Impact of buttermilk consumption on plasma
lipids and surrogate markers of cholesterol
homeostasis in men and women
V. Conwaya,b, P. Coutureb,c, C. Richardb, S.F. Gauthiera,b,
Y. Pouliota,b, B. Lamarcheb,*
aSTELA Dairy Research Center, Laval University, Quebec, Canada
bInstitute of Nutrition and Functional Foods, Laval University, 2440, Boulevard Hochelaga, Quebec
(QC), Canada G1V 0A6
cLipid Research Center, CHUQ Research Center, Quebec, Canada
Received 27 August 2012; received in revised form 5 March 2013; accepted 7 March 2013
Available online 17 June 2013
Milk polar lipids;
globule membrane (MFGM) found in buttermilk. While studies in animal models suggest that
dietary SL may have cholesterol-lowering properties, data in human are lacking. The aim of
this study was to investigate the impact of buttermilk consumption on plasma lipids and sur-
rogate markers of cholesterol (C) homeostasis in humans.
Methods and results: Menandwomen(nZ34)withserumLDL-C<5.0mmol/Latscreening(mean
LDL-C Z 3.8 mmol/L) were recruited in this double-blinded randomized crossover placebo
controlled study. Their diets were supplemented with 45 g/d of buttermilk and with 45 g/d of a
macro/micronutrient matched placebo (4 weeks each in random order). Serum lipid concentra-
ing mixed models for repeated measures. Consumption of buttermilk led to reduction in serum
cholesterol (?3.1%, P Z 0.019), LDL-C (?3.1%, P Z 0.057) and triacylglycerol (?10.7%,
P Z 0.007). Buttermilk consumption increased plasma lathosterol concentrations (þ12.1%,
(P Z 0.002) were the only significant predictor of the LDL-C response to buttermilk consumption.
Conclusion: Buttermilk consumption may be associated with reduced cholesterol concentrations
in men and women, primarily through inhibition of intestinal absorption of cholesterol.
Registration number: This trial is registered at clinicaltrials.gov as NCT01248026.
ª 2013 Elsevier B.V. All rights reserved.
Background and aims: Sphingolipids (SL) are important components of the milk fat
Abbreviations: ApoB, apoliporotein-B; BMI, body mass index; C, cholesterol; CHD, coronary heart disease; CRP, C-reactive protein; CDV,
cardiovascular disease; FA, fatty acids; FFQ, food-frequency questionnaire; FSH, follicle-stimulating hormone; MFGM, milk fat globule
membrane; PCSK9, protein convertase subtilisin kexin-9; SFA, saturated fatty acids; SL, sphingolipids; SM, sphingomyelin; TG,
* Corresponding author. Tel.: þ1 418 656 2131x4355; fax: þ1 418 656 5877.
E-mail address: firstname.lastname@example.org (B. Lamarche).
0939-4753/$ - see front matter ª 2013 Elsevier B.V. All rights reserved.
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/nmcd
Nutrition, Metabolism & Cardiovascular Diseases (2013) 23, 1255e1262
The role of low-density lipoprotein cholesterol (LDL-C) in
the pathogenesis of coronary heart disease (CHD) and the
clinical benefit of lowering LDL-C concentrations have both
been well established [1,2]. Cholesterol homeostasis and
hence plasma LDL-C concentrations are maintained by a
fine-tuned balance between intestinal cholesterol absorp-
tion, endogenous cholesterol synthesis and cholesterol
clearance . While the liver has long been recognized as a
key organ regulating plasma cholesterol concentrations,
the role played by the intestine in whole body cholesterol
homeostasis is being increasingly recognized. There is now
convincing evidence that diet-induced reduction in intes-
tinal cholesterol absorption has a significant impact on
plasma cholesterol and LDL-C concentrations. One good
example of this is the introduction of phytosterols into
regularly consumed foods, which consumption has been
reduction in LDL-C concentrations . This is important
because in low risk patients, most clinical guidelines
advocate the use of non-pharmacological approaches as the
first mean to lowering LDL-C concentrations .
