Vita OH All-trans retinol β Carotene Fig 3.1 Structures of all-trans-retinol and B-carotene 4: I(Gopalan et al, 1989) 6: 1(Report of a Joint FAO/WHO expert consultation 1988) 12: 1(Report of the Institute of Medicine 2001) 21: 1 and 26: 1(de Pee et al, 1998; Khan et al, 1998; van het Hof et al, 1999) Assuming that 100g DGLV contains 3000ug B-carotene and a child of 4 years needs an RNI of 400ug RE/day then using bioconversion factor 12:1 21: 1 or 26: 1, the child would need 160g, 280g or 360g of DGLv meet requirements 3.4.2 Equivalence factor as calculated by IoM In January 2001, the US National Academy of Sciences/Institute of Medicine (OM)announced a new equivalence factor(12: 1) for the conversion of B- carotene to retinol. The basis for formulating the equivalency factor was to relate the absorption of B-carotene in a principally mixed vegetable diet to that from oil in healthy and nutritionally-adequate individuals. The recommendations were based on a product of relative absorption of B-carotene in mixed vegetable diet 1: 6)(van het Hof et al, 1999)to the amount of retinol formed (1 ug) when 2ug B-carotene was fed in oil (Sauberlich et al, 1974). In the van het Hof study, the increase in serum B-carotene concentration after consumption of B-carotene-rich egetables was 1/7 or 14% of the increase after consumption of B-carotene in oil The Institute of Medicine (IOM)(2001)adjusted the value to 1/6 or 17%, because of the low fruit content in the diet used. From these two studies the iom con cluded that the bioefficacy of B-carotene in oil was(100%*0.94)/2=47%0 (Sauberlich et al, 1974)and 17% from mixed vegetables(van het Hof et al, 1999) Thus bioefficacy from vegetables in a mixed diet was 17%*0.47=8%0, that is 12ug B-carotene in food has the same vitamin A activity as l ug retinol. The
• 4 :1 (Gopalan et al, 1989) • 6 :1 (Report of a Joint FAO/WHO expert consultation 1988) • 12 :1 (Report of the Institute of Medicine 2001) • 21 :1 and 26 :1 (de Pee et al, 1998; Khan et al, 1998; van het Hof et al, 1999) Assuming that 100 g DGLV contains 3000mg b-carotene and a child of 4 years needs an RNI of 400mg RE/day then using bioconversion factors of 12 :1, 21 :1 or 26 :1, the child would need 160 g, 280 g or 360 g of DGLV each day to meet requirements. 3.4.2 Equivalence factor as calculated by IOM In January 2001, the US National Academy of Sciences/Institute of Medicine (IOM) announced a new equivalence factor (12 :1) for the conversion of bcarotene to retinol. The basis for formulating the equivalency factor was to relate the absorption of b-carotene in a principally mixed vegetable diet to that from oil in healthy and nutritionally-adequate individuals. The recommendations were based on a product of relative absorption of b-carotene in mixed vegetable diet (1 :6) (van het Hof et al, 1999) to the amount of retinol formed (1mg) when 2mg b-carotene was fed in oil (Sauberlich et al, 1974). In the van het Hof study, the increase in serum b-carotene concentration after consumption of b-carotene-rich vegetables was 1/7 or 14% of the increase after consumption of b-carotene in oil. The Institute of Medicine (IOM) (2001) adjusted the value to 1/6 or 17%, because of the low fruit content in the diet used. From these two studies, the IOM concluded that the bioefficacy of b-carotene in oil was (100% * 0.94)/2 = 47% (Sauberlich et al, 1974) and 17% from mixed vegetables (van het Hof et al, 1999). Thus bioefficacy from vegetables in a mixed diet was 17% * 0.47 = 8%, that is 12mg b-carotene in food has the same vitamin A activity as 1mg retinol. The Vitamins 39 Fig. 3.1 Structures of all-trans-retinol and b-carotene
40 The nutrition handbook for food processors 12: 1 ratio is referred to as the retinol activity equivalence(RAE)(Northrop Clewes, 2001a) In the calculation, the IOM used the mean ratio from only one ' oil study here were in fact 5'oil studies(Booher et al, 1939: Wagner, 1940: Hume and Krebs, 1949; Sauberlich et al, 1974; Tang et al, 2000)and if the data from all the studies had been used a bioconversion value of 3.5 ug B-carotene to l ug retinol would have been calculated giving an RAE of 21 instead of 12 as quoted. The lOM are at present reconsidering the published data(van Lieshout, 2001). How we interpret the bioconversion factors for B-carotene proposed by IOM or by others still needs much more thought before any of them can be put to practical use. However, even if the RaE is only 12: 1, there are probably not sufficient vegetables in developing countries to meet that required, therefore, there must be other, as yet undiscovered