Fd Chem Toxw. Vol 30, No. 7, pp 601-610, 1992 Pnnted m Great Britain All rights reserved
0278-6915/92$5.00+ 0.00 Copyright © 1992PergamonPress Ltd
COMPARISON OF CEREAL-BASED DIET WITH PURIFIED DIET BY SHORT-TERM FEEDING STUDIES IN RATS, MICE AND HAMSTERS, WITH EMPHASIS ON TOXICITY CHARACTERISTICS A. A. J. J L. RUTTENand A. P. DE GROOT Toxicology and NutnUon Institute TNO, Department of Biological Toxicology, PO Box 360, 3700 AJ Zelst, The Netherlands (Accepted 16 March 1992)
Abstract--Animal diets used m toxicity studies are prepared either from natural ingredients (cereal-based diet) or from more refined products (purified diet). The type of diet may mfluence both the outcome of the study and the values obtained with the various parameters in test and control animals. To detect the parameters sensmve to changes m diet composmon, short-term (4-wk) studies were conducted in rats, mice and hamsters fed either a cereal-based dtet or the AIN-76A purified diet supplemented with wtamms and minerals at the highest recommended levels for each of the species used. Although the purified diet was more palatable to rats and showed a higher protein quahty, growth rate and food intake were generally slightly higher with the cereal-based diet m each of the species exammed The haematological values of the two diet groups were generally comparable On the cereal-based diet the production of faeces was considerably higher than on the purified diet and was accompamed by a higher weight of the caecum. These findings were attributed to the relatively high level and mixed composition of the fibre fraction in the cereal-based diet. Blood levels of cholesterol and phosphohplds were clearly lower on the cereal-based diet than on the purified diet. Because the differenceswere probably due to the level and composition of the fibre fractton, they support the suggestion to replace the 5% celluloseof the AIN-76A diet by a higher level of a more composite but well defined source of dietary fibre.
INTRODUCTION
Diets for laboratory animals are traditionally prepared from natural ingredients such as cereal and protein concentrates from animal and plant origin. These diets, generally called cereal-based diets, are readily available, convenient, cheap and palatable to laboratory ammals. However, in addition to nutrients such diets contain large numbers of ill-defined or unknown chemicals, both naturally occurnng and man-made. The levels of nutrients and non-nutrients in different batches of natural ingredients and prepared diets may vary considerably and influence the experimental results. Numerous different ways in which variations in diet composition may influence the response of the body to drugs, nutrients and contaminants are mentioned in many reviews (Almeida et al., 1978; Basu, 1988; Boyd, 1969; Concon, 1988; Newberne, 1975; Newberne and McConnel, 1978; Wise, 1982). The incidence and growth of tumours may be affected by dietary levels of macronutrients (proteins, lipids and carbohydrates) and by several micronutrients as well as by dietary fibre and xenobioties (Almeida et aL, 1978; Carroll, 1980; Clayson, 1975; Concon, 1988; Abbreviations: ALP = alkaline phosphatase; ASAT = aspartate aminotransferase; Hb = haemoglobin; PCV packed cell volume; RBC = red blood cell count.
Newberne and Connor, 1986; Palmer, 1985; Welsburger, 1986; Willett and MacMahon, 1984). The activities of blotransformation enzymes are affected by many different diet components, nutrients and non-nutrients (Bidlack, 1982, Meydani, 1987; Parke, 1978; Rutten and Falke, 1987; Wade et aL, 1978). The toxicity of any test compound may depend on a variety of naturally occurring or man-made substances in the diet such as glycoalkaloids, mycotoxins, mtrates and heavy metals, pesticide residues, polycyclic aromatic hydrocarbons and nitrosamines (Newberne et al., 1978; Rader, 1989). To reduce varmbility in diet composition and thereby discrepancies in study results, d~ets have been developed with purified ingredients such as casein, sucrose and starch; they contain fewer non-nutrients than do cereal-based diets and show little batch-tobatch variation (OUer et aL, 1989). Although such purified d~ets are artificial, less readily available and more expensive than cereal-based diets, they are increasingly preferred in toxicity studies. The objective of the present study was to compare the effects of a purified and a cereal-based diet on small laboratory animals (rats, mice and hamsters) in a short-term (4-wk) feeding study, and to detect the biological parameters that are most sensitive to dietary changes. Because both types of diet are widely used in toxicity studies, the comparison was based mainly on the characteristics usually examined in
