Garlic and cardiovascular disease

The official wording from the Food and Nutrition Board of the National Academy, USA, defines functional foods as any food or ingredient that may provide a health benefit beyond the traditional nutrients it contains (Thomas and Earl, 1994). The European view states that a food can be said to be functional if it contains a component (whether or not a nutrient) that benefits one of a limited number of functions in the body in a way that is relevant to either the state of well-being and health or the reduction of the risk of a disease or it has a physiological effect (Clydesdale, 1997; Bellisle et al., 1998). Garlic clearly falls into the functional food category since it is normally eaten in small amounts either raw, cooked or as a salt. It is also sold as a supplement in the form of powdered tablets, capsules, steam-distilled oil or plant extract, all of which can be taken on a daily basis and guaranteed not to produce the strong smell associated with consumption of the raw garlic.

While the garlic may be taken as a pharmaceutical now, it has been used as a flavouring and food additive for many centuries, as described in detail by Rivlin (2001). The earliest reference is from 2600±2100 BC on Sumerian clay tablets. Its subsequent use was reported by the Egyptians and the Greeks to improve stamina generally, but it was the Chinese who prescribed it for specific illnesses, and in this case it was as an aid to digestion and respiration and more particularly for diarrhoea and worm infestation. In India early texts described its use in the treatment of heart disorders. In Europe it was of value in the treatment of a wide variety of ailments, which ranged from digestive disorders to worm infestation, kidney problems, toothache, constipation, dropsy and plaque (Rivlin, 2001). Fresh garlic is still prescribed for the treatment of many diseases, including the prevention of heart and circulatory problems (Grieve, 1998).

Garlic has received renewed prominence recently because of the popularity of the Mediterranean diet, which includes the extensive use of fresh and cooked garlic. The Mediterranean diet is thought to reduce the risk of cardiovascular disease through a reduction in cholesterol and its consequent reduction in blood pressure. There is extensive Web page documentation at for example http:// www.fair-flow .com and http://www.deliciousitaly .com/mediterraneancom1.htm of the benefits of the diet and in supporting clinical trials (de Lorgeril et al., 1996; Laino, 2003), but the exact cardioprotective mechanism of the diet is unknown. Since the Mediterranean diet includes a wide array of vegetables, fruit, fish and oil, the protective effect is likely to be multifactorial. Garlic is an essential component of the diet and is thought to make an important contribution to its medical benefits.

The current enthusiasm for herbalists’ remedies and healthy diets has led to detailed studies of the chemical composition of garlic and the medical importance of the chemicals it contains. The commercial direction is now away from the fresh plant and into a variety of garlic extracts. The long-term aim of the supplement industry is either to identify an active single compound in the extract that can then be used as a new pharmaceutical or to use the complete plant extract in concentrated form as a food supplement. It is assumed that since garlic has been prescribed for many centuries, there is no or less need for the supplements to undergo the rigorous (and very expensive) testing that is required for the introduction and acceptance of a new pharmaceutical. Garlic supplements and extracts are advertised extensively on the Web where the accompanying text claims that regular consumption of the supplement could delay the onset of cardiovascular disease. However, the supplement industry is now under increasing scrutiny from external regulators to justify their claims, and is looking towards the scientists to provide medical evidence that garlic can in fact prevent the early onset of specific diseases. There have been a large number of studies recently in which the effect of garlic on specific stages of cardiovascular disease has been explored. As an indication of the interest in the subject there has also been a number of reviews (Ernst, 1987; Orekhov and GruÈnwald, 1997; Bannerjee and Maulik, 2002; Rahman, 2001, 2003) describing the relationship between garlic and cardiovascular disease.

Raw garlic is eaten in small amounts largely as a flavouring, and although popular especially in the Mediterranean diet, it does lead to a strong odour on the breath. The flavour and pungency is a result of a high concentration of sulphur-related compounds in the cloves which amount to 1.0 per cent of their dry weight. In the undamaged clove, over 70 per cent of the sulphur compounds exists as (+)-S-allyl-L-cysteine sulphoxide or alliin, (+)-S-trans-1-propenyl-L-cysteine sulphoxide or isoalliin, and S-methylcysteine sulphoxide or methiin and -y glutamyl peptides, such as -y-glutamyl-S-allylcysteine and -y-glutamyl-S-trans­1-propenylcysteine. The sulphur is divided approximately 50 per cent between the cysteine sulphoxides and the -y-glutamyl peptides (MuÈtsch-Eckner et al., 1992). In the plant the function of the cysteine sulphoxides is probably one of protection since their breakdown products appear to have antibiotic and fungicidal properties (which also contributes to their medicinal value), while the -y-glutamyl peptides appear to have a storage function for N and S. The major alkylcysteine sulphoxide in garlic is alliin (85 per cent), with isoalliin (5 per cent) and methiin (10 per cent) occupying much more minor roles (Lawson, 1996).

Garlic cloves have a slight smell when intact but on cutting or maceration a strong smell is released rapidly. The volatile is only released on damage because there are different parts to the flavour mechanism: a degradative enzyme, alliinase, and its substrates, the alkylcysteine sulphoxides, which are stored in separate compartments in the plant. Unlike onion where there is an intracellular separation of the alliinase and the alkylcysteine sulphoxides (Lancaster and Collin, 1981), in garlic there is a spatial separation based on location within different tissues of the clove (Ellmore and Feldberg, 1994; Wen et al., 1995). The amounts of alliinase in the garlic are unusually large, amounting to at least 10 per cent of the total protein in the clove and would explain why there is such a rapid release of flavour and complete breakdown of the flavour precursors upon cutting the tissue.

