Balanced macronutrient diet

This invention relates to methods of reducing or preventing low-grade- systemic inflammation, e.g. treating or preventing lifestyle-related diseases, and to compositions for use in such methods.

There is growing evidence that some diseases can be linked to modern lifestyles. For example, the smoking of tobacco and consumption of alcohol or drugs, as well as a lack of exercise, is thought to increase the risk of developing certain diseases, especially later in life. Diet is also a factor thought to influence susceptibility to many diseases. Diseases which are thought to be linked in this way to lifestyle include Alzheimer's disease, asthma, cancer, type 2 diabetes,

cardiovascular disease and obesity. Worldwide, chronic disease account contribute to about 60% of deaths (based on statistics from 2001 - World Health Organisation) and the incidence of such diseases is expected to rise to 75% by 2020, i.e. it is a significant and growing problem.

Recent studies have investigated some of the biological and physiological mechanisms by which aspects of lifestyle, e.g. diet, can affect disease states. Whilst the effect of changing the type and amount of fats in the diet has received considerable attention (e.g. the ratio of polyunsaturated (PUFAs) to saturated fat (SFA) in the diet), the role of carbohydrate has not received much attention. Certain epidemiological studies have suggested that dietary glycemic index (GI), glycemic load (GL) and fibre may be implicated in various lifestyle diseases, and in particular metabolic syndrome, but the studies are not all consistent and the mechanisms behind this are not fully understood. Controversies regarding optimal dietary carbohydrate or fat intake and quality still exist.

Low-grade systemic inflammation is a condition characterised by low level inflammation throughout the body and has become recognised in recent years to precede, and indeed possibly to predict, lifestyle diseases such as type 2 diabetes and cardiovascular disease (CVD). Potential links between diet and low-grade systemic inflammation have been studied but these are still not well understood. In one recent study, Arbo et al. (Scandinavian Journal of Clinical and Laboratory Investigations (201 1) 71 :330-339) show that the expression of genes involved in immunological processes may be regulated by insulin levels, which suggests inflammation as a possible consequence of postprandial hyperinsuiinemia. However, whilst this study provides tantalising hints about how diet composition and the frequency of feeding might affect l ifestyle diseases, it provides no firm guidance about how to go about formulating diet compositions or about treating such diseases.

Given the prevalence of lifestyle diseases and their predicted increase as the world's population ages, there is an acute need to find ways to combat them. In particular there is a need to find ways to reduce low-grade systemic inflammation in order to prevent or treat diseases associated therewith, e.g. cardiovascular disease, type 2 diabetes and obesity.

The present inventors have now discovered that providing specific combinations of macronutrients in the diet can result in a marked reduction in low- grade systemic inflammation. In particular, prov iding a diet with a specific ratio of carbohydrate, protein and fat (on an energy basis) wherein the fat component has a defined rat io of omega 3 to omega 6 fatt y acids and, more part icularly, a defined monounsaturate content results in a marked reduction in inflammatory markers. This finding provides a way to influence low-grade systemic inflammation by modulation of the diet and therefore enables the prevention or treatment of a w ide range of l ifestyle-related diseases.

Accordingly, a first aspect the invention provides a method of treating or preventing low -grade systemic inflammation, in a subject wherein the diet of the subject is modified so as to provide in the region of (or around ) 27% of energy (27 kcal%) from carbohydrate, in the region of (or around ) 30% of energy (30 kcai%) from protein and in the region of (or around ) 43% of energy (43 kcal%) from fat and wherein the fat component of the diet has a ratio of omega 3 to omega 6 fatty acids of about 1 :3.

In a second, more particular, aspect of the invention, the method can be for treating or preventing a l ifestyle disease. The diet is typically provided by administration of a nutrient-containing composition which comprises the said carbohydrate, protein and fat, as well as other nutritional components such as insoluble fibre, vitamins and minerals. The nutrient containing composition is preferably a meal replacement composition.

Accordingly, in another aspect the invention provides a method of reducing or preventing low-level systemic inflammation in a subject, the method comprising administering to the subject a nutrient-containing composition comprising in the region of (or around ) 27% carbohydrate (kcal%), in the region of (or around ) 30% protein (kcal%) and in the region of (or around ) 43% fat (kcal%), wherein the fat component of the composition has a ratio of omega 3 to omega 6 fatty acids of about 1 :3.

The nutrient-containing composition is preferably designed to contain substantially all of the daily energy requirement of the subject when administered in sufficient daily dosages.

In another aspect the invention provides a nutrient-containing composition for use in reducing or preventing low-level systemic inflammation in a subject, the composition comprising in the region of (or around ) 27% carbohydrate (kcal%), in the region of (or around ) 30% protein (kcai%) and in the region of (or around ) 43 %> fat (kcal%), wherein the fat component of the composition has a ratio of omega 3 to omega 6 fatty acids of about 1 :3.

The invention also extends to the use of such a composition in the manufacture of a medicament for reducing or preventing low-level systemic inflammation in a subject.

The compositions of the invention may be used, or adapted for use, in the general consumer market (e.g. as a meal replacement formula or nutrition bar), in medical nutrition (e.g. pre- or -post surgery), infant nutrition (e.g. as a milk additive or replacement formula) and sports nutrition (e.g. in a muscle-building and/or bone- strengthening formula).

As noted above, such methods, compositions and uses can more particularly, additional ly or alternatively be specified or defined to be for, or for use in, treating or preventing a l i festyle disease. In a further aspect the invention provides a nutrient-containing composition (e.g. a meal-replacement composition ) which comprises a carbohydrate component providing in the region of (or around ) 27% of the energy in the composition, a protein component providing in the region of (or around ) 30% of the energy in the composition and a fat component providing in the region of (or around ) 43% of the energy in the composition, wherein the fat component of the composition has a ratio of omega 3 to omega 6 fatty acids of about 1 :3. The composit ions of the invent ion are suitable for, and typically adapted for, use in treating or preventing low-grade systemic inflammation, e.g. treating or preventing l ifestyle diseases, as described herein.

In a preferred embodiment the nutrient-containing composition is designed to provide substantial ly all of the dietary energy and nutritional needs of the recipient. The components of the composition, e.g. the carbohydrate component, the protein component and the fat component may be provided separately or in admixture.

Further, it also preferred that the fat component of the composition comprises at least about 28%, or more particularly, at least about 28.5%, by weight of monounsaturated fatty acids.

In a further preferred embodiment, the ratio of saturated fatty acids (SFA) to monounsaturated fatty acids (MUFA) in the fat component of the diet or nutritional composition is about 1 :3.

As discussed in more detail below, in further preferred embodiments of the methods, compositions or uses of the invent ion, the diet or composition is provided or administered 4 to 8, preferably 4-7 or 4-6, times a day. The compositions of the invention may accordingly be provided in a form suitable for administration multiple such times a day.

The present invention relates to the prevention or treatment of lifestyle- related diseases, referred to herein as "lifestyle diseases" w hich term includes in particular any condition that is linked to low-grade systemic inflammation, which also known as "metabolic inflammation". Low-grade systemic inflammation is a condition of sub-clinical systemic inflammation (that is inflammation that is not linked or confined to a particular tissue or organ of the body and which may accordingly occur in multiple ( i.e. at least 2, or at least 3) areas or tissues of the body ). Low-grade systemic inflammation may thus occur in the endothelium and other organ systems. It is typical ly characterised by upregulation or elevation of inflammatory processes in multiple locations in the body, e.g. in di fferent tissue types, and may be distinguished from acute and/or local inflammation by the absence of associated pain. Low-grade systemic inflammation does not typically involve pain but it may result in uncomfortable symptoms such as l ight swelling, reddening of the skin, water retention and mild flu-like symptoms. It is generally regarded as a chronic condition and may more particularly be referred to as chronic low-grade systemic inflammation. Commonly, low-grade systemic inflammation may be characterised by a two- to threefold increase in the systemic concentrations of cytokines such as TNF-a, IL-6 and GRP.

Specific conditions linked to lifestyle and low-grade systemic inflammation include obesity; metabol ic syndrome; insul in resistance and conditions or diseases associated therewith or more generally with insul in regulation, including notably diabetes, particularly type 2 diabetes; cardiovascular disease, e.g. atherosclerosis or conditions or disorders of the heart; and kidney disease, e.g. nephritis or diabetic or hypertensive nephropathy.

