Natural toxicants in human foods: psoralens in raw and cooked parsnip root.

Parsnip root contains three photoactive, mutagenic, and photocarcinogenic psoralens in a total concentration of about 40 parts per million. These chemicals are not destroyed by normal cooking procedures (boiling or microwave); thus humans are exposed to appreciable levels of psoralens through the consumption of parsnip and possibly other psoralen-containing foodstuffs. The toxicologic consequences to man of such exposure may be speculated on the basis of medicinal and laboratory studies, but epidemiologic data are not available.

Mercury content of common foods determined by neutron activation analysis.

The mercury contents in samples of flour, sugar, nonfat dry milk, potatoes, hamburger, chicken breast, shrimp, liver, eggs, and whole milk were determined by neutron activation analysis. The mercury was separated by anion exchange chromatography and precipitated as the sulfide. The mercury concentrations for all these foods were below 50 parts per billion.

Carotenoid analyses of selected raw and cooked foods associated with a lower risk for cancer.

We examined the carotenoid content of selected foods consistently found to be associated with a lower risk for various epithelial cancers in epidemiologic studies. Both raw and cooked samples of green, leafy vegetables and yellow or orange vegetables were quantitatively examined by high-performance liquid chromatography for individual carotenoid content. The results indicated that fresh, green, leafy vegetables were moderately high in beta carotene (0.5-14.6 mg/100 g) and very high in oxygenated carotenoids or xanthophylls, primarily lutein and its stereoisomers (2.3-63.0 mg/100g) [corrected]. The fresh, yellow or orange vegetables examined were very high in beta carotene (16.0-120.5 mg/100 g) [corrected] but had no detectable nonhydrocarbon carotenoids. Cooking differentially reduced the lutein content compared with the beta carotene content in green, leafy vegetables. These analyses suggest that consumption of carotenoids in addition to beta carotene may be associated with a lower risk for cancer.

Vitamin A in new food composition tables from Japan.

Standard Tables of Food Composition in Japan--Fourth Revised Edition--were issued to the public in October 1982. The data for vitamin A include chemically determined preformed vitamin A and provitamin A. Vitamin A values in the tables are shown as micrograms of retinol and/or micrograms of carotene and also as international units of retinol potency per 100 g edible portion of the foods. The term "carotene" is used for the total mixed carotenoids with vitamin A activity analyzed. In fact, the carotenes analyzed consisted of a large portion of beta-carotene and a small portion of alpha-carotene. One retinol potency (IU) is equivalent to 0.3 micrograms of retinol or 1.8 micrograms (0.6 micrograms X 3) of carotene, which means that the biologic activity of carotenoids is considered to be one-third that of retinol, mainly due to the inferior absorption of carotene. However, as an exception, in milk and dairy products 1 IU is considered to be equivalent to 0.6 micrograms of carotene. As for the foods for which both retinol and carotene contents are given in the table, the retinol potency is calculated by addition of values derived from both retinol and carotene. There are considerable differences in vitamin A values of some foods between the third revised edition and the fourth revised edition of the Standard Tables of Food Composition in Japan. The mean daily vitamin A intake of Japanese people as determined by the recent national nutrition survey was 2,120 IU when the calculation was made from the data in the fourth revised edition and 1,762 IU when the calculation was made from the data in the third revised edition, indicating approximately a 20% larger value in the former than the latter. This fact has to be taken into consideration in the comparison between new data and previous data from nutrition surveys.

Evaluation of vitamin A and carotenoid data in food composition tables.

Vitamin A and carotenoid data in food composition tables have been generated with the use of the Association of Official Analytical Chemists procedures. These procedures fractionate complex mixtures of retinoids and carotenoids into only three fractions, e.g., retinoids, carotenes, and xanthophylls. Food tables lack uniformity in the presentation of vitamin A and carotenoid data. For example, the revisions of Agriculture Handbook, No. 8 (Washington, D.C.: U.S. Govt Print Off, 1963), report both international units and retinol equivalents of vitamin A in foods. Food tables published in the United Kingdom (Paul AA, Southgate DA. McCance and Widdowson's: The composition of foods. Amsterdam: Elsevier/North-Holland, 1978) report preformed vitamin A as retinol and carotenoids as carotene on a weight basis (micrograms/100 g food). The vitamin A and carotenoid data currently available in food composition tables allow only the intake of total vitamin A activity to be estimated. Information about the intake of specific species of retinoids or carotenoids is not available and could not be calculated from existing food tables. Limited applications of new separation techniques and chemical instrumentation have demonstrated the potential for generating detailed analytic information in regard to the retinoid and carotenoid contents of foods.

Mutagenicity of extracts of pickled vegetables collected in Linhsien County, a high-incidence area for esophageal cancer in Northern China.