601
602
A A. J. J L. RUTTEN and A P. DE GgOOT
these studies to detect possible departures from normality in the treated animals. In addition, s t a n d a r d tests were p e r f o r m e d in rats to c o m p a r e the palatability (by preference testing) a n d the protein q u a h t y (by d e t e r m i n a t i o n o f net protein utflmation a n d digestibility) o f the two diets. MATERIALS AND METHODS
Animals, housmg and experzmental design. The two diets were c o m p a r e d by feeding two groups of 10 male a n d 10 female rats, m~ce a n d hamsters for 4 wk. Newly weaned W i s t a r albino rats ( C r I : W I ( W U ) B R , age 19-23 days, body weight 5 0 - 7 0 g ) a n d newly weaned albino mice (Crl: C D - I(ICR) BR, age 19-23 days, body weight 1 2 - 1 4 g ) were o b t a i n e d from outbred colonies maintained under specificpathogen-free conditions at Charles River (Sulzfeld, Germany) Syrian golden hamsters (Lak : L V G ( S Y R ) , a b o u t 3 wk of age, b o d y weight 51-57 g) were o b t a i n e d from a n o u t b r e d colony at Charles River (Montreal, C a n a d a ) . O n the day o f their arrival the a m m a l s were weighed, r a n d o m l y distributed to two groups o f l0 per sex, a n d assigned to the two diets. The rats were housed in suspended, stainless steel cages with wire-screen b o t t o m a n d front. The mice a n d h a m s t e r s were kept in m a c r o l o n boxes with grid cover containing a feed hopper. The mice were housed in&vtdually a n d the rats a n d h a m s t e r s m groups o f 5 separated by sex. Housing c o n d m o n s were conventional. The animal r o o m s were m a i n tained at a t e m p e r a t u r e between 20 a n d 25°C, a relative humidity between 50 a n d 70% a n d a ventilation rate o f at least 10 air changes/hr. Artificial h g h t was provided from fluorescent tubes for 1 2 h r / d a y continuously, from 7.30 a m until 7.30 pm. F o r rats Table 1 Composmon of the cereal-based stock diet given to rats, mice and hamsters Ingredient (%) Soyabean meal (defatted) II 0 Fish meal (Damsh) 70 Meat and bone scraps 40 Wheat (whole ground) 36 0 Maize (whole ground) 29 7 Brewer's yeast 30 Alfalfa meal 30 Whey powder 2.0 Defatted bone meal 0.4 Soyabcan od 30 Salt vath trace elements* 0.5 Vitamin B mlxturci" 01 Mixture of vitamins A, D, E and K:~ 03 *Salt with trace elements (g/kg mix) MnSO4 H20 (47 6), ZnCI2 (5 00), KIO3 (0 52), FeSO4 7H20 (25 0), CoC12-6H20 (0.40), CuSO4 5H20 (8.00) and NaCI to make up 1 0 kg tVltamm B mixture (g/kg mix) thtamm-HCl (2 50), nboflavln (3 00), pyndoxine-HCl (10.0), niacin (12 5), Ca-pantothenate (7 5), biotin (0 15), fohc acid (0.5), vitamin B,2 (0.035) and finely powdered sucrose to make up 1 0 kg. ,V~tamm A (2,112,517IU/kg mix), vitamin D (704,173 IU/kg mix), vitamin E-50% (30 1 g/kg mix), menadlon-Na-blsulphRe (1 00g/kg mix) and wheat starch to make up 1.0kg. The diet (powder) contained (by analysis) moisture 12.7%, crude protein 19 4%, crude fat 6.0%, crude fibre 2.9% (EEC, 1973); dietary fibre 11 5% (Hellendoorn et al, 1975), total ash 5.5%; calcium 0 60% and phosphorus 0.63%
a n d mice the experiment was started o n day 6 a n d day 9, respectively, after their arrival, a n d for h a m s t e r s o n the day of their arrival. After 28 days the studies were terminated by autopsy of the animals. Diets. T h e cereal-based diet was the o p e n - f o r m u l a stock diet for rats, mice a n d hamsters, which has been used in o u r Institute, w i t h o u t significant changes in composition, as the basal diet m m a n y toxicity studies in rodents for more t h a n 20 yr. The ingredient composition is s h o w n m Table I. The purified diet was based o n the A I N - 7 6 A diet for rodents, which has been developed a n d tested by a committee o f the A m e r i c a n Institute o f N u t r i t i o n (1977 a n d 1980) a n d the Institute of L a b o r a t o r y A n i m a l Resources (1979). However, the levels of several vitamins a n d minerals were increased, a n d salts of cobalt a n d fluorine were a d d e d to meet the s t a n d a r d s set for rats a n d mice as well as for h a m s t e r s ( N a t i o n a l Research Council, 1978). In addition, wheat starch instead o f sucrose was used as the source of c a r b o h y d r a t e , a n d the level of dibasic calcium p h o s p h a t e was lowered as proposed in a n A I N - 7 6 w o r k s h o p (Reeves, 1989). The ingredient c o m p o s i t i o n o f this modified A I N - 7 6 A diet ~s s h o w n in Table 2. The rats received the diets as powders, while for mice a n d h a m s t e r s the diets
