4 SYNTHESIS AND AVAILABILITY OF NIACIN IN ROASTED COFFEE
Jean Adrian and Regine Frangne Chaire de Biochimie industrielle et agro-alimentaire Conservatoire National des Arts et Metiers 292, rue Saint-Martin, 75003 Paris, France ABSTRACT The coffee bean contains about 1% of trigonelline that is demethylated at temperatures approaching 200 0 C ; it is partially converted into nicotinic acid. This operation is mainly proportional to the severity of dry heat treatment ; various other physico-chemical factors also influence the synthesis of niacin during the roasting. The niacin content of weakly roasted commercial coffee is about 10 mg/100 g (American coffee) and it reaches 40 mg in heavy roasted coffees, i.e. Italian coffee. Caffeine-free coffee is lower in niacin than the corresponding raw coffee. The drinking retains 85% of the niacin formed during roasting ; it is totally available for the organism and can constitute a noticeable part of the daily supply in niacin. INTRODUCTION Some observations attribute various deleterious physiological effects to high coffee consumption. These manifestations are due to the presence of particular constituents in coffee, such as chlorogenic acid (Teply et al., 1945 ; Chassevent, 1969 ; Challis and Bartlett, 1975) or caffeine (Feinberg et al., 1968 ; Naismith et al., 1969 ; Panigrahi and Rao, 1983 Pozniak, 1985 ; Stavric, 1988). However, an unfavourable incidence of coffee on the glycemia, the lipidemia and the occurrence of cardiovascular pathology is not obvious for the healthy person who consumes coffee normally (Callahan et al., 1979 ; Querat and Heraud, 1987 ; Kitts and Mathieson, 1989). Actually, the participation of coffee in the ethiology of cancer remains a point of question (Strubelt et al., 1973 ; Czok, 1977 ; Thomas, 1979 ; Sandler, 1983). The roasting process is responsible for the formation of characteristic molecules. Among them, such aromatic polycyclic hydrocarbons as the benzopyrene have a well established carcinogenic power (Kuratsune and Hueper, 1960 ; Fritz, 1969, 1975 ; Bracco, 1973 ; Soos and F8zy, 1974 ; Strobel, 1974) whereas others like harman and norharman (Adrian et al., 1985) are also likely to occur in the mechanism of the cancerization. These roasting components certainly represent the most probable risk bound to coffee consumption (Miller, 1983 ; Pozniak, 1985).
M. Friedman (ed.), Nutritional and Toxicological Consequences of Food Processing © Springer Science+Business Media New York 1991
49
=
A 30
~
40
,=
E
E 20
~
~
•uM
10
c 10 M
•
M
•
20
40
60
ROASTING TINE (NIN.)
B
,,,
~
~
,-
I I
~
~
c
I
=
,= 20
•MU
,
m30
230·
I
,,
I
10
20
30
40
WEIGHT LOSS (I)
Figure I. Niacin content in coffee according to the intensity of heat treatment. I-A: Roasting measured by the time and temperature of treatment (Hughes and Smith, 1946) ; I-B : Roasting measured by the weight loss of coffee (broken line : C. arabica ; heavy line: C. canephora var. robusta) (Adrian and Navellier, 1961).
However, roasting must not be considered as an operation with only negative consequences ; it develops the characteristic and enjoyed aroma of coffee and it provokes an abundant synthesis of nicotinic acid from trigonelline, which is its methylated precursor. This conversion during the roasting places coffee drinking among the most efficient food resources to satisfy the niacin requirement and to prevent or cure pellagra. The mechanism of the synthesis of niacin from trigonelline and the efficiency of the formed nicotinic acid have been described by different teams whose conclusions all demonstrate the interest of roasted coffee as a contribution to the niacin supply. It would be useful to review these works, often old, because they show that, besides undesirable components, strong roasting can also have favorable consequences in an accurate vitamin field. THE SYNTHESIS OF NICOTINIC ACID DURING COFFEE ROASTING Trigonelline is the N-methyl betaine of nicotinic acid. This compound is widely present in vegetal produce: it can be found in all cereal grains, in potatoes, beetroots, tomatoes, mushrooms and sweet peppers at rates of betwwen 10 and 80 mg/IOO g (Barbiroli, 1966). It is much more abundant in the coffee bean since C. arabica contains about 900 mg and C. canephora var. robusta, about 650 mg according to Hughes and Smith (1946). In a pure state, trigonelline heated in dry conditions is degraded in various derivated compounds, especially at temperatures ranging from 160°C to 230°C (Figure IA) : about 5% are converted to nicotinic acid by demethylation. The same phenomenon occurs in some food technology processes. During experimental burnings, the nicotinic acid content increases in all the grains and vegetals containing trigonelline (Barbiroli, 1966). Under more usual modalities, an increase of the niacin content is observed as a result of an intense heating carried out in dry conditions : for example, the synthesis of niacin is noticeable in bread crust but not in crumb (Gassmann and Schneeweiss, 1959). Coffee roasting is nevertheless the operation most favorable provoking this synthesis from trigonelline : the bean contains a large quantity of trigonelline, its wetness is very weak and the heat treatment easily reaches 200°C during the process.
50
Incidentally, the nicotinic acid is one of the most stable vitamins when met with the customary conditions in food processing, storage, cooking, etc. (Adrian, 1959). In particular, it is remarkably thermostable, which allows it to resist efficiently during a strong heat treatment, at least as long as it does not reach extremes. That is noticed during the commercial torrefaction of coffee. In practice, the importance of the nicotinic acid synthesis and the niacin content in roasted coffee depend on many factors and parameters such as the intensity of treatment, the technological preparation of the bean, etc. Table I. Influence of roasting intensity on the niacin content in coffee (Teply and Prier, 1957) Niacin content mg/IOO g relative value green coffee just before swelling just swelling New England roasting French roasting Italian roasting Heavy roasting (excessive roasting)
2.2 4.0 8.3 13.0 24.9 41.6 43.6
I
1.8 3.8 5.9 11.3 18.9 19.9
Intensity of coffee roasting The abundance of nicotinic acid in roasted coffee depends, first of all, on whether the intensity of torrefaction is estimated by the duration of heating, the browning of the bean or the weight loss (Hughes and Smith, 1946 ; Cravioto et al., 1955 ; Daum, 1955 ; Teply and Prier, 1957 ; Bressani and Navarrete, 1959 ; Adrian and Navellier, 1961 ; Bressani et al., 1962 ; Boddeker and Mishkin, 1963 ; Taguchi et al., 1985). When the roasting is very heavy, the quantity of nicotinic acid always exceeds 40 mg/IOO g of roasted coffee (Table I, Figure IB). This result corresponds to a value 20 to 30 times superior to that of the green bean, which is about 1.5 mg (Adamo, 1955 ; Cravioto et al., 1955 ; Daum, 1955 ; Teply and Prier, 1957 ; Bressani et al., 1959, 1962 ; Adrian and Navellier, 1961 ; Taguchi et al., 1985). According to Figure IB, at the beginning of the roasting, the trigonelline is easily converted into nicotinic acid and the formed molecule remains stable in these modalities. It is only damaged by extremely severe burnings (Adrian and Navellier, 1961 ; Tchetche, 1979) ; this evolution is not observed in all torrefactions : Bressani et ale (1962) proceed to heatings reaching 42% of weight loss without the treatment being accompagnied by any decrease in the niacin content of the roasted produce. Table 2. Niacin content of commercial roasted C. arabica and C. canephora types (a) Arabica Hughes and Smith (1946) Bressani et ale (1961) Adrian et ale (1967) (b)
