47
Atherosclerosis, 24 (1976) 47-62 0 Elsevier Scientific Publishing Company, Amsterdam - Printed in The Netherlands
PLASMA CHOLESTEROL LEVELS LOW CHOLESTEROL DIETS EFFECTS OF DIETARY PROTEINS, FROM DIFFERENT SOURCES *
R.M.G. HAMILTON 1 and K.K. CARROLL Department of Biochemistry, Ontario NtiA 5Cl (Canada) Received 7th May, 1975) (Accepted 12th February,
IN RABBITS
FED LOW FAT,
CARBOHYDRATES
AND FIBRE
2
University of Western Ontario, London,
1976)
Summary Rabbits become hypercholesterolemic when transferred from commercial feed to a low fat, cholesterol-free semisynthetic diet. The role of different dietary components in mediating this effect was investigated by varying the composition of the semisynthetic diet and of the commercial feed. Addition of alfalfa to the semisynthetic diet prevented the normal hypercholesterolemic response, but other plant products, including several with high fibre content, were ineffective. Increasing the content of powdered cellulose appeared to enhance the response. A commercially formulated diet in which the alfalfa and soybean meal components were replaced by ground corn and oats did not produce a significant elevation of plasma cholesterol. Substitution of different sugars or starches for the dextrose in the semisynthetic diet gave variable results, but of those tested, only potato starch prevented the hypercholesterolemia. The type of protein used in the semisynthetic diet had a marked influence on the level of plasma cholesterol. Semisynthetic diets containing proteins from animal sources gave higher plasma cholesterol levels than those containing proteins from plant sources. Very low levels were obtained with a low choline * Supported by the Ontario Heart Foundation and the Medical Research Council of Canada. Preliminary reports of this work were presented at the Third International Symposium on Atherosclerosis in West Berlin, Oct. 25-28, 1973; and at the 34th Annual Meeting of the Institute of Food Technologists in New Orleans, May 12-15.1974. 1 Recipient of a Medical Research Council of Canada Studentship. Present address: Animal Research Institute. Agriculture Canada. Ottawa, Ontario KlA 0C6, Canada. 2 Medical Research Associate of the Medtcal Research Council of Canada.
48
semisynthetic diet containing soy protein isolate, and supplementation with choline and methionine only raised the level to that normally obtained with commercial feed. Replacement of the salt mixture in the semisynthetic diet by one specially recommended for rabbits made no significant difference to the hypercholesterolemic response. Prevention of coprophagy did not significantly affect plasma cholesterol levels in rabbits on either commercial or semisynthetic diets. Growth performance was generally better on commercial feed than on semisynthetic diets but there was no direct correlation between weight gain and level of plasma cholesterol in rabbits on the different semisynthetic diets. Key words: Cholesterol-free
semisynthetic diets - Coprophagy - Dietary carbohydrate Dietary fibre -Dietary protein -Plasma cholesterol -Rabbits
-
Introduction It has been known for some time that hypercholesterolemia and atherosclerosis can be produced in rabbits by feeding semisynthetic diets without added cholesterol [l-17] , whereas this does not occur in rabbits fed commercial laboratory chow diets [8,13]. The first attempts to explain these effects of semisynthetic diets focussed primarily on the dietary lipids [l-5], since it was clear that they could be largely prevented by including polyunsaturated fats in the diet. It appears, however, that this provides only a partial answer to the question, since Kritchevsky and Tepper [13] showed that the protective effect of commercial feed was not due to the small amount of unsaturated fat present. Fat-extracted Purina chow did not produce hypercholesterolemia and atherosclerosis in rabbits, and these effects were not prevented by adding the extracted fat to a semisynthetic diet. Other studies have indicated that non-lipid components of the semisynthetic diet can influence the results obtained. Howard et al. [lo] altered the composition of the basic diet in various ways and reported that the hypercholesterolemia and atherosclerosis could be reduced by replacing the casein in the diet by whole soya flour or by hexane-extracted soya bean meal. Purified soya protein, however, was found to be ineffective. Moore [ 111 reported that diets containing cellulose gave higher plasma cholesterol levels in rabbits than similar diets containing wheat straw as source of fibre. The highest plasma cholesterol levels and greatest degree of aortic atherosclerosis were obtained with diets containing cellophane, but replacement of part of the cellophane by peat tended to neutralize its effects. The effect of varying the carbohydrate component of the semisynthetic diet was investigated by Kritchevsky et al. [14,17]. In their studies, starch, sucrose and fructose appeared to be more atherogenic than lactose or glucose. Malmros [ 151 reported on the effect of sucrose, glucose and starch fed to rabbits in fatfree, cholesterol-free semisynthetic diets. Sucrose gave the highest serum cholesterol levels and starch the lowest, and the sucrose diet was the most atherogenie of the three. Earlier studies in our own laboratory [ 161 showed that rabbits on a fat-free,
49
