169

Atherosclerosis, 39 (1981) 169-175

@Elsevier/North-Holland Scientific Publishers, Ltd.

EXPERIMENTAL ATHEROSCLEROSIS IN RABBITS FED CHOLESTEROL-

FREE DIETS

Part 9. Beef Protein and Textured Vegetable Protein

DAVID KRITCHEVSKY, SHIRLEY A. TEPPER, SUSANNE K. CZARNECKI,

DAVID M. KLURFELD and JON A. STORY *

The Wistar Institute of Anatomy and Biology, 36th and Spruce Streets, Philadelphia, PA
19104 (U.S.A.)

(Received 23 September, 1980)

(Revised, received 26 October, 1980)
(Accepted 26 October, 1980)

Summary

Rabbits were fed a semipurified diet containing 40% sucrose, 25% protein,
15% fiber and 14% tallow. The proteins fed were beef (B), textured vegetable
protein (TVP) and casein (C). One diet contained beef-TVP (1 : 1) and in
another a soy carbohydrate fraction, spent flakes, was added to the diet; it
provided 2.2% protein and 10.5% fiber. TVP provided 6.4% soluble carbohy-
drate and 10.5% fiber. The diets were fed from 8 months. Diets containing beef
protein or casein gave significantly higher serum cholesterol levels and atheroscle-
rosis and significantly lower serum HDL-cholesterol than did the other 3 diets.
The beef-TVP (1 : 1) diet gave low serum cholesterol (67% below beef) levels
and atherosclerosis (47% below beef). This effect is probably due to the pro-
tein. The diet containing spent flakes gave low serum cholesterol levels (44%
below beef) and atherosclerosis (45% below beef). This effect is attributed to
the different fiber, The lowest serum cholesterol levels and least severe athero-
sclerosis were observed in the rabbits fed TVP.

Key words: Beef protein - Casein - Experimental atherosclerosis - Rabbits - Semipuri-

fied diet - Textured vegetable protein

This work was supported by the National Institutes of Health, Grants HL-03299 and CA-09171 and
by a Research Career Award, HL-00734. It was also supported by a grant from the ADM Company. Peoria.
IL.
* Present address: Department of Foods and Nutrition, Purdue University. West Lafayette, IN 47907.
U.S.A.
170

Introduction

Although interest in the effects of animal or vegetable protein on atheroscle-

rosis dates back to Ignatowski [l] there has been little active research in this
field until recently. Newburgh et al. [2,3] studied the effects of beef on experi-
mental atherosclerosis in rabbits but used no other protein for comparison.
Meeker and Kesten [4] compared casein and soy protein in their rabbit studies
and most subsequent work has used casein as the animal protein [ 5-91.
We have tested the effects of dehydrated beef protein on experimental ath-
erosclerosis in rabbits fed a semipurified diet. The diet used was patterned
after that which has been shown to be atherogenic for rabbits [lo] and Ver-
vet monkeys [ 111 and sudanophilic for baboons [ 121. The basic formula is:
40% carbohydrate, 25% protein, 15% fiber, 14% fat, 5% salt mix and 1% vita-
min mix. Since the source of beef protein used contains fat we have used
adjusted amounts of beef tallow as the source of fat in this study.
We have compared the diet containing beef protein (diet B) with 4 others. In
diet TVP the source of protein is textured vegetable protein; in diet BT the pro-
tein consists of equal levels of beef protein and TVP and in diet C the protein
used is casein. Diet C is similar to the standard diet used in our earlier experi-
ments [8,9] in which the protein, carbohydrate, fat and fiber were casein,
sucrose, coconut oil and cellulose, respectively. The fat used in diet C is beef
tallow. The last diet, designated as BSF contains beef protein and tallow but
the principal source of fiber is a substance called spent flakes which is the
insoluble residue from preparation of soy isolate.
The levels of sucrose and cellulose used have been adjusted for the levels of
carbohydrate and fiber found in TVP and spent flakes. The level of tallow used
has been adjusted for the amount of fat found in beef and in the case of diet
BSF the level of beef protein has been adjusted for the amount of protein
remaining in spent flakes. Our findings are summarized in this report.