The chemistry of the Alliums is interesting, since from a small number of original flavour precursors a large number of compounds are produced, many of which have medicinal properties (Block, 1985; Lawson, 1996). In the alliinase­catalysed reaction, the first products of the reaction are the sulphenic acids, which have a very short half life before being condensed to form thiosulphinates. From the three original sulphoxides, eight or nine possible thiosulphinates are formed. The major one derived from alliin is diallyl thiosulphinate, or allicin, which represents about 70 per cent of the total thiosulphinates. Isoalliin does not form the lachrymatory compound found in onion but is rapidly converted into allyl 1­ propenyl thiosulphinate and the methiin finally degrades to largely allyl methane thiosulphinate. The thiosulphinates are colourless liquids with a pungent smell reminiscent of fresh cut garlic, of which allicin is the main contributor to the odour. The allicin yield is important since it is used as a measure of garlic quality in commercial preparations. It appears to be relatively stable especially in water, but in homogenates it is less stable, which is a major consideration when assessing the effectiveness of different garlic preparations.

The thiosulphinates are described as unstable and reactive and convert readily to more stable compounds, which contain the thioallyl (S-allyl) or thiomethyl group. In crushed garlic at room temperature the reactive thiosulphinates are transformed to diallyl sulphide, diallyl disulphide and allyl methyl trisulphide, which is also the reaction that occurs in the mouth. In vitro studies of the thiosulphinates, allicin, and methyl and propyl thiosulphinates show that they are capable of combining with the sulphydryl group of the amino acid cysteine to form S-allyl mercaptocysteine, S-methyl mercaptocysteine and S-propyl mercaptocysteine, thereby inhibiting the activity of enzymes which contain cysteine as the active site. This ability to react with the sulphydryl group of acetyl-CoA SH, the building block of cholesterol and triglyceride synthesis, may help to explain its biological effect (Wills, 1956). These reactions may have some significance in the intestinal tract during digestion as they may determine the form in which the breakdown products of garlic enter the bloodstream. There are no conclusive data that clearly identify the main metabolites in the bloodstream after garlic consumption (Amagase et al., 2001). Evidence has shown that allicin, which is an unstable product of alliin breakdown and could be present in the digestive system after consumption of garlic or garlic supplements, is rapidly converted to diallyl disulphide in the blood (Freeman and Kodera, 1995). This compound was found in micromolar concentrations in the rat plasma and liver tissues after oral administration of diallyl disulphide implicating it as a major garlic-derived metabolite in the body (Germain et al., 2002) but it does not seem to remain in the body for long after consumption of raw garlic. The water-soluble compound S-allylcysteine, itself a breakdown product of garlic, does seem to be a stable residue in the blood following oral consumption (Nagae et al., 1994).

Garlic is normally cooked, so the composition will differ from the raw garlic. Heat inactivates the alliinase and therefore inhibits the formation of allicin and other thiosulphinates, but some breakdown of the alliin occurs leading to the accumulation of small amounts of diallyl trisulphide and di- and tetrasulphides. In crushed garlic the cysteine sulphoxides are largely converted to the thiosulphinates (Lawson, 1996). After boiling in a closed container there was complete conversion of the thiosulphinates to sulphides, whereas in an open container over 90 per cent of the sulphides were lost. After cooking in hot fat, most of the allicin was lost but a much higher proportion of the sulphides remained. In an analysis of the volatiles released after standard periods of cooking such as oil frying, baking and microwaving, the dominant volatile was diallyl disulphide (Yu et al., 1993).

The commercial forms of garlic that are sold as a supplements include powdered dried garlic cloves, oils produced upon treating chopped garlic with steam, vegetable oil or ether, and aged extracts of chopped garlic in ethanol or water, as well as garlic cloves pickled in vinegar. An important aspect of the supplements is that the composition of the product will vary according to variety, the year of harvest and on the processing method. The preparation and composition of the major commercial products are described below.

Garlic powder such as Kwai is prepared by peeling the cloves, oven drying at 50±60 ëC then grinding to a powder. Some conversion of the flavour precursors by alliinase does occur on cutting but the loss is minimised by reducing the amount of cutting before drying. Alliinase retains its activity since the powder is very stable and will only lose about 10 per cent of its allicin yield after 5 years of storage. However it is important that the powder is stored at no more than 4-6 per cent water content to prevent the alliinase being activated. The powder can be stored as tablets when again the allicin potential is also very stable but with an average loss of 36 per cent allicin over 5 years is more variable than the powder. Garlic salt products generally have a lower yield of allicin than quality garlic, which probably reflects the greater degree of chopping prior to dehydration (Lawson and Hughes, 1992).

In order to standardise the various dried products, the quality is measured by the level of allicin. The values of some brands are 4 mg/g which compares favourably with the levels in fresh garlic. An alternative measure is the ability to form allicin from alliin (called the allicin potential), which depends on the presence of active alliinase in the tablet. Alliinase is inactivated by the acid conditions of the stomach. In order to provide the body with the pharmacologically active forms in the garlic, the activity of the alliinase must be maintained until the pills reach the non-acid environment of the intestine. The pills are protected from the stomach acid by being coated with cellulose esters which require the presence of intestinal enzymes for the coating to be removed. This has given rise to another measure, the effective allicin yield, which is obtained by using simulated gastrointestinal conditions, i.e. 1 hour in simulated gastric conditions and 2 hours in simulated intestinal conditions (Lawson and Hughes, 1992). Those brands with an effective enteric coating were those with the largest release after 1 hour. Other factors determining the amount of allicin released will depend on whether the tablets are consumed with or without food. A meal with a high protein content will see the pH of the stomach rise from 1.5 to 4.0, which will be less inhibitory to any alliinase activity. It is recommended that the tablets are taken with or just after a meal when the pH is higher and the alliinase-inactivating ability is lowered (Blania and Spangenberg, 1991).