Cholesterol is a highly hydrophobic molecule and for that
reason, its absorption is almost entirely dependent on its
in vitro studies from our laboratory have shown that butter-
milk, the by-product of butter manufacturing resulting from
. This phenomenon is likely due to the presence of polar
lipids from the milk fat globule membrane (MFGM). Frag-
ments of the MFGM end up in buttermilk along with most of
the water-soluble cream components such as lactose, min-
erals and milk proteins. Most of the research so far has
focused on phospholipids purified from MFGM, thereby over-
looking the complex and entire MFGM mixture of bioactive
proteins and polar lipids found in buttermilk. To the best of
our knowledge, no study has yet documented the impact of
whole buttermilk on plasma cholesterol concentrations in
humans, with considerations for potential underlying mech-
anism. The objective of this study was to investigate the
impact of buttermilk consumption on LDL-C concentration as
well as on surrogate markers of cholesterol homeostasis in
men and women with serum LDL-C <5.0 mmol/L. We have
used plasma phytosterols concentrations as a surrogate of
intestinal cholesterol absorption, plasma lathosterol con-
centrations as a surrogate of endogenous cholesterol syn-
thesis, and plasma protein convertase subtilisin kenin-9
(PCSK9) concentrations as a surrogate of LDL clearance
[8e10]. We hypothesized that buttermilk consumption re-
duces LDL-C concentrations, and that this occurs primarily
through inhibited intestinal cholesterol absorption.
Men and women were recruited in the Quebec City area via
newspaper, radio and electronic news letters. Recruitment
took place at the Institute of Nutraceuticals and Functional
Foods (INAF) between January 2011 and April 2011. To be
included in the study, participants had to be 18e65 years of
age with a body mass index (BMI) ?35 kg/m2. Subjects had
to have serum LDL-C concentrations below 5.0 mmol/L with
a 10-years calculated Framingham risk below 10% . Only
participants with a stable body weight for at least 6 months
prior to the study were eligible. Subjects were excluded if
they had a previous history of CHD, type 2 diabetes,
monogenic dyslipidemia, were
hyperlipidemia or hypertension or had endocrine disorders.
Among pre-menopausal women, only those with a regular
menstrual cycle for the last 3 months (25e35 days) were
eligible. All post-menopausal women were included, irre-
spective of their hormone supplementation status, as long
as it remained stable throughout the study. Individuals with
extreme nutritional habits such as vegetarianism, with
alcohol consumption >2 drinks/day and elite athletes were
not eligible. Smokers and women using contraceptive
agents were not excluded from the study. Follicular-
stimulating hormone (FSH) measurements were conducted
ifneeded to confirmthe
(FSH < 20 IU/L or FSH > 25 IU/L). Participants provided
their informed consent for the study. The study protocol
was approved by The Clinical Research Ethics Committees
of Laval University.
The study was carried out as a double-blind, randomized,
placebo controlled crossover study according to which
participants were subjected to 2 consecutive treatments of
4 weeks each, in a random order. For the total duration of
the study (8 weeks), participants were requested to main-
tain their usual diet, medication, weight, alcohol con-
sumption, and smoking and physical activity habits, except
for the 3 days preceding blood sampling, during which they
were asked to remain sedentary. Vitamins and natural
health product supplementation were strictly forbidden
during the entire experimental period. Consumption of tea
and coffee was allowed with a limit of 2 cups/day as long it
remained constant throughout the study. Participants with
any deviation to these recommendations were excluded
Buttermilk and placebo formulations
The 2 test formulations, artificially chocolate flavored
buttermilk and placebo, were designed to provide the
equivalent of 2 servings of low fat milk (45 g/day of skim
milk solids). The buttermilk powder was purchased from
Westland Milk Products (Hokitika, New Zealand) and was
fully characterized and formulated in ready-to-use pouches
(Table 1). Each pouch contained 22.5 g of formulated
buttermilk or placebo that subjects had to mix with 250 mL
of water using a shaker provided by the study center.
Sucralose was used to improve taste and acceptability. The
placebo was formulated using dairy ingredients (calcium
caseinate e DVM International, Veghel, Netherlands; whey
protein isolate e Davisco Food International Inc., Eden
Prairie, MN; whey protein permeate e Agropur, Longueuil,
Canada; butter powder e Kerry Inc., Beloit, WI) in order to
1256 V. Conway et al.
match the macro/micronutrient composition of buttermilk
with the exception of its MFGM components. This allowed
us to specifically study the impact of MFGM minor proteins
and phospholipids on study outcomes.
