Parameters for obtaining concentrates from rapeseed and sunflower meal

The use of various water-alcohol mixtures to remove antinutritional and technologically inconvenient factors from rapeseed and sunflower defatted meal is described. Meals obtained from dehulled seeds were repeatedly extracted with methanol, ethanol, propanol, isopropanol or isobutanol and water (50 or 60% V/V), at pH 6 at room temperature. Use of methanol, ethanol and isopropanol led to protein concentrates with very high nitrogen yields. These products exhibited low residual contents of soluble carbohydrates, and polyphenols and, for rapeseed, glucosinolates. Moreover, they were bland flavoured and white coloured.     

Growth,hormonal status and protein turnover in rats fed on a diet containing peas (Pisum sativum L.) as the source of protein

The inclusion of peas (Pisum sativum L.) as the source of protein in the diet of growing rats brings about a reduction in growth rate as well as the impairment in the liver, muscle and spleen weights as compared with casein fed controls. Also, a fall in plasma glucose, triglycerides and protein was observed in the legume fed animals, while no changes in cholesterol levels were found. Furthermore, the rats fed on the diet containing peas showed lower levels of plasma insulin, corticosterone, IGF-I and T₄ as compared with casein controls. Liver and muscle total protein (mg) and total DNA (mg) were markedly decreased in the legume fed animals, but DNA/g, protein/DNA and RNA/protein ratios were similar in both dietary groups. Likewise, liver and muscle fractional synthesis rates were similar in the casein and legume groups, while the whole body protein synthesis is assumed to be lower in the legume fed animals due to differences in body weights. It is concluded that animals fed on a diet containing peas (Pisum sativum L.) as the only source of protein showed less adverse effects than those found with other legumes such asVicia faba L. orPhaseolus vulgaris L., in which protein quality, antinutritional factors and nutrient availability could be involved.     

Growth rate,muscle protein composition,intestinal absorption of D-galactose and L-leucine and nitrogen balance in rats fed the raw legumeChamaecytisus proliferus L.

Growth rate, nitrogen balance, skeletal muscle nitrogen fractions and in vivo intestinal absorption of D-galactose (2 mM) and L-leucine (20 mM) have been measured in male growing rats (90–100 g initial body weight) fed 12% protein diets containing either casein (control) or the raw leafy legumeChamaecytisus proliferus L. (Western Canary Islands). As compared to control casein-fed rats, those fed the raw legume diet exhibited a significant reduction (P<0.01) in the rate of growth, nitrogen balance parameters, sarcoplasmic nitrogenous fraction of skeletal musculature and in vivo intestinal absorption of galactose and leucine; however, the non-protein nitrogenous fraction has been found significantly higher (P<0.01) than that of control animals, whereas the myofibrillar nitrogenous fraction remained unchanged. The possible nature of these findings is discussed.     

Isolation and chemical composition of protein concentrates from soya-bean,rice-bran and protelan

Protein concentrates, were isolated from three by-products, defatted soya-bean, rice-bran and protelan. The conditions of maximum protein isolation differed according to the material used. Concentrates of 55.70%, 22.41% and 57.45% protein could be obtained from soya-bean, rice-bran and protelan; respectively. Hydrogen peroxide was found to increase the isolated nitrogen from protelan and rice-bran, however, in soya-bean it was without any positive effect. Essential amino acid contents were found corresponding to the amino acid of the FAO recommended amino acid composition except for methionine in all cases and lysine in protelan.     

Protein,energy and mineral utilization in rats fed rice-legume diets

Protein, energy and mineral utilization were examined in growing rats fed cooked, dry legume seeds in combination with polished rice. The legumes tested included 3 varieties of common beans, black, white and brown (Phaseolus vulgaris), lentils (Lens esculenta) and peas (Pisum sativum). The rice:legume mixtures (1:1 N ratio) were the only dietary sources of protein, Zn, Fe and Cu. The rice:black bean mixture was also tested at a 4:1 N ratio. Nutrient utilization was studied by balance trials and mineral utilization was further assessed by tissue analyses. True protein digestibility (TD) and energy digestibility of the rice:legume (1:1) mixtures were high. Biological value (BV) varied only little but was lowest in the rice:lentil mixture. The rice:brown bean mixture had the lowest amount of utilizable protein (UP). The (1:1) rice:legume mixtures did not adversely affect Fe status, as measured by liver Fe content and blood hemoglobin and hematocrit, or Cu status, as measured by liver Cu content and plasma ceruloplasmin. However, femur zinc content indicated a sub-optimal Zn status. Increasing the dietary N from rice in combination with black beans resulted in a substantial improvement of BV, NPU and Zn status of the animals but had a negative effect on UP and Fe status of the rats.     

