Effects of Selective Breeding on the
Cholesterolemic Responses to Dietary Saturated Fat
and Cholesterol in Baboons
Henry C. McGill, Jr., C. Alex McMahan, Glen E. Mott, Yolan N. Marinez, and Thomas J. Kuehl
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Positive assortatlve mating of baboons {Paplo sp.) based on elevation of serum
cholesterol concentrations In response to a cholesterol- and saturated fat-enriched
diet produced 64 progeny (30 high line; 34 low line). When the animals were 3 to 4
years of age, we tested their llpoproteln cholesterol responses to dietary cholesterol
and fat In a factorial experiment with two levels of dietary cholesterol (1.7 and <0.01
mg/kcal) and two types of fat, coconut oil (P/S 0.1) and corn oil (P/S 3.5), each providing 40% of total calories from fat; we also tested their responses to chow. The high line
animals had significantly higher very low density plus low density llpoproteln (VLDL
+ LDL) and high density llpoproteln (HDL) cholesterol levels on all diets. The effects
of dietary cholesterol on VLDL + LDL cholesterol concentrations were greater In high
line animals than In low line animals, but dietary cholesterol's effects on HDL cholesterol were similar In both lines. The effects of saturated fat, compared to unsaturated
fat, on both VLDL + LDL and HDL cholesterol levels were similar in both lines.
Selective breeding produced lines diverging In llpoprotein cholesterol concentrations
by acting on several different genetically mediated processes that control serum
llpoprotein levels. At least one of these processes Involves responsiveness of serum
VLDL + LDL cholesterol concentration to dietary cholesterol.
(Arteriosclerosis 8:33-39, January/February 1988)
A
nimal species show a wide range in their serum lipoprotein concentration responses to diets enriched in
cholesterol and saturated fat. Among nonhuman primate
species, baboons (Papio sp.) are moderately sensitive to
dietary cholesterol, less than squirrel monkeys (Saimiri
sciureus), vervets (Cercopithecus aethiops), or rhesus
monkeys (Macaca mulatta), but more sensitive than spider
monkeys (Ateles sp.). 1 ' 2 The interspecies differences are
presumed to be genetically determined, but genetic control
of the serum cholesterol or lipoprotein cholesterol response to diet also has been demonstrated within several
species of nonhuman primates, including squirrel monkeys,3 cynomolgus monkeys,4 and baboons. 5 ' 6 ' 7
Dietary cholesterol affects primarily serum very low density plus low density lipoprotein (VLDL + LDL) cholesterol
levels, while saturated fat affects both VLDL + LDL and
high density lipoprotein (HDL) cholesterol levels.8 Previous genetic analyses did not distinguish between heritability of response to dietary cholesterol and heritability of
response to saturated fat. Therefore, we examined the
effects of dietary cholesterol and saturated fat in two diverging lines of pedigreed baboons. These lines were produced by positive assortatJve mating of parents based on
their serum cholesterol responses to a combined cholesterol and saturated fat dietary challenge.
Methods
Subjects
The subjects were the progeny of six sires and 64 dams
that had been selected from a large colony of baboons
(Papio sp.) on the basis of responses of serum cholesterol
to a cholesterol and saturated fat-enriched (challenge)
diet. The details of the original group, diet, serum cholesterol concentrations, and selection criteria have been presented in reports of the genetic analyses of selective
breeding.5 Briefly, six sires were selected from 15 fertile
males as the three highest and the three lowest responders, and 134 females were divided into high responders
and low responders. High-response females were randomly assigned to the three high-response males, and
low-response females were randomly assigned to the
three low-response males. These matings produced 96
live births over 32 months; 32 animals died or were excluded for health reasons before 3 years of age. Thus, 64
progeny (34 to 47 months old) were available for this experiment; 30 were derived from the high line and 34 from
the low line. Table 1 shows the serum cholesterol concentrations of each of the six sires and the mean values for the
two groups of females that contributed progeny for this
experiment.
From the Department of Physiology and Medicine, Southwest
Foundation for Biomedical Research, and the Department of Pathology, University of Texas Health Science Center, San Antonio,
Texas.
This research was supported by Grant HL-28972 from the National Heart, Lung, and Blood Institute, National Institutes of
Health. Part of this work was presented at the NHLBI Workshop
on the Impact of Dietary Cholesterol on Plasma Lipoproteins and
Atherogenesis, July 1-3, 1986, Bethesda, Maryland.