Table 2 Composition of the modifiedpurified diet given to rats, mice and hamsters lngredmnt (%) Casein* 20 0 DL-Methlonme 03 Wheat starcht 63 5 Corn ofl~ 50 Cellulose§ 50 Mineral mixture (based on AIN-76A)[I 35 Vitamin mixture (based on AIN-76A)¶ 10 CaHPO4 15 Chohne bltartrate 02 *Acid-precipitated, containing: protein 89 10% (N x 6 38), mmsture 89%, ash 467%, pH of a 10% aqueous suspension 4 5 t10% of native wheat starch was replaced by prcgelatmlzed wheat starch to improve the quality of the pellets :~No antloxldants were added. §Dmacel, highly purified and bleached fibrous filter powder, consisting of 87-90% pure ,,-cellulose, average length of fibres about 44 ,am, water 4%, ash 0 12~) 15%, and hgnm 0.04% [IMmeral mixture (g/kg mixture) N a C I (110), K~C6HsO7 H20 (394), K2SO4 (51 8), MgO (28.4), MnCO3 xH:O (3 5), FeC6HsO7 5H20 (24), 5ZnO 2CO3 4H20 (1 6), CuCO3(OH)2 • H20 (0 3), KIO3 (008), Na2S¢O3 5H20 (0.01), CrK(SO0 12H20 (0 55), NaF (0.063), COCI2-6H20 (0.127), and finely powdered sucrose to make up 1.0 kg. ¶Wltamm mixture (g or IU/kg mixture) thiamin-Ha (2 0 g), riboflavin (1 5 g), pyndoxme-HCl (0 7 g), nicotinic acid (9 0 g), Ca-D( + )-pantoth¢nate (4 0 g), fohc acid (0 2 g), D( + )biotin (0.06 g), vitamin Bn (0005g), momol (10g), retmyl palmltate/aeetat¢ (400,000IU), cholecaloferol (248,400IU), DL-atocopheryl acetate (5000 IU), menadton¢ Na bisulphlte (0 4 g), and finely powdered sucrose to make up 1 0 kg The diet (pellets) contained (by analysis), moisture 16 1%; crude protein 17 9%; crude fat 4.9%, crude fibre 3 6% (EEC, 1973), dietary fibre 4.3% (Hellendoorn et al, 1975); calcium 042%, phosphorus 0 50%, and total ash 3 1%
Comparison of cereal-based with purified diet were provided as pellets obtained by using a laboratory pelletizer (Monoroll Labor, Simon-Heesen, Boxtel, The Netherlands). Adequate pellets were obtained by adding 5% (w/w) of a mixture (1 : 1) of molasses and tap water to the cereal-based diet and by adding 10% (w/w) demineralized water to the purified diet, in which 10% of the original wheat starch had been replaced by pregelatinized wheat starch to ~mprove the quality of the pellets. Only one batch of the diets was prepared and stored in closed containers at c. - 20°C. Observations and measurements. The general condition and behaviour of all animals were checked daily. Individual body weights were recorded on arrival, at the start of the study and at weekly intervals thereafter. Food retake of the group-caged rats and hamsters and of the single-caged mice were recorded weekly. The efficiency of food utilization was calculated and expressed as g weight gain per g food consumed. Daily water consumption was measured per cage for rats and hamsters and individually for mice. Rat faeces were collected per cage during a 24-hr period and then dried and weighed. Haematology. On day 24, blood was collected from rats and hamsters, which had been depraved of water for 24 hr and of food for 16 hr. On day 26, blood was collected from mice kept w~thout food and water for 16 hr. Samples were drawn from the up of the tail of rats, and from the orbital plexus of mice and hamsters. The blood was examined for haemoglobin concentration (Hb), packed cell volume (PCV), red blood cell count (RBC), red blood cell distribution width, total white blood cell count, thrombocyte count, mean thromboeyte volume and thrombocyte distribution width, using a Sysmex K-1000 Haematology Analyzer (Tao Medical Electronics, Kobe, Japan). Differential white blood cell counts were made microscopically in stained blood smears Prothrombm time was determined in EDTA blood by a modified Normotest (Nyegaard and Co. A/S, Oslo, Norway). Clinical chemistry. The blood samples for haematology were also examined for fasting glucose concentration (Glucoquant kit No. 245-178; Boehringer Mannheim, Mannheim, Germany). At autopsy, blood of rats and hamsters was collected from the abdominal aorta in heparinized plastic tubes Plasma was prepared