Nutritive value of acetylated and hydrolysed rapeseed proteins in rats

Acetylated, hydrolysed and unmodified rapeseed proteins were compared to casein for nutritive value as determined by net protein ratio (NPR) and nitrogen apparent digestibility coefficient (ADC). The NPR and nitrogen ADC values of rapeseed proteins did not change significantly (P<0.05) after hydrolysis but decreased significantly (P<0.05) after 30% and 70% acetylation. Acetylation (30%) followed by protein hydrolysis decreased significantly (P<0.05) the NPR value but increased significantly (P<0.05) the nitrogen digestibility compared to the control and unmodified rapeseed proteins. Casein gave the highest value (P<0.05) for NPR and nitrogen ADC. Modification of rapeseed proteins significantly (P<0.05) decreased the nutritive value, and the effect was more pronounced with increasing level of acetylation.     

Food grade oilseed protein processing: sunflower and rapeseed

Industrial sunflower and rapeseed meals cannot be directly used as a food because of their high fiber content and because of the presence of some undesirable constituents (such as hulls, polyphenolic pigments, etc.) or precursors of toxic compounds (glucosinolates, etc.). Edible protein products (flours, concentrates, and isolates) from these two sources can be obtained by carrying out, to various degrees, and with different procedures, extraction operations of non-proteic and potentially toxic or antinutritional components. All the possible combinations of the single extraction operations (removal of fiber, lipids, polysaccharides, etc.) were studied by various authors in order to develop an optimum process both from the economic and the product quality points of view.In this report the problems related to the individual extraction operations, rather than to individual processes, are reviewed for reasons of recapitulation and to provide a common basis for comparison. Although it is impossible to reach a definitive conclusion it appears that some of the processes reviewed are able to produce very attractive raw materials for food manufacturing industries. Nevertheless, no food grade sunflower or rapeseed proteins have appeared on the market to date. However, sunflower and rapeseed protein sources will have to be taken into account in the near future as an added promising means for attacking food shortage problems.     

Effect of amino acid supplementation on protein quality of soy-based infant formulas fed to rats

The powder forms of soy-based infant formulas obtained from four manufacturers were fed to weanling rats for two weeks, as the sole source of protein in diets containing 8% protein, 20% fat, and adequate amounts of minerals and vitamins. The relative protein efficiency ratio (RPER) and the relative net protein ratio (RNPR) values (casein + methionine=100) of diets containing unsupplemented formulas were 71–81 and 78–85, respectively. Supplementation of the formula diets with lysine (0.2%), methionine (0.2%), threonine (0.1%) or tryptophan (0.05%) increased the level of the supplemental amino acid in rat serum but generally failed to improve the RPER or RNPR values. Addition of all four essential amino acids to the formula diets, however, caused a marked improvement in their protein quality (RPER or RNPR values=100). The data suggested that proteins in soy-based formulas could be marginally co-limiting in several indispensable amino acids.Presented in part at the 202nd annual meeting of the American Chemical Society, New York City, 25–30 August 1991.     

Evaluation of biological quality of protein concentrates from different vegetable protein sources

The quality of protein concentrates coming from conventional and non-conventional vegetable sources was evaluated. The protein quality was assessed by the multi-point assay RPV. Whole body weight gain, weight gain corrected for the gastrointestinal content (G.I.), total body nitrogen and total liver nitrogen were used as response parameters. The various protein sources ranked differently depending upon the response parameters utilized, but a great majority of the sources had a good nutritional value (over 0.5). The gastrointestinal content is one of the principal factors responsible for the differences found. However, even after correcting for the G.I. content, some differences remained. For the various protein sources, the nitrogen deposition in the liver correlates satisfactorily with the whole body nitrogen deposition, if in the calculation of the regression line, the data for faba bean and tobacco were omitted (respectivelyr=0.610 andr=0.925). On the whole, these data indicate that most plant protein concentrates in suitable combinations and properly processed, may become a promising alternative to animal proteins.     