Subjects to be treated according to the present invention will typical ly be humans, but the results described herein are believed to be applicable to other animals, especially to mammals (e.g. cats, dogs, cows, horses or pigs). In one embodiment, the sub ject to be treated is in need of general inflammatory

suppression, especial ly where the subject is awaiting or following surgical treatment and/or is a hospital inpatient. In another embodiment, the subject is a normal or healthy subject, e.g. a subject who does not present with chronic medical conditions or acute inflammatory conditions.

The present inv ention is based on the finding that control ling the diet can lead to rapid changes in expression of genes involved in inflammatory processes. These changes were observed within a week of changing from a high carbohydrate diet to the specific diet of the invention. The composition of the diet is believed to play a key role in the effects observed, as is the number of meals administered per day on the diet. Thus, in one embodiment the nutrient-containing composition of the invention is administered at least 4 times daily, e.g. at least 5, 6, 7 or 8 times daily, for example 4 to 8. 4 to 7 or 4 to 6 times daily. Preferably it is administered 5 or 6 times and especially 6 times daily.

As is clear from the foregoing, the individual components of the

compositions need not be administered or provided in admixture and can provided singly or individually, or in any desired or convenient combination. The term "composition" as used herein accordingly does not imply a single formulation. A composition according to the inv ention may accordingly contain, include or comprise multiple separate component parts (e.g. two or more, three or more, or four or more separate component parts for separate administration). Accordingly a composition of the invention may alternatively be defined as a kit of separate nutritional component parts. Each component part may comprise e.g. a fat, protein or carbohydrate component, or a mixture of different components e.g. fats and proteins, or a mixture of different fats or fatty acids etc. Other components or ingredients (e.g. vitamins or minerals as discussed further below) may be included as separate component parts or as part of a component part .

In a preferred embodiment the nutrient-containing composition of the invention represents essential ly all of the daily intake of food in the diet, i.e. the nutrient-containing composition replaces substantial ly all (or all) of the food in the diet of the recipient. The daily intake of the nutrient-containing composition is preferably chosen so as to be normocaloric, i.e. to provide the necessary amount of energy required by the recipient. However, in alternative embodiments the daily intake of the nutrient-containing composition may be chosen so as to be hypercaloric (i.e. to prov ide more energy that required by the recipient, especial ly if weight gain is desired, e.g. fol lowing illness or surgery), or to be hypocaloric ( i.e. to provide less energy that required by the recipient, especially if weight loss is desired ). The daily intake of energy required by a recipient may be determined by the skil led person, e.g. following general guidel ines or by determining the amount required for a given subject based on projected energy expenditure (e.g. using methods known in the art - Harris et al. , PNAS (1918) 4(12): 370-373; and Brooks et al, Am J Clin Nutr (2004 ) 79(suppl ):92 1 S-930S ). Typical v alues for the dai ly energy requirement of a human recipient are in the region of 2000-2500 kcal, e.g. around 2200 kcal, for adult women and 2500-3000 kcal, e.g. around 2700 kcal , for adult men (1 kcal is equivalent to 1 Calorie (Cai) and to 4.184 kilojoules (kJ)). Values for active individuals will typically be higher than this and for less active individuals, e.g. the elderly, and children it will typically be lower.

Administration of the nutrient-containing composition may be designed to provide multiple daily dosages having an equal nutritional content, e.g. 6 daily dosages each containing one sixth of the total daily requirement of nutrients. In other words the composition may be prov ided in isocaloric dosages. Alternatively, the administration profile may be designed to provide multiple daily dosages having an unequal nutritional content, e.g. 5 daily dosages wherein the first, third and fifth dosage each comprises 25% of the total daily requirement of nutrients and the second and fourth dosage each comprises 12.5% of the total daily requirement of nutrients.

Administration is typical ly enteral, especially oral (e.g. by normal ingestion ), although direct gastric administration via a feeding tube may also be employed.

The form of the nutrient-containing composition will depend on the intended use, e.g. cl inical use in hospital or use at home, but it will typical ly be administered as a solid meal (e.g. as a meal replacement bar or other shaped food item ) or in a liquid form (e.g. as a shake or a l iquid meal replacement drink ). The composition may come ready-prepared, e.g. separated into individual dosage forms, or may be provided in bulk for measuring into individual dosages.

l In a preferred embodiment, especially w hen prov ided in bulk ( i.e. in an amount sufficient to provide multiple doses, e.g. more than 100 g. especial ly more than 250 or 500 g, or more that 1 , 2, 5 or 10 kg), the composition is preferably in the form both of a powder, typically containing the dry ingredients (e.g. the protein, carbohydrate and any insoluble fibre, v itamins and minerals), and also an oil (typical ly comprising the fatty components). The powder and the oil are preferably provided separately, e.g. for admixture by the end user, although they may also be provided ready mixed so that each dosage may simply be made up in the required quantity w ith a l iquid (e.g. water) before administration. In one embodiment, the fat-containing components may be provided in powder form ( e.g. as encapsulated particlcs or as a l yophi lised powder, optionally in the presence of one or more stabilisers), in which case they are preferably provided in admixture with the other dry ingredients.

As defined above, the compositions of the invention comprise a carbohydrate component, a protein component and a fat component. The compositions also preferably comprise the other nutritional components required in the diet, for example insoluble fibre ( i.e. essentially indigestible complex carbohydrates ), salt (e.g. a source of chloride ions), vitamins and minerals. The fibre may be provided as part of the carbohydrate component of the diet. Where percentages by weight are mentioned hereinafter, they preferably relate to percentage by dry weight, i .e.

relative to a powder or solid composition substantially in the absence of water. The term "dry weight" may, however, include non-sol id components such as oils which are specifically defined as part of the composition.

The carbohydrate component of the composition is present at a level that prov ides in the region of (or around) 27% of the total (calorific) energy of the composition, i.e. at a level of about 24-30 kcal%, preferably about 25-29 or 26-28 kcal%, most preferably about 27 kcal%. In a typical composition of the invention, this would equate to in the region of (or around ) 30% by weight of carbohydrate, e.g. about 25-35% or 28-32% or 29-3 1 % by weight, especiall y about 31%> or about 30% by weight of carbohydrate. Preferred types of carbohydrate include one or more of fructose, glucose, maltodextrin (e.g. corn maltodextrin ) and ol igo fructose and preferably al l of these types are included. Other carbohydrates that may be included in the composition include mixtures of carbohydrates such as glucose syrup, e.g. corn syrup. In one embodiment, the carbohydrate component of the composition is essentially free from lactose.

In a preferred embodiment the carbohydrate component comprises (e.g. consists essentially of) one, more than one, or al l of maltodextrin MGX18

(Maltodextrin powder - HFG, Shandong, China), preferably at a level of in the region of (or around ) 40%o by weight (e.g. between 35 and 45%, and preferably about 40%) of the carbohydrate component: Raspberry 2251, (Obipektin AG, CH-9200 Bischofszel l, Switzerland) preferably at a level of in the region of (or around ) 37%> by weight (e.g. betw een 32 and 42%> and preferably about 40%>) of the carbohydrate component; fructose, preferably at a level of in the region of (or around ) 10% by weight (e.g. between 7 and 13% and preferably about 10%) of the carbohydrate component; and oligofructose, preferably at a level of in the region of (or around ) 13% by weight (e.g. between 1 0 and 16%, and preferably about 13%») of the carbohydrate component.

The protein component of the composition is present at a level that provides in the region of (or around ) 30% of the total (calorific) energy of the composition, i.e. at a level of about 27-33 kcal%, preferably about 28-32 or 29-31 kcal%, most preferably about 27 kcal%. In a typical composition of the invention, this would equate to in the region of (or around ) 42%o by weight of protein, e.g. about 38-46% or 40-44% by weight of protein, especial ly about 42% by weight of protein.