Address for reprints: Henry C. McGill, Jr., M.D., Department of
Pathology, The University of Texas Health Science Center, San
Antonio, Texas 78284.
Received June 12,1987; revision accepted September 2,1987.
33
34
ARTERIOSCLEROSIS VOL 8, No 1, JANUARY/FEBRUARY 1988
Animal Research Policies
The long-term selective breeding program, together with
this intensive characterization of the progeny, were conducted in accord with the United States Public Health Service policies for the use of laboratory animals and were
approved by the Animal Research Committee of the
Southwest Foundation for Biomedical Research.
Rearing
The breeders were housed in outdoor gang cages in
groups of about 20 females with each male. Infants were
breast fed and remained in the cages with their sires and
dams until weaning. When the infants were between 15
and 16 weeks of age, we removed them from their dams
and fed them a diet containing 1.7 mg cholesterol/kcal and
4 1 % of calories from fat, mainly lard, with a polyunsaturated/saturated fatty acid ratio of 0.42 (Table 2). They re-
mained on this diet until beginning the special dietary test
at between 34 and 47 months of age.
Diets
The rearing diet and the four special diets were prepared
by mixing the appropriate fat, salt, vitamins, and USP cholesterol with a special low-fat, low-salt meal prepared by
the Ralston Purina Company (St. Louis, Missouri) (Special
Monkey Chow 25-5045-6). The rearing diet was prepared
with lard as the added fat. The special diets were prepared
with coconut oil (Lou Ana Food Incorporated, Opelousas,
Louisiana) as the saturated fat and com oil (Best Foods
CPC International Incorporated, Englewood Cliffs, New
Jersey) as the unsaturated fat. The diet components were
mixed with a small amount of water, were pelleted and
stored frozen in plastic bags, and then were fed daily in
amounts to provide about 500 g per baboon. The chow
was regular Monkey Chow manufactured by the Ralston
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Table 1. Serum Cholesterol Concentrations of High and Low Line Sires and Dams and Numbers
of Progeny by Sire, Line, and Sex
Diet*
Parents
Sires
High line
X102
X116
X113
Low line
A956
X84
X98
Damsf
High line
Mean
Range
Low line
Mean
Range
Number of progeny
Chow
Challenge
Response
123
101
93
181
154
145
93
62
65
98
88
87
Female
Total
58
53
52
5
5
7
12
9
9
5
26
22
5
5
10
13
9
13
17
30
14
20
34
124
185
56
75-177
113-248
15-106
120
140
21
89-159
109-200
25-51
Male
12
Values are given in mg/dl.
'Chow values are single measurements; challenge values are geometric means of three measurements.
tincludes only the 64 dams that produced the offspring for this experiment.
Table 2. Selected Characteristics of Diets
Nutrient
Energy
Protein
Carbohydrate
Fat
Fatty addst
Saturated
Monounsaturated
Polyunsaturated
Cholesterol
Special dietst
Units
Chow*
Rearing
diet
HC-SF
HC-UF
LC-SF
LC-UF
kcal/100 g
%cal
%cal
%cal
329
28
62
10
381
21
38
41
377
21
39
40
377
21
39
40
377
20
40
40
377
20
40
40
%cal
%cal
%cai
mg/kcal
3.7
4.1
2.2
0.03
16.6
16.5
6.9
1.7
29.7
5.7
4.6
1.7
6.9
11.0
22.1
1.7
33.1
3.5
3.4
<0.01
5.9
10.2
23.8
<0.01
'Monkey Chow is an extruded, grain-based ration prepared by the Ralston Purina Company, St. Louis, Missouri.
fHC = high cholesterol; LC = low cholesterol; SF = saturated fat; UF = unsaturated fat.
^Analysis of mixed diet by gas-liquid chromatography.
SELECTIVE BREEDING AND DIETARY RESPONSE
Purina Company. The energy, nutrient, fatty acid, and cholesterol compositions of the six diets are shown in Table 2.