cholesterol-free semisynthetic diet had much higher plasma cholesterol levels than rabbits on fat-extracted commercial feed. Addition of one part casein to three parts commercial feed produced an elevation of plasma cholesterol, while addition of glucose (dextrose) at the same level had no effect. The present studies were undertaken to investigate in more detail the effects of different components of the commercial and semisynthetic diets on plasma cholesterol levels. Since it was easier to investigate the semisynthetic diet because of its better defined composition, most of the experiments consisted of feeding trials in which the protein, carbohydrate, fibre or other components of this diet were varied. One experiment was carried out, however, to see whether the alfalfa and soy bean components of commercially formulated rabbit diet might be responsible for the low plasma cholesterol levels on this type of diet, since both alfalfa [l&19] and soy bean meal [10,20] have been reported to produce hypocholesterolemic effects. For this purpose, a commercial-type diet containing no alfalfa or soy bean meal was fed. The effect of coprophagy was also investigated in rabbits on commercial and semisynthetic diets by fitting them with collars which prevented the practice of coprophagy. Material and methods Experimental
animals
Young male New Zealand rabbits, 8 to 12 weeks old, were used for these experiments. They were obtained from local breeders and usually weighed 0.75 to 1 kg at the beginning of the experiments. They were housed in galvanized cages (59 X 48 X 31 cm) with a wire mesh (1.5 cm) bottom. Normally they were housed individually, but in a few instances two animals were housed together. The cages were kept in an air-conditioned room maintained at 21-24°C with lights programmed to turn on at 07.00 h and off at 19.00 h. Fresh food and water were provided daily and the collection trays for excreta were cleaned three times per week. On arrival, all rabbits were initially fed ground Master Feed Rabbit Pellets for 3-4 days. They were then transferred gradually to the experimental diets over a 3-4 day period by feeding a mixture of ground rabbit pellets and experimental diet (2 : 1, v/v) for the first half and a 1 : 2 mixture for the second half of this period. The experimental diets were fed ad libitum for 28 days. Animals were weighed twice weekly and feed consumption was measured over 14-day periods. To minimize scattering of feed, each feeder was fitted with a metal cover containing an opening just large enough to permit access to the feed. Two groups of rabbits were fitted with plastic collars to prevent coprophagy. These collars weighed 70--80 g and were made from 3 mm thick acrylic plexiglass, with an outside diameter of 15 cm and an inside diameter of 4.4 cm for the animal’s neck. They were constructed in two halves held together by small bolts so that they could be easily removed. Diets and dietary
ingredients
The basal low fat, cholesterol-free semisynthetic diet used for these experiments consisted of casein 27, dextrose 60, salt mixture 4 and celluflour 5, expressed as parts by weight. The formulation was similar to that used in earlier
50
experiments [16] but was modified to include 2 parts by weight of molasses diluted (3 : 1, v/v) with water. This was done because some of the diets were quite dusty. The diet contained a supplement of water-soluble vitamins [ 121 as well as fat-soluble vitamins dissolved in corn oil. After some of the feeding trials had been completed, it was discovered that due to a typing error, choline was added at the level of 1.5 mg/kg rather than 1.5 g/kg. This applies to experiments for which results are given in Tables 2, 4 and 5. The fat-soluble vitamins consisted of 100 mg vitamin A acetate (200 units/mg) 110 mg DLQ-tocopherol acetate (1 I.U./mg) and 30 mg vitamin K (menadione) in 10 ml of corn oil per kg of diet. Other semisynthetic diets were prepared by replacement of part or all of various components of this basal diet. The commercial diet consisted of rabbit pellets (Master Feeds Division of Maple Leaf Mills, Ltd., Toronto, Ont.) which were ground and fed in powdered form, since the semisynthetic diets were also in powdered form. According to the manufacturer, the pellets contained crude protein (min.) l&O%, crude fat (min.) 3.0%, crude fibre (max.) 15.0%, salt 0.5%, calcium 1.5%, phosphorus 0.68%, vitamin A (min. I.U./lb) 5000, and added antibiotics (Aurofac 50, 112 g/ton, Lederle Laboratories, Cyanamid of Canada, Ltd., Montreal, Que.). “Vitamin-Free” casein, lactalbumin, raw egg white (spray dried), whole egg solids, wheat gluten, soybean protein (edible), zein, peanut meal, pectin and the various starches, lactose, Phillips-Hart salt mix [ 211, water-soluble and fatsoluble vitamins were purchased from Nutritional Biochemicals Corp., Cleveland, Ohio. Hydrous dextrose, U.S.P., was obtained from Ingram and Bell, Ltd., Toronto, Ont., and Celluflour from the Chicago Dietetic Supply House, Inc., Chicago, Ill. Powdered skim milk (Carnation Co. Ltd., Toronto, Ont.), granulated white sugar (Atlantic Sugar Refineries Co., Ltd., Montreal, Que.), corn oil (Mazola, Canada Starch Co., Ltd.) and butter were purchased from local food markets. Ground oat groats and ground oat hulls were obtained from Master Feeds Division of Maple Leaf Mills, Ltd. Ground wheat, ground oats and molasses were obtained from Arva Flour Mills, Ltd., Arva, Ont., and ground wheat straw from fall wheat, Tritium uulgare, was obtained from a local farmer. Sawdust, heat treated, was a product of Ab-Sorb-Dri, Incorp., Garfield, N.J. The alfalfa, which was pelleted and subsequently reground, was kindly provided by Dr. S. Fedoroff, Department of Anatomy, University of Saskatchewan, Saskatoon, Sask., and the Promine-R (isolated soy protein) was generously supplied by Dr. E.W. Meyer, Central Soya, Chemurgy Division, Chicago, Ill. A special rabbit ration (A 233) in which the alfalfa and soybean meal were replaced by ground corn and oats, was formulated and supplied by Dr. W.D. Morrison, Maple Leaf Mills, Ltd., Toronto, Ont. Lipids were removed from whole egg solids by extracting first with 6 volumes of chloroform-methanol (2 : 1) for 48 h, then with 4 volumes of the same solvent for 48 h and finally with 1.5 volumes of acetone for 3 h, filtering each time. The residue was dried in a vacuum oven at 35-40°C for 48 h. For preparation of beef and pork protein isolates, roasts of top round beef or pork tenderloin were stripped of excess fat and ground in a food chopper.