Extracts of pickled vegetables commonly consumed in Linhsien County, a high-incidence area for esophageal cancer in Northern China, were studied for mutagenicity. The liquid residue from ethereal extracts produced a dose-dependent increase of mutants in Salmonella typhimurium TA98 and TA100 strains; mutagenicity required the presence of a fortified liver microsomal activation system induced by Aroclor 1254 in adult male BD VI inbred rats. An amount of extract equivalent to 2.8 g fresh pickled vegetables produced sixfold (75 revertants/g) and twofold (45 revertants/g) increases in revertant frequencies in strains TA98 and TA100, respectively. Roussin's red methyl ester, a tetranitroso compound, [(NO)2Fe(CH3S)]2, not previously reported to occur in nature, was isolated and identified from the ethereal extracts. The synthetic compound was mutagenic in strain TA100 in the presence of a liver activation system, producing 25 revertants/mumol. Findings on the presence of mutagenic compounds in pickled vegetables were discussed in relation to their possible etiologic role in cancer of the esophagus in Linhsien County.

Quantitative analysis of Roussin red methyl ester in pickled vegetables.

In a study of the etiological factors in esophageal cancer, Roussin red methyl ester was isolated and identified in pickled vegetables of Linxian County, North China, where there is a high incidence of esophageal cancer. In this paper, a method is described for quantitative analysis of Roussin red methyl ester in pickled vegetables by gas chromatography-high-resolution mass spectrometry single-ion monitoring. The content of Roussin red methyl ester in pickled vegetables from Linxian has been found to be 0.1 to 4.5 mg/kg, and that from Beijing is below 0.005 mg/kg, which is the detection limit of the analytical method used. The marked difference between the contents might be one of the possible reasons for the difference in esophageal cancer incidence between the two regions.

Studies on plantacyanin. I. Distribution in the plant kingdom, subcellular localization and physicochemical properties.

The standard preparative procedure used earlier for the isolation of plantacyanin from cucumber peelings and spinach leaves was employed for the detection of this protein in 15 plants. It was found that four of these 15 species of plants contain plantacyanin. From the study of the subcellular distribution of plantacyanin in spinach leaves, it has been concluded that the protein is associated with chloroplasts. Macromolecular properties as well as optical, fluorescence and EPR spectra of plantacyanin from cucumber were studied. It has been shown that at alkaline pH, a short living state of the protein active site is formed. This state (state 3) has an absorption band at 575 nm. The incubation of the protein in alkaline media under anaerobic conditions brings about the bleaching of plantacyanin accompanied by the disappearance of the EPR signal. The effect of some protein-modifying agents was studied and the data obtained are discussed from the point of view of possible ligand amino acids of copper in plantacyanin. The protein copper can be removed by diethyldithiocarbamate and the procedure of reconstitution of holoprotein from apoprotein and Cu2+ was developed.

Dietary fiber: analysis and food sources.

The sources of dietary fiber are reviewed and the chemistry of the various components discussed in relation to their possible physiological properties and their analytical measurement in foods and the diet as a whole. Complete fractionation of all the polysaccharide species would be a time-consuming exercise; nevertheless some characterization of dietary fiber seems essential to an understanding of any physiological role. The composition of the dietary fiber in a range of foodstuffs is presented, and the relative contributions of the major food groups to the intake of total dietary fiber is outlined.

The lignin fraction of plant cell walls.

Methods are discussed for determining lignin in plant cell walls. The increase in apparent lignin content that may occur as a result of artifacts produced during food preparation is also discussed. The phenolic components, including lignin, of cell walls separated from 12 vegetable, fruit, and cereal foods are determined. Wheat bran, on a fresh weight basis, had a high cell wall content (48.6%) compared with the vegetables and fruit whose contents ranged from 1.0 to 6.0%. Wheat bran and spinach cell walls had the highest lignin contents (11.4 and 4.4%, respectively) and the lowest degradabilities (35.8 and 39.6%, respectively) as determined by a cellulase technique. Cell walls of spinach, beetroot, sweet corn, pineapple, and wheat bran contained the phenolic acids (ferulic, p-coumaric, and diferulic) bound to polysaccharide components. Cell walls of cabbage, Brussels sprout, celery, cauliflower, green bean, carrot, and pea contained only traces of these acids.

Dietary fiber content of selected foods.