Dietary Test
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Each juvenile baboon was assigned to one of four
groups so that each group included animals whose ages
were within 6 months of one another. Each group was
placed in a gang cage and was fed the same fat- and
cholesterol-enriched diet on which It had been reared from
weaning. The first set of observations was made while the
animals were consuming this diet. Thereafter, each group
was fed for 6 weeks on each of the four diets in the following sequence: high cholesterol, saturated fat (HC-SF); low
cholesterol, saturated fat (LC-SF); low cholesterol, unsaturated fat (LC-UF); and high cholesterol, unsaturated fat
(HC-UF). At the end of the last of the four special diet
periods, each group was fed chow and the observations
were repeated. The design did not permit adjustment for
carryover effects of previous diets, but work with similar
diets has indicated that these effects are small.9
Blood Collection Procedures
At 3-week intervals during the preweaning period, every
3 months during the juvenile period, and at 4, 5, and 6
weeks during each of the six diet periods, blood was drawn
under ketamine immobilization (Vetelar, Parke Davis and
Company, Detroit, Michigan) after an overnight fast. Serum was separated by centrifugation.
Llpoproteln Cholesterol Analyses
HDL cholesterol was measured in the supernatant after
precipitation of serum VLDL and LDL by the dextran sulfate-CaCI2 procedure.10 VLDL + LDL cholesterol was calculated as the difference between serum and HDL cholesterol concentrations. We measured cholesterol by an
enzymatic method using the ABA Bichromatic Analyzer
(Abbott Laboratories, South Pasadena, California). The
cholesterol measurements met the criteria of the Lipid
Standardization Program of the Centers for Disease Control, Atlanta, Georgia. The precipitation procedure was
validated by comparison with the results of preparative
ultracentrifugation and of heparin-MnCI2 precipitation.8
300 -I
-
Statistical Methods
For the statistical analyses, the data were transformed
logarithmically to normalize the data and equalize the variances. We then computed the means of the three serum
and lipoprotein cholesterol measurements made after 4,5,
and 6 weeks on each diet. Data were analyzed by analysis
of variance and analysis of covariance.11 The linear model11 contained terms for the overall mean, group, line, sire
within line, sex, interactions of sex by line, and sex by sire
within line. All effects were fixed. We also used the VLDL
+ LDL cholesterol or HDL cholesterol concentrations during the LC-UF diet period as a covariate.
The simple effects11 were the effect of cholesterol in the
presence of (denoted by @) unsaturated fat, the effect of
cholesterol in the presence of saturated fat, the effect of
saturated fat in the presence of low cholesterol, and the
effect of saturated fat in the presence of high cholesterol.
These differences in group means are given as:
Cholesterol @ Unsaturated Fat = (HC-UF) - (LC-UF)
Cholesterol @ Saturated Fat = (HC-SF) - (LC-SF)
Fat @ Low Cholesterol = (LC-SF) - (LC-UF)
Fat @ High Cholesterol = (HC-SF) - (HC-UF)
These effects are differences on the logarithmic scale, but
are multiplicative on the original scale.
Results
Line Differences to 3 Years of Age
The effects of line on serum and lipoprotein cholesterol
concentrations were reported in detail for 69 progeny up to
1 year of age,5 and in a preliminary communication for 20
progeny up to 2 years of age.12 Figure 1 shows the mean
preweaning and quarterly serum cholesterol concentrations computed from log-transformed data of the 64 progeny in this experiment up to 148 weeks of age. The high
and low lines were consistently different from birth. The
age-associated rise in serum cholesterol, which began immediately after birth, continued through the breast-feeding
period into the juvenile period in both high and low lines.
Other experiments have shown that serum cholesterol
concentrations stabilize after about 2 years of age.13 Thus,
the animals were relatively stable with regard to serum
cholesterol levels at the time of the special dietary test.
Table 3 shows the mean lipoprotein cholesterol concen-
250
200
Table 3. Mean Lipoprotein Cholesterol Concentrations of High and Low Responding Lines of Baboons
on Rearing and Chow Diets
O
I50-
Une
I00-
o
X
o
35
McGill et al.
Diet
50
20
40
60
80
I00
120
140
160
AGE (weeks)
Figure 1. Serum cholesterol concentrations of high (o) and low
(•) lines of progeny from birth to 148 weeks of age. Vertical lines
represent 95% confidence intervals.