In one embodiment, the protein source is chosen so as to minimise specific al lergic reaction. In other words the protein component(s) is (are), or is (are) selected to be, non-allergen ic, e.g. the protein is not derived from wheat and/or is not derived from milk and/or egg components which arc known to be allergenic. In a preferred embodiment, the protein component comprises proteins from multiple sources in order to minimise general allergic reactions, e.g. the protein is chosen from a plurality of sources selected from meat (e.g. animal meat such as beef, chicken and pork, and fish meat, such as cod, salmon, mackerel and tuna ), egg, milk (e.g. whey and/or casein ), vegetables (e.g. tubers such as potato and yam, and legumes such as pea, soy bean and peanut ), microorganisms (e.g. yeast) and flour (e.g. corn flour or wheat flour). Preferably, the protein component comprises one, two or al l of whey protein, egg white and pea protein. In one embodiment, the protein component of the composition is essential ly free from gluten, especial ly wheat gluten.

In a preferred embodiment the protein component comprises (e.g. consists essential ly of) one, more than one. or all of whey protein 80 (Tine BA, 0051 Oslo, Norway), preferably at a level of in the region of (or around ) 55% by weight (e.g. between 50 and 60%> and preferably about 55%) of the protein component; egg white, preferably at a level of in the region of (or around ) 13% by weight (e.g.

between 10 and 16%, preferably about 13%) of the protein component; and pea protein 90 ( Pisane M9, Provital. Industrie S.A., Belgium ), preferably at a level of in the region of (or around ) 32% by weight (e.g. between 27 and 37% and preferably about 32%) of the protein component.

The fat component ( i.e. the fat-containing component, typical ly including fats and/or oils) of the composition is present at a level that provides in the region of (or around ) 43% of the total (calorific ) energy of the composition, i .e. at a level of about 40-46 kcal%, preferably about 41 -45 or 42-44 kcal%, most preferably about 43 kcal%. In a typical composition of the invention, because fat contains roughly double the calorific content by mass compared to protein and carbohydrate, this equates to in the region of (or around) 26% by weight of fats, e.g. about 22-30% or 24-28% by weight of fats, preferably about 26%. The fat component is preferably derived from di fferent sources to allow for an optimum combination of fat and oil types in the final composition. Examples of suitable sources of fats include plant sources (e.g. flax seed (linseed) oil, hemp oil, soya oil, canola (rapeseed ) oil, vegetable oi l, olive oil, coconut oil, borage oil and evening primrose oil, and nut oils from nuts such as walnuts) and animal sources (e.g. marine animals such as seal and squid, fish and shellfish, especially fish liver oils such as cod liver oil). Preferably, the fat component is derived from two or more, e.g. al l, of the oils selected from rape seed oil, ol ive oil, coconut oil, flax seed oil and marine oils.

In a preferred embodiment, the fat component comprises (e.g. consists essential ly of) one, more than one, or all of rape seed oil, preferably at a level of in the region of (or around ) 9% by weight (e.g. between 6 and 12%, preferably about 9%) of the fat component; olive oil, preferably at a level of in the region of (or around ) 79%> by weight (e.g. betw een 70 and 88% and preferably about 79%) of the fat component; coconut oil, preferably at a level of in the region of (or around ) 9% by weight (e.g. between 6 and 12% and preferably about 9%) of the fat component; flax seed oil, preferably at a level of in the region of (or around ) 1 .5% by weight (e.g. between 1 and 2%, preferabl y about 1.5%) of the fat component; and marine omega 3/6/7/9 oils, preferably at a level of in the region of (or around ) 1 .5% by weight (e.g. between I and 2%, preferably about 1.5%) of the fat component.

Marine omega 3/6/7/9 oils produced by Vitomega AS (Hcimdal, Norway) are especial ly preferred. The fat component of the composit ion must comprise both omega 3 and omega 6 fatty acids. Typical ly these will be present in a sufficient amount to provide the recommended daily allowance (RDA) of fatty acids when the composition is administered, e.g. about 1-2 grams of omega 3 fatty acids per day and about 3-6 grams of omega 6 fatty acids per day. Thus, the fat component of the composition of the invention preferably comprises between about 0.5 and 2% by weight of omega 3 fatty acids, especial ly betw een 0.75 and 1.5% by weight, e.g. about 1% by weight. The fat component of the composition of the invention preferably comprises between about 1 .5 and 6% by weight of omega 6 fatty acids, especially between 2.5 and 4.5% by weight, e.g. about 3% by weight.

The fat component of the composition preferably comprises at least about 28 %, or more particularly at least about 28.5% by weight of monounsaturated fatty- acids, especial ly between 28 and 30% by weight. The ratio of monounsaturated to saturated fatty acids in the fat component of the composition is preferably about 3 : 1 .

Vitamins and/or minerals are preferably provided in the compositions of the invention at levels at or around those recommended for daily intake, i.e. around the RDA lev els.

Vitamins which may be included in the compositions of the inv ention preferably include one or more of the following, especial ly al l of the following: Vitamin C, Vitamin E, Thiamine, Riboflavin, Vitamin B6, Vitamin B12, Folic acid. Pantothenic acid. Niacin and Biotin. Other vitamins which may be included, e.g. in addition to those listed above, include one or more, and preferably all. of: Vitamin A, Vitamin D, Vitamin K, choline and beta-hydroxy-beta-methyibutyric acid (HMB). The v itamins are preferably present in the necessary quantity to provide the RDA of each when a daily dosage of the composition is administered. Where the composition is for administration in multiple daily dosages, the absolute amount of the vitamin will be typical ly reduced accordingly. For example, a single dosage of the composition of the invention which is intended for administration in six equal dosages per day may contain the fol lowing levels of each vitamin (one sixth of the total required daily amount ): Vitamin C - between 10 and 100 mg, e.g. about 80 mg; Vitamin E - between 2 and 5 mg, e.g. about 3.5 mg; Thiamine - between 0.2 and 0.6 mg, e.g. about 0.38 mg; Riboflav in - between 0.2 and 0.6 mg, e.g. about 0.42 mg; Vitamin B6 - between 0.2 and 0.8 mg, e.g. about 0.53 mg; Vitamin B12 - between 0.2 and 0.5 iig, e.g. about 0.26 iig: Folic acid - about 0.05 1 iig; Pantothenic acid - between 0.5 and 5 mg, e.g. about 1 .6 mg; Niacin - between 1 and 10 mg, e.g. about 4.8 mg; and Biotin - about 0.042 ug. Preferably the composit ion of the invention includes a vitamin C-containing extract of Acerola (e.g. Acerola C or Acerola 25-C supplied by Obipektin AG, CH-3400 Burgdorf, Switzerland) which comprises in the region of (e.g. around ) 17% by weight of vitamin C, e.g. between 1 2 and 30% or between 1 5 and 20%, preferably about 17% by weight vitamin C. The extract of Acerola preferably comprises about 0.5% by weight of the total composition.

In an especially preferred embodiment, the composition of the invention contains a vitamin component comprising (e.g. consisting essential ly of) about 97% by weight of an Acerola C extract (which extract preferably comprises about 17% by weight of v itamin C) and about 3% by weight of a multivitamin, the multiv itamin comprising (e.g. consisting essent ially of) about 25.5% by weight of Vitamin C, about 19.1% by weight of Vitam in E, about 2.7% by weight of Thiamine, about

3.0% by w eight of Riboflavin, about 3.8% by weight of Vitamin B6, about 0.0019% by weight ( 1 9 ppm ) of Vitamin B12, about 0.0004% by weight (4 ppm ) of Fol ic acid, about 1 1.4% by weight of Pantothenic acid, about 34.4% by weight of Niacin and about 0.0003% by weight (3 ppm ) of Biotin. The vitamin component is preferably present in the composition at a lev el of about 0.5% by weight.

Minerals which may be included in the compositions of the inv ention preferably include one or more of the following, especially all of the following: calcium, magnesium and potassium. Other minerals which may be included, e.g. in addition to those listed above, include one or more, and preferably all, of: iron. phosphorus, iodine, zinc, selenium, copper, manganese, chromium and

molybdenum. The term "mineral" as used herein is not intended to encompass sodium chloride. The minerals are in a physiologically tolerable, e.g. absorbable ionic, form. Such forms of minerals for human nutrition are wel l known, examples of which include citrates (e.g. potassium citrate), phosphates (e.g. calcium phosphate) and sulphates (e.g. zinc sulphate). Where the composition is for administration in multiple daily dosages, the absolute amount of each mineral will be typical ly reduced accordingly. For example, a single dosage of the composition of the invention which is intended for administration in six equal dosages per day may contain the fol low ing levels of each mineral (one sixth of the total required daily amount ): calcium - between 50 and 200 mg. e.g. about 64 mg; magnesium - between 20 and 80 mg, e.g. about 32 mg; and potassium - between 100 and 1000 mg. e.g. about 1 75 mg.