51
The ground meat was frozen (-20°C) in thin layers, lyophilized and reground to a fine powder in a coffee grinder. The meat powder was extracted with 6 volumes of ether for 48 h, 4 volumes for 48 h, and 3 volumes for 4 h at room temperature, filtering each time. The residue was spread in a thin layer on a tray and left in a fume hood for 48 h to remove residual ether. Peanut protein concentrate was prepared by extracting peanut meal with 5 volumes of methanol-water (1 : 1, v/v) for 48 h, then with 3 volumes of the same solvent for 48 h and finally with 2 volumes of acetone for 3 h. Solvents were removed by filtration and the residue was dried in a vacuum oven at 3540°C for 48 h. Analytical
methods
Egg, meat and peanut meal preparations were analyzed, before and after extraction, for crude protein, total lipid and cholesterol content, with the results shown in Table 1. Other protein-containing products were also analyzed for nitrogen by the micro-Kjehldahl method [ 221. Crude protein values were calculated from the total nitrogen content, using N, X 6.38 for milk proteins, N, X 5.70 for wheat protein and Nz X 6.25 for other proteins. Replacement of protein in the semisynthetic diets was done on an isonitrogenous basis. The crude fibre content of oat hulls and wheat straw was obtained from Tables of Feed Composition [ 231. Since the utilization of crude protein in oat hulls and wheat straw is reported to be very low, even by ruminant species [23], the casein content of the fibre-containing diets was not adjusted for nitrogen present in these fibre sources. The wood sawdust was assumed to consist entirely of crude fibre. Total plasma and liver lipids were determined gravimetrically after extraction by the method of Folch et al. [24]. Free and total cholesterol in plasma and liver were measured by the method of Sperry and Webb [ 251. Analysis of variance was performed according to a completely randomized TABLE 1 ANALYSIS
OF EGG, MEAT AND PEANUT PREPARATIONS
Material
Before
No. of samples analyzed
Crude protein a
Total lipid b
Cholesterol ’
(%)
(So)
(rngk)
50.6 70.9 71.9 56.9
41.5 19.6 19.5 1.5
17.4 2.0 2.5 ad. d
81.0 90.6 90.9 70.2
1.3 e 2.3 2.6 n.d. d
0.22 e 0.17 0.10 n.d. d
extraction
Whole egg powder Beef (freeze dried) Pork (freeze dried) Peanut meal Extracted Whole egg powder Beef Pork Peanut meal
a N2 X 6.25. Determined by micro-Kjehldahl [22J. b Lipid extractable with chloroform-methanol by the method of Folch et al. [ 241. ’ An aliauot of the total lipid extract was analyzed by the Spew-Webb procedure [25]. d Not determined. e Only two samples were analyzed for lipid.
block design [26]. Tests of significant Duncan’s New Multiple Range Test [27] for unequal numbers in different groups. cepted as indicating a significant biological
differences of means were based on as modified by Kramer [28] to allow The 5% level of probability was acdifference.
Results Rabbits fed the control low fat semisynthetic diet developed a hypercholesterolemia over a period of 2-4 weeks relative to those on commercial rabbit feed (Table 2). The semisynthetic diet was not as readily accepted as commercial feed and both feed intake and weight gain were reduced to about half on this diet, but most rabbits gained weight and appeared healthy. Since the commercial feed is largely made up of material derived from plant sources, a number of experiments were carried out in which different plant products were incorporated into the semisynthetic diet. The amount added was calculated on the basis that plant protein replaced l/4 of the casein in the control diet and the associated non-protein material was added at the expense of carbohydrate. Rabbits fed diets containing alfalfa, ground wheat, ground oats or oat groats generally had lower average plasma cholesterol levels than rabbits on the control semisynthetic diet, but the difference was only significant for the rabbits fed alfalfa for 28 days (P < 0.05). Addition of the plant materials improved the growth rate, which in most cases was as good as that obtained with commercial diet (Table 2). In other experiments, the crude fibre content of the semisynthetic diet was increased by adding plant materials which contained little or no available protein. These were added at the expense of the cellulose and part of the carbohydrate of the control diet so that the diets were isonitrogenous with respect to casein. Except for the group fed oat hulls, rabbits became hypercholesterolemic on these diets to a degree similar to that observed in animals on the control semisynthetic diet. Addition of extra cellulose in the form of Celluflour actually appeared to accentuate the hypercholesterolemia (Table 2). Growth rates on these diets were greater than on the control semisynthetic diet, but less than those obtained with commercial feed. Table 3 shows the plasma cholesterol levels and growth performance of rabbits fed a modified commercial diet in which the alfalfa and soybean meal of the regular diet were replaced by ground oats and corn. This diet had no significant effect on plasma cholesterol, although there was a reduction in weight gain (P < 0.05). Addition of 15% butter to this diet did produce a significant elevation in plasma cholesterol after 28 days (P < 0.01) but addition of 15% corn oil had no significant effect. In another series of feeding trials, the dextrose in the control semisynthetic diet was replaced by carbohydrates from various sources. The results are presented in Table 4. Each of these diets produced a significant hypercholesterolemia relative to commercial feed, with the exception of the one containing potato starch. Only wheat starch gave a higher mean plasma cholesterol level than dextrose. Except for potato starch there were no significant differences between the levels obtained with any of the carbohydrate diets, including the diet
d
6 6 5 4 4
5 6 6 5
22 20
FIBRE
917 814 957 925 877
1149 958 918 926 i f f + f
f f f f 30 30 41 61 32
119 41 75 9
917 f 33 994 f 46
Initial weight (9)
CRUDE
21 20 20 17 16
23 28 26 19 * f f f f
+ * f f 2 2 2 2 2
3 3 3 3
27 f 1 132 1
gain (g/day)
SOURCES
? + + f
6 7 11 3
78 f 11 88 * 3 93 + 13 103 19
84 89 87 71
loo+ 5 62 + 4
Feed intake (g/day)
DIFFERENT
Weight
FROM
CHOLESTEROL
56 29 40 62 99
16 38 48 59 f f f f f
+ f + + 1 5 5 14 4
7 2 I 10 170 110 100 160 27oi
f + t f
21 15 13 62 17
88 + 30 140 f 17 145 * 31 llO? 6
651r 5 140 * 21
67 2 40 + 57 f 96i 113 f
10 7 15 12 10
8+-2 41 f 2 42 + 8 63 f 18
26 f 2 72 * 7
Free
Free
25 f 2 54 f 7
28 days
LEVELS
14 days Total
Plasma cholesterol (mg/dl)
ON PLASMA
AND
195 145 185 280 355
60 135 125 160
16 11 20 38 t 26 C 16 ? 61 ? 60 ? 55
f f + +
IO * 5 200 f 22
Total
GROWTH
PERFOR-
a Results expressed as Mean f SEM. b Indicates source and amount of nutrient (figures in brackets) which distinguishes the diet from the semisynthetic control containing 27% casein. AII diets were &nitrogenous, excluding nitrogen in the fibrous products. which was assumed to be little utilized. See text for composition of low fat semisynthetic control diet. c Plant protein replaced l/4 of the casein. The associated non-protein plant material was added at the expense of dextrose. d The crude fibre content (16%) of the first 3 diets was equivalent to that of Master Feeds rabbit peUets.