The other factor determining the release of allicin is the ability of the alliinase to convert all the alliin to allicin. This is measured by the ratio of allicin to alliin. With a 100 per cent conversion this should appear as 0.46, whereas the range normally measures from 0.32 to 0.42 because allyl methyl and allyl 1-propenyl thiosulphinates are also formed. Ratios below this figure indicate a damaged alliinase. The usual reason for a reduced figure is due to excessive drying of the cloves at a high temperature which inactivates the alliinase. Significant amounts of -y-glutamylcysteines are also present in the powders and tablets (Block, 1996).

Steam-distilled garlic contains exclusively about 98 per cent allylmethyl and 1­ propenyl mono- and polysulphides. The commercial product is diluted by vegetable oil so that the final composition of sulphides represents about the same amount of allicin and other thiosulphinates present in a similar weight of crushed garlic. This dilution stabilises the polysulphides and decreases the extremely strong odour of the undiluted oil. The major compounds are the diallyl di-, tri­and tetrasulphides and the allyl methyl di-, tri- and tetrasulphides. The total content of different brands does vary as a result of the different levels of dilution whereas the percentage composition does not. The composition is also relatively stable over time as a 5-year study showed virtually no change in the amount of the main allyl sulphides (Lawson, 1996).

Garlic is macerated in a vegetable oil, and the oil is then filtered or the crushed garlic is left suspended in the oil. The product contains compounds not found in the fresh or powdered garlic, i.e. the ring-structured vinyldithiins and the oxygenated ajoenes. The composition seems not to vary but the total amount of sulphur compounds is dependent on the quality of the original garlic and the amount of dilution by the oil. If the garlic bulb is homogenised with an equal weight of oil, the highest amount of transformation products is about 3±5 mg/g. The yields rarely reach this value but can be increased by crushing the cloves first to produce the thiosulphinates more efficiently before addition of the oil. The oil-based compounds appear to be stable since vinyldithiins were stable over a 5-year period but the ajoenes much less so, even when enclosed in a gelatine capsule. It is recommended that oil-macerated products should be kept refrigerated to retain their ajoene concentration for more than 18 months (Iberl et al., 1990).

Chopped garlic is incubated in 15±20 per cent ethanol for up to 20 months at ambient temperatures as in aged garlic extract (AGE) (Lawson and Wang, 1995). The incubation medium is then filtered and evaporated to dryness and sold in dry form as tablets, powder or liquid forms. Analysis showed that allicin and other thiosulphinates were nearly depleted by 90 days, having been transformed into diallyl and allyl methyl tri-, di- and tetrasulphides, most of which was then lost to the atmosphere. There are considerable amounts of alliin in the extract, which had diffused out, whereas the alliinase had remained in the cells. Under these conditions the -y-glutamyl-S-allylcysteine and the -y-glutamyl­S-1-propenylcysteine were converted to S-allylcysteine and S-1-propenyl cysteine under the action of an enzyme -y-glutamyl peptidase. The quantity of these products will depend on the content of the original ry-glutamyl cysteine sulphoxides. The range in different sources varied from 1.6 to 6.8 mg/g fresh weight, which in turn produces an amount of S-allylcysteine content of 2.7± 11.3 mg/g dry weight which represents the approximate amount that should be found in commercial products. The amount of S-allylcysteine that is found in commercial products is much smaller than this figure. The alliin content of commercial aged extracts varies from 0.02±0.32 mg/g dry or fresh weight and only trace levels of the allyl sulphides have been found in commercial extracts. The total sulphur content of the liquid form of the commercial aged extract such as AGE was found to be 0.091 per cent in contrast to the 0.35 per cent typically found in garlic cloves.

Cardiovascular disease includes atherosclerosis, hypertension and myocardial infarction or heart attacks, and as such forms the major cause of death in industrial societies. Atherosclerosis is a disease of the arteries where the inner layer becomes thickened by fatty deposits and fibrous tissue, leading to a condition known as hardening of the arteries. Fatty streaks, which are the earliest indication of atherosclerosis, are areas of yellow discoloration on the inner surface of the artery, but do not protrude into the lumen or disturb the blood flow. The streaks are characterised by the subendothelial accumulation of large foam cells filled with intracellular lipid. The foam cells, which are derived from macrophages, and smooth muscle cells are the likely precursors of fibrous plaques, structures that form pale grey elevated lesions, which may project into the arterial wall and reduce the blood flow through the vessel. Calcification of the fibrous plaque leads to rigidity of the artery and hypertension while rupture of the plaque releases material into the bloodstream, causing a thrombus to form. Occlusion of the vessel locally or following transport to distant sites can lead to myocardial infarctions or strokes (Bhattacharyya and Libby, 1998).