Lipoprotein
High
Rearing VLDL + LDL 87(75-102)
112(105-121)
HDL
VLDL + LDL 35(30-40)
Chow
80(74-87)
HDL
Low
Ratio
high/low
60 (52-70)
96(89-102)
27(24-31)
61 (57-66)
1.45*
1.18*
1.26f
1.31*
Concentrations are given in mg/dl; 95% confidence intervals
are in parentheses.
•Ratio of high line to low line different from 1.0, p < 0.01.
fRatio of high line to low line different from 1.0, p < 0.05.
36
ARTERIOSCLEROSIS V O L 8, No 1, JANUARY/FEBRUARY 1988
trations of the two lines just before beginning the special
diet sequence, but while the animals were still consuming
the high cholesterol, saturated fat diet on which they were
reared. Table 3 also includes the lipoprotein cholesterol
concentrations during the chow diet period, which occurred after the four special diet periods. On both diets,
both VLDL + LDL and HDL cholesterol concentrations
were significantly greater in the high line progeny as compared to low line progeny. However, the line difference in
VLDL + LDL cholesterol was much greater on the rearing
diet than on the chow diet, while the line difference in HDL
cholesterol was essentially the same on the chow diet and
the rearing diet. Thus, selective breeding separated the
progeny lines on the basis of both VLDL + LDL and HDL
cholesterol, but it was particularly effective in separating
them on the basis of their VLDL + LDL cholesterol concentrations on the cholesterol and saturated fat rearing
diet.
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Effects of Sire and Sex
There were no significant effects of sire on either VLDL
+ LDL or HDL cholesterol within the high line, but there
Table 4. Mean Lipoprotein Cholesterol Concentrations of High and Low Responding Lines of Baboons
by Diet
Une
Lipoprotein
VLDL + LDL
HDL
Diet
High
HC-SF 103(92-117)
HC-UF 88(76-102)
81 (73-91)
LC-SF
LC-UF 39(35-45)
HC-SF 136(127-145)
HC-UF 85 (78-93)
LC-SF 115(105-125)
79 (73-85)
LC-UF
Low
74 (66-83)
51 (44-59)
64 (57-72)
31 (27-35)
120(112-128)
76 (70-83)
98(91-107)
63(58-68)
Ratio
high/low
1.40*
1.72*
1.27*
1.28*
1.13f
1.11t
1.17f
1.25*
Concentrations are given in mg/dl; 95% confidence Intervals
are in parentheses.
•Ratio of high line to low line different from 1.0, p < 0.01.
fRatlo of high line to low line different from 1.0, p < 0.05.
HC = high cholesterol; LC = low cholesterol; SF = saturated
fat; UF = unsaturated fat.
was a significant sire effect on HDL within the low line. The
progeny of one of the three low-line sires (X98) had considerably lower HDL cholesterol on all diets; the progeny of
the other two (X956 and X84) had nearly identical HDL
cholesterol levels (results not shown). There were no significant effects of sex, or sex by line interactions, on either
VLDL + LDL or HDL cholesterol concentration.
Differential Effects of Dietary Cholesterol and
Saturated Fat by Line
Table 4 shows the mean VLDL + LDL and HDL cholesterol concentrations while the animals were consuming
each of the four special diets. The HC-SF diet, which was
prepared with coconut oil, elevated both VLDL + LDL and
HDL cholesterol to higher levels in both the high and low
lines than did the long-term rearing diet, which was prepared with lard (Table 3), but the lines differed by about the
same ratio. The lines were significantly different in both
VLDL + LDL and HDL cholesterol concentrations while
animals were consuming each of the four diet combinations. However, the line difference in VLDL + LDL cholesterol was substantially greater while they were consuming
a high cholesterol diet, regardless of type of fat.
Table 5 shows the multiplicative effects of dietary fat and
cholesterol on each lipoprotein concentration in each line.
Both components significantly (p < 0.05) affected both
VLDL + LDL and HDL cholesterol concentrations (the
95% confidence intervals did not overlap 1.00).
Within each line, VLDL + LDL cholesterol had a greater
response to one dietary component when the other component was at its lower level; that is, the response to dietary cholesterol was greater when animals were fed unsaturated fat, while the response to saturated fat was
greater when tested in the presence of low cholesterol.
The lines differed in their response to cholesterol when
animals were fed unsaturated fat (high line, 2.23; low
line, 1.67; p < 0.01) and in their response to fat when fed
high cholesterol (high line, 1.18; low line, 1.45; p < 0.05)
(Table 5).