Materials and Methods

The composition of the diets, which have been discussed in the introduction,
is summarized in Table 1.
The diets were fed to groups of 12 male Dutch-belted rabbits whose starting
weight averaged 1600 g. The rabbits were fed the diets for 8 months and were
then bled by cardiac puncture and killed by injection of a barbiturate solution
[Euthanasia solution containing sodium 5-ethyl-5-isoamylbarbiturate and
sodium 5-ethyl-5-(1-methylbutylbarbiturate), F.E. Lentz Co., Inc., Philadel-
phia, PA].
The serum-free and total cholesterol levels were determined by the method
of Pearson et al. [ 131 using digitonin precipitation for separation of free choles-
terol [14]. Serum triglycerides were determined by the method of Levy and
Keyloun [15] and phospholipids by the method of Sokoloff and Rothblat
WI.
The low density lipoproteins (LDL) were precipitated with heparin and
MgClz [17] and the cholesterol content of the high density lipoprotein (HDL)
in the supematant was determined [ 131.
 171

TABLE 1

DIET COMPOSITION a

Diet Protein (%) Fat (%) Carbohydrate (I)

-
Soluble Fiber

B Beef (25) Tallow (9.3) Sucrose (40) Cellulose (15)

Beef (4.7)

TVP TVP (25) Tallow (14) Sucrose (33.6) Cellulose (4.5)

TVP (6.4) TVP (10.5)

BSF Beef (22.8) Tallow (9.8) Sucrose (39.4) Cellulose (4.5)

Spent flakes (2.2) Beef (4.2) Spent flakes (0.6) Spent flakes (10.5)

BT Beef (12.5) Tallow (11.7) Sucrose (36.8) Cellulose (9.8)

TVP (12.5) Beef (2.3) TVP (3.2) TVP (5.2)

C Casein (25) Tallow Sucrose (40) Cellulose (15)

-
a All diets contain 5% Salt Mix and 1% Vitamin mix.

Liver aliquots were homogenized in chloroform-methanol (2 : 1) and ali-

quots of the dried chloroform extract analyzed for free and esterified choles-
terol [13,14], triglyceride [15] and phospholipid [16]. Protein content of
serum and liver was determined by the method of Lowry et al. [ 181. An aque-
ous extract of liver was used for the protein determination.
The sterol content of the beef, tallow, TVP and spent flakes was determined.
Aliquots of the beef and tallow were extracted with chloroform-methanol
(2 : 1) and their cholesterol content was determined 1131. The TVP and spent
flakes were also extracted with chloroform-methanol (2 : 1) and their sterol
content was determined calorimetrically using a pure p-sitosterol standard
[19]. Cholesterol content of the beef and tallow was 18 and 76 mg/lOO g,
respectively. The sterol content of the TVP and spent flakes was 20 mg/lOO g
and 7 mg/lOO g, respectively.
Aortas were graded visually on a O-4 scale [20,21].

Results and Discussion

Our findings are summarized in Table 2. The data were analyzed using anal-
ysis of variance (ANOV) [ 221.
Weight gain was highest in the rabbits fed TVP and lowest in those fed casein
or beef (P < 0.001). In contrast liver weights were highest in the rabbits fed
casein or beef and lowest in those feed beef-TVP (1 : 1). The difference was
significant based on either absolute weight (P < 0.025) or on percentage of
body weight (P < 0.001).
Significant differences in serum cholesterol levels were observed (P < 0.001)
but there were no significant differences in levels of esterified cholesterol. Diets
containing beef protein or casein produced equally elevated serum cholesterol
levels which were 67-82% higher than those observed in rabbits fed vegetable
protein. Serum cholesterol levels of the rabbits fed TVP were in the normal
range for this strain [23]. The 50% dilution of beef protein with TVP reduced
serum cholesterol levels towards the normal range, 185-61 mg/dl.
TABLE 2

INFLUENCE OF PROTEIN SOURCE ON LIPID METABOLISM AND ATHEROSCLEROSIS IN RABBITS a (*SEM)