Wheat straw (42.8%) Oat huIIs (59.3%) Sawdust (16%) Cellulose (20%) CeIIuIose (40%)
Crude
fibs
material ’ Alfalfa (36.3%) Ground wheat (62.1%) Ground oats (62.5%) Ground oat groats (59.9%)
Plant
AND
No. of animals
MATERIAL
Ground rabbit pellets Semisynthetic
controls
Diet b
OF PLANT
IN RABBITS=
MANCE
2
EFFECT
TABLE
54
TABLE
3
PLASMA
CHOLESTEROL
COMMERCIAL
DIET
LEVELS OR
Diet
No.
of
animals
Regular
commercial
A233
’
A233
+ 15%
butter
A233
+ 15%
corn
a Results b Ground
expressed Master
c A233-Commercial
b
oil
diet
GROWTH
PERFORMANCE
COMMERCIAL
OF
RABBITS
FED
REGULAR
DIETa
Initial
Weight
Feed
Plasma
weight
gain
intake
____
(g)
(g/day)
(g/day)
14 days
k 5
cholesterol
Free
Total
11
6Of
+ 3
(mg/dl) 28
6
972
+ 45
24
+ 2
94
6
874
f 43
19
f 1
81?12
25k2
55
6
892
+ 53
15+
3
50
f 1
38
f 3
9Ok
7
893
f 29
15+
2
50
+ 3
32
k 9
7Oi-
as Mean Feeds
AND
MODIFIED
days
Free 10
+ 8
24k
Total 5
80
i- 12
33
k 3
90f
7
21
46
k 3
130+
9
18
26?
5
-____
75
+ 12
f SEM.
rabbit
pellets.
modified
by replacement
of alfalfa
and soybean
meal
with
ground
oats
and
corn.
in which 5% of the dextrose was replaced by an equal amount of pectin. Rabbits on the pectin diet excreted soft pasty feces and this was also observed sporadically in animals on the lactose diet; otherwise the feces were similar in appearance to those excreted on the control semisynthetic diet. Growth rates were also similar to those obtained with the control diet. In other feeding trials the casein in the semisynthetic control diet was substituted on an isonitrogenous basis by proteins from various animal or plant sources. The results are shown in Table 5. Rabbits fed diets containing proteins from animal sources for 28 days had higher mean plasma cholesterol levels, ranging from 105 to 235 mg/dl, than those fed diets containing proteins from plant sources, in which the averages ranged from 15 to 80 mg/dl. Among the animal proteins, the lowest values were obtained with egg white and pork protein concentrate, while lactalbumin, skim milk and extracted whole egg each gave average plasma cholesterol values of greater than 200 mg/dl. Diets containing plant proteins such as wheat gluten and peanut protein gave values similar to those obtained with commercial diet and even lower values were obtained with soy protein concentrate and isolated soy protein. The one diet containing a mixture of plant and animal protein (zein : albumin, 3 : 1) gave a higher level (170 mg/dl) than any of the diets containing plant protein alone. It can be seen from the weight gains on the different diets that there was no correlation between growth rate and level of plasma cholesterol in these experiments. The effects of supplementing casein and soy protein semisynthetic diets with choline and methionine are illustrated in Table 6. Supplementation improved growth on the soy protein diets, but addition of methionine to the casein diet had the reverse effect. The rabbits developed fatty livers on the low choline diets, particularly the diets containing soy protein. Total liver cholesterols also tended to be higher on the low choline diets but the increase was not statistically significant. It is evident that the differences in plasma cholesterol between rabbits on the casein and soy protein diets were not abolished by supplementation, although the levels were somewhat higher for the soy protein diets supplemented with methionine (Table 6). Very low levels were observed in rabbits on the low cho-
4
OF CARBOHYDRATE
6 6 5 6 6 5 5
22 20
SOURCES
904 809 896 930 944 1167 896
86 25 68 41 ??63 f 62 f 60
f * f f
917 + 33 994 f 46
(9)
Initial weight
DIFFERENT
No. of animals
FROM
13 * 15 ? 17 + 18 f lo* 15 * 13f
2 3 2 2 3 3 3
27 f 1 13 f 1
gain (g/day)
Weight
ON PLASMA
55 76 72 61 56 66 61
f f f f f f f
6 11 18 7 10 6 9
100 f 5 62 f 4
(g/day)
Feed intake
CHOLESTEROL
AND
GROWTH
37 f 45 * 79 + 60 + 39 f 355 64 +
8 7 21 8 9 5 13
100 140 220 170 100 110 150
f f 2 f f f; f
22 27 54 34 27 27 45
65* 5 140 + 21
63 44 85 59 53 30 47
+ 13 + 5 -t 22 f 13 * 8 f 5 t 10
26 + 2 72 + 7
Free
Free
25 f 2 54 f 7
28 days
PERFORMANCE
14 days Total
Plasma cholesterol (mg/dI)
LEVELS
185 135 260 185 140 95 140
f f f ? f + +
41 24 66 44 19 24 33
70 ? 5 200 + 22
Total
IN RABBITSa
a Results expressed as Mean f SEM. b Indicates type and amount (figures in brackets) of carbohydrate replacing dextrose in the semisynthetic control diet. Casein was the source of protein and all diets were isonitiogenous. See text for composition of low fat semisynthetic control diet.