factors, that influence bioefficacy. Despite this, intake of fruit and vegetable-sources of provitamin A should be encouraged, for although the bioefficacy of B-carotene is apparently so poor, it has to be remembered that he majority of the worlds children are not VAD 3.5 Function Vitamin A, its analogues and metabolites function in vision, cell differentiation embryogenesis, the immune response, reproduction and growth 3.5.1 Vision n vision, vitamin A is required in two forms for two processes 1. As 1l-cis-retinal in rhodopsin which on exposure to light in the retina iso- merises to a transoid intermediate, triggering a series of conformational changes in membrane potential which is transmitted to the brain 2. As retinoic acid to maintain normal differentiation of the cells of the con junctival membranes, cornea and other ocular structures hence preventin xerophthalmia(Ross, 1999) 3.5.2 Cell differentiation The role of vitamin a in cell differentiation has been clarified since the identifi- cation of two sets of nuclear receptors, RXR and RAR, which are activated by retinoic acid isomers(Olson et al, 2000). Each of the receptors has three distinct forms a, B and y and six domains involved in transcription of genes. RAR binds either all-trans or 9-cis retinoic acid, while rXR binds only 9-cis retinoic acid Both receptors are dimers, however, the rXR receptors form homodimers with themselves as well as heterodimers with RAR, the vitamin D receptor, the tri- iodothyronine receptor and several other nuclear transcription receptors(see also sections 3. 12.3, 3.20.2 and 3.21.3). These interactions usually activate gene expression but RAR can also be inhibitory
12 :1 ratio is referred to as the retinol activity equivalence (RAE) (NorthropClewes, 2001a). In the calculation, the IOM used the mean ratio from only one ‘oil study’. There were in fact 5 ‘oil’ studies (Booher et al, 1939; Wagner, 1940; Hume and Krebs, 1949; Sauberlich et al, 1974; Tang et al, 2000) and if the data from all the studies had been used a bioconversion value of 3.5mg b-carotene to 1mg retinol would have been calculated giving an RAE of 21 instead of 12 as quoted. The IOM are at present reconsidering the published data (van Lieshout, 2001). How we interpret the bioconversion factors for b-carotene proposed by IOM or by others still needs much more thought before any of them can be put to practical use. However, even if the RAE is only 12 :1, there are probably not sufficient vegetables in developing countries to meet that required, therefore, there must be other, as yet undiscovered factors, that influence bioefficacy. Despite this, intake of fruit and vegetable-sources of provitamin A should be encouraged, for although the bioefficacy of b-carotene is apparently so poor, it has to be remembered that the majority of the world’s children are not VAD. 3.5 Function Vitamin A, its analogues and metabolites function in vision, cell differentiation, embryogenesis, the immune response, reproduction and growth. 3.5.1 Vision In vision, vitamin A is required in two forms for two processes: 1. As 11-cis-retinal in rhodopsin which on exposure to light in the retina isomerises to a transoid intermediate, triggering a series of conformational changes in membrane potential which is transmitted to the brain. 2. As retinoic acid to maintain normal differentiation of the cells of the conjunctival membranes, cornea and other ocular structures hence preventing xerophthalmia (Ross, 1999). 3.5.2 Cell differentiation The role of vitamin A in cell differentiation has been clarified since the identifi- cation of two sets of nuclear receptors, RXR and RAR, which are activated by retinoic acid isomers (Olson et al, 2000). Each of the receptors has three distinct forms a, b and g and six domains involved in transcription of genes. RAR binds either all-trans or 9-cis retinoic acid, while RXR binds only 9-cis retinoic acid. Both receptors are dimers, however, the RXR receptors form homodimers with themselves as well as heterodimers with RAR, the vitamin D receptor, the triiodothyronine receptor and several other nuclear transcription receptors (see also sections 3.12.3, 3.20.2 and 3.21.3). These interactions usually activate gene expression but RAR can also be inhibitory. 40 The nutrition handbook for food processors