In an especial ly preferred embodiment, the composition of the invention contains a mineral component comprising (e.g. consisting essentially of) about 23.5% by weight of calcium, about 1 1 .6% by weight of magnesium and about 64.9% by weight of potassium. The mineral component is preferably present in the composition at a level of about 0.3% by weight. The values given herein for the

2+ amounts of minerals refer to the weight percent of the metal ion, e.g. Ca .

Salt ( i.e. a source of chloride ions, preferable NaCl) is typical l y present in the composition at a level of around 0.5 to 1% by weight, especially around or about 0.75% by weight, especial ly where the composition is a meal replacement composition. For example, a single dosage of the composition of the invention which is intended for administration in six equal dosages per day would typically contain about 0.68 g of salt (calculated as NaCl ). The compositions of the invention are preferably designed to prov ide a daily dosage which comprises a total of betw een 3 and 6 g of salt (calculated as NaCl ) per day, e.g. around 4 g or around 5 g per day.

In addition to the components listed above, other ingredients may be added to alter the properties of the composition. These include flavourings, which may be natural or synthetic, e.g. raspberry flavour, vanilla flavour, strawberry flavour or chocolate flavour; colourings, especially those w hich are designated with an "E" number betw een 1 00 and 199, e.g. riboflavin ( E 101 ) or cochineal (E120); and preservatives, especial ly those which are designated w ith an "E" number between 200 and 299, e.g. sodium benzoate (E21 1).

The compositions of the invention are preferably meal replacement compositions. As such, they will typically not comprise additional nutrient components. Preferably they are essentially free of other components besides those mentioned above. For example, the compositions of the invention may be free from corn flour, wheat flour, corn starch, potato starch, etc. In a further aspect the invention provides a nutrition-containing composition as defined herein.

In a preferred embodiment the composition comprises (e.g. consists essentially of):

a carbohydrate component which provides about 27% of the energy in the composition comprising maltodextnn MGX18 at a level of about 40% by weight, Raspberry 225 L at a level of about 37% by weight, fructose at a level of about 10%) by weight and oligo fructose at a level of about 13% by weight; a protein component which prov ides about 30% of the energy in the composition comprising whey protein 80 at a level of about 55% by weight ; egg white at a level of about 13% by weight and pea protein 90 at a level of about 32%o by weight;

a fat component which provides about 43% of the energy in the composition comprising rape seed oil at a level of about 9% by weight, olive oil at a level of about 79%) by weight, coconut oil at a level of about 9%> by weight, flax seed oil at a level of about 1.5% by weight and marine omega 3/6/7/9 oils at a level of about 1.5%> by weight;

optionally and preferably salt (e.g. NaCl) at a level of about 0.75% by weight of the composition; and

optional ly and preferably vitamins and minerals,

w herein the fat component of the composition has a ratio of omega 3 to omega 6 fatty acids of about 1 :3.

In a particularly preferred embodiment, the composition of the invention comprises:

a carbohydrate component which prov ides about 27% of the energy in the composition comprising maltodextnn MGX18 at a level of about 40% by weight. Raspberry 225 L at a level of about 37% by weight, fructose at a level of about 10%) by weight and oligo fructose at a level of about 13%> by weight; a protein component which provides about 30% of the energy in the composition comprising whey protein 80 at a level of about 55% by weight; egg white at a level of about 13% by weight and pea protein 90 at a level of about 32%o by weight; a fat component which provides about 43% of the energy in the composition comprising rape seed oil at a level of about 9% by weight, olive oi l at a level of about 79% by weight, coconut oil at a level of about 9% by weight, flax seed oil at a level of about 1.5% by weight and marine omega 3/6/7/9 oils at a level of about 1 .5% by weight;

NaCi at a level of about 0.75% by weight of the composition;

a v itamin component at a level of about 0.5% by weight of the composition, the vitam in component consisting essentially of about 97% by weight of an Acerola C extract (which extract preferably comprises about 17% by weight of vitamin C) and about 3% by weight of a multiv itamin, the multivitamin consisting essentially about 25.5% by weight of Vitamin C, about 19.1% by weight of Vitamin E, about 2.7% by weight of Thiamine, about 3.0% by weight of Riboflavin, about 3.8% by weight of Vitamin B6, about 0.0019% by weight ( 1 9 ppm ) of Vitamin B12, about 0.0004% by weight (4 ppm ) of Fol ic acid, about 1 1 .4% by weight Pantothenic acid, about 34.4% by weight Niacin and about 0.0003% by weight (3 ppm ) Biotin; and

a mineral component at a level of about 0.3% by weight of the composition, the mineral component consisting essentially of about 23.5% by weight of calcium, about 1 1 .6% by weight of magnesium and about 64.9% by weight of potassium,

wherein the fat component of the composition has a ratio of omega 3 to omega 6 fatty acids of about 1 :3.

The invention will now be further described with reference to the follow ing non-limiting Examples and Figures in which:

Figure 1 shows - Mean glucose response to a single isocaloric meal of diet HC and diet MC respectively, n = 7. Paired samples two-sided t-tests for: the comparison of the area under the glucose response curve (AUC) 0 1 20 min after a diet A meal (drawn as solid l ine) to after a diet B meal (drawn as dashed l ine) (the AUC is limited by a basel ine draw n between the 0 and 1 20 min values for the respective meal test) (P = 0.032 ), the glucose concentration at 30 min after intake of a diet AHC meal versus 30 min after a diet BMC meal (P = 0.038), and the glucose concentration at 120 min after intake of a diet AHC meal versus 120 min after a diet BMC meal (P = 0.002 );

Figure 2 shows - Summary of the filtration of gene probes, selection of differential ly expressed genes in response to diet HC, diet MC, and overlapping genes between the two diets and the number of up- and down-regulated genes in response to each diet;

Figure 3 shows - Pathways significantly regulated by the separate effect of the HC (red/light-grey upper bars ) and/or the MC diet (blue/dark-grey lower bars). The pathways arc grouped into Metabolic pathways, Apoptosis, Cell cycle, prol iferation and growth, Stress/immunity, and Other. Bars indicates -log(p-values), where the vertical l ine at 1 .3 corresponds to a P = 0.05 cut-off value for pathway significance level; and

Figure 4 shows - Regulation of differentially expressed genes downstream of a selection of transcription factors in response to the HC (AHC) and MC (BMC) diets separately. Genes indicated by green or red labels, respectively, are

significantly down- or up-regulated on transcriptional level for the actual before diet to after diet comparison. Grey labels represent genes not significantly changed in the actual comparison. Genes outlined with blue are members of pathways related to apoptosis, proliferation/cell cycle regulation, or stress/immunity. All relations are curated from human studies and registered in the I PA library ( Ingenuity Pathway Analysis version 8.0; Ingenuity Systems Inc, Redwood City, USA ).

Example 1 - Anti-inflammatory meal replacement diet composition (referred to as diet "MC")

A meal replacement diet was prepared with a defined quantity and qual ity of each of the three main macronutrient components (carbohydrate, protein and fat ). The diet was formulate to contain adequate amounts of fibre, minerals and vitamins as well as to provide specific levels of fats (e.g. a defined ratio of omega 3 to omega 6 fatty acids of 1 :3) and carbohydrates (e.g. a distribution of monosaccharides, disaccharides and complex carbohydrates).

Table 1 below shows the amount of composition required (in grams) to prov ide a 3000 kcal/day diet. Tablc 1 - Composition of MC diet.