Dietary fiber measurements are essential to assessment of the potential therapeutic and preventive effects of fiber intake. Ideally, dietary fiber analyses should measure all components--soluble polysaccharides, noncellulosic polysaccharides, cellulose, and lignin--and the constituent sugars of the soluble and noncellulosic polysaccharides. We modified existing techniques to measure reproducibly the total dietary fiber, polysaccharide, and lignin components and the sugar constituents of selected foods. Soluble-fiber content as percentage of total dietary fiber for groups of foods averaged 32% for cereal products, 32% for vegetables, 25% for dried beans, and 38% for fruits. Lignin content, estimated gravimetrically, was approximately 1.4 g/100 g dry wt for 24 foods. Detailed fiber measurements are critical for evaluating the potential health benefits of dietary fiber intake.

Estimation of per capita crude and dietary fiber supply in 38 countries.

Per capita availability of crude and dietary fiber was estimated from 1972-74 FAO Food Disappearance Tables and factors derived from data on the fiber composition of foods. Total dietary and crude fiber, as well as dietary fiber from total cereals, wheat, rice, maize, other cereals, roots and tubers, pulses, vegetables, and fruits, was calculated for 38 countries. Total dietary fiber correlated highly with total crude fiber (r = 0.92). Cereals were the major source of fiber for most countries. With a few exceptions the estimated total dietary fiber and dietary fiber from cereals appear to be in reasonable agreement with other values obtained by a variety of methods for a number of developed countries. The reasons for some of the discrepancies found and the need for additional, more precise information on dietary fiber intake are discussed.

Relationship between calculated dietary and crude fiber intakes of 200 college students.

Two hundred college students recorded dietary intakes for 2 consecutive days. Records were analyzed for 12 nutrients, and crude (CF) and dietary fiber (DF). Means of the 2 days were calculated to represent the daily intake of each student. The crude and dietary fiber provided by cereals and legumes, vegetables, fruits, and miscellaneous sources were determined for a subpopulation of 50 students. Mean (+/-SD) daily CF intake of the population was 4.07 +/- 2.17 g, of DF, 15.36 +/- 7.96 g. When related to energy intake, women (n = 143) consumed significantly more CF and DF than the men (n = 57). Only the cereal and legume group provided appreciably more DF than CF, 32.4 versus 18.8%, of the total daily intake of the subpopulation. The mean DF: CF ratio for the population was 3.9 +/- 1.0 and intakes of the two measures of fiber were highly correlated, r = 0.86. Intakes of CF/1000 kcal and the number of servings of cereal products may indicate when the daily DF intake is less than or equal to 3 or greater than or equal to 5 times the daily CF intake. These data suggest that for the majority of the college-aged population we studied, who consumed a wide variety of foods, a daily DF intake ranged from 3 to 5 times the daily CF intake.

Soluble and insoluble plant fiber in selected cereals and vegetables.

Certain soluble plant fibers have hypocholesterolemic effects. However, information about soluble fibers of most foods is not available. A modification of Southgate's method was used to separate plant fiber into soluble and insoluble fractions. Both fractions were hydrolyzed with dilute sulfuric acid or trifluoracetic acid. Comparison of these two hydrolysis methods was discussed. The hydrolyzed sugars were measured by gas-liquid chromatography. Fiber content of selected cereals and vegetables were examined. Oat, corn, and cooked beans have greater proportions of soluble fiber than wheat and green vegetables. Soluble fractions of cereals and beans were mainly glucose whereas in green vegetables uronic acids were the major components. The differing fiber composition of various plant food sources probably is important in determining their physiological effects.

Folate conjugase activity in fresh vegetables and its effect on the determination of free folate content.

The levels of naturally-occurring folate conjugates in fresh asparagus, broccoli, Brussels sprouts, cabbage, cauliflower lettuce, and spinach were determined together with the effect on the hydrolysis of pteroylpolyglutamates during the extraction phase of the Lactobacillus casei assay. The levels of conjugate activity vary widely among the vegetables tested, from a low of 0.30 +/- 0.21 in lettuce to a high of 106.2 +/- 6.6 mumoles/hr per 100 mg of fresh weight in spinach. The free folate values (L. casei activity before chicken pancreas conjugase treatment) in the fresh vegetables boiled 1 hr after homogenization were significantly higher than those in vegetables boiled before homogenization, while there were no significant differences in total folate (L. casei activity after treatment with conjugase from chicken pancreas). The increase in free folate values must have been due to the action of endogenous conjugases on the pteroylpolyglutamates from the time the vegetables were homogenized until the conjugase was inactivated by heat. It is concluded that reliable free folate values in fresh vegetables can be obtained only if the naturally present conjugases are inactivated before homogenization of sample in order to prevent the hydrolysis of the pteroylpolyglutamates in the vegetables.

Estimated fluoride intake of 6-month-old infants in four dietary regions of the United States.