15.2 (4) 22.65 (29) 18.45 (18)
Robusta 18.8 18.1 15.4
( 4) ( 2)
(45)
(.) the values in brackets represent the number of samples (b) the roasting time is 7.1 min. for Arabica and 8.2 min. for Robusta
51
In fact, coffee torrefaction primarily provokes a nicotinic acid synthesis and secondarily that of nicotinamide (Taguchi et al., 1985) commercial roasting heavy roasting in mg/100 g : nicotinic acid 2.9 - 4.5 28.8 nicotinamide 0.48 - 0.84 2.3 The small proportion of nicotinamide is probably due to the partial degradation of aminoacids with releasing of amino groups, which react on nicotinic acid. Influence of geographical and botanical factors The Arabica and Robusta coffees constitute the two largest species of cultivated coffees. The former tends to contain more trigonelline but requires a less intense torrefaction than the Robusta types. Moreover, during heating, the possibilities of conversion can be compared in the both species. That is why the commercially roasted products have similar amounts of niacin (Figure IB, Table 2). On the other hand, within the species C. canephora, particularities appear according to the geographical area of cultivation. In Angola, a notion of "local variety" has been established to describe specific characteristics of Robusta cultivated in distinct regions. When samples of these "local varieties" are submitted to roasting, the influence of geographical data becomes especially evident : a series of 9 samples roasted under standardized modalities shows the following results : Ambriz "variety" hold 19.85 mg of niacin/lOO g, Cazengo "variety" 15.35 mg and Amboim 13.15 mg (Adrian et al., 1967). These differences are attributed to the combination of geographical, pedologic and climatic factors on the development and the physiology of the coffee tree. They reveal a more important influence on the niacin synthesis than the differences observed between C. arabica and C. canephora. On the other hand, the year of the crop is not a determining factor to the rate of niacin in roasted coffees (Adrian et al., 1969). Table 3. Influence of the bean size and of the swelling during the roasting on the niacin content (Adrian et al., 1967) Number of samples coarse beans (oversize of sifter 17/64") middle beans (oversize of sifter 15/64") small beans
Apparent swellina (%)
Niacin content (ma/ IOO a)
21
69.3 (100)
14.60 (100)
21
74.0 (107)
16.43 ( 113)
21
80.8 ( 117)
17.40 ( 119)
Influence of cultural and technological factors Technological factors involving in the conversion of trigonelline can be divised into two main groups : those which facilitate the heat transfer within the bean and are favorable to vitamin synthesis, and those which modify the concentration of trigonelline. The smallness of the bean and its swelling capacity during roasting increase the capacity for heat transfer and, consequently, raise the synthesis rate in niacin. The data of Table 3 correspond to reproducible roastings of moderate intensity like that of American torrefaction. The differences observed would probably be unlike in the case of very intense treatments, of the Italian type. It must be pointed out that the part of swelling is difficult to define : the amount of niacin does not bear a direct relation to the importance of swelling in all the studies. However, the swelling 52
makes the grain texture less dense and should have a positive effect on the penetration of heat during the process. Some modalities seem to decrease the concentration of trigonelline in the bean ; they are the cause of lower niacin synthesis. First, most coffee plantations are established under high trees which supply shade and increase the productivity of the cultivation. According to Carvalho (1962). the shaded culture produce a roasted coffee holding 15% less niacin than unshaded coffee trees. which let us suppose that the synthesis of trigonelline is controlled by a photosynthetic mechanism. In a general way. the technological preparation of the bean can indirectly modify the rate of niacin in roasted coffee : modalities which avoid a damping or soaking of the grain preserve its potential of trigonelline and result in a coffee richer in niacin. Thus, the cherries dried in parchment produce a coffee with more niacin (+19%) than the sun-dried cherries (Carvalho, 1962). Bressani et al. (1961) examine some technics of demucilagination (depulping) by either wet or dry procedure, the dry treatment allows more important syntheses during the roasting. This observation is confirmed by Adrian et al. (1967) who note a superiority of 15% with roasted coffee.demucilaginated by a dry process. Table 4. Niacin content of French commercial coffee. raw or caffeine-free samples (Adrian and Navellier, 1962) Raw coffee Coffee in bean number of samples Total niacin (mg/IOO g) mean extremes Percentage of free niacin Instant coffee number of samples Total niacin (mg/IOO g) Percentage of free niacine
Caffeine-free coffee
8
8
24.0 ± 2.55 19.8 - 28.4 93
16.75 ± 6.0 6.6 - 26.4 94
2
31.5 - 55.5 100
52.8
2 - 64.5 97
The most detrimental operation to the niacin production is the decaffeination : the caffeine-free products contain much less niacin than the raw coffees (Teply et al •• 1945 ; Adrian and Navellier. 1962). This inferiority can be easily imputed to a partial solubilization of trigonelline when the caffeine is extracted (Trugo et al., 1983). Moreover, the niacin rates are more variable in caffeine-free samples : the amount varies in a proportion of 1 to 4 in French coffees (Table 4). Finally. the commercial quality (estimated by the number of defects, color and smell of the green bean) has no repercussion on the richness in niacin in roasted coffees (Adrian et al., 1967). NIACIN CONTENT IN COMMERCIAL COFFEES Some countries, such as the U.S. and Portugal prefer slightly roasted coffee, while others such as France and Italy prefer a more heavily roasted coffee. As the intensity of torrefaction holds the main responsible for the niacin responsible for the niacin synthesis. the vitamin rate of commercial coffees varies in notable proportions according to consumer habits (Table I). Possibly. apart from Italian coffees, torrefaction is stopped before producing the maximal quantity of nicotinic acid. 63
In Anglo-american countries, the most usual niacin content approaches 10 mg/IOO g in U.S. (Teply et al., 1945 ; Teply and Prier, 1957), although Barton-Wright (1944) as well as Hughes and Smith (1946) indicate values between 13 and 18 mg for English products. In Latin America, the coffees are roasted more and have from 22 to 34 mg/IOO g (Daum, 1955 ; Bressani et al., 1961). In Europa, the torrefaction is very often high and niacin is abundant in roasted goods. In France, coffee supplies between 19.8 and 26.0 mg of niacin (Table 4). Italian coffees, known for their strength, contain 32 to 50 mg, with an average of 42 mg (Adamo, 1955). Schematically, Italian products provide 4 times more niacin than American coffees. Table 5. Niacin content of instant coffee from various origin Niacin content (mg/IOO g) U.S.