by centnfugation (at 1250g for 15mm) and analysed for alkaline phosphatase (ALP), aspartate aminotransferase (ASAT), alanine aminotransferase, y-glutamyl transferase, total protein, albumin, total bilirubin, urea, creatinine, cholesterol, tr~glycerides, phospholipids, calcium and inorganic phosphate, using a Cobas-Bio Centrifugal Analyzer (Hoffman-La Roche, Basle, Switzerland); sodium and potassmm using an Electrolyte-2-Analyzer (Beckman Instruments, Brea, CA, USA); and chloride by coulometric titration using a Chloro-counter (Marius, Utrecht, The Netherlands). Urine and faeces analysis. On day 23, the faeces of rats were collected per cage during a 24-hr period, dehydrated at 105-110°C for 24 hr and weighed. The
603
rats were then placed individually in metabohsm cages and were deprived of water (for 24 hr) and food (for 16 hr). The pH of the urine excreted during the first 2 hr of the collection period was measured with a Philips PW9410 pH meter. The urine was collected in cahbrated tubes over the last 16 hr of the deprivation period. Urinary density was measured by a refractometer (Bellingham and Stanley, London, UK). After centnfugatmn, microscopic examination of the unnary sediment was conducted, and the supernatant was examined for the presence of protein, glucose, occult blood, ketones, bilirubin and urobilinogen, using Combur test stnps (Boehringer Mannheim, Mannhelm, Germany). Pathology. At the end of the 4-wk period, the animals were anaesthetized by ether and killed by exsangumation from the abdominal aorta. The pancreas, testes or ovaries, adrenals, kidneys, liver and caecum (full and empty) were weighed, and the organ:body weight raUos were calculated. Tissue samples of the liver and kidneys were fixed in a 4% aqueous phosphate buffered (pH 7.0) formaldehyde solution. After processing and embedding in paraffin, sectmns were cut (5 #m), stained with haematoxylin and eosm and examined microscopically. Palatability assay. Four male newly weaned Wistar albino rats (CrI:WI(WU)BR, age 18-22 days, body weight 35-50 g), were housed two per cage and given free access to two weighed food containers, one containing the cereal-based diet and the other the purified d~et The posiuon of the two containers m each cage was changed dally to exclude a posluon effect. The amount of food consumed daily from each container was recorded during an 8-day feeding period. Net protein utilization and digestibility assay. Net protein utilization and digestibility were determined by the method of Miller and Bender (1955) with three groups of six male and six female newly weaned Wistar albino rats (Crl:WI(WU)BR, age 18-22 days, body weight 35-50 g). Each group was housed in three cages of two males and two females per cage. After an acclimatization period of 6 days on the stock diet and tap water, they were fed a protein-free diet or a diet containing 10% crude protein (purified diet or cereal-based diet). The protein-free d~et was derived from the purified diet by replacing casein and DLmethionine by wheat starch. The 10% crude-protein diets were obtained by diluting the purified diet and the cereal-based diet w~th calculated amounts of the protein-free diet. After a feeding period of 10 days, the amount of nitrogen consumed from the diet, the amount of nitrogen excreted m the faeces and that present in the carcasses were determined by the Kjeldahl method from each of the two 10% protein groups and the protein-free controls. From these data protein digestibility and utilization were calculated as described by Miller and Bender (1955). Statistical analysis. All data were analysed by comparing the values of the two diet groups with
604
A. A J J L. RUTTEN and A. P. DE GROOT Table 3 Mean results of the palatabdlty test and of the net protein ut]hzauon and d~gesUbdltyassay in rats Protein quality Palatabdlty Net protein Diet Food consumedt utdizaUon Dlgest]blhty Bmloglcal:~ group (g/rat/day) (%) (%) value Cereal-based 5 8 + 1 86 58 6 -I- I 00 86 1 + 0 82 68 I -I-0 60 Purified 102+088 789_+211"* 994_+006** 794_+216"* tMean dally retake of each diet when rats were offered both dints stmultaneously ~Blologmal value = net protein utlhzatton/dlgest]b]htyx 100 Values are the means_ SD for two groups of two male rats m the palatabthty test and for three groups of two male and two female rats m the protein quahty assay Values marked wtth astensks d~ffer significantly (Student's t-test) from the correspondmg control value (*P