¹ HFG, Shandong, China

² Obipektin AG, CM 9200 Bischofszeii, Switzerland

³ Tine BA, 005 1 Oslo, Norway

4 Pi sane M9 - Provital Industrie S.A., Belgium

⁵ capsules from Vi to mega, Heimdal, Norway

⁶ Obipektin AG, CH-3400 Burgdorf Switzerland

Example 2 - Control meal replacement diet composition (referred to as diet "HC")

A control meal replacement diet was prepared to represent a typical western diet, based on Norwegian and US official dietary recommendations. The diet contained the same ingredients as the diet set out in Example 1 , but the amounts of carbohydrate, protein and fat were altered to provide 65 kcal% from carbohydrates, 1 5 kcal% from protein and 20 kcal% from fat. Diet HC contained half the amount of saturated fat and hal f the amount of monounsaturatcd fat compared to diet MC. The amount and type of polyunsaturated fatty acid (PUFA ) in diets MC and HC was exactly the same.

Example 3 - Feeding study

Background

The aim of present study was to investigate if and how the quantity of the macronutrients in an iso- and normocaloric diet would affect chronic disease risk in modestly overweight, otherwise healthy young subjects. In order to do so, two meal replacement diets (MRDs) as defined in Examples 1 and 2 tested in a randomized crossover fashion, meaning that al l subjects consumed both diets where diet order was randomized. Both diets were similar in number of meals per day (6), quality and quantity of m inerals, vitamins, fibre, PUFA (n-6/n 3 ), and ratio

S F A/monou n sat u rated fat (MUFA). Variable components were quantity of proteins and SFA/MUFAs to compensate for difference in carbohydrate calories between the two diets. Transcriptomic, proteomic and metabol ic data col lected from blood cells and blood serum or plasma were analyzed and combined in a systems biological fashion.

Subjects and methods

All subjects gave their w ritten informed consent to participate in the study. The study protocol and the ethical standards employed w ere approved by the Regional Committee for Medical and Health Research Ethics, Central Norway (REK ID 4.2007.5 1 5 ).

A total of 32 healthy male (19) and female (13) volunteers aged 18-30 from the Norwegian University of Science and Technology student populat ion participated in the study (Table 2 ). Study inclusion criteria were: body mass index (BMI; in kg m ) >24.5 or <27.5; blood pressure less than 135 mmHg systol ic and 80 mmHg diastolic; urine negative for protein or glucose dip-stix values; fasting blood glucose level inside normal range, high sensitiv ity (hs)CRP concentration <5 mg/L; w hile exclusion criteria were regular use of prescription medication; receipt of inoculations within 2 month prior to the study or the intention to receive such during the study ; diagnosis of chronic medical condition (e.g. diabetes, cardiovascular disease, anaemia, gastrointestinal disease); symptoms of allergy; planned or current pregnancy or lactation; acute inflammation as assessed on the basis of w hite blood cell count, platelet count or hsCRP; abnormal kidney, l iver or metabolic functions. Tablc 2 - Anthropometric and biochemical measures at baseline (n fasting values).

serum

² EDTA-plasma.

We performed a randomized, controlled, cross-over diet intervention trial. The study design consisted of two randomly assigned study arms stratified by sex. age (< 25 years or > 25 years), and waist circumference (females < 77 cm and males < 86 cm, or female > 77 cm and males > 86 cm at the most narrow part of the waist) using WebCRF (Unit for Appl ied Clinical Research, Norwegian University of Science and Technology, Trondlieim, Norway).

The method of Jorstad et al. (TRENDS in Plant Science (2007 ) 12(2 ):46-50 ) was used to perform sample size estimation for detecting differences in gene expression of diet (measurements 6 days apart ) from pilot data of Brattbakk et al. (OMICS: A Journal of Integrative Biology, June 201 1 , ePub ahead of print ). We found that a sample size of around 20 subjects (calculation performed on men only ) were needed to achieve an average power (expected proportion of correct rejections ) of 0.8 and a positive false discovery rate ( FDR ) of 0.05. This means that we would expect to detect at least 80 % of the regulated genes with control of the multiple type I error, as defined by the FDR, at 5 %. Effect sizes for regulated genes were estimated from the pilot data, using a mixture model, that is, we assumed that effect sizes of interest in the current study are in the same order as in the pilot study.

Two iso- and normocaloric l iquid M RDs composed of 65 : 15 :20 (diet HC, high-earbohydratc, see Example 2) and 27:30:43 (diet MC, m o d e ra t e-c a rbo h yd ra t e . see Example 1) energy percent (E %) of carbohydrates, proteins and fats, respectively, were given as six isocaloric meals per day. This constituted al l nutrient intakes during the study periods. The two study arms started with either study diet HC first or study diet MC first. We col lected fasting blood samples at four time points, before and after each of the two diet periods. Diet periods were separated by eight days with normal eating habits. The study is registered in ClinicalTrials.gov with identifier NCT00733018. Microarray data are submitted to the Array Express (accession number E-TABM-1073)

Composition of meal replacement diets

Two dietary powder-mixes containing the carbohydrate and protein components of the diets of Examples 1 and 2 were produced for the purpose of the study by Food Innovations AS, Lillestrom, Norway. Two different oil mixtures, as discussed above, were mixed by the study coordinator to obtain the desired value of SI As, MUFAs and PUFAs. Other dietary factors such as fibre, PUFAs, vitamins, and minerals were kept constant at the level of the recommended daily allowances (PDA) in both diets. The fibre content in both diets was 25 g per 3000 kcal . The glycemic load (GL) was calculated to be 2.71 times higher in diet HC than in diet MC. We adjusted the diets to be normocaloric to each individual based on calculations of resting energy expenditure using Harris-Benedict equation (Harris et al, supra) multipl ied by the sel f reported average daily physical activity level - PA L (Brooks et al, supra).

Al l food was provided; individually pre-packed with ID-code, diet date and meal number, as six isocaloric meals per day to be mixed with water in a provided shaker. The diets were flavoured w ith raspberry flavour to optimize compliance. Nevertheless, the diets were not blinded because the taste and colour of the diet pow ders and the amount of oil in the two sets of meals were easy to distinguish. However, analyses of blood samples were performed blinded without knowledge of diet code. To the knowledge of the inventors, compliance was about 99.5%, as registered in frequent, at least bi-weekly, personal dialogue between subjects and the study coordinator. Anyone who skipped more than 3 meals was excluded from the study.

Body composition measurements

At each appearance, measures of body composition (body mass, body fat percent, body water percent, v isceral fat rating, muscle mass, physical rating, basal metabolic rate, metabolic age, and bone mass) were recorded by bioelectric impedance analysis (BIA), using Tanita innerscan BC 545 segmental body composition monitor (Tanita Corporation, Tokyo, Japan). We also measured body weight at day 3 or 4 during each diet period, upon the subjects ^' collection of the second half of the pre-packed study meals. An eventual weight deviation within the diet period could thus be compensated half-w ay in the diet period by adjusting the size and caloric content of the remaining pre-packed meals.

Blood sampling

From all study participants venous blood samples were collected between 7.30 and 10.30 a.m. after an overnight fast. An initial sample was col lected as part of the health check screening. Next, we collected samples at four time points;

before the start of each diet intervention period and the mornings after completing six days on the diets. We col lected blood in appropriate blood col lecting tubes for subsequent RNA isolation (K ₂ EDTA-tubes, Greiner Bio-One, Kremsmiinster, Austria) and biochemical analyses (clot separation tubes, Greiner Bio-One). Tubes for serum separation were left to coagulate at room temperature for 30 min before cent rifugat ion at room temperature for 10 min at 2200 x g. Anti-coagulatcd blood for plasma separation was immediately placed on ice and centrifuged at 4°C ( 1 500 x g, 1 0 min ) within 30 min. Unless analyzed the same day, plasma and serum samples were kept on ice, al iquoted, and frozen at -80°C within 2-4 h, until further analysis. Biochemical analyses

The cl inical chemistry analyses were performed at The Department of Medical Biochemistry at St. Olavs University Hospital, Trondheim. Norway. Of these biomarkers, triglycerides, total- and H DL-cholesterol, glucose, haemoglobin. total leukocytes and differential count of leukocytes, platelets, hsCRP, and uric acid were analyzed before and after each diet period. LDL-choiesterol, the ratio of triglycerides to HDL-cholesterol, the ratio of total- to HDL-cholesterol, and the atherogenic index (AIP, the logarithm of the ratio of triglycerides to HDL- cholesterol ) were calculated.