U.K. Japan France
instant caffeine-free coffee (Baker et al., 1976) instant raw coffee (Trugo et al., 1985) instant raw coffee (Taguchi et a1., .1985) instant raw coffee instant caffeine-free coffee (Adrian and Navellier, 1962)
7.0 20.6 - 46.B 13.3 - 70.6 35.5 - 55.5 52.B - 64.5
The situation is very different for caffeine-free coffee, the niacin content remaining always very inferior to that of raw products, whatever the modality of the torrefaction : Difference Caffeine-free coffees Raw coffees (%) (mg/IOO g)(a) (mg/100 g)(a) American coffees (Teply et al., 1945) French coffees Adrian et al., 1962}
9.5
(3)
24.0 (8)
4.5
(J)
- 53
16.75 (B)
- 30
(a) the values in brackets indicate the number of samples The American caffeine-free products becoming at once weaker in trigonelline and undergoing a moderate heat treatment, the niacin content is inevitably very weak. The decaffeination may be less detrimental when beans are heavily roasted, like in Italian technology. This hypothesis seems to follow from the analyses of French samples. Instant coffee seems to have a high content in niacin (Table 5). Only, the American instant caffeine-free product is very poor in niacin because of the decaffeination process and of weak roasting. All the other instant coffees contain a high niacin concentration, that shows the solubility of the vitamin synthetized during the torrefaction. But the technological modalities of production contribute to determine the richness of instant coffee; this offers a wide variability of data (Table 5). Lastly, the quantities used for drinking preparations are weak and variable, consequently these products have a hardly calculable nutritional interest. BIOLOGICAL EFFICIENCY OF NIACIN COFFEE Almost all the determinations of niacin in coffees are carried out by microbiological method using Lactobacillus arabinosus (L. plantarum). Only the nicotinic acid, the nicotinamide and the nicotinuric acid are measured by this procedure (in Adrian, 1959). Therefore, the vitamin potential esti54
mated by L. arabinosus is likely to correspond to the available amount of niacin. This hypothesis has been checked with many methodologies. First, it must be noted that the niacin formed during roasting is in a free state and that a very high proportion is found in the drink. According to the results of numerous studies (Hughes and Smith, 1946 ; Daum, 1955 ; Cravioto et a1., 1955 ; Bressani et a1., 1961 ; Adrian and Nave11ier, 1961), it is admitted that an average of 85% of the niacin contained in roasted products are found in drinking, either raw or caffeine-free coffee. The nutritional value of a cup of coffee can be estimated thus : 10 g of coffee (sometimes 12 g) are used and 85% of the vitamin content are extracted by simmering water. Under these conditions, a cup of American coffee must provide about 0.85 mg of niacin, a cup of English product about 1.5 mg and a cup of Italian coffee about 3. I mg. If the niacin requirement is assessed at 15 niacin-equiva1ent* per day, these rates supply respectively 7, 10 and 21 per cent of the daily need. Table 6. Effect of coffee consumption on pellagra symptomatology (28 adult humans [al, 4 cups of coffee per day [bl, 2 months) (Adrian et a1., 1969) prevalence before treatment (%) Cutaneous symptoms : hyperpigmentation depigmentation hyperkeratosis Casal I s collar erythema membranous desquamation Merk's seam Buccal symptoms : glossitis fissured lip scrotal tongue cheilitis Intestinal symptoms diarrhea intestinal colic Psycho-nervous symptoms asthenia cephalgia insomnia buzzing in the ears
disappearing after treatment (%)
97 75 61 61 60 57 25
48 71 59 82 75 81 86
54 21 18 18
32 68
36
100
32
82 64
47 47
o
60
89
100
100 100
100
[a] customary diet : 8500 kJ/day, with 55 g of protein, the 4/5 of which are derived from corn. [b] C. arabica (24 mg of niacin/IOO g) whose 4 cups supply II mg/day. Various studies have been undertaken on animals and on humans to reveal the efficiency of coffee niacin. Teply and Prier (1957) with rat and Bressani et a1. (1962) with chicken demonstrate the overall efficiency of this vitamin. Goldsmith et al. (1959) confirm its availability for the human by establishing the rates of urinary outputs of niacin metabolites. It seems obvious that coffee could be a means of prevention or recovery against pellagra, * I niacin-equivalent corresponds to a supplying of I mg of nicotinic acid or a furnishing of 60 mg of tryptophan, which are converted into 1 mg of nicotinic acid by metabolic way.
55
as was already supposed by Teply and Prier (1957), Goldsmith et al. (1959) as well as Bressani and Naverrete (1959). This has been demonstrated in the Central Angola, where corn is the main basis of the customary diet. Table 7. Effect of coffee consumption on blood niacin content on pellagrous humans (72 adult humans, 4 cups of coffee per day, 3 months) (a) (Adrian et a1., 197 I) number of subjects primary pellagra chronic pellagra control subjects (a)
blood serum niacin (~g/ml) before treatment after treatment
34
38 J3
0.120 0.118 0.260
0.267 0.255 0.230
See experimental procedure in Table 6.
Pellagra has been cured by distributing 2 cups of coffee in the morning and 2 in the evening to humans revealing clinical symptoms of deficiency. This treatment lasts for periods of I month, interspersed with one week without coffee. The coffee used was C. arabica because of its lower percentage in caffeine (Gounelle de Pontanel and Astier-Dumas, 1969). The 4 daily cups supplied II mg of niacin and only 0.4 g of caffeine. The pellagra symptoms were recorded before the cure and after 2 months of coffee consumption (Table 6). Most specific signs of deficiency (Casal's collar, Merk's seam, diarrhea and intestinal colic) have been resorbed in a large part. These favorable results do not yet constitute formal proof because the symptomatology of pellagra undergoes a spontaneous annual cycle, with a resorption of lesions during the winter period. That is why a second biochemical experiment has dealt with the restoration of the blood content in niacin in pellagrous adults: within 3 months, the blood concentration has risen to 120% (Table 7). It even became slightly superior to those of control subjects, in good health. Thus, the niacin of roasted coffee is fully available and this drink could play an useful part in the prevention or cure of the pellagra. It is due to eating habits that Bressani and Navarrete (1959) attribute the absence of pellagra in Central America, where the consumption of coffee is high : they have noticed that in many regions of Guatemala people drank on average 3 cups of coffee per day and that concurrently pellagra remains non-existent in spite of the predominant part of corn in the diet. The results obtained with living organisms entirely confirm the above hypothesis. The richness of niacin in coffee is particularly noticeable in Africa and Central and Latin America because of the frequent proximity of corn cultivations and of coffee plantations. It would be justified to incite local consumption of coffee in the producting countries to prevent any pellagra risk (Davis, 1978). ACKNOWLEDGEMENTS We thank deeply M. Joaquim Xabregas and The Instituto do Caf~ de Angola for their help and keen interest in the work related to pellagra. REFERENCES Adamo, G. (1955). 11 contenuto in acido nicotinico nel caffe. Boll. Soc. Ital. BioI. Sper., l.!., 79-82.
56
Adrian, J. (1959). Le dosage microbiologique des vitamines du groupe B. Cahier technique du CNERNA, 183 p., CNRS, Paris. Adrian, J. and Navellier, P. (1961). Int~r@t nutritionnel du source de vitamine PP. Caf~, Cacao, Th~, 1, 263-268.
caf~
comme
Adrian, J. and Navellier, P. (1962). Teneur en vitamine PP de differents ~chantillons de cafe du commerce. Cafe, Cacao. The. ~. 224-227. Adrian. J. (1963). Synthese de la niacine au cours de la torrefaction du caf~ et son efficacite biologique. Cafe. Cacao. The. I. 359-365. Adrian. J., Frangne, R•• Xabregas, J., and Corte dos Santos, A. (1967). Teneur en vitamine PP des cafes grill~s de l'Angola : ra1e des facteurs botaniques et technologiques. 3eme ColI. Chimie Cafe. Trieste, 427-435. Adrian, J •• Pena, J •• Morais de Carvalho. J •• Miranda. A•• Xabregas. J •• and Corte dos Santos. A. (1969). La boisson de cafe dans Ie traitement de la pellagre humaine. 4eme ColI. Chimie Cafe. Amsterdam. 232-242 ; instalment Instituto do cafe de Angola, Luanda, 20 p. Adrian. J •• Xabregas. J •• Pena, J., Morais de Carvalho. J •• and Gomes. N. (1971). La restauration en vitamine PP par la consommation de cafe. Etude chez Ie pellagreux. 5erne Colloque Chimie Cafe. Lisboa. 371-374 Adrian. J. (1972). La consommation de cafe et la pellagre. Med. et Nutr.,.!!., 71-80. Adrian, J., Guillaume, J.L., and Rabache. M. (1985). Occurrence of aminoacid pyrolysis products in roasted coffees and overgrilled cereal products. Sci. Aliments, 1. hors s~rie. 199-203. Baker. D.H •• Yen. J.T., Jensen. A.H •• Teeter, R.G •• Michel, E.N., and Burns. J.H. (1976). Niacin activity in niacin amide and coffee. Nutr. Repts. Int •• J!. 115-120. Barbiroli. G. (1966). Ricerche sperimentali sulla tostatura degli alimenti 1a trssformazione della trigonellina in acido nicotinico. Univ. Messina, atti del V convegno della qualita, 10-12 septembre. Barton-Wright. E.C. (1944). The microbiological assay of nicotinic acid in cereal and other products. Biochem. J •• ~, 314-319. Boddeker, H. and Mishkin. A.R. (1963). Determination of nicotinic acid in coffee by paper chromatography. Anal. Chem •• 11. 1662-1663. Bracco. U. (1973). Determination des hydrocarbures polycyeliques aromatiques : technique et application aux huiles de cafe. Riv. Ital. Sost. Grasse., 50, 166-176. Bressani, R. and Navarrete. D.A. (1959). Niacin content of coffee in Central America. Food Res., 24. 344-351. Bressani. R•• Fiester, D•• Navarrete, D.A •• and Scrimshaw. N.S. (1961). Effect of processing method and variety on niacin and ether extract content of green and roasted coffee. Food Technol •• 11. 306-308. Bressani, R•• Gomez-Brenes. R•• and Conde. R. (1962). Cambios de la composieion quimica del grano y de la pulpa del cafe durante el proceso de tostacion. y actividad biologica de la niaeina del cafe. Arch. Venez. Nutr., E, 93-104.