HDL cholesterol responded more to type of fat than to
dietary cholesterol. In contrast to the foregoing observation about VLDL + LDL response, the HDL response to
one component depended little on the level of the other,
Table 5. Effects of Dietary Cholesterol and Fat on Lipoprotein Cholesterol Levels In High and
Low Responding Lines of Baboons
Une
Lipoprotein
Effect
High
Low
VLDL + LDL
Cholesterol @ saturated fat
Cholesterol @ unsaturated fat
Fat @ high cholesterol
Fat @ low cholesterol
Cholesterol @ saturated fat
Cholesterol @ unsaturated fat
Fat @ high cholesterol
Fat @ low cholesterol
1.27(1.15-1.40)
2.23(1.98-2.51)
1.18(1.04-1.33)
2.06(1.90-2.23)
1.19(1.11-1.27)
1.08(1.01-1.16)
1.60 (1.48-1.73)
1.46 (1.37-1.55)
1.15(1.05-1.27)
1.67(1.49-1.87)
1.45(1.28-1.63)
2.09 (1.93-2.25)
1.22(1.14-1.30)
1.21 (1.14-1.30)
1.57(1.46-1.69)
1.56(1.47-1.66)
HDL
The 95% confidence intervals are given in parentheses.
'Ratio of high line to low line different from 1.0, p < 0.01.
tRatlo of high line to low line different from 1.0, p < 0.05.
Ratio
high/low
1.10
1.34*
0.811
0.99
0.97
0.89t
1.02
0.93
SELECTIVE BREEDING AND DIETARY RESPONSE
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
that is, the response to type of fat was similar regardless of
cholesterol level, and response to cholesterol was similar
regardless of type of fat. The lines responded slightly
differently to dietary cholesterol when animals were fed
unsaturated fat (high line, 1.08; low line, 1.21; p < 0.05)
(Table 5).
Regardless of which special diet was fed, the two lines
were different in both VLDL + LDL and HDL cholesterol
concentrations (Table 4). To separate a differential response to diet from a uniform difference between lines, we
adjusted each lipoprotein cholesterol concentration on
each diet for the concentration attained while animals were
fed the LC-UF diet (Table 6). The adjustment should have
removed any uniform separation between the lines so that
a differential response could be detected. The adjusted
VLDL + LDL concentrations in the two lines were similar
when animals consumed the LC-SF diet, but the concentrations were significantly different when the animals consumed the two high cholesterol diets. The ratios of high
line to low line were 1.23 (p < 0.05) on the HC-SF diet and
1.43 (p < 0.01) on the HC-UF diet. The adjusted HDL
concentrations were similar in the two lines regardless of
the diet they were consuming.
Table 6. Lipoprotein Cholesterol Concentration
after Adjusting for Corresponding Concentration during the Low Cholesterol, Unsaturated Fat Diet Period
Line
Lipoprotein
Diet
High
VLDL + LDL HC-SF 99(88-110)
HC-UF 82 (73-92)
LC-SF
76 (70-82)
LC-UF 36
HDL
HC-SF 130(122-138)
HC-UF 79 (74-85)
LC-SF 106(99-113)
LC-UF
71
Low
. Ratio
high/low
80 (72-89)
57(51-65)
72 (66-77)
36
127(119-135)
83(78-89)
108(101-115)
71
1.23t
1.43*
1.06
1.00
1.02
0.95
0.99
1.00
Concentrations are given in mg/dl; 95% confidence intervals
are in parentheses.
•Ratio of high line to low line different from 1.0, p < 0.01.
tRatio of high line to low line different from 1.0, p < 0.05.
37
McGill et al.
Table 7 shows the multiplicative effects of the dietary
components after adjusting for lipoprotein concentration
while the animals were fed the LC-UF diet. Within each
line, each dietary component had a significant (p < 0.05)
effect regardless of the level of the other factor (the 95%
confidence intervals did not overlap 1.00). The pattern of
response was similar to that observed in Table 5. HDL
cholesterol responded more to type of fat than to dietary
cholesterol, while the response of VLDL + LDL cholesterol to one of the dietary components depended on the level
of the other component. The two lines had similar responses for HDL cholesterol while dietary cholesterol significantly elevated VLDL + LDL cholesterol more in the
high line than in the low line. The lines were best separated
by their VLDL + LDL response to dietary cholesterol when
tested in the presence of unsaturated fat.
