) adiponectin, whereby decreased levels or expression of proinflammatory cytokines and/or increased levels or expression of adiponectin identifies a said agent.

41. A method as claimed in claim 40, wherein the pro-inflammatory cytokines are adipocytokines.

42. A method as claimed in any of claims 37 to 41 , wherein the pro-inflammatory cytokines or adipocytokines are selected from one or more of IL-1, IL-6, TNFα

43. A method as claimed in any of claims 37 to 42, wherein the innate immune pathway is monitored or additionally monitored by determining the level or expression of CD 14, preferably soluble CD 14 and/or plasminogen activator inhibitor type 1 (PAI-1).

44. A method as claimed in any of claims 37 to 43, wherein the innate immune pathway is monitored or additionally monitored by determining activation of NFkB and/or translocation of NFkB to the nucleus.

45. A method as claimed in any of claims 37 to 44, wherein the animal is a rodent, preferably a rat or a mouse.

46. A method as claimed in any of claims 37 to 45, wherein the adipose tissue is subcutaneous abdominal adipose tissue, preferably human subcutaneous abdominal adipose tissue.

47. A method as claimed in any of claims 37 to 45, wherein the adipocyte is a cultured adipocyte, preferably a mature human adipocyte, more preferably a human subcutaneous abdominal adipocytes.

48. A method as claimed in any of claims 37 to 47, wherein the metabolic or cardiovascular disease is selected from: obesity, non-alcoholic fatty liver disease (NAFLD), atherosclerosis, coronary artery disease, metabolic syndrome, hyperinsulinaemia, insulin resistance and type 2 diabetes mellitus (T2DM).