Sucrose (60%) Lactose (60%) Wheat starch (60%) Corn starch (60%) Rice starch (60%) Potato stamh (60%) Pectin (5%)
Carbohydrate b
Ground rabbit pellets . Semisynthetic (Dextrose)
Controls
Diet
EFFECT
TABLE
5
AND
protein
PLANT
6 6 4 6 6
6
4 6 5 5 6 6
22 20
No. of animals
PROTEIN
f f f + f f
33 26 38 50 72 51
979 1025 1058 924 952 * f f f *
59 37 74 60 96
945 * 50
980 879 872 1033 1130 875
917 + 33 994 f 46
3fl 15f 24 ? 12f 4*
2+-l
2 2 2 1
-1 f 0.7 18 + 3 9?2 20 f 2 25+ 1 9+2
27 * 1 13 f 1
(g/day)
(9)
SOURCES
Weight gain
DIFFERNT
Initial weight
FROM
11 5 7 6 4 4
61 74 81 84 81
? f + ? f
8 4 6 4 5
69 i 13
57 ? 71? 53 f 58 * 67 ? 38 ?
105 t 5 62 + 4
(g/day)
Feed intake
ON PLASMA
LEVELS
* f f f f +
10 5 14 15 7 14
25 ? 40? 25 * 24 + 15 r
4 10 6 8 7
35 + 4
65 59 46 54 45 40
25 + 2 54 f 7
Free
14 days
65 90 85 75 30
f 11 f 23 f 18 .+ 23 f; 14
95 f 14
150 ? 16 185 f 14 115 * 92 150*38 140 fr 21 110 f 36
65 f 5 140 ? 21
Total
Plasma cholesterol (mg/dl)
CHOLESTEROL
AND
70 10 26 21 5 7
25 ? 30 + 25 ? 11* 10 +
6 7 9 4 4
60 + 25
130 f 67 t 92 + 60 C 35 t 35+
26 f 2 72 t 7
Free
28 days
GROWTH
+ ? t + + *
89 40 69 60 17 28
80 + 80 * 75 f 25 + 15*
21 10 27 5 5
170 + 66
235 230 215 160 110 105
70 * 5 200 f 22
Total
PERFORMANCE
IN
a Results given as Mean f SEM. See text for composition of low fat semisynthetic control diet. b Indicates source and amount (figures in brackets) of nutrient which distinguishes the diet from the semisynthetic control diet that contained casein (27%). All diets are isonitrogenous and the carbohydrate was supplied by dextrose. c The zein and lactalbumin were in the ratio of 3 : 1.
Peanut protein cont. (“36%) Peanut meal (44.5%) Soy protein cont. (28.7%) Promine-R (27.5%)
Wheatgluten (32.2%)
Plant protein
protein
(29%) c
and plant
la&albumin
animal
Zein:
Mixed
Extracted whole egg (?33%) Skim milk (73.2%) Lactalbumin (31.3%) Beef protein cont. (=28%) Pork protein cont. (=28%) Raw egg white (30.6%)
Animal
Ground rabbit pellets Semisynthetic (Casein)
controls
Diet b
EFFECT OF ANIMAL RABBITS a
TABLE
OF
a Results
expressed
as Mean
k SEM.
-
See
text
for
6
5
-
1.5
isolate
5
4
1.5
5
0.0015
5.4
7
7
isolate
5.4
-
-
protein isolate
1.5
concentrate
Commercial
Soy
1.5
0.0015
concentrate
SOY protein concentrate
7
4
5.4
-
(g/kg)
of
1255
1313
1233
995
1338
1157
1068
1414
1176
1146?
w
weight
Initial
AND
composition
animals
No.
CHOLINE
1.5
0.0015
CZX4Il
DL-Meth-
ionine
WITH
6
(g/kg)
a
1.5
Choline
Dietary
DIETS
SUPPLEMENTATION
SEMISYNTHETIC
6
Protein
FAT
EFFECT
TABLE
62
53
of basal
+ 58
+ 25
+ 54
* 94
+ 50
+ 50
f
+ 178
? 36
5 + 1
2
f 2
* 1
3
* 3
? 1
low
25?
22
12
12f
16
13
4t1
4+3
14
12+ + 6 5
4
? 4
+ 7
f 8
+ 5
f 3
+ 3
f
f 1 3.8
4.4
5.6
3.8
4.2
3.6
8.1
3.8
4.7
f 0.2
+ 0.2
f 0.2
+ 0.7
* 0.3
* 0.5
k 2.4
k 0.2
Diets
!z)
f 0.4
diet.
w100
were
0.23
0.27
0.22
0.17
0.25
0.24
0.20
0.22
0.27
fed
for
+ 0.01
f 0.03
f 0.02
? 0.06
+ 0.01
i 0.01
+_ 0.05
* 0.02
?- 0.02
i 0.04
days.
0.25
0.33
0.26
0.37
0.31
0.34
0.50
0.29
0.41
0.69
w100
0.23
Total
AND
Frt?l?
28
LIVER
e/100
rz)
Cholesterol
PERFORMANCE,
+ 0.5
Total lipid
13.6
fat semisynthetic
1OOC
76
49
68
68
67
66
49
58
17
(g/day)
(g/day)
Liver
GROWTH
intake
Feed
ON
gain
Weight
METHIONINE
+ 0.01
f 0.05
+ 0.02
+ 0.09
-r 0.03
+_ 0.03
+_ 0.23
? 0.03
+ 0.04
? 0.17
g)
PLASMA
361
390
446
133
480
283
160
607
522
460
f 55
‘40
k 45
f 12
+ 45
+_ 23
+ 22
+ 167
? 103
? 103
(mg/dl)
lipid
Total
Plasma
LIPIDS
OF
? 3
t
21
26
26
3
+ 23
5 f 4
t
k 3
7+6
29
14
14 + 11
6?2
71
60
68i
kg/d0
Free
Cholesterol
RABBITS
191*
55+
76
67
11
80?