Discussion
Genetic Control of Serum and Lipoprotein
Cholesterol Concentrations
These results confirm and extend the conclusions derived from genetic analyses of these progeny at one year
of age.5 At that time, there were significant differences
between the lines in serum and HDL cholesterol concentrations, but not in VLDL + LDL cholesterol concentration.
The line difference in serum cholesterol concentration continued to 3 years of age (Figure 1); and, between 3 and 4
years of age, there were significant differences between
the two lines in both VLDL + LDL and HDL cholesterol
levels on all diets, including chow (Tables 3 and 4). These
consistent differences, no matter what diet the animals
were consuming, indicate that the genetic control is manifested regardless of the diet. Dietary manipulation results
in a greater expression of genetic differences in VLDL +
LDL cholesterol levels and facilitates detection, but does
not result in a greater expression of genetic differences in
HDL cholesterol levels. The line differences in both VLDL
+ LDL and HDL cholesterol levels are consistent with the
results of genetic analyses of three other groups of baboons which included both selectively bred and randomly
bred animals.6'7
Table 7. Effects of Dietary Cholesterol and Fat on Lipoprotein Cholesterol Concentration after
Adjusting for Corresponding Concentration during the Low Cholesterol, Unsaturated Fat Diet
Period
Line
Lipoprotein
VLDL + LDL
HDL
Cholesterol @ saturated fat
Cholesterol @ unsaturated fat
Fat @ high cholesterol
Fat @ low cholesterol
Cholesterol @ saturated fat
Cholesterol @ unsaturated fat
Fat @ high cholesterol
Fat @ low cholesterol
Ratio
High
Low
high/low
1.30(1.18-1.43)
2.29 (2.03-2.57)
1.20(1.06-1.37)
2.12(1.96-2.29)
1.22(1.14-1.31)
1.11 (1.03-1.19)
1.64(1.51-1.78)
1.49(1.40-1.59)
1.12(1.01-1.23)
1.60(1.42-1.80)
1.39(1.23-1.58)
2.00(1.85-2.16)
1.18(1.10-1.26)
1.17(1.09-1.26)
1.52(1.40-1.65)
1.51 (1.42-1.62)
1.16f
Dietary component
The 95% confidence intervals are given in parentheses.
'Ratio of high line to low line different from 1.0, p < 0.01.
fRatio of high line to low line different from 1.0, p < 0.05.
1.43*
0.87
1.06
1.04
0.95
1.08
0.99
38
ARTERIOSCLEROSIS VOL 8, No 1, JANUARY/FEBRUARY 1988
Dietary Components and Selective Breeding
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Our major question in this experiment with the high and
low line progeny was whether selective breeding, which
was based on the serum cholesterol response to a combined cholesterol and fat challenge, had produced the diverging lines through selection for response to dietary cholesterol, response to type of dietary fat, or response to
both; and whether the effect was on VLDL + LDL cholesterol, HDL cholesterol, or both. The question is answered
in Table 5, which shows that both VLDL + LDL and HDL
cholesterol responded to saturated fat (as compared to
unsaturated fat) in both high and low lines to about the
same degree, whereas VLDL + LDL cholesterol responded to dietary cholesterol in the high line to a much greater
degree than in the low line. Expressed in another way, the
high line is more susceptible to elevation of VLDL + LDL
cholesterol by dietary cholesterol than is the low line, but
the high line is not more susceptible to elevation of either
VLDL + LDL or HDL cholesterol by saturated fat. The
greater susceptibility of high line animals to elevation of
VLDL + LDL cholesterol by dietary cholesterol is seen
most dramatically in the effect of dietary cholesterol combined with unsaturated fat (Table 5,2.23 versus 1.67; ratio,
1.34, p < 0.01; or in Table 7, 2.29 versus 1.60; ratio, 1.43,
p < 0.01). In contrast, the effect of dietary cholesterol on
HDL is less in the high line than in the low line.
Selective breeding may have produced divergence between the two lines by several different mechanisms that
regulate serum lipoprotein cholesterol levels. The mechanism affected most by selective breeding is that controlling
responsiveness of VLDL + LDL cholesterol to dietary cholesterol. The other mechanisms affected by selective
breeding regulate maintenance of both VLDL + LDL and
HDL cholesterol levels, but not their responsiveness to
saturated fat. Another way of expressing this interpretation
is that these baboons have substantial genetic variability in
their response to dietary cholesterol, but less genetic variability in their responses to saturated fat.