0278-6915/92$5.00+ 0.00 Copyright © 1992PergamonPress Ltd
COMPARISON OF CEREAL-BASED DIET WITH PURIFIED DIET BY SHORT-TERM FEEDING STUDIES IN RATS, MICE AND HAMSTERS, WITH EMPHASIS ON TOXICITY CHARACTERISTICS A. A. J. J L. RUTTENand A. P. DE GROOT Toxicology and NutnUon Institute TNO, Department of Biological Toxicology, PO Box 360, 3700 AJ Zelst, The Netherlands (Accepted 16 March 1992)
Abstract--Animal diets used m toxicity studies are prepared either from natural ingredients (cereal-based diet) or from more refined products (purified diet). The type of diet may mfluence both the outcome of the study and the values obtained with the various parameters in test and control animals. To detect the parameters sensmve to changes m diet composmon, short-term (4-wk) studies were conducted in rats, mice and hamsters fed either a cereal-based dtet or the AIN-76A purified diet supplemented with wtamms and minerals at the highest recommended levels for each of the species used. Although the purified diet was more palatable to rats and showed a higher protein quahty, growth rate and food intake were generally slightly higher with the cereal-based diet m each of the species exammed The haematological values of the two diet groups were generally comparable On the cereal-based diet the production of faeces was considerably higher than on the purified diet and was accompamed by a higher weight of the caecum. These findings were attributed to the relatively high level and mixed composition of the fibre fraction in the cereal-based diet. Blood levels of cholesterol and phosphohplds were clearly lower on the cereal-based diet than on the purified diet. Because the differenceswere probably due to the level and composition of the fibre fractton, they support the suggestion to replace the 5% celluloseof the AIN-76A diet by a higher level of a more composite but well defined source of dietary fibre.
INTRODUCTION
Diets for laboratory animals are traditionally prepared from natural ingredients such as cereal and protein concentrates from animal and plant origin. These diets, generally called cereal-based diets, are readily available, convenient, cheap and palatable to laboratory ammals. However, in addition to nutrients such diets contain large numbers of ill-defined or unknown chemicals, both naturally occurnng and man-made. The levels of nutrients and non-nutrients in different batches of natural ingredients and prepared diets may vary considerably and influence the experimental results. Numerous different ways in which variations in diet composition may influence the response of the body to drugs, nutrients and contaminants are mentioned in many reviews (Almeida et al., 1978; Basu, 1988; Boyd, 1969; Concon, 1988; Newberne, 1975; Newberne and McConnel, 1978; Wise, 1982). The incidence and growth of tumours may be affected by dietary levels of macronutrients (proteins, lipids and carbohydrates) and by several micronutrients as well as by dietary fibre and xenobioties (Almeida et aL, 1978; Carroll, 1980; Clayson, 1975; Concon, 1988; Abbreviations: ALP = alkaline phosphatase; ASAT = aspartate aminotransferase; Hb = haemoglobin; PCV packed cell volume; RBC = red blood cell count.
Newberne and Connor, 1986; Palmer, 1985; Welsburger, 1986; Willett and MacMahon, 1984). The activities of blotransformation enzymes are affected by many different diet components, nutrients and non-nutrients (Bidlack, 1982, Meydani, 1987; Parke, 1978; Rutten and Falke, 1987; Wade et aL, 1978). The toxicity of any test compound may depend on a variety of naturally occurring or man-made substances in the diet such as glycoalkaloids, mycotoxins, mtrates and heavy metals, pesticide residues, polycyclic aromatic hydrocarbons and nitrosamines (Newberne et al., 1978; Rader, 1989). To reduce varmbility in diet composition and thereby discrepancies in study results, d~ets have been developed with purified ingredients such as casein, sucrose and starch; they contain fewer non-nutrients than do cereal-based diets and show little batch-tobatch variation (OUer et aL, 1989). Although such purified d~ets are artificial, less readily available and more expensive than cereal-based diets, they are increasingly preferred in toxicity studies. The objective of the present study was to compare the effects of a purified and a cereal-based diet on small laboratory animals (rats, mice and hamsters) in a short-term (4-wk) feeding study, and to detect the biological parameters that are most sensitive to dietary changes. Because both types of diet are widely used in toxicity studies, the comparison was based mainly on the characteristics usually examined in