37
13
245
204
f
11
12
+ 9
? 8
11
f 8
+ 3
+ 67
+ 44
39
LOW
(mz/dI)
Total
ON
a
b
SALT
MIXTURE
recommended
for rabbits
’
for
treat-
1106
1396
Prevented
group
contained
1327
Prevented
a Each
1167
Practiced
Semisynthetic
-
ON
D.
+ 70
* 175
Results
f
2
?r 4
* 2
expressed
6?2
12
24
20
f 70
(g/day)
gain
FEED
Weight
GROWTH,
control
is given
as Mean
69
f 2
+ 10
f SEM.
52 f 2
72
f
5
f 6
108
intake
105
(g/day)
Feed
f 20 t 10
55
* 3
15
60
+ 11
30
Free
days
Plasma 14
26
section.
21
* 2
cholesterol
+ 56 + 27
165
+ 9
150
55
75*
27
(mg/dI)
Total
+ 14
? 10
f 11
LEVELS
60
55 55
* 6 f 8
f 19
? 3
+ 21
days
80+
60
15?
20
12
+ 28
4
* 5
Free
28
RABBITS
14
71
73
Free
29
FED
AND
days
Fret?
IN
(mg/dl)
LEVELS
28 Total
cholesterol
CHOLESTEROL
Initial
Plasma
PLASMA
CHOLESTEROL
Methods
* 2
? 5
+ 8
PLASMA
and
61 64
(g/day)
f 3
AND
ON
intake
Feed
DIET
f 2
in Materials
CONSUMPTION
diet
78
17
1236
i
5
12
13
t 70
1250
* 107
weight
Diet
4 rabbits.
Initial
(B)
Practiced
Commercial
coprophagy
and
Diet
ment
[291
COPROPHAGY
OF
DIETa
THETIC
8
et al.
EFFECT
TABLE
’ See Gaman
75
f
1175
5
(g/day)
(9)
7
gain
weight
Weight
SEMISYNTHETIC
Initial
MODIFIED
of
OF
animals
No.
AND
a Results expressed as Mean +_ SEM. b Composition of the low fat semisynthetic
Diet
Casein-control with recommended salt mix ’
Casein-control
Diet
OF
RABBITS
I
EFFECT
TABLE
55
? 9
f 51
f
* 34 40
210
* 39
170*80
+ 16 60
Total
_
OR
SEMISYN-
PERFORMANCE
COMMERCIAL
46
190
222
Total
GROWTH
IN
59
line (0.0015 g/kg) diets containing soy protein concentrate or soy protein isolate. Two of the animals in the latter group had no measurable plasma total cholesterol after 28 days on diet and this observation has since been repeated in other rabbits fed this diet. Gaman et al. [29] investigated the nutrient requirements of the rabbit and formulated a salt mixture and a semisynthetic diet which were reported to give better growth performance and to prevent the loss of hair sometimes seen in rabbits on semisynthetic diets. The effect of these formulations on growth performance and on plasma cholesterol levels was tested and the results are presented in Table 7. Use of the recommended salt mixture in our regular semisynthetic diet had no demonstrable effect on either growth rate or plasma cholesterol level, and did not prevent hair loss. The recommended semisynthetic diet gave a low plasma cholesterol level, probably because it was based on isolated soy protein rather than casein. Growth was not significantly improved on this diet. [ 30-321 and it seemed possible Rabbits are known to practice coprophagy that cholesterol metabolism might be altered by recycling of fecal material through the intestinal tract. Collars were therefore constructed to prevent coprophagy and these were worn by rabbits consuming commercial or semisynthetic diets. The prevention of coprophagy had no significant effect on plasma cholesterol levels (Table 8), but weight gain was depressed in the rabbits with collars, especially those receiving the semisynthetic diet. Discussion Alfalfa and soybean meal, two of the ingredients commonly used in commercial feed for rabbits, have been reported to have hypocholesterolemic and antiatherogenic properties. More than 30 years ago, Meeker and Kesten [20] reported that diets containing soybean flour gave lower plasma cholesterols and less atherosclerosis than diets containing casein. Howard et al. [lo] have also called attention to the hypocholesterolemic and antiatherogenic effects of soya bean meal. The hypocholesterolemic properties of alfalfa have similarly been well documented [ 18,191. The nature of the substances in soybeans and alfalfa responsible for these effects is still unknown. Although it seemed possible that the presence of soybean meal and alfalfa in the feed might be responsible for maintaining normal plasma cholesterol levels in rabbits on commercial diets, this was not borne out in our experiments (Table 3). A commercially-formulated diet in which these ingredients were replaced by ground oats and corn did not produce a hypercholesterolemia when fed to rabbits. On the other hand, replacement of the casein in the semisynthetic diet by soy protein concentrate or soy protein isolate prevented the hypercholesterolemit response (Table 5). Alfalfa was also capable of preventing the hypercholesterolemia produced by a semisynthetic diet, when added at a level sufficient to replace one-quarter of the casein in the diet (Table 2). Other plant products such as ground wheat and ground oats were less effective. It has been suggested that dietary fibre can protect against hyperlipidemia and ischaemic heart disease [ 33-351, and the effects on plasma cholesterol lev-
60
els of adding various fibrous materials to the control semisynthetic diet were also investigated in our experiments (Table 2). These did not prevent the hypercholesterolemia, and increasing the level of the cellulose in the diet even seemed to enhance it. Moore [ 111 also observed that cellulose gave higher plasma cholesterol values than wheat straw, although the levels observed in his experiments were considerably lower than ours in each case. As in earlier studies by Kritchevsky et al. [ 14,171 and Malmros [ 151, our results showed that the carbohydrate component of the semisynthetic diet had some influence on plasma cholesterol levels, and the variation in results obtained with different starches was greater than that between starches and sugars (Table 4). The lowest average value was obtained with potato starch and the highest with wheat starch. The differing effects of starches were unexpected since they presumably all yield glucose on digestion. Vijayagopal et al. [36] suggested on the basis of their experiments with rats fed high fat, high cholesterol diets that the hypolipidemic action of different purified starches is correlated with their fibre content. Earlier studies in our laboratory [16] showed that addition of casein to commercial