Genetic Basis of Variability
Segregation analyses of lipoprotein cholesterol levels of
about 650 pedigreed baboons have indicated the presence of a major gene regulating HDL cholesterol levels on
the chow diet, and a major gene regulating VLDL + LDL
cholesterol levels on the combined cholesterol and saturated fat diet.14 The lipemic responses to dietary cholesterol and saturated fat separately were not available. It is
premature to speculate on the relationship of the major
genes identified by segregation analyses to the results of
this experiment.
molecular or cellular mechanisms responsible for this difference.
Comparison with Observations on Humans
The responses of humans to dietary cholesterol also are
highly variable.18'17 This variability may be responsible for
the conflicting results often obtained from experiments
with dietary cholesterol in humans, and also may be partly
responsible for the controversy regarding the relative importance of dietary cholesterol and saturated fat in human
hyperlipidemia, atherosclerosis, and arterlosclerotic heart
disease.18
Katan and Beynen19 have compared a number of characteristics of high responding (to dietary cholesterol) humans with those of low responding humans.16 Responsiveness to dietary cholesterol was correlated negatively
with habitual cholesterol consumption (r = -0.62), body
mass index (r = -0.40), and endogenous cholesterol
synthesis (r = -0.40). Responsiveness was correlated
positively with HDL2 cholesterol levels (r = 0.41) and with
serum cholesterol level on the high cholesterol diet (r =
0.31). With multiple regression analysis, only habitual cholesterol intake and HDL2 cholesterol levels contributed significantly to explaining variation in response. The baboons
in our experiment were reared on identical diets, and therefore there should have been little differences in lifetime
cholesterol intakes. We did not measure the HDL subclasses and were unable to test for the correlation of HDLj
with responsiveness. We cannot find similar data regarding variability of cholesterolemic responses of humans to
saturated fat.
Conclusion
Positive assortative mating based on serum cholesterol
response to a combined saturated fat and cholesterol enriched diet produced lines of progeny that differed in both
VLDL + LDL and HDL cholesterol concentrations on chow
and on diets with two types of fat and two levels of cholesterol. The high line had a greater responsiveness to dietary
cholesterol than the low line, and both lines had a similar
responsiveness to saturated fat. The responsiveness to
dietary cholesterol was due almost entirely to elevation of
VLDL -I- LDL cholesterol concentrations. The results are
consistent with observations on humans showing a wide
variability in cholesterolemic responsiveness to dietary
cholesterol.
References
Metabolic Basis of Responses
Kushwaha et al. 15 compared the metabolism of VLDL
and LDL apolipoprotein B (apo B) among baboons with
high and low VLDL + LDL cholesterol produced by selective breeding. The rate of conversion of VLDL apo B to LDL
apo B was higher and the disappearance of VLDL apo B by
other routes was lower in the high LDL animals than in the
low LDL animals. As yet, however, we do not know the
1. Eggen DA. Cholesterol metabolism in rhesus monkey, squirrel monkey, and baboon. J Upid Res 1974;15:139-145
2. Srlnlvasan SR, Radhakrlshnamurthy B, Smith CC, Wolf
RH, Berenson GS. Serum lipid and lipoprotein responses of
six nonhuman primate species to dietary changes in cholesterol levels. J Nutr 1976;106:1757-1767
3. Clarkson TB, Lofland HB Jr, Bullock BC, Goodman HO.
Genetic control of plasma cholesterol. Studies on squirrel
monkeys. Arch Pathol 1971 £2:37-45
4. Clarkson TB, Kaplan JR, Adams MR. The role of individual
SELECTIVE BREEDING AND DIETARY RESPONSE
5.
6.
7.
8.
9.
10.
11.