Atherosclerosis has a genetic component, which is difficult to overcome, and an environmental and diet-related component, which is susceptible to modification. Regular consumption of garlic as in the so-called Mediterranean diet or the use of garlic supplements is thought to prevent early onset of atherosclerosis and consequently delay the possibility of hypertension, stroke and heart attack. There are a number of stages during the development of atherosclerosis at which garlic consumption has a delaying effect. These stages are the synthesis of cholesterol, oxidation of cholesterol, platelet aggregation, modification of the arterial cell walls and hypertension. Investigations of the role of garlic on these stages have been based on an in vitro approach using the response of blood cells or arterial cell cultures to garlic supplements and an in vivo approach involving animal and human trials where the supplement has been consumed regularly. The humans trials are difficult to undertake since there is always the problem of achieving a satisfactory level of replication. Decisions have to be made about the number of volunteers, their age, sex, level of atherosclerosis, duration of trial and type of garlic supplement. The next part of the review provides examples of the in vitro and in vivo approach used to examine the effect of garlic on specific stages of the disease.

The level of cholesterol in the blood is an important factor in the development of atherosclerosis. When fats are ingested as part of the diet, cholesterol and triglycerides are absorbed in the intestine and finally transferred to the venous circulation. These large molecules are hydrolysed by the enzyme lipoprotein lipase, which releases fatty acids into peripheral tissues while the metabolic remnants composed largely of cholesterol remain in the circulation. The liver in an endogenous cycle of cholesterol production and metabolism releases very low-density lipoprotein (VLDL) into the circulation. Lipoprotein lipase acts on VLDL at muscle cells and adipose tissue to release free fatty acids into the cells as before and the residue, intermediate-density lipoprotein (IDL), which contains esterified cholesterol remains in circulation. Further processing results in cholesterol-rich low-density lipoprotein (LDL) which is largely taken up by the liver. Cholesterol released back into circulation is transported by high-density lipoprotein (HDL) which returns the cholesterol to the liver via IDL and LDL for recycling into lipoproteins or excretion in the bile. The HDL appears to act in a protective role, while elevated levels of LDL correlates with a high incidence of atherosclerosis.

The level of cholesterol, particularly LDL, is critical (Bhattacharyya and Libby, 1998) The control of cholesterol synthesis is determined by the enzyme hydroxylmethyl glutaryl CoA reductase (HMG CoA reductase) in the liver and other enzymes in the blood. Inhibition of these enzymes, possibly by the presence of garlic-related compounds provides an important route for slowing down development of atherosclerosis. The consumption of garlic may inhibit the HMG CoA reductase for instance by interfering with the signal transduction pathway such as the AMP-dependent kinase pathway, or adenosine-induced signalling (Gebhardt and Beck, 1996). Much of the approach has been based on establishing an atherosclerotic effect in the blood of whole garlic extracts or separate components of the extracts using in vitro and whole organisms studies.

In vitro studies have concentrated on the ability of blood cells to accumulate cholesterol, as this provides the vehicle for the movement of cholesterol about the body. The response of rats to a high-cholesterol diet with and without the garlic supplement AGE showed that plasma concentrations of total cholesterol and triacylglycerol of the AGE supplemented rats were 15 and 30 per cent lower respectively than those of the non-supplemented rats (Yeh and Liu, 2001). Subsequently, rat hepatocytes were used to determine the role of garlic in cholesterol biosynthesis in vitro by measuring the incorporation of C14 acetate into cholesterol in the presence of different crude fractions from raw garlic (Yeh and Yeh, 1994). These were the water extractable fraction (WEF), methanol-extractable fraction (MEF) and petroleum ether-extractable fraction (PEF), Kyolic, a liquid form of AGE and SAC (S-allylcysteine). The rates of C14 acetate incorporation into cholesterol were depressed 44, 56, and 64 per cent by MEF, PEF and WEF respectively, suggesting that water- and lipid-soluble components were effective in inhibiting cholesterol production.

The results showed that water-soluble compounds such as SAC were more effective than the sulphides but maximum activity was exerted by the complete extract, Kyolic. The effect of the individual compounds, i.e. SAC, SEC (S-ethylcysteine), SPC (S-propylcysteine), SMC (S-methylcysteine), GSAC (y-glutamyl S-allylcysteine), GSMC (y-glutamyl S-methylcysteine), GSPC (y-glutamyl S-propylcysteine), SAAC (S-allylacetylcysteine), SASA (S-allyl sulphonylalanine), SAMC (S-allyl mercaptocysteine) and alliin were assessed. Lipid-soluble compounds were diallyl sulphide (DAS), diallyl disulphide (DADS), diallyl trisulphide (DATS), dipropyl sulphide (DPS), dipropyl disulphide (DPDS) and methyl allyl sulphide (MAS). The cells were treated with C14 acetate in the presence or absence of the test compound at 0.05±4.0 mmol/L for measurement of cholesterol synthesis as before. Among the water soluble compounds, SAC, SEC and SPC exhibited dose-dependent inhibition on the rate of cholesterol synthesis with maximal 40±60 per cent inhibition achieved at 2.0±4.0 mmol/L. Glutamate derivates, GSAC, GSMC and GSPC, depressed the synthesis by 20±35 per cent. Alliin SAAC and SASA had no inhibitory effect whereas the sulphides, DADS, DATS and DPDS diminished the rate of acetate incorporation into cholesterol at 1.0, 2.0 and 4.0 mmol/L (Yeh and Liu, 2001). On the basis of these results, SAC was regarded as a major factor responsible for the cholesterol-lowering effect seen in the human intervention and animals studies.