The composition and nutritive value of protein concentrates from potato haulms

Summary Protein concentrates were prepared from potato haulms by standard, methods. There were distinct differences between potato varieties in chemical composition and nutritive value of their concentrates. The amino acid composition of proteins was similar and deficiency of sulphur amino acids was observed. The biological value of proteins of the best preparations was nearly as good as that of soya bean meal protein but their digestibility was lower.

---

Zusammenfassung Aus Kartoffelkraut von 8 Kartoffelsorten wurden Eiweisskonzentrate hergestellt. Eine Parzelle der Sorte Uran wurde mit Stickstoff gedüngt (180 kg N/ha). Frisch geschnittenes Kartoffelkraut wurde gemahlen, der Saft ausgepresst, daraufhin das Safteiweiss thermisch koaguliert und nach Filtration getrocknet. Die H?lfte des feuchten Eiweisskoagulats der Sorte Merkury wurde vor dem Trocknen mit Wasser gespült, dat mit HCl auf pH 4,0 anges?uert wurde. Bei Anwendung dieser Methode konnte bis zu 155 kg Eiweiss pro Hektar erzielt werden. Die Konzentrate enthielten ungef?hr 45% Eiweiss, 12 % Fett und 18 % Asche in der Trockenmasse (Tabelle 1). Der Karotingehalt schwankte zwischen 85 mg und 335 mg und der des Solanins zwischen Spuren und 1015 mg pro Kilogramm des Konzentrates. Die Zusammensetzung der Aminos?uren aller Eiweisskonzentrate war ?hnlich (Tabelle 2). Sie enthielten verh?ltnism?ssig viel Lysin, aber wenig Schwefelaminos?uren. Dennoch enthielten sie mehr Methionin als das Sojabohnenprotein, das als Standard benutzt wurde. Es wurden deutliche Unterschiede im N?hrwert der mit Hilfe der Bilanzmethode von Thomas-Mitchell an Ratten bestimmten Proteine festgestellt (Tabelle 3). Der biologische Eiweisswert betrug 39 bis 62 und war für die besten Konzentrate fast so hoch wie für das Sojaprotein, die Verdaulichkeit war jedoch geringer. Wahrscheinlich war dies auf den hohen Aschegehalt in den Konzentraten zurückzuführen. Die Stickstoffdüngung hatte eine Steigerung der vom Hektar erzielten Eiweissmenge zur Folge. Sie hatte keinen Einfluss auf die Zusammensetzung des Eiweisses, aber erh?hte seine Verdaulichkeit. Dies konnte das Resultat eines h?heren Gehaltes an leicht l?slichen, Protein nicht enthaltenden Substanzen im Konzentrat der gedüngten Pflanzen sein. Das Spülen mit Wasser erh?hte den Gehalt aller Aminos?uren im Konzentrat und den biologischen Wert des Eiweisses, vielleicht als Folge des Auswaschens von Stickstoff aus dem Konzentrat, der nicht an Eiweiss gebunden war. Die erzielten Ergebnisse zeigen, da? das Kartoffelkraut einen guten Rohstoff für die Produktion von Eiweisskonzentraten bilden kann. Diese Konzentrate enthalten eine zufriedenstellende Menge an Eiweiss, Karotin und einzelnen Aminos?uren mit Ausnahme der Schwefelaminos?uren. Der biologische Wert des Eiweisses der besten Konzentrate ist nur et was schlechter als der des Sojabohnenproteins.