Protein analyses

Bio-Plex Diabetes Panel assay ( Bio-Rad Laboratories Inc., Hercules, CA, USA) was performed using Luminex xMAP™ technology on fasting EDTA-plasma samples from before and after both diet periods from all subjects, with a Bio-Plex 200 suspension array reader and analysis was done with the Bio-Plex Manager 5.0 software ( Bio-Rad Laboratories Inc. ). The analysis principle combines parallel bead-based sandwich immunoassay with flow cytometry in a 96-well microplate format. The diabetes panel is assembled to determine twelve diabetes related biomarkers in the same assay using the same dilution of plasma samples. The panel biomarkers can be grouped into cytokines (interleukin-6 ( I L-6), tumour necrosis factor-a (TNF-a), plasminogen activator inhibitor- ! ( PA 1- 1 )) adipokines (leptin, resist in, ivsfatin ( nicotinamide pehosphoribosyl transferase. Nampt)), gut hormones and incretins (ghrelin (appetite-regulating hormone), glucagon-likc peptide- 1 (GLP- 1), glucose-dependent insul inotropic polypeptide (GIP, Gastric inhibitory polypeptide)), and glucose disposal hormones ( insulin, glucagon, insulin connecting peptide (C-peptide)). We performed the analyses according to the manufacturers ^* protocol with exception of the fol lowing modification : a fifty percent reduction in the number of antibody label led magnetic beads was used. We analyzed all samples from each subject in duplicate in the same assay to minimize inter-assay variation. Low- and high-level controls for all biomarkers were included in al l assays.

Adiponectin ( Mercodia, Uppsala, Sweden ) and Scrum Amyloid A (SAA, human ) ( Invitrogen, Camaril lo, CA, USA ) were analyzed by EL ISA technique in separate assays using EDTA-plasma, on a Dynex DS2 automatic EL ISA instrument (Dynex Technologies, Magel lan Biosciences, Chantilly, VA, USA ).

Postprandial meal test glucose responses

A subgroup of the subjects representing both sexes and intervention regimes performed a postprandial meal challenge on the seventh day following each of the two six-day diet intervention periods. At sampling time point, fasting samples for determination of serum glucose were collected between 7.30 a.m. and 9.30 a.m. Thereafter one study meal identical to all meals in the immediate previous study period was ingested within 5 min. After 30 and 120 min, venous blood samples were collected for subsequent serum glucose determination. We calculated the area under the glucose response curve (AUC) to each of the study meals by applying the trapezoid rule for the intervals 0 to 30 min, and 30 to 120 min. We performed a baseline correction to adjust for the 120 min glucose concentration being lower than baseline, by subtracting the AUC for the 0 to 120 min interval from the sum of

AUCs 0 to 30 min and 30 to 1 20 min. Both the corrected AUCs and the 30 min and 1 20 min postprandial serum glucose concentrations were compared using paired samples t-test. Assessment of insulin sensitivity and resistance

The homeostasis model assessment (HOMA) determines insulin sensitivity (% S), β-cell function (% B) and i nsulin resistance ( IR ) based on fasting blood glucose (in mmol/L) and ( ^' -peptide (in nmol/L). We used the HO VIA 2 calculator version 2.2.2 © ( Diabetes Trials Units, University of Oxford,

http://w-ww.dtu.ox.ac.uk/homacalculator/index.php).

Isolation and quality determination of RNA from blood leukocytes

Leukocytes from EDTA-blood (K ₂ EDTA-tubes, 9 mL) were captured immediately after sampling on LeukoLOCK™ filters using the LeukoLOCK™ Total RNA Isolation System (Applied Biosystems Ambion, Austin, TX, USA ) according to the manufacturers' instructions

(http://w ^, ww ^f . ambion.com/techlib/prot/fm_l 923.pdf). Plasma, erythrocytes and platelets were removed by filtering, while the leukocytes captured on the

LeukoLOCK™ filters were washed, and stabilized in RNAIater (Applied

Biosystems/Ambion) at -80°C until RNA extraction using the alternative protocol http://www.arnbion.com/techlib/misc/leuko_iso.pdf) (version 0602). The isolated RNA was quantified and checked for protein and organic contaminations using NanoDrop® ND- 1 000 spectrophotometer (NanoDrop Technologies, Delaware, USA) at OD 260/280 and OD 260/230 respectively. RNA integrity was checked on an Agilent 2 100 Bioanalyzer (Agilent Technologies Inc., Santa Clara, CA, USA). Microarray analyses

The gene expression analysis was performed by the Norwegian Microarray Consortium in Trondheim, Norway. Briefly, 300 ng of total RNA was processed into cRNA using lllumina* Total Prep™ RNA Ampl ification Kit (Applied

Biosystems/Ambion, Austin, TX, USA ). The cRNA concentration of each sample was quantified using NanoDrop® N D- 1000 spectrophotometer, to ensure that equal amounts of each sample were used for hybridization. Then, 750 ng cRNA was hybridized onto HumanHT- 1 2 Expression BeadChip v3.0 ( ll lumina, San Diego, CA, USA ) using Whole-Genome Gene Expression Direct Hybridization Assay

( lllumina). These BeadChips contain 48,804 probes derived from the National Center for Biotechnology Information (NCBI ) Reference Sequence (RefSeq ) ( Build 36.2, Re! 22 ) and UniGene (Build 1 99) databases, of which 27,455 were annotated in Entrez Gene. The complete set of four samples from each subject was hybridized to the same 1 2-sample BeadChip to minimize intra-subject variation due to inter- array variation. A total of 128 samples from 32 subjects were hybridized on 1 1 BeadChips in randomized three-subject groups. After scanning, the results were imported into ll lumina BeadStudio where the qual ity of each array and scan was tested.

Preprocessing of microarray data

Leukocyte gene expression profiling was done on the HumanHT- 1 2

Expression BeadChip v3.0 ( ll lumina). Analyses of microarray data were performed in the R statistical analysis framework (R Development Core Team, 201 1 ), using the lumi ( Bead Array Specific Methods for ll lumina Microarrays) and the limma library (Smyth et al, Bioinformatics (2005) 21(9):2067-2075) add-on packages from Bioconductor (http://www.bioconductor.org) (Gentleman et al , 2004). and the lllumina add-on package (An R Library for pre-processing ll lumina Whole Genome Expression BeadChips) available from the Walter and Eliza Hal l Institute for Medical research www-pages at http://bioinf.wehi.edu.au/software/iridex.html. After removal of two outl ier samples, background correction based on negative controls, quantilc-quantiic normal ization, signal iog2 -transformation, and removal of not detected or bad probes, 27 372 unique probes were left in the gene expression dataset. Data were analysed using moderated t- tests with indiv idual as block effect in the l imma package. Covariates used were time, diet and the time-diet interaction, sex and which diet that was administered first.

Three contrasts, i.e. the difference between the start-point and end-point of both diet periods, were tested to determine the effect of the HC diet, the effect of diet MC diet, and the difference between the effect of the two diets. FDR adjusted p-values using the Ben j a m i n i - I I ochberg step-up algorithm was produced for each of the three contrasts. Using a FDR cut-off of 5% this did not identify any

differentially expressed genes comparing the effect of the diets. Responding to the HC diet and the MC diet, a total of 3225 and 1370 differentially expressed genes were found, respectively, where 843 genes overlapped.

Pathway analysis was performed using Ingenuity Pathways Analysis 8.5 ( IP A, Ingenuity Systems®, Redwood City, CA, USA, w w w .ingenuity.com ).

Statistical analyses

Analyses were performed using the R statistical analysis framework, and the nlme library (Pinheiro et al , 201 1 ). For each response a linear mixed effects model was fitted with a four levels fixed factor (before and after diet MC and HC ). A random effect for each individual was used. Which sex and w hich diet that was administered first were also included as covariates in the initial analyses, but was not found to be significant and not included in the final analyses. Analyses were conducted either on the original scale (model 1), the log-scale (model 2 ) or the square root scale (model 3). The scale chosen was based on assessment of the normality of residuals using qq-plots and the Anderson-Darling test for normality.

The same three contrasts as for the microarray analysis were investigated.