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
12.
differences in lipoproteln, artery wall, gender, and behavioral
responses in the development of atherosclerosis. Ann NY
Acad Sci 1985;45428-45
Flow BL, Cartwrlght TC, Kuehl TJ, et al. Genetic effects on
serum cholesterol concentrations in baboons. J Hered 1981;
72:97-103
Flow BL, Mott GE, Kelley J L Genetic mediation of lipoprotein cholesterol and apoprotein concentrations in the baboon
(Papto sp). Atherosclerosis 1982;43:83-94
Kammerer CM, Mott QE, Carey KD, McGIII HC Jr. Effect of
selection for serum cholesterol concentrations on serum lipid
concentrations and body weight in baboons. Am J Med Genet
1984;19:333-345
McGIII HC Jr, McMahan CA, Kruski AW, Kelley JL, Mott
QE. Responses of serum lipoproteins to dietary cholesterol
and type of fat in the baboon. Arteriosclerosis 1981; 1:337344
McGIII HC Jr, McMahan CA, Wene JD. Unresolved problems in the diet-heart issue. Arteriosclerosis 1981 ;1:164-176
Sakagaml T, Zllveramrt DB. Separation of dog serum lipoproteins by ultracentrifugation, dextran sulfate precipitation,
and paper electrophoresis. J Lipid Res 19612:271-277
Cochran WG, Cox GM. Experimental designs. Wiley: New
York, 1957:127-139
McGIII HC Jr, Mott GE, Kuehl TJ. Effects of diet and heredity
on serum cholesterol and triglyceride concentrations and on
cholesterol absorption of Infant and juvenile baboons (Papio
Index Terms:
lipoproteins
selective breeding
13.
14.
15.
16.
17.
18.
19.
McGill et al.
39
cynocephalus). In: Lauer RM, Shekelle RB, eds. Childhood
prevention of atherosclerosis and hypertension. New York:
Raven Press, 1980:113-120
Wlssler RW, McGIII HC Jr. Conference on Blood Upids In
Children: Optimal levels for early prevention of coronary artery disease. Prev Med 1983:12:868-902
MacCluer JW, Kammerer CA, VandeBerg JL, Cheng M-L,
Mott GE, McGIII HC Jr. Detecting genetic effects on lipoprotein phenotypes in baboons: a review of methods and preliminary findings. GenetJca 1987;73:159-168
Kushwaha RS, Barnwell GM, Carey KD, McGIII HC Jr.
Metabolism of apoprotein B in selectively bred baboons with
low and high levels of low density lipoproteins. J Lipid Res
1986:27:497-507
Katan MB, Beynen AC, Do Vrlet JHM, Nobela A. Existence
of consistent hypo- and hyperresponders to dietary cholesterol in man. Am J Epidemiol 1986:123221-234
Zannl EE, Zannla VI, Blum CB, Herbert PN, Breslow J L
Effect of egg cholesterol and dietary fats on plasma lipids,
lipoproteins, and apoproteins of normal women consuming
natural diets. J Upid Res 1987:28:518-527
McGIII HC Jr. The relationship of dietary cholesterol to serum
cholesterol concentration and to atherosclerosis in man. Am J
Clin Nutr 1979:32:2664-2702
Katan MB, Beynen AC. Characteristics of human hypo- and
hyperresponders to dietary cholesterol. Am J Epidemiol
1987;125:387-399
baboon (Papio sp.)
• dietary cholesterol
• dietary fat
Downloaded from http://atvb.ahajournals.org/ by guest on June 16, 2017
Effects of selective breeding on the cholesterolemic responses to dietary saturated fat and
cholesterol in baboons.
H C McGill, Jr, C A McMahan, G E Mott, Y N Marinez and T J Kuehl
Arterioscler Thromb Vasc Biol. 1988;8:33-39
doi: 10.1161/01.ATV.8.1.33
Arteriosclerosis, Thrombosis, and Vascular Biology is published by the American Heart Association, 7272 Greenville
Avenue, Dallas, TX 75231
Copyright © 1988 American Heart Association, Inc. All rights reserved.
Print ISSN: 1079-5642. Online ISSN: 1524-4636
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://atvb.ahajournals.org/content/8/1/33
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Arteriosclerosis, Thrombosis, and Vascular Biology can be obtained via RightsLink, a service of the Copyright
Clearance Center, not the Editorial Office. Once the online version of the published article for which permission
is being requested is located, click Request Permissions in the middle column of the Web page under Services.
Further information about this process is available in the Permissions and Rights Question and Answerdocument.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Arteriosclerosis, Thrombosis, and Vascular Biology is online
at:
http://atvb.ahajournals.org//subscriptions/