---

Résumé On a réussi à produire des concentrés protéiques à partir du feuillage de 8 variétés, une partie de la culture d’Uran ayant re?u une fumure azotée (180kg N/ha). Les fanes fra?chement récoltées ont été broyées et pressées. Les protéines contenues dans le jus ont été coagulées par la chaleur et séchées après filtration. La moitié du coagulum protéique humide de la variété Merkury a été rincé avec une eau acidifiée par HCl (pH 4,0) avant séchage. Grace à cette méthode 155 kg de protéines ont été obtenus par hectare. Par rapport à la matière sèche, les concentrés contenaient environ 45% de protéines, 12% de lipides extractibles par l’éther et 18% de cendres. La teneur en carotène variait de 85 à 335 mg/kg et celle de la solanine de quelques traces à 1015 mg/kg de concentré (tableau 1). Les protéines avaient une composition en amino acides semblable (tableau 2). Elles étaient relativement riches en lysine mais pauvre en amino acides soufrés. Cependant elles contenaient plus de méthionine que la protéine du soja utilisée comme témoin. La valeur nutritive des protéines a été déterminée sur des rats par la méthode de bilan de Thomas-Mitchell. Des différences ont été constatées (tableau 3). La valeur biologique variait de 39 à 62 et pour les meilleurs concentrés était presque aussi haute que pour la protéine du soja (VB=66). La digestibilité des protéines de concentrés était relativement basse. Cela était probablement d? à leur teneur élevé en cendres. La fumure azotée a augmenté la quantité de protéines produites à l’ha. Elle n’a pas eu d’effet sur la composition en acides aminés mais a augmenté la digestibilité des protéines. Le lavage augmenterla teneur en amino acides et la valeur biologique de la protéine, peut être parce que le rin?age élimine du coagulum l’azote non protéique. Les résultats obtenus indiquent que les fanes de pommes de terre peuvent être une bonne matière première pour la production de concentrés protéiques. Ces concentrés contiennent des quantités satisfaisantes de protéines, carotène et acides a minés à l’exception des acides aminés soufrés. La valeur biologique de la protéine des meilleurs concentrés est seulement légèrement inférieure à celle de la protéine soja.

     

Nutritive value of protein fractions extracted from soybean,rapeseed and wheat flours in the rat

The nutritive value of various protein fractions was studied. Fractions 2S and 12S from rapeseed, 2S, 7S and 11S from soybean were obtained by dissolution in ammonium sulfate solutions. Albumin-globulin, gluten, glutenin and gliadin fractions from wheat were obtained by dissolution in salted water (albumin-globulin), acetic acid (glutenin) and alcohol (gliadin). Liveweight gains, protein efficiency ratio (PER) and apparent digestibility coefficient (ADC) were used as measures of the nutritive value. The protein fractions had a lower nutritive value than the unfractionated proteins except for the albumin-globulin fraction of wheat which had a nutritive value higher than that of the unfractionated wheat protein. PER obtained with the rapeseed 2S and 12S fractions were 2.49 and 2.21, respectively, as compared to 2.64 for unfractionated rapeseed. With soybean fractions, PER were 0.92 for 2S, — 0.007 for 7S and 1.47 for 11S, as compared to 2.19 for the original protein. The wheat albumin-globulin fraction gave a PER of 2.78, as compared to 1.45 for the unfractionated wheat protein. Gluten, glutenin and gliadin fractions had a lower PER than that of unfractionated wheat protein. ADC of all fractions were higher than those of the original proteins. The difference in liveweight gains and PER observed between protein fractions can be partially explained on the basis of the essential amino acid content.     

Nutritive value of soybean,rapeseed and wheat proteins,and various blends of these vegetable proteins and their fractions in rats

Blends of vegetable proteins were prepared to improve the nutritive value of these proteins. Soybean flour, rapeseed protein concentrate, whole wheat flour, soybean 2S+11S extract, wheat albumin-globulin + glutenin (AG+G) functions, and some of their blends were compared to casein for protein efficiency ratio (PER) and apparent digestibility coefficient (ADC). Autoclaving (1 h) soybean proteins was also studied. Casein and rapeseed protein concentrate gave the highest weight gains and PER. Other protein sources gave lower values for both PER (P<0.05) and weight gain. The digestibility of all vegetable proteins was lower (P<0.05) than that of casein. PER of soybean 2S+11S extract was significantly lower (P<0.05) than that of soybean flour. Autoclaving significantly (P<0.05) improved the PER of both soybean flour and 2S+11S extract. The ADC of autoclaved 2S+11S extract was similar (P<0.05) to that of autoclaved soybean flour and significantly higher (P<0.05) than that of 2S+11S extract. Soybean flour had the lowest (P<0.05) ADC. Heat treatment destroyed antinutritional factors, probably trypsin inhibitors and/or haemagglutinins of soybean. This was accompanied by improvement in the nutritional value of protein. The four blends were chosen on the basis of amino acid composition, chromatography of proteolyzates and nutritive value of each fraction. The PER of wheat albumin-globulin and glutenin (AG+G) blend was similar (P<0.05) to that of wheat flour and lower (P<0.05) than that of all other blends used. Wheat AG+G+rapeseed protein concentrate and wheat AG+G+soybean flour blends gave the highest (P<0.05) PER. Wheat AG+G improved the PER of 2S+11S extract and of soybean flour but decreased the PER of rapeseed protein concentrate. Wheat AG+G and wheat AG+G+rapeseed protein concentrate blends gave an ADC significantly higher (P<0.05) than that of wheat AG+G blend containing soybean flour of 2S+11S extract. However, blending wheat AG+G with either 2S+11S extract of soybean flour improved ADC.     