3.5.3 Embryogenesis Both retinoic acid and retinol are essential for embryonic development, however, retinoic acid does not seem to be involved in vertebrate embryogenesis before gastrulation. After gastrulation specific genes of the Hox family are expressed in a wave starting about 7.5 days postcoitus in the mouse. The presence of retinoids together with their binding proteins and receptors in a temporally precise manner provides strong circumstantial evidence that retinoic acid-activated RARs regu late Hox gene expression(Ross, 1999). The Hox family consists of 38 genes arranged in four chromosomal clusters, which code for transcription factors that regulate development along the posterior axis. Retinoic acid is also thought to act in limb development and formation of the heart, eyes and ears. Experimental VAD has demonstrated that major target tissues of retinoic acid include the heart, central nervous system and structures derived from it, the circulatory, urogenital and respiratory systems and the development of skull, skeleton and limbs. Home ostasis of retinoic acid is maintained by enzyme systems, which are develop- mentally regulated by vitamin A. Inadequate vitamin A nutrition during early pregnancy may account for some paediatric congenital abnormalities(zile, 2001) However, the study of this aspect of vitamin A is in its ' infancy'and much more work is still to be done. A high incidence of spontaneous abortions and birth defects has been observed in the foetuses of women taking therapeutic doses of 13-cis retinoic acid. The drugs Accutane(13-cis retinoic acid) and etretinate, an aromatic analogue of all-trans retinoic acid have been most implicated in such effects(Armstrong et al, 1994). The dysmorphogenic effects caused by retinoids depend on dosage(exposure), the form of the retinoid, its rate of metabolism and tage of foetal development at the time the retinoid is taken. Retinoids are ter atogenic during the period of foetal organogenesis(first trimester)(Ross, 1999) 3.5.4 Immunity One of the first names for vitamin a was the anti-infective vitamin, which was based on the increased number of infections noted in VAd animals and humans (RoSs, 1996). In VAD the humoral response to bacterial, parasitic and viral infec- tions, cell-mediated immunity, mucosal immunity, natural killer cell activity and phagocytosis are all impaired. The primary immune response to protein antigens is markedly reduced but the response to bacterial lipopolysaccharides or the process of immunological memory essential for secondary response does not seem to be affected(Ross and Hammerling, 1994) A major site for vitamin A action in the immune response is the T-helper cell nd studies have shown that the activity of T-helper cells variety 1 (Thl)pre- cedes that of Th variety 2(Th2) and that vitamin A supports Th2 development (Cantorna et al, 1994). Hence, in the presence of VAd the ratio of Thl: Th2 may become elevated during an immune response. Retinol, probably in the form of 14-hydroxy-retroretinol(HRR), is thought to be involved in the proliferation of normal B-and T-cells, an action which can be modulated by some cytokines Retinoic acid, the usual active form of vitamin A, appears to be inactive in thes
3.5.3 Embryogenesis Both retinoic acid and retinol are essential for embryonic development, however, retinoic acid does not seem to be involved in vertebrate embryogenesis before gastrulation. After gastrulation specific genes of the Hox family are expressed in a wave starting about 7.5 days postcoitus in the mouse. The presence of retinoids together with their binding proteins and receptors in a temporally precise manner provides strong circumstantial evidence that retinoic acid-activated RARs regulate Hox gene expression (Ross, 1999). The Hox family consists of 38 genes arranged in four chromosomal clusters, which code for transcription factors that regulate development along the posterior axis. Retinoic acid is also thought to act in limb development and formation of the heart, eyes and ears. Experimental VAD has demonstrated that major target tissues of retinoic acid include the heart, central nervous system and structures derived from it, the circulatory, urogenital and respiratory systems and the development of skull, skeleton and limbs. Homeostasis of retinoic acid is maintained by enzyme systems, which are developmentally regulated by vitamin A. Inadequate vitamin A nutrition during early pregnancy may account for some paediatric congenital abnormalities (Zile, 2001). However, the study of this aspect of vitamin A is in its ‘infancy’ and much more work is still to be done. A high incidence of spontaneous abortions and birth defects has been observed in the foetuses of women taking therapeutic doses of 13-cis retinoic acid. The drugs Accutane (13-cis retinoic acid) and etretinate, an aromatic analogue of all-trans retinoic acid have been most implicated in such effects (Armstrong et al, 1994). The dysmorphogenic effects caused by retinoids depend on dosage (exposure), the form of the retinoid, its rate of metabolism and stage of foetal development at the time the retinoid is taken. Retinoids are teratogenic during the period of foetal organogenesis (first trimester) (Ross, 1999). 