Estimated effects and 95% confidence intervals on the original scale for the three contrasts are presented in Table 3 below together with p-values and False Discovery Rate (FDR) adjusted p-vaiues. P-vaiues were calculated using t-tests, and the FDR adjusted p-vaiues were calculated using the B e n j a m i n i - H o c h berg step-up algorithm (Benjamini et al, J. Roy. Statist. Soc. Ser. B (1995) 57:289-300) for controlling the FDR for all the 32 responses and 3 contrasts (96 tests). Using a cut-off of 5% for the FDR 22 of the 96 tests are found to be significant. The linear mixed effects degrees of freedom (DF) are reported in Table 3.

Table 3 - Diet responses of markers regulating glucose, adipogenesis, insulin resistance and HOMA indices (Mean for change day 6 - day 0 with 95% CI; n = 32 for al l analyses except glucagon and GLP-1 n = 31 ; paired samples t-test within each diet (AHC - Example 2; and BMC - Example 1) between day 0 and day 6; bold indicates P < 0.05 between time points within diet).

sample matrix is serum

2

sample matrix is EDTA-plasma

sample matrix is EDTA-blood

4 p-values are calculated using logio transformed data because data are not norm a 11 y-d i stri bu t ed .

Results

Anthropometric and biochemical measurements

Thirty-two subjects (13 women and 19 men) completed the randomized cross-over diet intervention of two six-day intervention periods separated by eight days normal dieting. Baseline biochemical and anthropometric characteristics for al l subjects who completed the study are found in Table 2 above. Observed weight changes of more than 1 kg hal fway during any of the diet periods was reason to adjust the quantities of the remaining meals. Even though the mean body mass was reduced (P < 0.001 ) by 1 .02 kg after six days on diet HC (80.3 ± 8.8 to 79.3 ± 8.6 kg) and 0.95 kg (80.3 ± 8.5 to 79.4 ± 8.8 kg) after six days on diet MC, or 0.91 kg during the first diet period, and 1.06 kg during the second diet period, the w eight changes w ere not significantly different between the two diets or intervention regimes. The total w eight loss during the w hole diet intervention (20 days ) was in average 1 .56 kg ( 1 .96 %).

Microarray analysis

Microarray hybridization was performed on leukocyte RNA collected from all 32 subjects before and after each of the interventions, six days each on diet HC and MC. Changes in gene expression w ere determined by comparison of the microarray results of the samples from before the diet intervention (day 0) with those after day 6 of the intervention. Dieting on the HC diet resulted in changes in expression of 3225 genes, whereas the MC dieting resulted in changes in expression of 1370 genes; of these, 843 genes overlapped between the groups. Among the latter, al l except 10 genes changed in the same direction (see Figure 2).

Gene expression changes

In order to determine the role of the various genes that changed expression in response to each diet, pathway analysis was performed. Significant enrichment (P < 0.05 ) of differentially expressed genes was observed in several pathways related to metabolism, apoptosis, cell c y c 1 e/p ro I i fe ra t i o n/gro vv t h , stress/immunity and a variety of other pathways (see Figure 3). In general, the II C diet induced changes in gene expression to a much larger extent (number of genes and higher iog2 ratio) than consumption of the MC diet, including both up- and down-regulation of genes within the same pathway. HC diet consumption led to stimulation of a number of genes including NF-κΒ signal ing (RELA), STAT3, STATS A, STAT5B, SRC, AHR, and TRAIL family member (TNFSF10) regulating both apoptosis,

prol i feration/cancer and stress/immunity, but also more specific regulators of apoptosis such as CASP2, BAX, cIAP (BIRC3), ICAD (DFFA), JAK2, MAP3K14, and DAXX. Upregulated genes relevant to prol iferation and cancer are genes involved in β-catenin signaling (CTNNB1), LEF/TCF (TCF7L2, TCF4), frizzled (FZDl ), CEBPA, and adenylate cycla.se/protein kinase A signaling (PRKACA, PRKACB, ADCY7). Apoptosis-relevant genes downregulated after HC dieting included CASP8, BID, MDM2, FASLG, and CEBPB; while the prol iferation and cancer relevant genes in this category were NOTCH 1, c-myc (MYC), raf (A RAF, RAFl), BRCAl, and SHIP (INPP5D). SHIP and raf are also mediators in the insul in signaling pathway. Together with the PI3-kina.se (PIK3R3, PIK3R2, PIK3R4), atypical protein kinase C (PRKCZ), Akt (RAC2), PPP1CC, ERK1/2 (MAPK1, MAPK3), syntaxin (STX4). GSK3 (GSK3B), and RAPGEF1 they are mediating glucose uptake and glycogen synthesis in insul in target cells. These genes are all down-regulated during HC dieting. MC dieting showed upregulation of genes including gene coding for G protein subunit βγ (GNG2), CDC42, FOX03, VCAMl, of relevance to apoptosis, proliferation/cancer; while down-regulated genes in this category included NF-κΒ signaling (NFKB1), ICAM1, TNFSF12, CCR5, CD40LG, and MMP3, al l belonging to the atherosclerosis signal ling pathway, and gene coding for G protein subunit a (GNA15), IL15, and PAIP2 with relevance to apoptosis, proliferation, cancer, stress, and immunity. Very few genes showed differential regulation by both diets. However, among them were genes for two growth factors; thymidine phosphorylase (TYMP) where a high TYMP expression at tumour sites is correlated with tumour growth, induction of angiogenesis, and metastasis ( Bijnsdorp et a!. , 2008); and granul in (GRN) which contributes to the regulation of early embryogenesis, to adult tissue repair, inflammation, dementia and elevated levels are observed in cancers and dementia ( Bateman et a!., Bioessays (2009)

31(1 1): 1245-1254). Both genes were up-regulated in response to the HC diet, and down-regulated in response to the MC diet. Moreover, RNF4, a regulator of DNA demethyiation whose expression is regulated down following the HC diet and up following the MC diet. Several transcription factor members of the pathways in Figure 3 were identified showing differential expression exclusively in response to either HC or MC diet. The interrelationship among these and their respective downstream effects on apoptosis, prol iferation, cancer and stress immunity is suggested in Figure 4. Approved HUGO (http ://www. genenames.org/) gene nomenclature of all gene symbols is described here in italics, and is presented in Figure 4. Glucose, insulin and lipid metabolism

A significant decrease in fasting blood (serum ) C-peptide levels during the MC diet was observed, compared to baseline, along with glucagon levels. No significant changes in serum glucose or insulin levels were observed upon completion of each of the two six-day diet periods, although a trend for reduction of serum insulin after the MC diet could be observed, see Table 3. A trend in hsCRP concentration increase in response to the HC diet was evident, while the antiinflammatory adiponectin concentration showed a decreasing trend during the same diet. PA I- 1 , and GLP-1 concentrations showed a decreasing trend in response to the MC diet, see Table 3.

Based on changes in fasting blood glucose and C-peptide concentrations in response to the two diet interventions, we applied the homeostasis model assessment ( HOMA2 computer model ), allowing for an insight into the underlying pathophysiological disorders regulating insulin resistance, β-cell dysfunction and adipogenesis. The HO MA 2 computer model for calculation of insulin resistance (HOMA2 IR), β-ce!l function (HOMA2 % B) as well as the insul in sensitivity (HOMA2 % S) is cal ibrated to a reference population, where the HO MA 2 I R is normal ly 1 , and the HOMA2 % B as well as the HOMA2 % S are normal ly 100 %. In fasting samples, responses to six days on diet MC assessed as the mean HOMA2- IR and HOMA2 % B was decreased, while the HOMA2 % S was increased compared to the respective fasting baseline samples (Table 3). The % S and % B must be seen in relation to each other, as an increased insulin sensitivity may be compensated by decreased β-cel l function.

In order to determine the glycemic response to a single meal from each of the two diets, a meal cliallengc test was performed using a random subgroup of the participants at the day follovving each of the two six-day isocaloric diet intervention periods. As shown in Figure 1 , the postprandial serum glucose at 30 min was significantly higher in response to an HC diet meal than after a MC diet meal . At 1 20 min the serum glucose was significantly lower following both meal s than in fasting samples, but the hypoglycaemia was significantly greater in response to diet HC. The corrected AUC for the glucose response was larger for a diet HC meal compared to a diet MC meal.