Nutritive value of rapeseed proteins and fractions supplemented with methionine and tyrosine in rats

Rapeseed protein concentrate (RPC), rapeseed 2S and 12S fractions, a blend of 2S and 12S fractions, retentate (RET) and precipitate (PRE) proteins obtained during rapeseed fractionation were compared to casein for nutritive value as determined by net protein ratio (NPR) and apparent digestibility coefficient (ADC). Tyrosine concentration in rapeseed 2S fraction was the lowest compared to all other protein sources and methionine was the first limiting amino acid. The NPR and ADC of RPC, RET, and fraction 12S did not change significantly (P<0.05) following supplementation with methionine, but for PRE the NPR was decreased significantly (P<0.05). However, the rapeseed 2S fraction supplemented with tyrosine had a NPR significantly (P<0.05) higher than that of RPC and not significant (P<0.05) different from that of casein. The NPR of the 2S and 12S blend was significantly (P<0.05) higher than that of RET and PRE; however, no significant (P<0.05) difference was observed in NPR and ADC of the blend when supplemented with tyrosine. The nutritive value and digestibility of rapeseed 2S fraction were improved by supplementation with tyrosine and by blending with the 12S fraction.     

Nutritional responses of rats fed mixtures of plant and animal proteins

Twenty four male rats (45–65 g) were used to determine the nutritional responses to mixed plant and animal protein diets. Rats were fed various mixtures of sorghum (S), pigeon pea (PP), bread fruit (TAP) plus crayfish (CR) to provide 1.6 g N/100g of daily diet for a 35-day study period. Three different diets were fed to 6 rats assigned to each diet on the basis of body weight. Combination of PP:CR caused increases in N intake and retention, weight gain, and PER higher than for those of the other test groups. Its values for N intake and retention, and BV were higher than for those of the control except for food intake, NPU, weight gain and PER. Substitution of CR (20%) with TAP and PP as supplements to S decreased food intake, weight gain and PER and increased N intake, digested and retained N, NPU and BV. The results appear to indicate that (a) PP:CR blend was better than other blends as judged by the parameters, and (b) TAP and PP were superior to CR alone as supplement to S.     

Iron status and growth of rats fed different dietary iron sources

The present study was carried out to investigate the availability of iron from spirulina, whole wheat, whole egg and standard ferrous sulphate in terms of haemoglobin formation, serum and tissue iron levels. Male albino Wistar rats were first depleted of iron by giving low-iron diet (9 ppm) and bleeding 1–2 ml blood at weekly intervals for a period of 21 days. The anaemic rats were repleted with iron sources at a level of 35 ppm for 21 days. Rats receiving whole egg gained significantly (p<0.01) higher weight than the rest of the three groups. The increase in haemoglobin was significantly higher with ferrous sulphate than with whole wheat (p<0.05), spirulina and whole egg (p<0.01). Feeding of ferrous sulphate, whole egg and spirulina produced significantly higher tissue iron levels than feeding of whole wheat. Thus, availability of iron from spirulina and whole egg were found to be comparable to that of the standard.     

Preparation and antinutritional characteristics of dry bean (Phaseolus vulgaris L.) protein concentrates

Protein concentrates and starches were prepared by a wet extraction process from five dry bean (Phaseolus vulgaris L.) cultivars. The protein contents ranged from 69.7–76.4%. Concentrates prepared from dehulled beans under similar conditions had higher protein contents (80.6–87.9%). Each additional washing of the concentrates with distilled water increased their protein content. However, the protein recovery progressively decreased. The yield of starch ranged from 48.0–51.1% of the starting material. The solubility of bean proteins was minimal at pH 4.0, and under alkaline conditions, it was influenced by the tannin contents of the concentrates. Protein concentrates had lower trypsin, chymotrypsin, and amylase inhibitory activities as well as lower phytic acid and tannin contents compared to whole bean flours.     