B TVP BSF BT C ANOVA b

P<

NO. 12/12 9112 12/12 11/12 11/12

weight gain 348 f 67 788 * 115 522 * 60 527 f 66 166 f 54 0.001

Liver weight (g) 48.4 f 2.9 43.9 f 3.3 44.5 f 1.4 40.4 f 2.9 54.6 f 2.5 0.025
Liver (% body weight) 3.08 f 0.29 2.65 * 0.19 2.08 * 0.80 1.84 f 0.10 3.11 f 0.15 0.001

Serum lipids (mgldl)

cholesterol 185 f 24 37 * 4 103 *4 61 *6 200 t 18 0.001
96 ester 72.3 * 1.8 64.3 * 3.3 70.1 f 1.4 63.9 * 3.1 66.8 f 1.9 NS
% HDL-cholesterol 20.1 f 2.1 38.8 f 5.1 33.9 * 3.9 43.4 * 4.2 29.6 f 3.6 0.001
a/P 0.25 f 0.03 0.72 f 0.15 0.56 * 0.08 0.87 f 0.15 0.36 f 0.08 0.001
triglyceride 60 f 8 59 f 7 54 f 8 70 f 13 92 * 10 NS
phospholipid 92 f 8 67 f 5 88 *8 70 f 5 126 *8 0.001
Serum protein (g/dl) 5.53 f 0.16 5.09 + 0.21 4.86 f 0.14 5.24 f 0.16 5.39 f 0.20 NS
Liver lipids (g/100 g)
cholesterol 0.77 * 0.09 0.28 f 0.01 0.47 f 0.04 0.38 f 0.06 0.97 * 0.05 0.001
% ester 68.0 f 1.4 65.1 f 0.90 61.8 f 1.63 58.7 f 1.6 67.4 f 1.3 0.05
trigIyceride 0.77 f 0.08 1.16 f 0.39 1.11 f 0.63 0.72 i 0.12 0.51 +. 0.08 NS
phospholipid 1.46 f 0.06 1.18 f 0.07 1.64 f 0.61 1.58 f 0.04 1.41 f 0.06 0.001

Liver protein (g/100 g) 20.3 * 0.7 19.2 f 1.2 22.9 f 1.2 19.9 f 1.2 17.3 f 1.00 0.001

Atherosclerosis
arch 1.29 + 0.23 0.78 f 0.12 0.83 f 0.09 0.73 f 0.10 1.32 f 0.22 0.05
thoracic 0.75 f 0.12 0.22 It 0.09 0.29 f 0.07 0.36 f 0.12 0.86 f 0.12 0.005

a Diets, described in Table 1, fed for 8 months.