601
602
A A. J. J L. RUTTEN and A P. DE GgOOT
these studies to detect possible departures from normality in the treated animals. In addition, s t a n d a r d tests were p e r f o r m e d in rats to c o m p a r e the palatability (by preference testing) a n d the protein q u a h t y (by d e t e r m i n a t i o n o f net protein utflmation a n d digestibility) o f the two diets. MATERIALS AND METHODS
Animals, housmg and experzmental design. The two diets were c o m p a r e d by feeding two groups of 10 male a n d 10 female rats, m~ce a n d hamsters for 4 wk. Newly weaned W i s t a r albino rats ( C r I : W I ( W U ) B R , age 19-23 days, body weight 5 0 - 7 0 g ) a n d newly weaned albino mice (Crl: C D - I(ICR) BR, age 19-23 days, body weight 1 2 - 1 4 g ) were o b t a i n e d from outbred colonies maintained under specificpathogen-free conditions at Charles River (Sulzfeld, Germany) Syrian golden hamsters (Lak : L V G ( S Y R ) , a b o u t 3 wk of age, b o d y weight 51-57 g) were o b t a i n e d from a n o u t b r e d colony at Charles River (Montreal, C a n a d a ) . O n the day o f their arrival the a m m a l s were weighed, r a n d o m l y distributed to two groups o f l0 per sex, a n d assigned to the two diets. The rats were housed in suspended, stainless steel cages with wire-screen b o t t o m a n d front. The mice a n d h a m s t e r s were kept in m a c r o l o n boxes with grid cover containing a feed hopper. The mice were housed in&vtdually a n d the rats a n d h a m s t e r s m groups o f 5 separated by sex. Housing c o n d m o n s were conventional. The animal r o o m s were m a i n tained at a t e m p e r a t u r e between 20 a n d 25°C, a relative humidity between 50 a n d 70% a n d a ventilation rate o f at least 10 air changes/hr. Artificial h g h t was provided from fluorescent tubes for 1 2 h r / d a y continuously, from 7.30 a m until 7.30 pm. F o r rats Table 1 Composmon of the cereal-based stock diet given to rats, mice and hamsters Ingredient (%) Soyabean meal (defatted) II 0 Fish meal (Damsh) 70 Meat and bone scraps 40 Wheat (whole ground) 36 0 Maize (whole ground) 29 7 Brewer's yeast 30 Alfalfa meal 30 Whey powder 2.0 Defatted bone meal 0.4 Soyabcan od 30 Salt vath trace elements* 0.5 Vitamin B mlxturci" 01 Mixture of vitamins A, D, E and K:~ 03 *Salt with trace elements (g/kg mix) MnSO4 H20 (47 6), ZnCI2 (5 00), KIO3 (0 52), FeSO4 7H20 (25 0), CoC12-6H20 (0.40), CuSO4 5H20 (8.00) and NaCI to make up 1 0 kg tVltamm B mixture (g/kg mix) thtamm-HCl (2 50), nboflavln (3 00), pyndoxine-HCl (10.0), niacin (12 5), Ca-pantothenate (7 5), biotin (0 15), fohc acid (0.5), vitamin B,2 (0.035) and finely powdered sucrose to make up 1 0 kg. ,V~tamm A (2,112,517IU/kg mix), vitamin D (704,173 IU/kg mix), vitamin E-50% (30 1 g/kg mix), menadlon-Na-blsulphRe (1 00g/kg mix) and wheat starch to make up 1.0kg. The diet (powder) contained (by analysis) moisture 12.7%, crude protein 19 4%, crude fat 6.0%, crude fibre 2.9% (EEC, 1973); dietary fibre 11 5% (Hellendoorn et al, 1975), total ash 5.5%; calcium 0 60% and phosphorus 0.63%
a n d mice the experiment was started o n day 6 a n d day 9, respectively, after their arrival, a n d for h a m s t e r s o n the day of their arrival. After 28 days the studies were terminated by autopsy of the animals. Diets. T h e cereal-based diet was the o p e n - f o r m u l a stock diet for rats, mice a n d hamsters, which has been used in o u r Institute, w i t h o u t significant changes in composition, as the basal diet m m a n y toxicity studies in rodents for more t h a n 20 yr. The ingredient composition is s h o w n m Table I. The purified diet was based o n the A I N - 7 6 A diet for rodents, which has been developed a n d tested by a committee o f the A m e r i c a n Institute o f N u t r i t i o n (1977 a n d 1980) a n d the Institute of L a b o r a t o r y A n i m a l Resources (1979). However, the levels of several vitamins a n d minerals were increased, a n d salts of cobalt a n d fluorine were a d d e d to meet the s t a n d a r d s set for rats a n d mice as well as for h a m s t e r s ( N a t i o n a l Research Council, 1978). In addition, wheat starch instead o f sucrose was used as the source of c a r b o h y d r a t e , a n d the level of dibasic calcium p h o s p h a t e was lowered as proposed in a n A I N - 7 6 w o r k s h o p (Reeves, 1989). The ingredient c o m p o s i t i o n o f this modified A I N - 7 6 A diet ~s s h o w n in Table 2. The rats received the diets as powders, while for mice a n d h a m s t e r s the diets