Preparation and properties of flours and protein concentrates from raw, fermented and germinated fluted pumpkin (Telfairia occidentalis Hook) seeds

In vitro protein digestibility, chemical composition and selected functional properties of flours and protein concentrates prepared from raw, fermented and germinated fluted pumpkin (Telfairia occidentalis Hook) seeds were studied. Protein concentrates prepared by an alkaline extraction process had increased crude protein contents (61.5–70.8%) compared to flour samples (46.4–52.7%). The yields of protein concentrates ranged from 24.5% to 29.4% while values for protein recoveries varied between 64.8% and 65.2%. Protein concentrates also had increased foam volume and decreased foam stability (100% decrease over a 2 h period), compared to flour samples. Fermentation and germination were observed to significantly (p<0.05) lower polyphenol and phytic acid contents, but increased protein digestibility of fluted pumpkin seed flours and concentrates. Both raw flour and concentrate were significantly (p<0.05) higher in water absorption capacity than germinated or fermented flours and concentrates. Protein concentrates had comparatively better fat absorption properties than the flour samples. Hence protein concentrates may prove to have useful applications in ground meat formulations.     

The relationship between amino acid composition and nutritive value for plant and animal protein concentrates

Evaluation by a chick growth test of mixtures of protein concentrates with cereals did not always confirm quality predictions based upon amino acid composition, but predictions based upon the content of the first limiting amino acid—the Chemical Score—were more generally useful when the concentrates were fed to chicks or rats as a sole protein source. Predictions involving a number of essential amino acids were no more useful than Chemical Score. A knowledge of amino acid deficiencies alone is not sufficient for the prediction of performance under practical conditions.     

Hypocholesterolemic effect of blue-green alga,ishikurage (Nostoc commune) in rats fed atherogenic diet

Nutritional quality of the castor meal protein isolate detoxified using boiling and lime cum heat treatments was evaluated in experiments with rats. Chemical scores of both the treated isolates were similar, threonine being the first limiting amino acid. The calculated nutritional indices (essential amino acid index) and PER were higher for the boiled isolate (1.3) than that was for lime-cum-heat treated (0.86). Though, necropsy examination of organs did not reveal any abnormalities, histopathological changes were observed in the organs — liver, kidney, intestine that could be attributed to the deficiency of essential amino acids in the detoxified castor protein isolates.     

Serum and liver lipids in rats fed diets containing corn oil,cottonseed oil,or a mixture of corn and cottonseed oils

Previous research has demonstrated that the total replacement of corn oil (CO) with cottonseed oil (CSO) in the diets of growing male rats results in a lowering of the total cholesterol (TC) concentration in the serum. It is not, however, known if a partial replacement of dietary CO with CSO would be sufficient to lower TC levels, nor is the effect of replacing dietary CO with CSO on other indices of cholesterol status (e.g., the levels of high-density lipoprotein cholesterol [HDL-C], non-HDL-C [i.e., very-low and low-density lipoprotein-cholesterol], or the TC/HDL-C ratio). Growing male rats were fed diets having CO, CSO, or a CO/CSO (1:1, w/w) oil mixture (OM) as the source of dietary oil (diet groups CO, CSO, and OM, respectively). There were no differences between CO and OM groups for TC, HDL-C, non-HDL-C, or the TC/HDL-C ratio. However, TC was significantly lower (p 0.05)for the CSO group than the CO group (2.85 vs. 3.23 mmol/l, or 110 vs. 125 mg/dl, respectively) as was HDL-C (1.83 vs. 2.13 mmol/l, or 71 vs. 82 mg/dl, respectively), but there were no differences between these groups for non-HDL-C or the TC/HDL-C ratio. Diet had no effect on serum triglycerides (TGs) or on total liver content of either cholesterol or TGs. Thus, partial (50%) replacement of CO with CSO was without effect on any of these serum indices of cholesterol status, but total replacement resulted in lower TC and HDL-C, without affecting non-HDL-C or the TC/HDL-C ratio. Because, relative to CO, CSO has higher levels of saturated fatty acids but lower levels of monounsaturated and polyunsaturated fatty acids, the cholesterol-lowering effect of CSO appears to be unrelated to its fatty acid composition, and may be caused by a component of its nonsaponifiable fraction.