Eleven composite food groups comprising the infant "market basket" food collections for 1977 or 1978 from each of four dietary regions of the United States were analyzed for their fluoride content. Based upon the determined fluoride content of each composite and Food and Drug Administration estimates of food consumption the daily fluoride intake of an average 6-month-old infant residing in each of the dietary regions was calculated. The daily fluoride intake varied from 0.207 mg/day in Grand Rapids, Mich. (north central dietary region) to 0.541 mg/day in Orlando, Fla. (south dietary region). Flouride intakes of 0.272 and 0.354 mg/day were calculated for Philadelphia, Pa. (northeast dietary region) and Los Angeles, Calif. (west dietary region), respectively. The fluoride content of the water supplies ranged from 0.37 ppm (Los Angeles) to 1.04 ppm (Grand Rapids). Drinking water, dairy products and substitutes (other than milk), and grain and cereal products contributed 44 to 80% of the daily fluoride intake. In three of the four dietary regions the daily fluoride intake was less than the optimum level of 0.05 mg/kg body weight.

Fluoride intakes of young male adults in the United States.

The total daily fluoride intake for young male adults living in four geographical areas of the United States has been estimated by the analysis of "market basket collections" obtained in 1975 and 1977. The fluoride intake in 1975 vaired from 0.912 mg/day in an unfluoridated city (Kansas City, Mo.) to 1.720 mg/day in a fluoridated city (Atlanta, Ga.). The 1977 collection from San Francisco, Calif. contained more fluoride (1.636 mg/day) than the 1975 collection (1.213 mg/day). The level of intake found in this study is less than that reported by San Filippo and Battistone (Clin. Chem. Acta 31: 453, 1971) who analyzed similar collections from Baltimore, Md. in 1967 to 1968 (2.09 to 2.34 mg/day). This study, as well as the earlier one, indicates that relatively low levels of fluoride are being consumed in the United States by the young adult male 16 to 19 years of age.

An in vitro method for predicting the bioavailability of iron from foods.

An in vitro method for the determination of availability of nonheme iron from foods and diets was investigated. Food was extracted with pepsin-HCl at pH 1.35 and subsequently the pH was adjusted to pH 7.5 and filtered. Ionizable iron was determined in the pH 7.5 filterate by the alpha, alpha-dipyridyl method. The percent iron absorption from the same diets observed in the adult males. Ionizable iron at pH 7.5 was shown to increase in presence of ascorbic acid and meat extract while it decreased in presence of phytate and tannins, similar to the effects of these factors on iron absorption in human subjects. Based on these observations it is proposed that ionizable iron at pH 7.5 determined as described in this study can be used as a reliable measure of bioavailability of nonheme iron in foods.

Serotonin content of foods: effect on urinary excretion of 5-hydroxyindoleacetic acid.

Using a highly specific radioenzymatic assay we determined the serotonin concentration in 80 types of foods. The following fruits had a high serotonin concentration (mean +/- SEM) expressed in micrograms/g weight: plantain 30.3 +/- 7.5; pineapple 17.0 +/- 5.1; banana 15.0 +/- 2.4; Kiwi fruit 5.8 +/- 0.9; plums 4.7 +/- 0.8; and tomatoes 3.2 +/- 0.6. Only nuts in the walnut or hickory family had a high serotonin concentration expressed in micrograms/g weight; butternuts 398 +/- 90; black walnuts 304 +/- 46; English walnuts 87 +/- 20; shagbark hickory nuts 143 +/- 23; mockernut hickory nuts 67 +/- 13; pecans 29 +/- 4; and sweet pignuts 25 +/- 8. Ingestion of these fruits and nuts resulted in an increase in urinary 5-hydroxyindoleacetic acid excretion with no change in platelet serotonin concentration. The above foods should not be eaten while a urine is being collected for 5-hydroxyindoleacetic acid analysis.

Myo-inositol content of common foods: development of a high-myo-inositol diet.

Since virtually no information is available concerning the myo-inositol content of dietary constituents, we have measured the amount of this material present in 487 foods by gas-liquid chromatography. We observed that the greatest amounts of myo-inositol were present in fruits, beans, grains, and nuts. Fresh vegetables and fruits were found to contain more myo-inositol than did frozen, canned, or salt-free products. The data provided in this report were used to develop diets that contained varying, but known amounts of myo-inositol. The myo-inositol intake that could be provided by such diets ranged from 225 to 1500 mg/day per 1800 kcal and within this range the agreement between the calculated and measured amounts of this material was excellent (r = 0.98). Since abnormalities in the metabolism of myo-inositol have been speculated to play a role in the pathogenesis of the polyneuropathies associated with diabetes mellitus and chronic renal failure, it is possible that the natural history of these neuropahties can be altered by modifying the amount of dietary myo-inositol that is ingested by patients with these diseases.