3.5.4 Immunity One of the first names for vitamin A was ‘the anti-infective vitamin’, which was based on the increased number of infections noted in VAD animals and humans (Ross, 1996). In VAD the humoral response to bacterial, parasitic and viral infections, cell-mediated immunity, mucosal immunity, natural killer cell activity and phagocytosis are all impaired. The primary immune response to protein antigens is markedly reduced but the response to bacterial lipopolysaccharides or the process of immunological memory essential for secondary response does not seem to be affected (Ross and Hammerling, 1994). A major site for vitamin A action in the immune response is the T-helper cell and studies have shown that the activity of T-helper cells variety 1 (Th1) precedes that of Th variety 2 (Th2) and that vitamin A supports Th2 development (Cantorna et al, 1994). Hence, in the presence of VAD the ratio of Th1 :Th2 may become elevated during an immune response. Retinol, probably in the form of 14-hydroxy-retroretinol (HRR), is thought to be involved in the proliferation of normal B- and T-cells, an action which can be modulated by some cytokines. Retinoic acid, the usual active form of vitamin A, appears to be inactive in these Vitamins 41
42 The nutrition handbook for food processors systems(Olson et al, 2000). The enzyme responsible for the conversion of retinol to hrr has not been characterised 3.6 Health-related roles of B-carotene 3.6.1 B-Carotene as an antioxidant The ability of carotenoids to act as antioxidants can be measured in vitro, ex vivo, or in vivo. LDL isolated from an individual who has been supplemented with carotenoids and then evaluated for its antioxidant activity is an extension of an in vivo study, i.e. ex vivo. However, when carotenoids are added to plasma and then the oxidisable value of the ldl is measured it is more like an in vitro model Krinsky, 2001). Many studies report using the ex vivo method of measuring the oxidisability of the Ldl particles after coning ncr reased amounts of car containing foods. However, when using fruits and vegetables the outcome is vari- able and difficult to interpret because they also contain vitamin C, polyphenols and flavonoids, which are also potential antioxidants. One study which gave additional dietary fruits and vegetables to subjects reported an increase in the resistance of LDL to oxidation(Hininger et al, 1997) while two other studies found no effect (Chopra et al, 1996; van het Hof et al, 1999). The differing results obtained may be due to different time periods on the diets, different degrees of change in the plasma carotenoids or to different study populations(Krinsky, 2001) 3.6.2 B-Carotene and protection from cancer The strongest epidemiological evidence suggesting high intake of fruits and veg- etables might give protection against lung cancer came from prospective studies tive studies and 18 out of 20 retrospective studies(Zeigler et al, 1996). Based q ia which low plasma B-carotene was associated with a higher incidence of lung cancer. Carotenoid intake was associated with reduced cancer risk in 8 prospec the results of these studies, three major intervention studies investigated the pro- tective effect of B-carotene in the prevention of lung cancer: (a)The alpha-tocopherol, beta-carotene(ATBC) Cancer Prevention study, was a randomised-controlled trial that tested the effects of daily doses of 50mg(50 IU) vitamin E(all-racemic a-tocopherol acetate), 20 mg of B-carotene, both or placebo in a population of more than 29000 male smokers for 5-8 years. No reduction in lung cancer or major coronary events was observed with any of the treatments. What was more startling was the unexpected increases in risk of death from lung cancer and ischemic heart disease with B-carotene supplementation (ATBC Cancer Prevention Study Group, 1994) (b) Increases in risk of both lung cancer and cardiovascular disease mortality were also observed in the beta-Carotene and Retinol Efficacy Trial ( CArEt). which tested the effects of combined treatment with 30mg/d B-carotene and retinyl palmitate(25 000IU/d)in 18000 men and women with a history of ciga- rette smoking or occupational exposure to asbestos(Hennekens et al, 1996)