57
Callahan, M.M., Rohovsky, M.W., Robertson, R.S., and Yesair, D.W. (1979). The effect of coffee consumption on plasma lipids, lipoproteins and the development of aortic atherosclerosis in rhesus monkey fed an atherogenic diet. Am. J. Clin. Nutr., ~, 834-845. Carvalho, A. (1962). Variability of the niacin content in coffee. Nature, ~, 1096. Challis, B.C. and Bartlett, C.D. (1975). Possible cocarcinogenic effects of coffee constituents. Nature, 254, 532-533. Chassevent, F. (1969). L'acide chlorogenique : ses actions physiologiques et pharmacologiques. Ann. Nutr. Alim.,~, 1-14. Cravioto, R.O., Guzman, J.G., and Suarez, M.L.S. (1955). Increments del contenido de niacina durante la torrefaccion del cafe y su significado. Ciencia, ji, 24-26. Czok, G. (1977). Kaffee und Gesundheit. Z. Ernahrungsw., ji, 248-255. Daum, M.G. (1955). La niacina en el cafe y su importancia nutricional en Venezuela. Arch. Venez. Nutr., ~, 61-70. Davis, R.G. (1978). Increased bitter taste detection thresholds in Yucatan inhabitants related to coffee as a dietary source of niacin. Chem. senses flavor, i, 423-429. Feinberg, L.J., Sanberg, H., de Castro, 0., and Bellet, S. (1976). Effects of coffee ingestion on oral glucose tolerance curves in normal human subjects. Metabolism, lI, 916-922. Fritz, W. (1969). Zum Losungsverhalten der Polyaromate beim Kochen von Kaffee-Ersatzstoffen und Bohnenkaffee. Deut. Lebensm. Rdsch., ~, 83-85. Fritz, W. (1975). Entstehen bei der Zubereitung von Lebensmitteln krebserzeugende Stoffe ? Ernahrungsforschung. Nahrung, ~, 83-87. Gassmann, B. and Schneeweiss, R. (1959). The vitamins Bl, B2 and PP in the profile of normally baked breads and those baked with infrared radiation. Nahrung, i, 42-54. Goldsmith, G.A., Miller, O.N., Unglaub, W.G., and Kercheval, K. (1959). Human studies of biological availability of niacin in coffee. Proc. soc. Exp. BioI. Med., ~, 579-580. Gounelle de Pontanel, H. and Astier-Dumas, M. (1969). Cafe et Sante: la place du cafe en dietetique. Bull. Acad. Nat. Medecine, ~, 628-640. Hughes, E.B. and Smith, R.F. (1946). The nicotinic acid content of coffee. J. Soc. Chern. Ind., 65, 284-286. Kitts, D.D. and Mathieson, R. (1989). Effects of caffeine on ovine maternal glucose-insulin response and fetal metabolite levels. Nutr. Repts. Int., 40, 673-684. Kuratsune, H. and Hueper, W.C. (1960). Polycyclic aromatic hydrocarbons in roasted coffee. J. Nat. Cancer lnst., ~, 463-469. Miller, A.B. (1983). Coffee and Cancer. Carcin. Hutag. Environm.,
58
i,
13-20.
Naismith, D.J., Akinyanju, P.A., and Yudkin, J. (1969). Influence of caffeine-containing beverages on the growth, food utilization and plasma lipids of the rat. J. Nutr., ~, 375-381. Panigrahi, G.B. and Rao, A.R. (1983). Influence of caffeine on arecolineinduced SCE in mouse bone-marrow cells in vivo. Mutation Res., Jl!, 347-353. Pozniak, P.C. (1988). The carcinogenicity of caffeine and coffee. J. Am. Diet. Assoc., 85, 1125-1133. Querat, F. and Heraud, G. (1987). Pathologie du cafe. Gaz. Medicale, 94, nO 33, 37-40. Sandler, R.S. (1983). Diet and cancer Nutr. Cancer, !, 273-279.
food additives, coffee and alcohol.
Slattery, M.L., West, D.W., and Robinson, L.M. (1988). Fluid intake and bladder cancer in Utah. Intern. J. Cancer, 42, 17-22. Soos, K. and Fazy, I. (1974). The content of polyaromatic hydrocarbon carcinogens in various coffee types. Edisipar., 25, 7-10 ; 37-40 ; 65-69. Stavric, B. (1988). Methylxanthines : toxicity to human. 2 Caffeine. Food Chem. Toxic., 26, 645-662. Strobel, R.G.K. (1974). The determination of 3-4 benzopyrene of coffee products. 6eme Coli. Chimie Cafe, Bogota, 128-134. Strubelt, 0., Siegers, C.P., Breining, H., and Steffen, J. (1973). Tierexperimentelle Untersuchungen zur chronischen Toxizitit von Kaffee und Coffein. Z. Ernihrungsw., j1, 252-260. Taguchi, H., Sakaguchi, M., and Shimabayashi, Y. (1985). Trigonelline content in coffee beans and the thermal conversion of trigonelline into nicotinic acid during the roasting of coffee bean. Agric. Bioi. Chem., 49, 3467-3471. Tchetche, A.G. (1979). Dosage quantitatif de la vitamine PP dans Ie Coffea canephora var. robusta par une methode microbiologique utilisant Lactobacillus arabinosus. 8th Intern. Coli. on Coffee, 147-152. Teply, L.J., Krehl, W.A., and Elvehjem, C.A. (1945). Studies on the nicotinic acid content of coffee. Arch. biochem.,!, 139-1. Teply, L.J. and Prier, R.F. (1957). Nutritional evaluation of coffee including niacin bioassay. J. Agric. Food Chem., 1, 375-7. Thomas, H.E.jr (1979). The relationship of coffee drinking to death and cardiovascular disesse. 8th Intern. ColI. on Coffee, 305-310. Trugo, L.C., Macrae, R., and Dick, J. (1983). Determination of purine alkaloids and trigonelline in instant coffee and other beverages using high performance liquid chromatography. J. Agric. Food Chem., 34, 300-306. Trugo, L.C., Macrae, R., and Trugo, N.M.F. (1985). Determination of nicotinic acid in instant coffee using high-performance liquid chromatography. J. Micronutrient anal., 1, 55-63.