b ANOVA. analysis of variance using F-test.
NS = not significant.
 173

Diets TVP and BSF contained similar amounts of cellulose and soy fiber but
in the latter diet most of the protein (91%) was beef. Serum cholesterol levels
of rabbits fed diet BSF were 178% higher than those of rabbits fed diet TVP
but 44% lower than those of rabbits fed diet B. The difference between choles-
terol levels of groups BSF and B may be attributed largely to the type of fiber
present in the diet and the difference between groups BSF and TVP to the type
of protein.
The level of HDL-cholesterol was significantly (P < 0.01) higher in rabbits
fed vegetable protein. The lowest levels of HDL-cholesterol were observed in
the rabbits fed beef protein or casein. HDL-cholesterol levels in sera of rabbits
fed diets BSF, TVP or BT were 69, 93, and 116% higher than those of rabbits
fed beef protein and 15, 31, and 47% higher than those of rabbits fed casein.
We have shown [24] that altering the type of protein in the diet of rabbits can
effect significant changes in lipoprotein pattern. It is interesting to note that
the highest HDL-cholesterol levels were observed in rabbits fed beef protein-
TVP (1 : 1).
The differences in serum triglycerides were not significant by ANOV but the
casein diet produced the highest levels again suggesting differences in lipopro-
tein type and composition.
Phospholipid levels were significantly different (P < 0.001) and were highest
in those fed beef or casein. The level of serum protein, in decreasing order, was
beef protein, casein, beef protein-TVP, TVP and BSF. The data suggest that
serum protein levels increase with increasing quality of protein. The spent
flakes may have limited absorption of protein.
Liver cholesterol levels were significantly elevated in rabbits fed casein or
beef protein. Liver protein levels of TVP-fed rabbits were within the normal
range. The effect of diluting beef protein with TVP is remarkable since it
results in a 51% lowering of liver cholesterol. Levels of esterified cholesterol
were also lower in the rabbits fed diets BSF and BT.
Triglyceride levels were elevated in the rabbits fed TVP or BSF but the dif-
ferences were not significant by ANOV. Phospholipid levels were significantly
different but the levels showed no pattern which could be related to the protein
type. The concentration of total liver lipid (cholesterol plus triglyceride plus
phospholipid) was lowest in rabbits fed diets TVP or BT.
Liver protein levels were highest in rabbits fed diet BSF, no relationship
between protein quality and liver protein level could be adduced.
The levels of atherosclerosis in the rabbits fed beef protein or casein were sim-
ilar and significantly more severe than that observed in the other groups: The
differences in severity of atherosclerosis were significant in both the aortic arch
(P < 0.05) and the thoracic aorta (P < 0.01). Addition of spent flakes to the
beef protein diet reduced arch and thoracic atherosclerosis by 36 and 67%,
respectively. These reductions are probably due to the replacement of 70% of
the cellulose with another type of dietary fiber. We have shown [9] that the
type of fiber present in the diet affects the atherogenicity of both animal and
vegetable protein. Levels of atherosclerosis in rabbits fed diets TVP or BT were
similar and about 40% less severe in the arch and 60% less severe in the thoracic
aorta. These data show that dilution of beef protein with TVP can markedly
affect the course of experimental atherosclerosis. It could be argued that the
174

effects observed in groups B and C were due to the cholesterol content of beef
and/or tallow or that reduced levels of serum lipids and less severe atherosclerosis
in groups TVP, BT and BSF were due to the possible protective action of plant
sterols present in TVP or spent flakes. Diet B contained 29.7% beef plus beef
fat and 9.3% tallow. Assuming the rabbits ate 100 g of diet daily, their choles-
terol intake was 12.4 mg. The cholesterol intake of the animal fed casein was
10.6 mg/day. Rabbits fed diet BSF ingested 12.3 mg of cholesterol and 0.15
mg of plant sterol daily and those fed diet TVP ate 10.6 mg of cholesterol and
5.0 mg of plant sterol per day. Finally, rabbits fed beef-TVP (1 : 1) consumed
11.6 mg of cholesterol and 2.5 mg of plant sterol daily.
The cholesterol intake of the rabbits fed TVP or casein was identical but the
results were disparate. The plant sterol content of TVP (5.0 mg) could not have
exerted much effect since it has been shown [25] that for soy protein to be a
really effective inhibitor of cholesterolemia its concentration in the diet must
exceed that of cholesterol by a factor of 2 or 3, preferably the latter. We con-
clude that the observed effects were due to the dietary protein and not the
dietary sterol.
Another interesting observation was the number of rabbits exhibiting gall-
stones. The stones were readily soluble in organic solvents. Six of the rabbits
fed beef had gallstones but none of those fed TVP did. We [26] have shown that
substitution of soy protein for casein significantly lowers the incidence of gall-
stones in hamsters fed a lithogenic diet. Among the other groups, one of the
rabbits fed diet BSF had gallstones; in the group fed diet BT, 2 rabbits had gall-
stones and in the case&fed group, 5 had gallstones. The effect is most pro-
bably due to the tallow since we have not seen gallstones in rabbits fed casein
and coconut oil [9,10]. Fiber has been shown to inhibit the effects of litho-
genie diets in rabbits [27] and this could explain the relative absence of gall-
stones in rabbits fed diets BT and BSF. The absence of gallstones in rabbits fed
TVP could be due to the combined effects of vegetable protein and fiber.
We have hypothesized [24,28,29] that the ratio of lysine to arginine of a
protein may affect its atherogenicity. Thus casein with a lysine/arginine ratio of
2.00 is significantly more atherogenic than soy protein with a lysine/arginine ratio
of 0.86. The lysine/arginine ratio of beef is 1.41. Dilution of beef protein
with TVP would necessarily lower the lysine/arginine ratio of the diet.
Our earlier experiments [9] showed that the atherogenic effect of specific
proteins could be influenced by the type of fiber in the diet. The current study
has shown that the atherogenicity of beef protein can be reduced by altering
the fiber (from 15% cellulose to cellulose-soy, 1 : 2.33) or by diluting it with
an equal weight of a less atherogenic protein (TVP).
Extrapolation of these findings to man is unwarranted but it is noteworthy
that Sirtori et al. [30] were able to reduce lipemia and lipoproteinemia in hu-
man subjects simply by changing from animal to vegetable protein in diets of
otherwise virtually similar composition.
Acknowledgements
We are indebted to Drs. F. Horan and R. Daftary of the ADM Company for
many helpful discussions and for generously providing the dried beef, tallow,
TVP and spent flakes.
 175