Table 2 Composition of the modifiedpurified diet given to rats, mice and hamsters lngredmnt (%) Casein* 20 0 DL-Methlonme 03 Wheat starcht 63 5 Corn ofl~ 50 Cellulose§ 50 Mineral mixture (based on AIN-76A)[I 35 Vitamin mixture (based on AIN-76A)¶ 10 CaHPO4 15 Chohne bltartrate 02 *Acid-precipitated, containing: protein 89 10% (N x 6 38), mmsture 89%, ash 467%, pH of a 10% aqueous suspension 4 5 t10% of native wheat starch was replaced by prcgelatmlzed wheat starch to improve the quality of the pellets :~No antloxldants were added. §Dmacel, highly purified and bleached fibrous filter powder, consisting of 87-90% pure ,,-cellulose, average length of fibres about 44 ,am, water 4%, ash 0 12~) 15%, and hgnm 0.04% [IMmeral mixture (g/kg mixture) N a C I (110), K~C6HsO7 H20 (394), K2SO4 (51 8), MgO (28.4), MnCO3 xH:O (3 5), FeC6HsO7 5H20 (24), 5ZnO 2CO3 4H20 (1 6), CuCO3(OH)2 • H20 (0 3), KIO3 (008), Na2S¢O3 5H20 (0.01), CrK(SO0 12H20 (0 55), NaF (0.063), COCI2-6H20 (0.127), and finely powdered sucrose to make up 1.0 kg. ¶Wltamm mixture (g or IU/kg mixture) thiamin-Ha (2 0 g), riboflavin (1 5 g), pyndoxme-HCl (0 7 g), nicotinic acid (9 0 g), Ca-D( + )-pantoth¢nate (4 0 g), fohc acid (0 2 g), D( + )biotin (0.06 g), vitamin Bn (0005g), momol (10g), retmyl palmltate/aeetat¢ (400,000IU), cholecaloferol (248,400IU), DL-atocopheryl acetate (5000 IU), menadton¢ Na bisulphlte (0 4 g), and finely powdered sucrose to make up 1 0 kg The diet (pellets) contained (by analysis), moisture 16 1%; crude protein 17 9%; crude fat 4.9%, crude fibre 3 6% (EEC, 1973), dietary fibre 4.3% (Hellendoorn et al, 1975); calcium 042%, phosphorus 0 50%, and total ash 3 1%
Comparison of cereal-based with purified diet were provided as pellets obtained by using a laboratory pelletizer (Monoroll Labor, Simon-Heesen, Boxtel, The Netherlands). Adequate pellets were obtained by adding 5% (w/w) of a mixture (1 : 1) of molasses and tap water to the cereal-based diet and by adding 10% (w/w) demineralized water to the purified diet, in which 10% of the original wheat starch had been replaced by pregelatinized wheat starch to ~mprove the quality of the pellets. Only one batch of the diets was prepared and stored in closed containers at c. - 20°C. Observations and measurements. The general condition and behaviour of all animals were checked daily. Individual body weights were recorded on arrival, at the start of the study and at weekly intervals thereafter. Food retake of the group-caged rats and hamsters and of the single-caged mice were recorded weekly. The efficiency of food utilization was calculated and expressed as g weight gain per g food consumed. Daily water consumption was measured per cage for rats and hamsters and individually for mice. Rat faeces were collected per cage during a 24-hr period and then dried and weighed. Haematology. On day 24, blood was collected from rats and hamsters, which had been depraved of water for 24 hr and of food for 16 hr. On day 26, blood was collected from mice kept w~thout food and water for 16 hr. Samples were drawn from the up of the tail of rats, and from the orbital plexus of mice and hamsters. The blood was examined for haemoglobin concentration (Hb), packed cell volume (PCV), red blood cell count (RBC), red blood cell distribution width, total white blood cell count, thrombocyte count, mean thromboeyte volume and thrombocyte distribution width, using a Sysmex K-1000 Haematology Analyzer (Tao Medical Electronics, Kobe, Japan). Differential white blood cell counts were made microscopically in stained blood smears Prothrombm time was determined in EDTA blood by a modified Normotest (Nyegaard and Co. A/S, Oslo, Norway). Clinical chemistry. The blood samples for haematology were also examined for fasting glucose concentration (Glucoquant kit No. 245-178; Boehringer Mannheim, Mannheim, Germany). At autopsy, blood of rats and hamsters was collected from the abdominal aorta in heparinized plastic tubes Plasma was prepared by centnfugation (at 1250g for 15mm) and