A typical human trial to show the cholesterol-lowering effect of garlic supple­ments involved a double-blind randomised, placebo-controlled intervention study of free-living hypercholesterolemic men (34; 48.2 Ô 0.8-years-old). The subjects were divided into two groups to receive garlic extract or placebo as dietary supplement for 5 months. The garlic group consumed nine capsules a day, each containing 800 mg of AGE, whereas the placebo subjects took nine capsules, each containing 800 mg of a common food ingredient. Lipids were assessed after 2, 4, and 5 months (Yeh and Liu, 2001). Plasma concentrations remained unchanged 2 months after the supplements. After 5-months supple­mentation reduced the mean plasma LDL-cholesterol concentration by 10 per cent from its baseline value. Plasma concentrations of HDL-cholesterol and triacylglycerol remained constant. Examination of food intake and weight of subjects indicated that the changes were not due to a change in diet or life style and were therefore due to the garlic supplement.

In a further trial using a diffent source (Kwai), garlic appeared to have a short-term and a long-term effect (Orekhov et al., 1996). Two and four hours after a single dose of Kwai (one tablet containing 300 mg garlic powder), the atherogenicity of sera as measured by cholesterol levels taken from patients with coronary atherosclerosis was decreased. After 3±4 weeks of long-term Kwai therapy blood serum atherogenicity was significantly lower compared with the initial level. In the control group on a placebo the atherogenicity of sera obtained from patients on a placebo was unchanged. The atherogenicity was considerably reduced after 4 weeks, indicating an initial rapid reduction and a long-term decline. From this evidence, garlic does appear to have an effect on lowering the cholesterol level in the blood.

The risk factor in atherosclerosis, such as high LDL or low HDL concentrations can lead to excess cholesterol available being taken up by the intimal layer, which is the inner layer lining the lumen of the arteries. High LDL predisposes the arteries to endothelial dysfunction by making them more permeable to the transport of LDL. Once within the intima, LDL accumulates in the subendothelial space by binding to components of the extracellular matrix. This trapping increases the residence time of LDL within the vessel wall where the lipoprotein may undergo chemical modifications. The LDL becomes oxidised by local free radicals and as oxidised LDL it attracts circulating monocytes to the vessel wall. The modified or oxidised LDL can be ingested by macrophages contributing to the development of foam cells. Following oxidation of the LDL, the next stage is the attraction of leucocytes, primary monocytes and T lymphocytes. After the monocytes have adhered to the luminal surface, they may penetrate the subendothelial space by slipping between the junctions. Once localised beneath the endothelium, monocytes differentiate into macrophages, the phagocytic cells that are able to ingest oxidised LDL. The macrophages then become lipid-laden foam cells ± the primary constituent of the fatty streak. More recently, oxidised LDL has been recognised as playing a more important role in vascular dysfunction leading to atherosclerosis than native LDL (Bhattacharyya and Libby, 1998). Garlic seems to play a key role in this stage of the disease.

The approach has been to use an in vitro system to show whether garlic supplement can prevent or reduce the oxidation of LDL. In the in vitro cell-free system, CuSO4 was used to oxidise LDL and the product, thiobarbituric acid (TBARS), measured after 24 hours incubation in the presence and absence of the garlic supplement, AGE (Lau, 2001). The supplement exerted a concentration-dependent inhibition of Cu2+ induced oxidation of LDL. All four water-soluble compounds derived from garlic, N-acetyl-S allylcysteine, S-allylcysteine, alliin and allyl mercaptocysteine showed significant inhibition of LDL oxidation. In a further in vitro test vascular endothelial cells, exposed to Ox-LDL showed signs of cell membrane damage by a significant increase of lactic acid dehydrogenase (LDH) release and a decrease of methylthiazol tetrazolium (MTT) absorbance, indicating mitochondrial injury. Pretreatment of vascular cells with AGE and SAC minimised these Ox-LDL induced parameters of cellular injury. These garlic compounds also inhibited Ox-LDL induced lipid peroxidation implicating lipids as the principal target in Ox-LDL mediated cellular injury.

In a small-scale study (a double-blind placebo-controlled crossover study involving eight subjects, four men and four women, mean age 68), four participants took 1.2 g AGE three times a day for 2 weeks, then 2 weeks of no garlic (washout period) followed by 2 weeks of placebo. The remaining four subjects took a placebo for the first 2 weeks, followed by 2 weeks washout and 2 weeks of 1.2 g AGE three times a day (Lau, 2001). Blood was drawn at the beginning of the experiment, at 2, 4 and 6 weeks and when the experiment was completed. Plasma LDL was isolated and the CuSO4 test repeated. The use of garlic supplements was found to significantly increase the resistance of LDL to oxidation.

One of the consequences of the formation of foam cells and subsequently the fibrous plaque is rupture of the plaque-releasing material into the bloodstream, causing a thrombus to form and platelet aggregation to occur. Aggregation is a consequence of exposure of fibrinogen receptors on the surface of cells. These receptors bind fibrinogen in the presence of extracellular Ca2+ and crosslink the platelets to form aggregates. The fibrinogen is a heterodimer of the membrane glycoproteins (GP) IIb and IIIa and although unstimulated platelets express the GPII-III complex at their surface, this complex is unable to bind fibrinogen until platelets are activated for example by ADP. The GPIIb-IIIa receptor has a high content of ±SH groups and binding of fibrinogen is inhibited by the organosulphur compound ajoene, which is a component of some garlic extracts. A reduction in aggregation would decrease the risk of thrombosis. The approach has been to test the blood of trial volunteers for the ability to aggregate in response to a number of chemical compounds known to stimulate this process.