References

1 Ignatowski, A., tier die Wirkung des tierischen Eiweisses auf die Aorta und die parenchymat6sen 0~
gane der Kamuchen. Virchows Arch. Path. Anat. Physiol. Klin. Med., 198 (1909) 248.
2 Newburgh. L.H. and Squier. T.L., High protein diets and atherosclerosis in rabbits - A Preliiinary
report, Arch. Intern. Med., 26 (1920) 38.
3 Newburgh. L.H. and Clarkson, S., The production of atherosclerosis in rabbits by feeding diets rich in
meat, Arch. Intern. Med., 31 (1923) 653.
4 Meeker, D.R. and Kesten, H.D., Effect of high protein diets on experimental atherosclerosis of rab-
bits, Arch. Path., 31 (1941) 147.
5 Nath, N., Harper, A.E. and Elvehjem. C.A.. Dietary protein and serum cholesterol. Arch. Biochem.
Biophys.. 77 (1958) 234.
6 ~~~~~ A.N., Gresh~, G:A., Jones, D. and Jennings, I.W.. The prevention of rabbit atherosclerosis
by soya beau meal, J. Atheroscler. Res.. 5 (1965) 330.
7 Loflsnd, H.B., Clarkson, T.B., Rhyme, L. and Goodman, H.O., Interrelated effects of dietary fats and
proteins on atherosclerosis in the pigeon, J. Atheroscler. Res.. 6 (1966) 395.
8 C~OU, K.K. and Hamilton, R.M.G., Effects of dietary protein and carbohydrate on Plasma choles-
terol levels in relation to atherosclerosis, J. Food Sci.. 40 (1975) 18.
9 Kntchevsky, D., Tepper. S.A., Williams, D.E. and Story, J.A.. Experimental atherosclerosis in rabbits
fed cholesterol-free diets, Part 7 (Interaction of animal or vegetable protein with fiber). Atheroscle-
rosis, 26 (1977) 397.
10 Kritchevsky, D. and Tepper, S.A.. Experimental atherosclerosis in rabbits fed cholesterol-free diets -
Influence of chow components, J. Atheroscler. Res.. 8 (1968) 357.
11 Kntchevsky, D., Davidson, L.M.. Kim, H.K., Krendal. D.A., Malhotra, S., Vander Watt. J.J.. Du
Fled, J.P., Winter, P.A.D., Ipp. T.. Mendelsohn. D. and Bersohn. I.. Influence of semi-purified diets
on atherosclerosis in African Green Monkeys, EXP. Mol. Path., 26 (1977) 28.
12 Kritchevsky, D., Davidson, L.M.. Shapiro, I.L.. Kim, H.K., Kltagawa. M.. Malhotra. S.. Nair. P.P..
Clarkson, T.B., Bersohn, I. and Winter, P.A.D., Lipid metabolism and exPerimenta atherosclerosis in
baboons - Influence of cholesterol-free, semi-synthetic diets, Amer. J. Clin Nutr., 27 (1974) 29.
13 Pearson, S., Stem, S. and McGavack, T.H.. A rapid accurate method for the determination of total
cholesterol in serum, Anal. Chem.. 25 (1953) 813.
14 Sperry, M.W. and Webb,M., Arevisionof the Schoenheimer-Sperry method for cholesterol determina-
tion, J. Biol. Chem.. 187 (1950) 97.
15 Levy, A.L. and Keyloun, C., Measurement of triglycerides using nonane extraction and colofimetry.