feed produced a hypercholesterolemic response and experiments reported in the present communication indicate clearly that the level of plasma cholesterol is strongly influenced by the nature of the protein preparation used in the semisynthetic diet (Table 5). A number of different protein preparations derived from animal sources gave a hypercholesterolemic response similar to that obtained with casein. On the other hand, none of the diets formulated with vegetable proteins produced a significant hypercholesterolemia relative to commercial feed, although they contained less polyunsaturated fat than the commercial diet. Work in other laboratories has also provided evidence that casein is more hypercholesterolemic than plant proteins such as soybean protein [ 10,201 or gluten [ 71 , From these results it appears that the hypercholesterolemia observed in earlier studies on rabbits fed semisynthetic diets without added cholesterol was due primarily to the use of casein as source of protein in the diets. Conversely, it seems probable that the relatively low level of plasma cholesterol in rabbits on commercial diets is due to the fact that the protein in these diets is derived mainly from plant sources. Our results with semisynthetic diets containing different plant proteins and the experiment in which the alfalfa and soybean meal in our commercial diet were replaced by ground corn and oats (Table 3) suggest, however, that the low levels obtained with commercial feeds are not due to any specific plant protein. The possible influence of dietary salts and vitamins was not investigated in any detail in these experiments and indeed, the requirements of the rabbit in this respect are not well defined [ 32,371. The very low plasma cholesterols observed with soy protein diets (Tables 5 and 6) may be related in part to fatty livers due to low choline intake, but the levels were still relatively low when the diets were supplemented with choline and methionine (Table 6). It is of interest that rabbits on the casein diet did not develop fatty livers, even on the low choline intake. Perhaps the somewhat higher level of methionine in casein prevented this from happening. Coprophagy might be expected to influence the availability of various vitamins, but it seemed to have little influence on plas-
61
ma cholesterol levels (Table 8). A change to a different type of salt mixture likewise had no apparent effect on plasma cholesterol (Table 7). It is interesting to note that Ignatowski [38], who first showed that atherosclerosis could be produced in rabbits by feeding diets containing meat, milk and eggs, attributed the lesions to toxic effects of animal protein. This view was largely abandoned after Anitschkow and Chalatow [39] showed that the typical picture of rabbit atherosclerosis could be produced by feeding pure cholesterol dissolved in vegetable oil, although some subsequent workers provided suggestive evidence that animal protein was also capable of giving rise to hypercholesterolemia and atherosclerosis in rabbits [ 20,40,41] . Cholesterol feeding has been a useful method of producing atherosclerotic lesions in the rabbit and other animals, but the role of dietary cholesterol, and indeed of dietary fat, in the etiology of hypercholesterolemia and atherosclerosis in humans is still being debated [42,43] . Epidemiological data on human populations show that mortality from cardiovascular disease is correlated as strongly with animal protein intake as with dietary fat [44,45] . Although some previous studies have failed to demonstrate an effect of dietary protein on serum cholesterol levels in humans, its role in atherosclerosis may be worthy of closer scrutiny [ 461. Acknowledgements Technical assistance by R.I. Hill, L.F. McPhee, S.A. Hill and R.M. Rasmussen is gratefully acknowledged. References 1 Lambert. G.F.. Miller, J.P., Olsen, R.T. and Frost, D.V., Hypercholesteremia and atherosclerosis induced in rabbits by purified high fat rations devoid of cholesterol, Proc. Sot. EXP. Biol. Med.. 97 (1958) 544. 2 Wigand, G., Production of hypercholesterolemia and atherosclerosis in rabbits by feeding different fats without supplementary cholesterol, Acta Med. Stand., Suppl. 1 (1959) 351. 3 Malmros, H. and Wigand, G.. Atherosclerosis and deficiency of essential fatty acids, Lancet. 2 (1959) 149. 4 Funch, J.P.. Krogh, B. and Dam, H., Effects of butter, some margarines and arachis oil in purified diets on serum lipids and atherosclerosis in rabbits, Brit. J. Nutr., 14 (1960) 355. 5 Funch. J.P.. Kristensen, G. and Dam. H., Effects of various dietary fats on serum cholesterol, liver lipids and tissue pathology in rabbits, Brit. J. Nutr.. 16 (1962) 497. 6 Gresham, G.A. and Howard, A.N.. Atherosclerosis produced by semisynthetic diet with no added cholesterol, Arch. Pathol., 74 (1962) 1. 7 Enselme. J., Cottet, J. and Fray. G., Etude de diverses influences alimentaires sur l’ath8rosclbros.e provoquee par une alimentation privee de cholesterol. Arch. Mal. Coeur, 56 (1963) (Suppl. 3), 52. 8 Kritchevsky, D.. Experimental atherosclerosis in rabbits fed cholesterol-free diets, J. Atheroscler. Res., 4 (1964) 103. 9 Moore, J.H. and Williams, D.L.. The effect of diet on the level of plasma cholesterol and the degree of atheromatous degeneration in the rabbit, Brit. J. Nutr., 18 (1964) 253. 10 Howard. A.N.. Gresham, G.A.. Jones. D. and Jennings, I.W.. The prevention of rabbit atherosclerosis by soya bean meal, J. Atheroscler. Res., 5 (1965) 330. 11 Moore. J.H., The effect of type of roughage in the diet on plasma cholesterol levels and aortic atherosis in rabbits. Brit. J. Nutr., 21 (1967) 207. 12 Carroll, K.K., Diet. cholesterol metabolism. and atherosclerosis, J. Amer. Oil Chem. Sot., 44 (1967) 607. 13 Kritchevsky, D. and Tepper, S.A.. Experimental atherosclerosis in rabbits fed cholesterol-free diets Influence of chow components, J. Atheroscler. Res., 8 (1968) 357.