59
Jean Adrian and Regine Frangne Chaire de Biochimie industrielle et agro-alimentaire Conservatoire National des Arts et Metiers 292, rue Saint-Martin, 75003 Paris, France ABSTRACT The coffee bean contains about 1% of trigonelline that is demethylated at temperatures approaching 200 0 C ; it is partially converted into nicotinic acid. This operation is mainly proportional to the severity of dry heat treatment ; various other physico-chemical factors also influence the synthesis of niacin during the roasting. The niacin content of weakly roasted commercial coffee is about 10 mg/100 g (American coffee) and it reaches 40 mg in heavy roasted coffees, i.e. Italian coffee. Caffeine-free coffee is lower in niacin than the corresponding raw coffee. The drinking retains 85% of the niacin formed during roasting ; it is totally available for the organism and can constitute a noticeable part of the daily supply in niacin. INTRODUCTION Some observations attribute various deleterious physiological effects to high coffee consumption. These manifestations are due to the presence of particular constituents in coffee, such as chlorogenic acid (Teply et al., 1945 ; Chassevent, 1969 ; Challis and Bartlett, 1975) or caffeine (Feinberg et al., 1968 ; Naismith et al., 1969 ; Panigrahi and Rao, 1983 Pozniak, 1985 ; Stavric, 1988). However, an unfavourable incidence of coffee on the glycemia, the lipidemia and the occurrence of cardiovascular pathology is not obvious for the healthy person who consumes coffee normally (Callahan et al., 1979 ; Querat and Heraud, 1987 ; Kitts and Mathieson, 1989). Actually, the participation of coffee in the ethiology of cancer remains a point of question (Strubelt et al., 1973 ; Czok, 1977 ; Thomas, 1979 ; Sandler, 1983). The roasting process is responsible for the formation of characteristic molecules. Among them, such aromatic polycyclic hydrocarbons as the benzopyrene have a well established carcinogenic power (Kuratsune and Hueper, 1960 ; Fritz, 1969, 1975 ; Bracco, 1973 ; Soos and F8zy, 1974 ; Strobel, 1974) whereas others like harman and norharman (Adrian et al., 1985) are also likely to occur in the mechanism of the cancerization. These roasting components certainly represent the most probable risk bound to coffee consumption (Miller, 1983 ; Pozniak, 1985).
M. Friedman (ed.), Nutritional and Toxicological Consequences of Food Processing © Springer Science+Business Media New York 1991
49
=
A 30
~
40
,=
E
E 20
~
~
•uM
10
c 10 M
•
M
•
20
40
60
ROASTING TINE (NIN.)
B
,,,
~
~
,-
I I
~
~
c
I
=
,= 20
•MU
,
m30
230·
I
,,
I
10
20
30
40
WEIGHT LOSS (I)
Figure I. Niacin content in coffee according to the intensity of heat treatment. I-A: Roasting measured by the time and temperature of treatment (Hughes and Smith, 1946) ; I-B : Roasting measured by the weight loss of coffee (broken line : C. arabica ; heavy line: C. canephora var. robusta) (Adrian and Navellier, 1961).
However, roasting must not be considered as an operation with only negative consequences ; it develops the characteristic and enjoyed aroma of coffee and it provokes an abundant synthesis of nicotinic acid from trigonelline, which is its methylated precursor. This conversion during the roasting places coffee drinking among the most efficient food resources to satisfy the niacin requirement and to prevent or cure pellagra. The mechanism of the synthesis of niacin from trigonelline and the efficiency of the formed nicotinic acid have been described by different teams whose conclusions all demonstrate the interest of roasted coffee as a contribution to the niacin supply. It would be useful to review these works, often old, because they show that, besides undesirable components, strong roasting can also have favorable consequences in an accurate vitamin field. THE SYNTHESIS OF NICOTINIC ACID DURING COFFEE ROASTING Trigonelline is the N-methyl betaine of nicotinic acid. This compound is widely present in vegetal produce: it can be found in all cereal grains, in potatoes, beetroots, tomatoes, mushrooms and sweet peppers at rates of betwwen 10 and 80 mg/IOO g (Barbiroli, 1966). It is much more abundant in the coffee bean since C. arabica contains about 900 mg and C. canephora var. robusta, about 650 mg according to Hughes and Smith (1946). In a pure state, trigonelline heated in dry conditions is degraded in various derivated compounds, especially at temperatures ranging from 160°C to 230°C (Figure IA) : about 5% are converted to nicotinic acid by demethylation. The same phenomenon occurs in some food technology processes. During experimental burnings, the nicotinic acid content increases in all the grains and vegetals containing trigonelline (Barbiroli, 1966). Under more usual modalities, an increase of the niacin content is observed as a result of an intense heating carried out in dry conditions : for example, the synthesis of niacin is noticeable in bread crust but not in crumb (Gassmann and Schneeweiss, 1959). Coffee roasting is nevertheless the operation most favorable provoking this synthesis from trigonelline : the bean contains a large quantity of trigonelline, its wetness is very weak and the heat treatment easily reaches 200°C during the process.
50
Incidentally, the nicotinic acid is one of the most stable vitamins when met with the customary conditions in food processing, storage, cooking, etc. (Adrian, 1959). In particular, it is remarkably thermostable, which allows it to resist efficiently during a strong heat treatment, at least as long as it does not reach extremes. That is noticed during the commercial torrefaction of coffee. In practice, the importance of the nicotinic acid synthesis and the niacin content in roasted coffee depend on many factors and parameters such as the intensity of treatment, the technological preparation of the bean, etc. Table I. Influence of roasting intensity on the niacin content in coffee (Teply and Prier, 1957) Niacin content mg/IOO g relative value green coffee just before swelling just swelling New England roasting French roasting Italian roasting Heavy roasting (excessive roasting)
2.2 4.0 8.3 13.0 24.9 41.6 43.6
I
1.8 3.8 5.9 11.3 18.9 19.9
Intensity of coffee roasting The abundance of nicotinic acid in roasted coffee depends, first of all, on whether the intensity of torrefaction is estimated by the duration of heating, the browning of the bean or the weight loss (Hughes and Smith, 1946 ; Cravioto et al., 1955 ; Daum, 1955 ; Teply and Prier, 1957 ; Bressani and Navarrete, 1959 ; Adrian and Navellier, 1961 ; Bressani et al., 1962 ; Boddeker and Mishkin, 1963 ; Taguchi et al., 1985). When the roasting is very heavy, the quantity of nicotinic acid always exceeds 40 mg/IOO g of roasted coffee (Table I, Figure IB). This result corresponds to a value 20 to 30 times superior to that of the green bean, which is about 1.5 mg (Adamo, 1955 ; Cravioto et al., 1955 ; Daum, 1955 ; Teply and Prier, 1957 ; Bressani et al., 1959, 1962 ; Adrian and Navellier, 1961 ; Taguchi et al., 1985). According to Figure IB, at the beginning of the roasting, the trigonelline is easily converted into nicotinic acid and the formed molecule remains stable in these modalities. It is only damaged by extremely severe burnings (Adrian and Navellier, 1961 ; Tchetche, 1979) ; this evolution is not observed in all torrefactions : Bressani et ale (1962) proceed to heatings reaching 42% of weight loss without the treatment being accompagnied by any decrease in the niacin content of the roasted produce. Table 2. Niacin content of commercial roasted C. arabica and C. canephora types (a) Arabica Hughes and Smith (1946) Bressani et ale (1961) Adrian et ale (1967) (b)