systems (Olson et al, 2000). The enzyme responsible for the conversion of retinol to HRR has not been characterised. 3.6 Health-related roles of b-carotene 3.6.1 b-Carotene as an antioxidant The ability of carotenoids to act as antioxidants can be measured in vitro, ex vivo, or in vivo. LDL isolated from an individual who has been supplemented with carotenoids and then evaluated for its antioxidant activity is an extension of an in vivo study, i.e. ex vivo. However, when carotenoids are added to plasma and then the oxidisable value of the LDL is measured it is more like an in vitro model (Krinsky, 2001). Many studies report using the ex vivo method of measuring the oxidisability of the LDL particles after feeding increased amounts of carotenecontaining foods. However, when using fruits and vegetables the outcome is variable and difficult to interpret because they also contain vitamin C, polyphenols and flavonoids, which are also potential antioxidants. One study which gave additional dietary fruits and vegetables to subjects reported an increase in the resistance of LDL to oxidation (Hininger et al, 1997) while two other studies found no effect (Chopra et al, 1996; van het Hof et al, 1999). The differing results obtained may be due to different time periods on the diets, different degrees of change in the plasma carotenoids or to different study populations (Krinsky, 2001). 3.6.2 b-Carotene and protection from cancer The strongest epidemiological evidence suggesting high intake of fruits and vegetables might give protection against lung cancer came from prospective studies in which low plasma b-carotene was associated with a higher incidence of lung cancer. Carotenoid intake was associated with reduced cancer risk in 8 prospective studies and 18 out of 20 retrospective studies (Zeigler et al, 1996). Based on the results of these studies, three major intervention studies investigated the protective effect of b-carotene in the prevention of lung cancer: (a) The alpha-tocopherol, beta-carotene (ATBC) Cancer Prevention study, was a randomised-controlled trial that tested the effects of daily doses of 50 mg (50 IU) vitamin E (all-racemic a-tocopherol acetate), 20 mg of b-carotene, both or placebo in a population of more than 29 000 male smokers for 5–8 years. No reduction in lung cancer or major coronary events was observed with any of the treatments. What was more startling was the unexpected increases in risk of death from lung cancer and ischemic heart disease with b-carotene supplementation (ATBC Cancer Prevention Study Group, 1994). (b) Increases in risk of both lung cancer and cardiovascular disease mortality were also observed in the beta-Carotene and Retinol Efficacy Trial (CARET), which tested the effects of combined treatment with 30 mg/d b-carotene and retinyl palmitate (25 000 IU/d) in 18 000 men and women with a history of cigarette smoking or occupational exposure to asbestos (Hennekens et al, 1996). 42 The nutrition handbook for food processors
Vitamins 43 A.(c) The third study was the Physicians Health Study, in which 22071 US male lysicians were randomised to get 50mg B-carotene or aspirin(325 mg), or both or neither every other day for 12 years. There was no evidence of a significant beneficial or harmful effect on cancer or cardiovascular disease. but the number of smokers in the study was too small to be certain whether B-carotene was harmful in the group or not(Hennekens et al, 1996) One other study should also be mentioned. The Cancer Prevention Study Il,a prospective study on more than one million US adults investigated the effect of ommercially available multivitamins and/or vitamins A, C, and/or E on mortal ity during a 7 year follow-up. The use of multivitamins plus A, C and/or E sig nificantly reduced the risk of lung cancer in former smokers and in those who never smoked but increased the risk in men who smoked and used vitamins a C and/or E compared with men who reported no supplement use. Thus the antioxidant vitamins A, C and E only appear to benefit male non-smokers. No association with smoking was seen in women(Watkins et al, 2000). 