Participants were asked to consume 2 pouches every
day, just before breakfast and dinner, for a total con-
sumption of 45 g/day of buttermilk or placebo, which
corresponded to 5%e10% of their daily energy intake based
on a 2500 kcal/day diet. Participants received specific
instruction on how to introduce the test formulations in
their diet by substituting specific food items. Subjects
were asked to restrain and to maintain their intake of
dairy product to a maximum of 2 servings/day. Usual food
intake was assessed using a validated in-house food-fre-
quency questionnaire (FFQ)  along with a question-
naire specific to dairy foods on three occasions during the
study: 1 e at study entry, 2 e after the first treatment and
3 e after the second treatment. The first evaluation was
used as a teaching tool on how to integrate the supple-
ments into the subject’s nutritional habits. The 2nd and
3rd evaluations were used to generate nutritional profiles
during each treatment and to adjust recommendations,
when necessary, for any variations that may have occurred
between treatments. Data from these questionnaires were
analyzed using the Nutrient Data System software based
ona mixof Canadianand
Compliance was measured by counting the pouches
returned by subjects to the research staff. A check list
provided to all participants was also used for tracking un-
used products. Compliance ?85% was defined as acceptable
to include participants’ data into the study analyses.
Subjects had to notify the physician in charge of the clinical
aspects of the study before initiating any new medication.
Risk factor assessment
Blood samples were taken at screening as well as on two
consecutive days at week 4 of each treatment. For each
subject and for all study outcomes described below, the
average of the two post-treatment values was used. Total
cholesterol (total-C), triacylglycerol (TG), and HDL-C con-
centrations in serum were measured using commercial re-
Diagnostics, Mannheim, Germany). Serum LDL-C concen-
trations were obtained by calculation using the Friedewald
equation. Plasma apolipoprotein-B (apoB-100 and apoB-48)
concentrations were measured with ELISA kits (Alerchek
Inc., Springvale, ME). Plasma concentrations of C-reactive
protein were measured using a commercial high sensitivity
ELISA kit (BioCheck Inc., Foster City, CA).
Plasma concentrations of lathosterol, b-sitosterol and
campesterol after each treatment were quantified using gas
chromatography (GC) as described previously . Since
non-cholesterol sterols are transported in plasma by lipo-
proteins, their concentrations have also been expressed
(102mmol/mmol C of the same GC run) to correct for the
plasma PCSK9 levels were measured by a commercially
available ELISA kit (CircuLex Co., Ltd, Ina, Nagano, Japan).
Anthropometric measures including waist and hip circum-
ferences were taken at the beginning and at the end of
each test diet according to standardized procedures .
Composition of the ready-to-use buttermilk and placebo pouches (22.5 g).a
Total proteins, g
Total fat, g
Total phospholipids, mg
aLactose, total proteins, total fat, ashes and water values are presented as means and standard deviations (n Z 3).
Cholesterol-lowering activity of buttermilk1257
The study has been designed to have adequate power to
test the main hypothesis that consumption of 45 g/day of
buttermilk leads to significant reduction in plasma LDL-C
concentrations. Sample size estimation was performed to
allow the detection of a minimum of 5% reduction in plasma
LDL-C concentrations. An SD of 0.39 mmol/L for the LDL-C
change in response to dietary modifications has been used
for sample size calculations . At a mean baseline LDL-C
concentration of 4.0 mmol/L, the required number of
participants completing the study for optimal statistical
power was n Z 32 (alpha Z 0.05 two-tailed, power Z 90%).
The anticipated dropout rate was 10%.
All data were analyzed for statistical differences using
the PROC MIXED procedure for repeated measures in SAS
9.2 (SAS Institute Inc., Cary, NC). Normal distribution and
homogeneity of variance were checked before further
analysis. Variables with a skewed distribution were log10
transformed. The impact of buttermilk on serum lipid
concentrationsand surrogatemarkersof cholesterol
homeostasis was adjusted for sex through multivariate
modeling. The impact of treatment sequence was investi-
gated using appropriate interaction terms in the MIXED
model; no significant interaction was noted for any of the
outcome measures. The impact of baseline values (at
screening) on the response to buttermilk was also investi-
gated using individual and interaction terms in the MIXED
model. When interaction terms were significant, subgroups
were created using the median values for the selected
variable to present the interaction graphically. Univariate
correlation analysis and multivariate stepwise regression
models were used to investigate associations among
outcome measures. Data are presented as means ? SDs and
as percentages of change compared to the placebo unless
stated otherwise. P values < 0.05 were considered statis-
Of the 40 subjects who met the eligibility criteria and who
were recruited to start the study, 6 individuals did not
complete both treatments and were excluded from further
analysis (Fig. 1). Adherence to the test diets was very good,
with a self-reported compliance above 97%. None of the
subjects were excluded based on their lack of compliance.