15.2 (4) 22.65 (29) 18.45 (18)
Robusta 18.8 18.1 15.4
( 4) ( 2)
(45)
(.) the values in brackets represent the number of samples (b) the roasting time is 7.1 min. for Arabica and 8.2 min. for Robusta
51
In fact, coffee torrefaction primarily provokes a nicotinic acid synthesis and secondarily that of nicotinamide (Taguchi et al., 1985) commercial roasting heavy roasting in mg/100 g : nicotinic acid 2.9 - 4.5 28.8 nicotinamide 0.48 - 0.84 2.3 The small proportion of nicotinamide is probably due to the partial degradation of aminoacids with releasing of amino groups, which react on nicotinic acid. Influence of geographical and botanical factors The Arabica and Robusta coffees constitute the two largest species of cultivated coffees. The former tends to contain more trigonelline but requires a less intense torrefaction than the Robusta types. Moreover, during heating, the possibilities of conversion can be compared in the both species. That is why the commercially roasted products have similar amounts of niacin (Figure IB, Table 2). On the other hand, within the species C. canephora, particularities appear according to the geographical area of cultivation. In Angola, a notion of "local variety" has been established to describe specific characteristics of Robusta cultivated in distinct regions. When samples of these "local varieties" are submitted to roasting, the influence of geographical data becomes especially evident : a series of 9 samples roasted under standardized modalities shows the following results : Ambriz "variety" hold 19.85 mg of niacin/lOO g, Cazengo "variety" 15.35 mg and Amboim 13.15 mg (Adrian et al., 1967). These differences are attributed to the combination of geographical, pedologic and climatic factors on the development and the physiology of the coffee tree. They reveal a more important influence on the niacin synthesis than the differences observed between C. arabica and C. canephora. On the other hand, the year of the crop is not a determining factor to the rate of niacin in roasted coffees (Adrian et al., 1969). Table 3. Influence of the bean size and of the swelling during the roasting on the niacin content (Adrian et al., 1967) Number of samples coarse beans (oversize of sifter 17/64") middle beans (oversize of sifter 15/64") small beans
Apparent swellina (%)
Niacin content (ma/ IOO a)
21
69.3 (100)
14.60 (100)
21
74.0 (107)
16.43 ( 113)
21
80.8 ( 117)
17.40 ( 119)
Influence of cultural and technological factors Technological factors involving in the conversion of trigonelline can be divised into two main groups : those which facilitate the heat transfer within the bean and are favorable to vitamin synthesis, and those which modify the concentration of trigonelline. The smallness of the bean and its swelling capacity during roasting increase the capacity for heat transfer and, consequently, raise the synthesis rate in niacin. The data of Table 3 correspond to reproducible roastings of moderate intensity like that of American torrefaction. The differences observed would probably be unlike in the case of very intense treatments, of the Italian type. It must be pointed out that the part of swelling is difficult to define : the amount of niacin does not bear a direct relation to the importance of swelling in all the studies. However, the swelling 52
makes the grain texture less dense and should have a positive effect on the penetration of heat during the process. Some modalities seem to decrease the concentration of trigonelline in the bean ; they are the cause of lower niacin synthesis. First, most coffee plantations are established under high trees which supply shade and increase the productivity of the cultivation. According to Carvalho (1962). the shaded culture produce a roasted coffee holding 15% less niacin than unshaded coffee trees. which let us suppose that the synthesis of trigonelline is controlled by a photosynthetic mechanism. In a general way. the technological preparation of the bean can indirectly modify the rate of niacin in roasted coffee : modalities which avoid a damping or soaking of the grain preserve its potential of trigonelline and result in a coffee richer in niacin. Thus, the cherries dried in parchment produce a coffee with more niacin (+19%) than the sun-dried cherries (Carvalho, 1962). Bressani et al. (1961) examine some technics of demucilagination (depulping) by either wet or dry procedure, the dry treatment allows more important syntheses during the roasting. This observation is confirmed by Adrian et al. (1967) who note a superiority of 15% with roasted coffee.demucilaginated by a dry process. Table 4. Niacin content of French commercial coffee. raw or caffeine-free samples (Adrian and Navellier, 1962) Raw coffee Coffee in bean number of samples Total niacin (mg/IOO g) mean extremes Percentage of free niacin Instant coffee number of samples Total niacin (mg/IOO g) Percentage of free niacine
Caffeine-free coffee
8
8
24.0 ± 2.55 19.8 - 28.4 93
16.75 ± 6.0 6.6 - 26.4 94
2
31.5 - 55.5 100
52.8
2 - 64.5 97
The most detrimental operation to the niacin production is the decaffeination : the caffeine-free products contain much less niacin than the raw coffees (Teply et al •• 1945 ; Adrian and Navellier. 1962). This inferiority can be easily imputed to a partial solubilization of trigonelline when the caffeine is extracted (Trugo et al., 1983). Moreover, the niacin rates are more variable in caffeine-free samples : the amount varies in a proportion of 1 to 4 in French coffees (Table 4). Finally. the commercial quality (estimated by the number of defects, color and smell of the green bean) has no repercussion on the richness in niacin in roasted coffees (Adrian et al., 1967). NIACIN CONTENT IN COMMERCIAL COFFEES Some countries, such as the U.S. and Portugal prefer slightly roasted coffee, while others such as France and Italy prefer a more heavily roasted coffee. As the intensity of torrefaction holds the main responsible for the niacin responsible for the niacin synthesis. the vitamin rate of commercial coffees varies in notable proportions according to consumer habits (Table I). Possibly. apart from Italian coffees, torrefaction is stopped before producing the maximal quantity of nicotinic acid. 63
In Anglo-american countries, the most usual niacin content approaches 10 mg/IOO g in U.S. (Teply et al., 1945 ; Teply and Prier, 1957), although Barton-Wright (1944) as well as Hughes and Smith (1946) indicate values between 13 and 18 mg for English products. In Latin America, the coffees are roasted more and have from 22 to 34 mg/IOO g (Daum, 1955 ; Bressani et al., 1961). In Europa, the torrefaction is very often high and niacin is abundant in roasted goods. In France, coffee supplies between 19.8 and 26.0 mg of niacin (Table 4). Italian coffees, known for their strength, contain 32 to 50 mg, with an average of 42 mg (Adamo, 1955). Schematically, Italian products provide 4 times more niacin than American coffees. Table 5. Niacin content of instant coffee from various origin Niacin content (mg/IOO g) U.S.