62 14 15 16 17
18 19 20 21 22 23 24 25 26 27 28 29 30 31 32
33 34 35
Kritchevsky, D.. &Rata, P. and Tepper. S.A.. Experimental atherosclerosis in rabbits fed cholesterolfree diets, Part 2 (Influence of various carbohydrates), J. Atheroscler. Res., 8 (1968) 697. Malmros, H.. Dietary prevention of atherosclerosis, Lancet. 2 (1969) 479. Carroll, K.K., Plasma cholesterol levels and liver cholesterol biosynthesis in rabbits fed commercial or semisynthetic diets with and without added fats or oils. Atherosclerosis, 13 (1971) 67. Kritchevsky. D., Tepper, S.A. and Kitagawa. M., Experimental atherosclerosis in rabbits fed cholesterol-free diets, Part 3 (Comparison of fructose and lactose with other carbohydrates), Nutr. Rep. bit.. 7 (1973) 193. Cookson, F.B., Altschul, R. and Fedoroff, S.. The effects of alfalfa on serum cholesterol and in modifying or preventing cholesterol-induced atherosclerosis in rabbits, J. Atheroscler. Res., 7 (1967) 69. Barichello, A.W. and Fedoroff, S., Effect of ileal bypass and alfalfa on hypercholesterolemia, Brit. J. Exp. Path&, 52 (1971) 81. Meeker, D.R. and Kesten, H.D., Effect of high protein diets on experimental atherosclerosis in rabbits, Arch. Path&. 31 (1941) 147. Phillips. P.H. and Hart, E.B., The effect of organic dietary constituents upon chronic fluorine toxicosis in the rat, J. Biol. Chem.. 109 (1935) 657. Assoc. Offic. Agr. Chemists, Official Methods of Analysis. 10th edition, Section 38.012-38.014 A.O.A.C., Washington, D.C.. 1965. NRC, Joint United States -Canadian Tables of Feed Composition. Publ. 1232. National Academy of Sciences -National Research Council, Washington, D.C., 1964. Folch. J., Lees, M. and Sloane Stanley, G.H.. A simple method for the isolation and purification of total lipids from animal tissues. J. Biol. Chem.. 226 (1957) 497. Sperry, W.M. and Webb, M., A revision of the Schoenheimer-Sperry method for cholesterol determination, J. Biol. Chem., 187 (1950) 97. Steele, R.D.G. and Torrie. J.H.. Principles and Procedures of Statistics, McGraw-Hill Inc., New York, N.Y., 1960. Duncan, D.B., Multiple range and multiple F tests, Biometrics, 11 (1955) 1. Kramer, C.Y., Extension of multiple range tests to group means with unequal numbers of replications. Biometrics, 12 (1956) 307. Gaman, E.. Fisher, H. and Feigenbaum, A.S.. An adequate purified diet for rabbits of ail ages, Nutr. Rep. Int.. 1 (1970) 35. Eden, A.. Coprophagy in the rabbit, Nature (London), 145 (1940) 36. Thacker, E.J. and Brandt, C.S.. Coprophagy in the rabbit, J. Nutr. 55 (1955) 375. Hunt, C.E. and Harrington, D.D., Nutrition and nutritional diseases of the rabbit. In: S.H. Weisbroth, R.E. Flatt and A.L. Krans (Eds.) The Biology of the Laboratory Rabbit. Academic Press, New York, N.Y., 1974, p. 403. Trowell. H.. Dietary fibre, ischaemic heart disease and diabetes mellitus. Proc. Nutr. Sot., 32 (1973) 151. Burkitt, D.P., Walker. A.R.P. and Painter, N.S., Dietary fiber and disease, J. Amer. Med. Ass., 229 (1974) 1068. Kritchevsky. D., Tepper, S.A. and Story, J.A., Nonnutritive fiber and lipid metabolism, J. Food Sci.,
40 (1975) 8. Vijayagopal, P.. Saraswathi Devi. K. and Kuru~. P.A., Fibre content of different dietary starches and their effect on lipid levels in high fat-high cholesterol diet fed rats, Atherosclerosis, 17 (1973) 156. 37 Nutrient Requirements of Rabbits (Nutrient requirements of domestic animals, No. 9). Publ. No. 1194 of National Academy of Sciences -National Research Council, Washington, D.C., 1966. 38 Ignatowski, A.. tier die Wirkung der tierischen Eiweisses auf die Aorta und die parenchymatiisen Organe der Kaninchen. Virchows Arch. Pathol. Anat. Physiol. Klin. Med., 198 (1909) 248. 39 Anitschkow. N. and Chalatow, S.. Ueber experimentelle Cholesterinsteatose und ihre Bedeutung fiir die Entstehung einiger pathologischer Prozesse, Cbl. Allg. Path& Pathol. Anat., 24 (19131 1. 40 Newburgh, L.H. and Clarkson. S., The production of atherosclerosis in rabbits by feeding diets rich in meat, Arch. Int. Med., 31 (1923) 653. 41 Clarkson, S. and Newburgh, L.H.. The relation between atherosclerosis and ingested cholesterol in the rabbit, J. Exp. Med., 43 (1926) 595. 42 Reiser, R., Saturated fat in the diet and serum cholesterol concentration - A critical examination of the literature, Amer. J. Clin. Nutr., 26 (1973) 524. 43 Keys, A., Grande. F. and Anderson, J.T.. Bias and misrepresentation revisited: “Perspective” on saturated fat, Amer. J. Clin. Nutr., 27 (1974) 188. 44 Yudkln, J., Diet and coronary thrombosis -Hypothesis and fact. Lancet, 2 (1957) 155. 45 Connor, W.E. and Connor, S.L., The key role of nutritional factors in the Prevention of coronary heart disease, Prev. Med., 1 (1972149. 46 Carroll, K.K. and Hamilton, R.M.G., Effects of dietary protein and carbohydrate on plasma cholesterol levels in relation to atherosclerosis, J. Food Sci.. 40 (1975) 18. 36