In: Advances in Automated Analysis, Vol. 1, Thurman Assoc., Miami, FL, 1972, PP. 497-502.
16 Sokoloff. L. and Rothblat. G.H.. Sterol to phospholipid molar ratios of L cells with qualitative and
quantitative variations of cellular sterol. Proc. Sot. Exp. Biol. Med., 146 (1974) 1166.
17 Lipid Research Clinics Program. Manual of Laboratory Operations, Vol. 1 (Lipid and Lipoprotein
Analysis) (DHEW Publications 75-628). 1974. pp. 56-59.
18 Lowry, 0-H.. Rosebrough. N.J., Farr. A.L. and Randall, R.J., Protein measurement with the Folin
phenol reagent, J. Biol. Chem., 193 (1951) 265.
19 Kritchevsky, D. and Tepper, S.A.. The free and ester sterol content of various foodstuffs, J. Nutr.. 74
(1961) 441.
20 Duff, G.L. and McMillsn. G.C.. The effect of alloxsn diabetes on experimental cholesterol atheroscle-
rosis in rabbits, J. Exp. Med., 89 (1949) 611.
21 Kritchevsky. D., Tepper. S.A. and Story, J.A.. Cholesterol vehicle in experimental atherosclerosis, Part
16 (Effect of peanut oil on pre-established lesions), Atherosclerosis, 31 (1978) 365.
22 Sokal, R.R. and Rohlf. F.J.. Introduction to Biostatiitics. W.H. Freeman and Co., San Francisco, CA,
1973. pp. 134-184.
23 Kritchevsky, D., Moynihan, J.L.. Langan. J.. Tepper, S.A. and Sachs, M.L., Effects of D- and Gthy-
roxine and D- and L-3.5.3’.triiodothyronine on development and regression of experimental athero-
sclerosis in rabbits, J. Atheroscler. Res., 1 (1961) 211.
24 Czamecki, S.K. and Krltchevsky, D.. The effect of dietary proteins on lipoprotein metabolism and
atherosclerosis in rabbits, J. Amer. Oil Chem. Sot.. 56 (1979) 388A.
25 Pollak, O.J., Successful prevention of experimental hypercholesterolemia and cholesterol atheroscle-
rosis in the rabbit, Circulation, 7 (1953) 696.
26 Kntchevsky. D. and Klurfeld, D.M.. Influence of vegetable protein on gallstone formation in hamsters.
Amer. J. Clin. Nutr.. 32 (1979) 2174.
27 Borgman, R.F. and Wardlaw, F.B.. Serum cholesterol and cholelithiasis in rabbits treated with pectin
and cholestyramine, Amer. J. Vet. Res.. 35 (1974) 1445.
28 Kritchevsky, D. Tepper, S.A. end Story, J.A., Influence of soy protein and casein on atherosclerosis in
rabbits, Fed. Rot., 37 (1978) 747.
29 Kritchevsky, D., Vegetable protein and atherosclerosis, J. Amer. Oil Chem. Sot., 56 (1979) 136.
30 Sirtori, C.R.. Gatti, E., Mantcro. 0.. Conti. F.. Agradi. E.. Tremoli, E., Sirtori. M., Fraterrigo, L.,
Tavazzi, L. and Kritchevsky, D.. Clinical experience with the soybean protein diet in the treatment of
hypercholesteremia, Amer. J. Clin. Nutr., 32 (1979) 1645.