U.K. Japan France
instant caffeine-free coffee (Baker et al., 1976) instant raw coffee (Trugo et al., 1985) instant raw coffee (Taguchi et a1., .1985) instant raw coffee instant caffeine-free coffee (Adrian and Navellier, 1962)
7.0 20.6 - 46.B 13.3 - 70.6 35.5 - 55.5 52.B - 64.5
The situation is very different for caffeine-free coffee, the niacin content remaining always very inferior to that of raw products, whatever the modality of the torrefaction : Difference Caffeine-free coffees Raw coffees (%) (mg/IOO g)(a) (mg/100 g)(a) American coffees (Teply et al., 1945) French coffees Adrian et al., 1962}
9.5
(3)
24.0 (8)
4.5
(J)
- 53
16.75 (B)
- 30
(a) the values in brackets indicate the number of samples The American caffeine-free products becoming at once weaker in trigonelline and undergoing a moderate heat treatment, the niacin content is inevitably very weak. The decaffeination may be less detrimental when beans are heavily roasted, like in Italian technology. This hypothesis seems to follow from the analyses of French samples. Instant coffee seems to have a high content in niacin (Table 5). Only, the American instant caffeine-free product is very poor in niacin because of the decaffeination process and of weak roasting. All the other instant coffees contain a high niacin concentration, that shows the solubility of the vitamin synthetized during the torrefaction. But the technological modalities of production contribute to determine the richness of instant coffee; this offers a wide variability of data (Table 5). Lastly, the quantities used for drinking preparations are weak and variable, consequently these products have a hardly calculable nutritional interest. BIOLOGICAL EFFICIENCY OF NIACIN COFFEE Almost all the determinations of niacin in coffees are carried out by microbiological method using Lactobacillus arabinosus (L. plantarum). Only the nicotinic acid, the nicotinamide and the nicotinuric acid are measured by this procedure (in Adrian, 1959). Therefore, the vitamin potential esti54
mated by L. arabinosus is likely to correspond to the available amount of niacin. This hypothesis has been checked with many methodologies. First, it must be noted that the niacin formed during roasting is in a free state and that a very high proportion is found in the drink. According to the results of numerous studies (Hughes and Smith, 1946 ; Daum, 1955 ; Cravioto et a1., 1955 ; Bressani et a1., 1961 ; Adrian and Nave11ier, 1961), it is admitted that an average of 85% of the niacin contained in roasted products are found in drinking, either raw or caffeine-free coffee. The nutritional value of a cup of coffee can be estimated thus : 10 g of coffee (sometimes 12 g) are used and 85% of the vitamin content are extracted by simmering water. Under these conditions, a cup of American coffee must provide about 0.85 mg of niacin, a cup of English product about 1.5 mg and a cup of Italian coffee about 3. I mg. If the niacin requirement is assessed at 15 niacin-equiva1ent* per day, these rates supply respectively 7, 10 and 21 per cent of the daily need. Table 6. Effect of coffee consumption on pellagra symptomatology (28 adult humans [al, 4 cups of coffee per day [bl, 2 months) (Adrian et a1., 1969) prevalence before treatment (%) Cutaneous symptoms : hyperpigmentation depigmentation hyperkeratosis Casal I s collar erythema membranous desquamation Merk's seam Buccal symptoms : glossitis fissured lip scrotal tongue cheilitis Intestinal symptoms diarrhea intestinal colic Psycho-nervous symptoms asthenia cephalgia insomnia buzzing in the ears
disappearing after treatment (%)
97 75 61 61 60 57 25
48 71 59 82 75 81 86
54 21 18 18
32 68
36
100
32
82 64
47 47
o
60
89
100
100 100
100
[a] customary diet : 8500 kJ/day, with 55 g of protein, the 4/5 of which are derived from corn. [b] C. arabica (24 mg of niacin/IOO g) whose 4 cups supply II mg/day. Various studies have been undertaken on animals and on humans to reveal the efficiency of coffee niacin. Teply and Prier (1957) with rat and Bressani et a1. (1962) with chicken demonstrate the overall efficiency of this vitamin. Goldsmith et al. (1959) confirm its availability for the human by establishing the rates of urinary outputs of niacin metabolites. It seems obvious that coffee could be a means of prevention or recovery against pellagra, * I niacin-equivalent corresponds to a supplying of I mg of nicotinic acid or a furnishing of 60 mg of tryptophan, which are converted into 1 mg of nicotinic acid by metabolic way.
55
as was already supposed by Teply and Prier (1957), Goldsmith et al. (1959) as well as Bressani and Naverrete (1959). This has been demonstrated in the Central Angola, where corn is the main basis of the customary diet. Table 7. Effect of coffee consumption on blood niacin content on pellagrous humans (72 adult humans, 4 cups of coffee per day, 3 months) (a) (Adrian et a1., 197 I) number of subjects primary pellagra chronic pellagra control subjects (a)
blood serum niacin (~g/ml) before treatment after treatment
34
38 J3
0.120 0.118 0.260
0.267 0.255 0.230
See experimental procedure in Table 6.
Pellagra has been cured by distributing 2 cups of coffee in the morning and 2 in the evening to humans revealing clinical symptoms of deficiency. This treatment lasts for periods of I month, interspersed with one week without coffee. The coffee used was C. arabica because of its lower percentage in caffeine (Gounelle de Pontanel and Astier-Dumas, 1969). The 4 daily cups supplied II mg of niacin and only 0.4 g of caffeine. The pellagra symptoms were recorded before the cure and after 2 months of coffee consumption (Table 6). Most specific signs of deficiency (Casal's collar, Merk's seam, diarrhea and intestinal colic) have been resorbed in a large part. These favorable results do not yet constitute formal proof because the symptomatology of pellagra undergoes a spontaneous annual cycle, with a resorption of lesions during the winter period. That is why a second biochemical experiment has dealt with the restoration of the blood content in niacin in pellagrous adults: within 3 months, the blood concentration has risen to 120% (Table 7). It even became slightly superior to those of control subjects, in good health. Thus, the niacin of roasted coffee is fully available and this drink could play an useful part in the prevention or cure of the pellagra. It is due to eating habits that Bressani and Navarrete (1959) attribute the absence of pellagra in Central America, where the consumption of coffee is high : they have noticed that in many regions of Guatemala people drank on average 3 cups of coffee per day and that concurrently pellagra remains non-existent in spite of the predominant part of corn in the diet. The results obtained with living organisms entirely confirm the above hypothesis. The richness of niacin in coffee is particularly noticeable in Africa and Central and Latin America because of the frequent proximity of corn cultivations and of coffee plantations. It would be justified to incite local consumption of coffee in the producting countries to prevent any pellagra risk (Davis, 1978). ACKNOWLEDGEMENTS We thank deeply M. Joaquim Xabregas and The Instituto do Caf~ de Angola for their help and keen interest in the work related to pellagra. REFERENCES Adamo, G. (1955). 11 contenuto in acido nicotinico nel caffe. Boll. Soc. Ital. BioI. Sper., l.!., 79-82.
56
Adrian, J. (1959). Le dosage microbiologique des vitamines du groupe B. Cahier technique du CNERNA, 183 p., CNRS, Paris. Adrian, J. and Navellier, P. (1961). Int~r@t nutritionnel du source de vitamine PP. Caf~, Cacao, Th~, 1, 263-268.
caf~
comme
Adrian, J. and Navellier, P. (1962). Teneur en vitamine PP de differents ~chantillons de cafe du commerce. Cafe, Cacao. The. ~. 224-227. Adrian. J. (1963). Synthese de la niacine au cours de la torrefaction du caf~ et son efficacite biologique. Cafe. Cacao. The. I. 359-365. Adrian. J., Frangne, R•• Xabregas, J., and Corte dos Santos, A. (1967). Teneur en vitamine PP des cafes grill~s de l'Angola : ra1e des facteurs botaniques et technologiques. 3eme ColI. Chimie Cafe. Trieste, 427-435. Adrian, J •• Pena, J •• Morais de Carvalho. J •• Miranda. A•• Xabregas. J •• and Corte dos Santos. A. (1969). La boisson de cafe dans Ie traitement de la pellagre humaine. 4eme ColI. Chimie Cafe. Amsterdam. 232-242 ; instalment Instituto do cafe de Angola, Luanda, 20 p. Adrian. J •• Xabregas. J •• Pena, J., Morais de Carvalho. J •• and Gomes. N. (1971). La restauration en vitamine PP par la consommation de cafe. Etude chez Ie pellagreux. 5erne Colloque Chimie Cafe. Lisboa. 371-374 Adrian. J. (1972). La consommation de cafe et la pellagre. Med. et Nutr.,.!!., 71-80. Adrian, J., Guillaume, J.L., and Rabache. M. (1985). Occurrence of aminoacid pyrolysis products in roasted coffees and overgrilled cereal products. Sci. Aliments, 1. hors s~rie. 199-203. Baker. D.H •• Yen. J.T., Jensen. A.H •• Teeter, R.G •• Michel, E.N., and Burns. J.H. (1976). Niacin activity in niacin amide and coffee. Nutr. Repts. Int •• J!. 115-120. Barbiroli. G. (1966). Ricerche sperimentali sulla tostatura degli alimenti 1a trssformazione della trigonellina in acido nicotinico. Univ. Messina, atti del V convegno della qualita, 10-12 septembre. Barton-Wright. E.C. (1944). The microbiological assay of nicotinic acid in cereal and other products. Biochem. J •• ~, 314-319. Boddeker, H. and Mishkin. A.R. (1963). Determination of nicotinic acid in coffee by paper chromatography. Anal. Chem •• 11. 1662-1663. Bracco. U. (1973). Determination des hydrocarbures polycyeliques aromatiques : technique et application aux huiles de cafe. Riv. Ital. Sost. Grasse., 50, 166-176. Bressani, R. and Navarrete. D.A. (1959). Niacin content of coffee in Central America. Food Res., 24. 344-351. Bressani. R•• Fiester, D•• Navarrete, D.A •• and Scrimshaw. N.S. (1961). Effect of processing method and variety on niacin and ether extract content of green and roasted coffee. Food Technol •• 11. 306-308. Bressani, R•• Gomez-Brenes. R•• and Conde. R. (1962). Cambios de la composieion quimica del grano y de la pulpa del cafe durante el proceso de tostacion. y actividad biologica de la niaeina del cafe. Arch. Venez. Nutr., E, 93-104.