In contrast to earlier trials referred to, the importance of aggregation was investigated in 34 normolipodaemic individuals (Steiner and Li, 2001). These volunteers were recruited for a 44-week-long double-blind crossover study. No supplements were given for the first 6 weeks, then participants were selected to receive either AGE or a placebo in a dosage of three capsules/day (each 800 mg) for a period of 6 weeks. The dose was then raised to six capsules/day for 6 weeks and finally to nine capsules for another 6 weeks. The first intervention period was followed by a 2-week washout period. The subjects were then switched to the supplement they had not received during the first arm of the study. A final 2­ week washout period concluded the study. Blood was sampled every 2 weeks and processed for platelet aggregation and adhesion studies. Aggregation was tested with the following agonists which would normally encourage aggregation: arachidonic acid, ADP, collagen and epinephrine. Each platelet stimulant was used in a range of concentrations to determine the threshold levels required to induce complete aggregation. Aggregation studies using ADP as a stimulant showed minimal increases. Collagen and epinephrine-induced aggregation was significantly inhibited by AGE. All individuals consuming AGE showed an increase in SAC in the blood, which dropped when AGE ceased to be consumed.

In a contrasting investigation on platelet aggregation in humans, 23 healthy and normolipodaemic participants (12 men and 11 women) consumed 5 ml AGE daily for 13 weeks, blood samples were taken before and after the last dose. Platelet aggregation was then measured (Rahman and Billington, 2000). The extent and rate of aggregation in response to ADP were reduced after dietary supplementation with 5 ml AGE. Serum lipid concentrations were not affected. It is thought that the effect of AGE was to inhibit the ADP-induced rise in cytosolic Cu2+ concentrations. Both trials suggested that regular consumption of garlic supplements would reduce the level of aggregation in the blood and therefore the risk of thrombosis.

The most important consequence of plaque formation is that it leads to a restricted blood flow and hardening of the arteries. Resistance in the arterial circulation is increased and there is a risk that the plaques might break, releasing material into the bloodstream and causing a thrombus to form and strokes to follow. Garlic consumption appears to have a direct effect on the artery wall by reversing or reducing wall hardening. The approach has been to examine the response of cell cultures of arterial cells to compounds such as LDL, then the arteries of animals fed a high-cholesterol diet for lesion development.

Free cholesterol and cholesterol esters are present in arterial cells and contribute to the eventual formation of plaques. Uptake and metabolism of cholesterol by arterial cells are critical to the final stages of atherosclerosis. The control of this process has been examined in vitro by using the response of cultured smooth muscle cells to cholesterol. Cultures of such cells derived from a fibrous plaque of an atherosclerotic human aorta loaded with cholesterol were incubated for 24 hours with an aqueous extract of garlic powder. Free cholesterol decreased by 30 per cent, cholesterol esters by 30±40 per cent and triglycerides by 20 per cent (Orekhov et al., 1995; Orekhov and Tertov, 1997). This change is achieved because garlic appears to suppress lipid synthesis in the cells by inhibiting the activity of acetyl-CoA:cholesterol acyltransferase (ACAT), the enzyme involved in the formation of cholesteryl esters, the main component of the excessive fat accumulated by cells. In the untreated atherosclerotic cells, ACAT activity was three-fold higher than in normal cells. The aqueous extract of garlic cells decreased this enzyme activity to normal levels. Garlic extract also stimulated cholesterol ester hydrolase, an enzyme that degrades and therefore reduces cholesterol esters in atherosclerotic cells (Orekhov and Tertov, 1997).

In addition to a drop in cholesterol concentration and retardation of atherosclerotic lesion formation, garlic enhanced the fibrinolytic activity of the blood plasma, which is lowered as a result of cholesterol feeding. A reduction in fibrinolytic activity may accelerate atherosclerosis by exposing the vascular walls surfaces to recurrent thrombi and clot association mitogens. The effect of garlic may result from an increased fibrinolytic activity (Bordia et al., 1975). In a further correlation it was found that in cultured cells there was a strong correlation between atherogenicity of LDL (measured by the lipoprotein’s capacity to induce lipid accumulation in cultured arterial cells and stimulate cell proliferation) and sialic acid content of LDL. Desialylated LDL caused cholesterol accumulation in arterial cells via increased uptake, through interaction with both the B,E-receptor and an unregulated scavenger receptor, whereas non-desialylated LDL is taken up through the B,E-receptors only. Desialylated LDL may be responsible for the onset and further development of atherosclerotic lesions since atherogenic LDL isolated from patients’ blood had a lower content of sialic acid compared with native LDL of healthy subjects. Long-term therapy with garlic tablets increased the LDL sialic content up to normal (Orekhov et al., 1992).

In an animal study atherosclerosis was induced in rabbits by feeding them a cholesterol-rich diet for 2±4 months when lesions were found to have formed in the aorta of a sample of the animals. Cholesterol feeding was withdrawn, then the rabbits given essential oil of garlic for 8±9 months or garlic homogenate for 3 months. After cholesterol feeding was discontinued, serum cholesterol concentration decreased. In the garlic-fed animals the fact that the area of atherosclerotic lesions had decreased more than 50 per cent and the lipid content of the artery wall decreased by 69 per cent indicated a direct effect of garlic on the arterial wall and a regression of the disease (Bordia and Verma, 1980). Garlic supplementation also partially suppressed biosynthesis and accumulation of collagen in the aorta of the cholesterol-fed rabbits.