analysed for alkaline phosphatase (ALP), aspartate aminotransferase (ASAT), alanine aminotransferase, y-glutamyl transferase, total protein, albumin, total bilirubin, urea, creatinine, cholesterol, tr~glycerides, phospholipids, calcium and inorganic phosphate, using a Cobas-Bio Centrifugal Analyzer (Hoffman-La Roche, Basle, Switzerland); sodium and potassmm using an Electrolyte-2-Analyzer (Beckman Instruments, Brea, CA, USA); and chloride by coulometric titration using a Chloro-counter (Marius, Utrecht, The Netherlands). Urine and faeces analysis. On day 23, the faeces of rats were collected per cage during a 24-hr period, dehydrated at 105-110°C for 24 hr and weighed. The
603
rats were then placed individually in metabohsm cages and were deprived of water (for 24 hr) and food (for 16 hr). The pH of the urine excreted during the first 2 hr of the collection period was measured with a Philips PW9410 pH meter. The urine was collected in cahbrated tubes over the last 16 hr of the deprivation period. Urinary density was measured by a refractometer (Bellingham and Stanley, London, UK). After centnfugatmn, microscopic examination of the unnary sediment was conducted, and the supernatant was examined for the presence of protein, glucose, occult blood, ketones, bilirubin and urobilinogen, using Combur test stnps (Boehringer Mannheim, Mannhelm, Germany). Pathology. At the end of the 4-wk period, the animals were anaesthetized by ether and killed by exsangumation from the abdominal aorta. The pancreas, testes or ovaries, adrenals, kidneys, liver and caecum (full and empty) were weighed, and the organ:body weight raUos were calculated. Tissue samples of the liver and kidneys were fixed in a 4% aqueous phosphate buffered (pH 7.0) formaldehyde solution. After processing and embedding in paraffin, sectmns were cut (5 #m), stained with haematoxylin and eosm and examined microscopically. Palatability assay. Four male newly weaned Wistar albino rats (CrI:WI(WU)BR, age 18-22 days, body weight 35-50 g), were housed two per cage and given free access to two weighed food containers, one containing the cereal-based diet and the other the purified d~et The posiuon of the two containers m each cage was changed dally to exclude a posluon effect. The amount of food consumed daily from each container was recorded during an 8-day feeding period. Net protein utilization and digestibility assay. Net protein utilization and digestibility were determined by the method of Miller and Bender (1955) with three groups of six male and six female newly weaned Wistar albino rats (Crl:WI(WU)BR, age 18-22 days, body weight 35-50 g). Each group was housed in three cages of two males and two females per cage. After an acclimatization period of 6 days on the stock diet and tap water, they were fed a protein-free diet or a diet containing 10% crude protein (purified diet or cereal-based diet). The protein-free d~et was derived from the purified diet by replacing casein and DLmethionine by wheat starch. The 10% crude-protein diets were obtained by diluting the purified diet and the cereal-based diet w~th calculated amounts of the protein-free diet. After a feeding period of 10 days, the amount of nitrogen consumed from the diet, the amount of nitrogen excreted m the faeces and that present in the carcasses were determined by the Kjeldahl method from each of the two 10% protein groups and the protein-free controls. From these data protein digestibility and utilization were calculated as described by Miller and Bender (1955). Statistical analysis. All data were analysed by comparing the values of the two diet groups with
604
A. A J J L. RUTTEN and A. P. DE GROOT Table 3 Mean results of the palatabdlty test and of the net protein ut]hzauon and d~gesUbdltyassay in rats Protein quality Palatabdlty Net protein Diet Food consumedt utdizaUon Dlgest]blhty Bmloglcal:~ group (g/rat/day) (%) (%) value Cereal-based 5 8 + 1 86 58 6 -I- I 00 86 1 + 0 82 68 I -I-0 60 Purified 102+088 789_+211"* 994_+006** 794_+216"* tMean dally retake of each diet when rats were offered both dints stmultaneously ~Blologmal value = net protein utlhzatton/dlgest]b]htyx 100 Values are the means_ SD for two groups of two male rats m the palatabthty test and for three groups of two male and two female rats m the protein quahty assay Values marked wtth astensks d~ffer significantly (Student's t-test) from the correspondmg control value (*P