Blood lipid profiles were changed in response to both diet periods, where triglyceride concentrations were reduced during HC and further reduced during MC. Total cholesterol was decreased during both HC and MC diet periods. LDL- cholesterol was decreased during the MC diet, while HDL-choiesterol was decreased in response to the HC diet. The trigl yceride H D L-chol esterol-rat io was decreased in response to the MC diet, and the atherogenic index (AIP,

Iog l 0(triglycerides HDL-cholesterol-ratio)) was reduced by 53 % in response to the MC diet (Table 3). AIP has been shown to correlate with LDL-particle diameter, and a more negative value is suggested to be a predictor of a marker of a lipoprotein phenotype w ith decreased cardiovascular risk. Blood cytokines and subsets of leukocytes

The pro-inflammatory cytokines TNF-a and IL-6 showed trends of reduced concentrations in response to the MC diet, while resistin was significantly increased in response to both MC and HC diets. The blood concentration of uric acid, which at elevated levels forms crystals accumulating in synovial fluid triggering a cytosolic sensor, the in flam masonic leading to inflammation via activation of NF-κΒ, is significantly decreased in response to both diets. The total number of leukocytes decreased during both diets, while the monocyte subtraction of leukocytes increased during the 11 C diet. The cosinophile subtraction diminished during diet MC (Table 3).

Discussion

We performed a controlled, randomized crossover trial and show that six days consumption of a meal replacement diet with reduced carbohydrate, higher protein and fat (BMC) significantly improves the atherogenic index, main blood cytok i nes ad i poki nes (TNF-a (TNF), 11.6, PA 1- 1 , adiponectin ), and also improves HOMA indices (reduces insulin resistance and improves insulin sensitivity, improves β-cell function ) in sl ightly overweight subjects, compared to a high carbohydrate (ARC) diet. The diet interventions resulted in diet-specific changes in leukocyte gene expression profiles characterized by differential effects on metabolic pathways, apoptosis, proliferation cel l cycle regulation, and stress/immunity. This includes activation of genes for transcription factors NF-κΒ, STAT3/5, ( 'ΤΝΝβ- LEF/TCF, CEBPA in response to ARC and inhibition of gene for NF-κΒ while activation of FOX03 during BMC dieting.

The strength of our study resides in its randomized crossover design, defined diet macronutrient qual ity and quantity, the inclusion of both sexes, and a frequent eating pattern (six meals per day) on a normocaloric diet. The latter is important, since overeating, which refers to the overconsumption of energy, both per meal and day, that is inappropriately large for a given energy expenditure, thus, leads itsel f to obesity. Moreover, the combination of data on blood lipids, cytokines, adipokines and leukocyte transcriptomics allow the identification and a more comprehensive understanding of mechanisms underlying disease risk associated with diet carbohydrate quantity. To the knowledge of the inventors, the current study is unique in that respect.

The meal replacement diet used in this study al lowed for a highly defined meal composition, in contrast to the use of conventional diets, and enabled the definition of the mechanisms involved in progression of lifestyle diseases, which is essential to study specific aspects like macronutrient quantity and quality independently, and in a control led manner, hence the choice of replacement diets and a cross-over design. The number of subjects was chosen based on power calculations from transcriptome expression data in a pilot study.

A health promoting trend was to some extent observed in response to the control (HC) diet, possibly partly due to common beneficial features between the diets (e.g. high fibre, high meal frequency with even distribution of caloric intake, adequate micronutrient and essential fatty acid intake, and weight reduction ).

However, this common health promoting trend was enforced and reached significance in response to the experimental (MC) diet. Weight reduction, increase in PUFA, and optimal n 3 to n 6 balance, share the common property of reducing inflammation. Although the intervention was intended to be normocaloric, a small but significant weight reduction was observed on both diets. However, there was no significant difference in weight reduction between the diets, suggesting that the weight reduction itself does not influence the health promoting trend. Reduction in triglycerides is highly correlated with increase in PUFA, especial ly the iv 3 variety. The daily intake of fish oil capsules and other PUFAs were common in both diets. Also, there was a significant reduction in triglycerides in response to both diets, suggesting a common beneficial effect of PUFAs. Reduction of the pro- inflammatory adipokine, v isfatin and uric acid, was observed in response to both diets. The mean number of meals before entering the intervention was 3.8; during intervention it was 6, so an increase in meal number can be expected to have a beneficial effect on postprandial hyperglycaemia independent of meal composition.

In summary, the two diets share several common inflammation lowering properties which may explain the common trend. However, the difference between the diets in ability to reach significance in improved HO MA indices and atherogenic index, and improv ement in main c y t o k i n c ad i po k i n c concentrations, may be attributed to a shift in relative macronutrient composition. Inflammation is an important early variable in the metabol ic response to diet via postprandial hypcrglycaemia. Compared to the control diet, the experimental diet significantly decreased the 30 min postprandial hypcrglycaemia and postprandial hypoglycaemia at 2 hours. This correlated with improvements in HOMA indices, demonstrating increased insulin sensitivity, reduced insulin resistance and improved β-cell function. Leukocytes have active metabolism and insulin receptors, and we have shown that monocytes are responsive to insul in stimulation (Arbo et al. ,

Scandinavian Journal of Clinical and Laboratory Investigations (201 1) 71(4):330- 339). However, knowledge of insul in action in leukocytes is sparse, although it has been speculated whether impaired presence of GLUT4 presence on the monocyte ceil surface could be an indicator of systemic insulin resistance. During HC dieting, the gene expression of several insulin signal ing mediators was down-regulated, suggesting a change in the insulin sensitivity homeostasis in leukocytes.

Accompanying these diet differences, we observed a significant improvement in the most important cytokine/adipokines regulating insulin resistance (TNF-a, 11.-6, PA I- 1 , and adiponectin ) in response to MC dieting. Metabolic inflammation in adipose tissue is regulated by subsets of leukocytes, most importantly monocytic cells, but also by T lymphocytes. Among the few changes in leukocyte subsets in response to dieting, we note a significant increase in monocytes in response to the HC diet, a pro-inflammatory property. Al l these changes suggest that compared to the HC diet, the MC diet elicits a more anti-inflammatory response. At the transcriptional level this was reflected in that pathways belonging to the processes apoptosis,

proliferation 'cell cycle regulation, and stress/ immunity were among the most notable pathways significantly exhibiting gene expression changes due to the two diets.

Also, several transcription factors showed differentially regulation including NF-KB, STAT3 and 5, CEBPA/B. Among these, the most prominent regulator involved in all processes is NF-κΒ. It was activated in response to the HC diet and inhibited in response to the MC diet, suggesting an important role in regulating the early diet specific changes in the current study. FOX03 has important, but cell type specific regulatory roles in leukocytes. In mice, FOX03 inhibits NF-κΒ and thereby T cel l activity. How ever, in neutrophil ic inflammation FOX 03 is needed to sustain the pro-inflammatory environment, through suppression of FAS LG and neutrophil apoptosis. FOX03 was activated in response to the MC diet.

Inflammatory conditions in selected organs increase the risk of cancer.

Inflammatory components in the microenvironment of tumours include leukocytes, cytokines, and complement components, orchestrated by transcription factors, such as NF-KB and STAT3. As cancer may be seen as the consequence of deregulation of the balance between proliferation and apoptosis, there may be important roles for diet composition in cancer development in certain types of cancer. The current study showed that genes for the transcriptions factors NF-κΒ, STAT3, and other tumour promoting factors (SRC. TP. CRN) are activated and BRCA1 (a tumour suppressor gene regulating breast cancer development ) is inhibited in response to the HC diet. The transcription factor CEBPA has the capacity to induce monocytic maturation and was induced in response to the HC diet suggesting that it may contribute to the monocyte driven inflammation in circulation and remote tissue, e.g. adipose tissue. Thus, there are reasons to bel ieve that the metabolic changes taking place in response to dieting include transcription factors, proteins, lipids and other metabolites and that these may be used to identify the link between postprandial hyperglycemia, inflammation and insul in resistance.

The results presented above show that a reduction in diet carbohydrate quantity towards a caloric macronutrient balance seems to be an important single factor in reducing postprandial h y p e rg I y c e m i a - i n d u ced systemic low-grade inflammation, associated with lifestyle disease development and supports the use of the compositions of the invention in the treatment of lifestyle diseases.