When the aorta of rabbits fed a standard diet was examined directly there were no fatty lesions on the intimal surface in the presence or absence of Kyolic. In the cholesterol-fed rabbits 70 per cent of the luminal aortic surface was covered by lipid-filled lesions and the neointima was greatly increased with a concentrated thickening of fibro-fatty plaques. The Kyolic-treated group had only 25 per cent coverage of the aortic surface by lesions, the accumulated cholesterol in the aorta was significantly reduced and the neointima was only half the size (Campbell et al., 2001). The function of smooth muscle cells is to contract but during atherosclerosis there is a change to a cell that can accumulate lipid, migrate, proliferate and synthesise appreciable extracellular matrix. This change to smooth muscle cells also occurs in culture but it was inhibited by the presence of Kyolic supplement in the culture medium. Inhibition of the conversion of the smooth muscle cell to this altered phenotype may represent a mechanism by which garlic achieves its protective effect against atherosclerosis under in vivo conditions.

Hypertension is a very common condition and one that results from atherosclerosis. One result of the changes during the development of atherosclerosis is that the arterial wall tissue hardens and loses its ability to relax and contract. In order to test the effect of garlic on the contractility of aorta, rats were fed a high-cholesterol diet with and without garlic extract, then killed after 16 weeks of this treatment. The arteries were cut into rings. These were suspended between two hooks, so the contractile response could be measured. Aortic rings were contracted with noradrenalline and exposed to acetylcholine or sodium nitroprusside. When contractions reached a plateau, the endothelium-dependent relaxation and independent relaxation were tested. Contractions were higher in those rings treated with garlic (Slowing et al., 2001), suggesting that garlic consumption can offset the hardening effect of atherosclerosis.

Chronic hypertension is a major contributor to the development of myocardial ischaemia infarction and strokes, but hypertension can be induced by factors other than age-related changes. One of the causes of induced hypertension is stress, with its accompanying depressive effect on the immune system and reduced control of oxidative stress. Activation of the transcription factor NF—?B, which is associated with the regulation of numerous genes encoding proteins in immune function, is an important part of this stress response (Grimme and Baeuerle, 1993). Oxidised LDL is a product of high blood cholesterol, which can act as a second messenger in the activation of NF—?B. In in vitro studies using cultured arterial endothelial cells, activation of NF—?B by oxidised LDL led to the expression of cell adhesion factors, VCAM-1 and ICAM-1, formation of which can accelerate the development of atherogenic lesions and cell death. Anti oxidants such as garlic derived compounds inhibited oxidant-induced NF—?B (Geng et al., 1997). The transcription factor NF—?B can also be activated by the tumour necrosis factor (TNF)-?. When endothelial cells from the human umbilical vein (HUVEC) were incubated with TNF-?, the cells showed activation of NF—?B expression, whereas preincubation of HUVEC with S­allylcysteine, a major products of alliin breakdown in fresh garlic and a component of some supplements, inhibited the activation (Ide and Lau, 2001).

Antihypertensive medications are the standard means to lower chronically elevated blood pressure. These are diuretics, whose side effects include elevation of cholesterol and triglyceride levels, beta-blockers, through a reduction in heart rate and a mild decrease in contractility with side effects that may include a rise in serum triglycerides, and a reduction of good cholesterol (HDL), and Ca2+ channel blockers, which have fewer side effects but are costly (Deshmukh et al., 1998). Although the influence of garlic on the risk factors of atherosclerosis and ultimately hypertension is weaker than that of synthetic drugs, the broad spectrum of garlic is an important advantage. Of equal importance is safety, and here garlic has the advantage of having being taken as a supplement for many centuries with no obvious ill effects except bad breath and the occasional gastric complaint.

An obvious method of increasing the uptake of active organosulphur compounds into the body is simply to eat more garlic or consume more supplements. Another strategy is to optimise the levels of the precursor compounds such as alliin in the existing garlic. This both improves the amounts of precursor compounds and breakdown products in the raw and cooked garlic and ensures the highest possible concentration in the garlic supplements. Increases in levels of organic sulphur compounds in garlic can be achieved in a number of ways. These are by the addition of sulphur fertiliser to the soil, by classical breeding techniques of selection and crossing and by genetic manipulation.

In an example of the first approach when sulphur fertiliser was added to the soil as 3.4?10?2 kg.m2 CaSO4, it increased the total sulphur in the fresh garlic from 96.4 mmol/kg to 190 mmol/kg. In the dried garlic powder from these bulbs, there was an increase in concentration of organosulphur compounds, alliin, allicin, -y-glutamyl propylcysteine and -y-glutamyl-phenylalanine. The effect of increasing soil sulphur led indirectly to a change in levels of NF?nB in blood cells The nuclear factor, NF?nB, is an important regulator of the immune system and is involved in atherosclerosis through activation of adhesion factors. Lipopolysaccharide are able to induce the inflammatory cytokines interleukin (IL)-1i and TNF-a in human blood which can activate production of NF-KB (Li and Verma, 2002). Pretreatment of the blood cells with garlic powder extract reduced the lipopolysaccharide production of cytokines and in turn reduced the production of the nuclear factor NF?nB. The effect of garlic was to reduce the induction of the nuclear factor by 25 per cent, whereas samples treated with sulphur-fertilised garlic lowered the NF?nB by 41 per cent (Keiss et al., 2003). The effect of the garlic was probably through diallyl disulphide, a breakdown product of alliin, since Keiss et al. (2003) found this compound reduced the lipopolysaccharide activation of interleukin (IL)-1i and TNF-a levels in blood.