3.6.3 Reasons for increased cancer risk associated with B-carotene supplementation The mechanism for the increased risk associated with B-carotene supplementa tion in smokers is unclear. One suggestion is that the subjects of the studies already had a ' high risk ' of developing lung cancer and many might have had undetected tumours at the start. The stage of carcinogenesis that B-carotene might affect is not known but if mediated by the immune system the effect might be at the promotional stages preceding the formation of a malignant tumour(Hughes, Immune cells rely heavily on cell-to-cell communication, particularly via mea s 2001). The immune system appears to be particularly sensitive to oxidative stress brane bound receptors, to work effectively. Cell membranes are rich in polyun saturated fatty acids and if peroxidised, can lead to a loss of membrane integrity, altered membrane fluidity and result in alterations in intracellular signalling and cell function. It has been shown that exposure to reactive oxygen species(ROS) can lead to a reduction in cell-membrane expression(Hughes, 2001). In addition, the production of Ros by phagocytic immune cells can damage the cells them selves if not adequately protected by antioxidants such as B-carotene, lycopene and lutein One of the major unresolved dilemmas of B-carotene research is the required to optimise immune function and provide other health benefits(Hr nd it is not clear whether different intakes are associated with different outcomes It is also possible that supplemental B-carotene might be interfering with intesti- nal absorption of other possible chemopreventive nutrients e.g. B-carotene can inhibit absorption of lutein, a-carotene and canthaxanthin, all of which show good antioxidant properties(Olson, 1999). Another explanation might be that B- carotene is acting as a pro-oxidant in the presence of high oxygen tension in
(c) The third study was the Physicians Health Study, in which 22 071 US male physicians were randomised to get 50 mg b-carotene or aspirin (325 mg), or both or neither every other day for 12 years. There was no evidence of a significant beneficial or harmful effect on cancer or cardiovascular disease, but the number of smokers in the study was too small to be certain whether b-carotene was harmful in the group or not (Hennekens et al, 1996). One other study should also be mentioned. The Cancer Prevention Study II, a prospective study on more than one million US adults investigated the effect of commercially available multivitamins and/or vitamins A, C, and/or E on mortality during a 7 year follow-up. The use of multivitamins plus A, C and/or E significantly reduced the risk of lung cancer in former smokers and in those who never smoked, but increased the risk in men who smoked and used vitamins A, C and/or E compared with men who reported no supplement use. Thus the ‘antioxidant’ vitamins A, C and E only appear to benefit male non-smokers. No association with smoking was seen in women (Watkins et al, 2000). 3.6.3 Reasons for increased cancer risk associated with b-carotene supplementation The mechanism for the increased risk associated with b-carotene supplementation in smokers is unclear. One suggestion is that the subjects of the studies already had a ‘high risk’ of developing lung cancer and many might have had undetected tumours at the start. The stage of carcinogenesis that b-carotene might affect is not known but if mediated by the immune system the effect might be at the promotional stages preceding the formation of a malignant tumour (Hughes, 2001). The immune system appears to be particularly sensitive to oxidative stress. Immune cells rely heavily on cell-to-cell communication, particularly via membrane bound receptors, to work effectively. Cell membranes are rich in polyunsaturated fatty acids and if peroxidised, can lead to a loss of membrane integrity, altered membrane fluidity and result in alterations in intracellular signalling and cell function. It has been shown that exposure to reactive oxygen species (ROS) can lead to a reduction in cell-membrane expression (Hughes, 2001). In addition, the production of ROS by phagocytic immune cells can damage the cells themselves if not adequately protected by antioxidants such as b-carotene, lycopene and lutein. One of the major unresolved dilemmas of b-carotene research is the intake required to optimise immune function and provide other health benefits (Hughes, 2001). Most studies have been done using pharmacological doses of b-carotene and it is not clear whether different intakes are associated with different outcomes. It is also possible that supplemental b-carotene might be interfering with intestinal absorption of other possible chemopreventive nutrients e.g. b-carotene can inhibit absorption of lutein, a-carotene and canthaxanthin, all of which show good antioxidant properties (Olson, 1999). Another explanation might be that bcarotene is acting as a pro-oxidant in the presence of high oxygen tension in the lung. Vitamins 43