57
Callahan, M.M., Rohovsky, M.W., Robertson, R.S., and Yesair, D.W. (1979). The effect of coffee consumption on plasma lipids, lipoproteins and the development of aortic atherosclerosis in rhesus monkey fed an atherogenic diet. Am. J. Clin. Nutr., ~, 834-845. Carvalho, A. (1962). Variability of the niacin content in coffee. Nature, ~, 1096. Challis, B.C. and Bartlett, C.D. (1975). Possible cocarcinogenic effects of coffee constituents. Nature, 254, 532-533. Chassevent, F. (1969). L'acide chlorogenique : ses actions physiologiques et pharmacologiques. Ann. Nutr. Alim.,~, 1-14. Cravioto, R.O., Guzman, J.G., and Suarez, M.L.S. (1955). Increments del contenido de niacina durante la torrefaccion del cafe y su significado. Ciencia, ji, 24-26. Czok, G. (1977). Kaffee und Gesundheit. Z. Ernahrungsw., ji, 248-255. Daum, M.G. (1955). La niacina en el cafe y su importancia nutricional en Venezuela. Arch. Venez. Nutr., ~, 61-70. Davis, R.G. (1978). Increased bitter taste detection thresholds in Yucatan inhabitants related to coffee as a dietary source of niacin. Chem. senses flavor, i, 423-429. Feinberg, L.J., Sanberg, H., de Castro, 0., and Bellet, S. (1976). Effects of coffee ingestion on oral glucose tolerance curves in normal human subjects. Metabolism, lI, 916-922. Fritz, W. (1969). Zum Losungsverhalten der Polyaromate beim Kochen von Kaffee-Ersatzstoffen und Bohnenkaffee. Deut. Lebensm. Rdsch., ~, 83-85. Fritz, W. (1975). Entstehen bei der Zubereitung von Lebensmitteln krebserzeugende Stoffe ? Ernahrungsforschung. Nahrung, ~, 83-87. Gassmann, B. and Schneeweiss, R. (1959). The vitamins Bl, B2 and PP in the profile of normally baked breads and those baked with infrared radiation. Nahrung, i, 42-54. Goldsmith, G.A., Miller, O.N., Unglaub, W.G., and Kercheval, K. (1959). Human studies of biological availability of niacin in coffee. Proc. soc. Exp. BioI. Med., ~, 579-580. Gounelle de Pontanel, H. and Astier-Dumas, M. (1969). Cafe et Sante: la place du cafe en dietetique. Bull. Acad. Nat. Medecine, ~, 628-640. Hughes, E.B. and Smith, R.F. (1946). The nicotinic acid content of coffee. J. Soc. Chern. Ind., 65, 284-286. Kitts, D.D. and Mathieson, R. (1989). Effects of caffeine on ovine maternal glucose-insulin response and fetal metabolite levels. Nutr. Repts. Int., 40, 673-684. Kuratsune, H. and Hueper, W.C. (1960). Polycyclic aromatic hydrocarbons in roasted coffee. J. Nat. Cancer lnst., ~, 463-469. Miller, A.B. (1983). Coffee and Cancer. Carcin. Hutag. Environm.,
58
i,
13-20.
Naismith, D.J., Akinyanju, P.A., and Yudkin, J. (1969). Influence of caffeine-containing beverages on the growth, food utilization and plasma lipids of the rat. J. Nutr., ~, 375-381. Panigrahi, G.B. and Rao, A.R. (1983). Influence of caffeine on arecolineinduced SCE in mouse bone-marrow cells in vivo. Mutation Res., Jl!, 347-353. Pozniak, P.C. (1988). The carcinogenicity of caffeine and coffee. J. Am. Diet. Assoc., 85, 1125-1133. Querat, F. and Heraud, G. (1987). Pathologie du cafe. Gaz. Medicale, 94, nO 33, 37-40. Sandler, R.S. (1983). Diet and cancer Nutr. Cancer, !, 273-279.
food additives, coffee and alcohol.
Slattery, M.L., West, D.W., and Robinson, L.M. (1988). Fluid intake and bladder cancer in Utah. Intern. J. Cancer, 42, 17-22. Soos, K. and Fazy, I. (1974). The content of polyaromatic hydrocarbon carcinogens in various coffee types. Edisipar., 25, 7-10 ; 37-40 ; 65-69. Stavric, B. (1988). Methylxanthines : toxicity to human. 2 Caffeine. Food Chem. Toxic., 26, 645-662. Strobel, R.G.K. (1974). The determination of 3-4 benzopyrene of coffee products. 6eme Coli. Chimie Cafe, Bogota, 128-134. Strubelt, 0., Siegers, C.P., Breining, H., and Steffen, J. (1973). Tierexperimentelle Untersuchungen zur chronischen Toxizitit von Kaffee und Coffein. Z. Ernihrungsw., j1, 252-260. Taguchi, H., Sakaguchi, M., and Shimabayashi, Y. (1985). Trigonelline content in coffee beans and the thermal conversion of trigonelline into nicotinic acid during the roasting of coffee bean. Agric. Bioi. Chem., 49, 3467-3471. Tchetche, A.G. (1979). Dosage quantitatif de la vitamine PP dans Ie Coffea canephora var. robusta par une methode microbiologique utilisant Lactobacillus arabinosus. 8th Intern. Coli. on Coffee, 147-152. Teply, L.J., Krehl, W.A., and Elvehjem, C.A. (1945). Studies on the nicotinic acid content of coffee. Arch. biochem.,!, 139-1. Teply, L.J. and Prier, R.F. (1957). Nutritional evaluation of coffee including niacin bioassay. J. Agric. Food Chem., 1, 375-7. Thomas, H.E.jr (1979). The relationship of coffee drinking to death and cardiovascular disesse. 8th Intern. ColI. on Coffee, 305-310. Trugo, L.C., Macrae, R., and Dick, J. (1983). Determination of purine alkaloids and trigonelline in instant coffee and other beverages using high performance liquid chromatography. J. Agric. Food Chem., 34, 300-306. Trugo, L.C., Macrae, R., and Trugo, N.M.F. (1985). Determination of nicotinic acid in instant coffee using high-performance liquid chromatography. J. Micronutrient anal., 1, 55-63.
59