A classical and genetic engineering approach to plant breeding to enhance the yields of organosulphur compounds in garlic has been undertaken by participants of the Garlic and Health grant, an EU-financed project (http:// www.plant.wageningen- ur .nl). Garlic is vegetatively propagated and none of the current varieties shows any seed formation, which makes it difficult to attempt varietal improvement by standard selection and breeding techniques. The approach undertaken was to return to the country of origin of the garlic and obtain plant varieties that showed seed formation, in combination with high yields of alliin in the bulb. It should now be possible to select high-yielding individuals, which can be put through a standard breeding programme involving crossing and seed production. In a genetic engineering approach an attempt was made to alter biosynthesis of the flavour precursors by genetic manipulation. This approach required identification of the pathway intermediates and the controlling enzymes. The mechanism of alliin biosynthesis is largely speculative but it is thought that there are two possible routes: one via serine and an allyl source and the other via glutathione and an allyl source (Fig. 12.1) (Lawson, 1996). These two stages represent the beginning of the secondary pathways and probably where the control of the secondary pathways occurs. Genetic manipulation of the control enzymes would require identification of the enzymes and the genes responsible, followed by transformation of the garlic and overexpression of the control enzymes to overcome any metabolic limitations on the pathway.

While the large array of different garlic supplements available to the customer can be confusing, the Web is now a valuable reference for both products and medical advice. Many of the statements, however, have not been evaluated by regulatory bodies such as the American Food and Drug Aministration, and the European Union (Thomas and Earl, 1994; Clydesdale, 1997; Bellisle et al., 1998), so the garlic cannot be seen as an official cure, or treatment or for the prevention of cardiovascular disease. The evidence only suggests that garlic may have a beneficial effect on cardiovascular disease. An example of a helpful website is http://illness.altmedangel. com/heart.htm which is a non-profit site offering educational information and broad-based research on various health conditions, medications, supplements and therapies. It advocates the use of garlic supplement to regulate blood pressure and to thin the blood. It claims that consumption of the garlic will also preserve the elasticity of the arteries and delay their stiffening. In another site, http://www2.vitaminconnection .com, garlic as fresh garlic (up to one clove daily) and 400 mg tablet (standardised to 1 per cent total allicin potential) was shown to cause a lowering of cholesterol. There were also beneficial effects on blood pressure, fibrinogen and LDL. In the more commercially orientated sites the descriptions of the effects of garlic consumption are sometimes vague, using such language as cleansing cholesterol and normalising blood pressure, helping to reduce harmful LDL-cholesterol and triglycerides, reducing blood clotting and strengthening blood vessels.

Another site (http://www.americanheart .org/presenter.jtml) was more critical since it made an objective analysis of a number of clinical trials on the effect of garlic consumption on cardiovascular disease. The analysis showed that there were a number of modest short-term benefits of garlic supplements. It maintained that the main problem with making a comparison between the trials was the variation in the composition of the different commercial preparations. In a larger meta-analysis the effect of garlic on blood pressure reduction and lipid lowering was examined by Silagy and Neil (1994a,b). The conclusions from analysis of 11 randomised controlled trials for blood pressure and 25 randomised controlled trials for serum cholesterol was that garlic therapy may reduce both blood pressure and serum lipids. Silagy and Neil (1994a,b) suggested that this result should be treated with caution because of shortcomings in methodology of some of the trials examined. In a later meta-analysis (Stevinson et al., 2000), the effect of garlic on total cholesterol in persons with an elevated level of cholesterol of at least 200 mg/mL was analysed in a number of randomised, double-blind, placebo-controlled trials. The results showed that compared with the placebo, garlic reduces the total cholesterol level in those persons whose levels were elevated. The findings confirm the results of earlier studies but the effect may be smaller than originally thought with a reduction of 4±6 per cent of cholesterol due to garlic. The meta-analysis included only 13 trials since it was only these that satisfied the strict requirements of the test and of these there were serious variations in the methodology used. One of the most significant was the unblinding of the test because of the difficulty of disguising the garlic odour. Empirical research suggested that trials that are not wholly double-blind tend to exaggerate the treatment effects. The other major problem in trying to establish a significant effect was the variation in the garlic source. The test used Kwai, garlic oil and spray-dried garlic powder. Of these there was known variation in the allicin content of the samples in addition to the known variation between the different sources of garlic. According to Stevinson et al. (2000), what is required is to have well-designed placebo-controlled trials comparing the efficiency of the different types of garlic.

The trials took essentially a short-term view of the effect of garlic consumption, i.e. the increase in cholesterol was measured, whereas it is more important to take a long-term view in terms of clinical effects, such as heart disease. Many of the other protective effects of garlic such as reduced blood pressure, platelet inhibition and stimulation to blood flow have been assessed in vitro and in vivo, but the clinical effects have been implied but not measured. The results of the various meta-analyses suggested that garlic is superior to a placebo in reducing elevated cholesterol levels but the size of the effect is moderate. Compared with the use of pharmaceutical drugs such as the modern statins, which can cause a reduction of 17-32 per cent, the reduction in cholesterol from garlic is small. Even dietary modifications can cause a 5.3 per cent reduction over a 6-month period. The implication for clinical practice is that garlic use is not an efficient way to decrease serum cholesterol level. Large-scale and long-term trials are required to establish the association between garlic consumption and the clinical outcomes. The one test that attempted this involved 152 patients over a 4-year period when it was shown that there was deceleration of the development of atherosclerosis in those who included garlic in the diet (Koscielny et al., 1999). The other more long-term test for cardiovascular health is perhaps the Mediterranean diet, which contains a significant amount of garlic and which appears to have a protective effect on the cardiovascular system.