The characteristics at screening of subject included in the
analyses are presented in Table 2. Among the 34 subjects
(15 men and 19 women), body weight remained stable
throughout the 8 week study period and no change was
observed in hip and waist circumference measures.
Plasma lipids and lipoproteins
As shown in Table 3, buttermilk supplementation signifi-
cantly reduced serum levels of total-C (?3.1%, P Z 0.019)
and TG (?10.7%, P Z 0.007) compared with placebo. The
reduction in serum LDL-C concentrations (?3.1%, P Z 0.057)
was borderline significant. However, there was a significant
interaction between LDL-C values at screening and treat-
ment on the LDL-C response (P for interaction Z 0.028).
Figure 2 illustrates the individual changes (%) in LDL-C con-
centrations (buttermilk vs. placebo) according to LDL-C
values at screening, < or > the median 3.7 mmol/L.
Figure 3 illustrates how LDL-C status at screening modified
the LDL-C response to buttermilk. Serum LDL-C concentra-
tions were significantly reduced after buttermilk consump-
tion among subjects with screening LDL-C > 3.7 mmol/L
(?5.6%, P Z 0.004) but not among those with screening LDL-
had no impact on serum HDL-C concentrations (Table 3).
There was no apparent interaction between BMI (or waist
circumference) and treatment on any of the study outcomes
Cholesterol homeostasis markers
As shown in Table 3, buttermilk consumption did not
significantly affect surrogate markers (mg/L) of intestinal
patients who chose to stop the study for medical issues, health
problems were not linked to the treatments.
Flow of patients through the clinical study. Of the 2
at screening (n Z 34).a
Baseline characteristics of the study participants
Subject characteristics MeanSD
Waist circumference, cm
Systolic blood pressure, mm Hg
Diastolic blood pressure, mm Hg
aBMI, body mass index; C, cholesterol.
1258V. Conway et al.
P Z 0.527 and b-sitosterol ?7.3%; P Z 0.096). Figure 3
illustrates how baseline LDL-C influenced the response
of plasma markers of cholesterol absorption to buttermilk
consumption. The change in plasma concentrations of
(?11.2%, P Z 0.004) were significantly reduced with
buttermilk consumption compared with placebo only in
subjects with high LDL-C values at screening. Buttermilk
increased plasma lathosterol concentrations (þ12.1%;
P Z 0.001), a surrogate marker of endogenous choles-
terol synthesis. Baseline LDL-C did not modify the plasma
lathosterol, campesterol and PCSK9 response to butter-
milk (P for interaction Z 0.783, 0.067 and 0.732
Changes (%) in plasma phytosterol, b-sitosterol and
PCSK9 concentrations with buttermilk vs. placebo were
positively correlated with concomitant change (%) in total-
C and LDL-C concentrations (Table 4). There was no such
correlation between the change (%) in plasma lathosterol
or campesterol concentrations and the change (%) in total-
C and LDL-C levels (Table 4). In multiple stepwise regres-
sion analysis, variations (%) in b-sitosterol concentrations
with buttermilk treatment were the only significant pre-
dictor of the reduction (%) in serum LDL-C (R2Z 27%,
P Z 0.002) and total-C concentrations (R2
P < 0.001, results not shown). There was no correlation
between the change in TG and the change in LDL-C with
buttermilk (not shown).
This study indicates that short-term buttermilk consump-
tion significantly reduces serum total-C and TG concentra-
tions in men and women. The 3.1% reduction in serum LDL-
C concentrations with buttermilk did not quite reach sta-
tistical significance in the entire study sample (P Z 0.057),
but was highly significant among subjects with higher LDL-C
concentration at screening (?5.6%, P Z 0.004). Plasma
lathosterol concentrations were increased after buttermilk
consumption but the magnitude of the change in this esti-
mate of endogenous cholesterol synthesis did not correlate
with the LDL-C response to buttermilk. Only changes (%) in
surrogate markers of intestinal cholesterol absorption
(plasma phytosterols and b-sitosterol) were correlated to
the total-C and LDL-C response to buttermilk in multivar-
To the best of our knowledge, this is the first study to
have investigated the impact of buttermilk consumption on
lipid levels in humans, with considerations for potential
underlying mechanism. Ohlsson et al. have investigated the
acute  and chronic  impact of sphingolipids (SL)-
enriched buttermilk supplementation on blood lipids in
healthy human subjects and have reported no significant
effect on fasting and postprandial plasma lipids concen-
trations. These two previous studies had smaller sample
sizes and involved subjects with relatively normal LDL-C
concentrations, thereby limiting the ability to observe sig-
nificant effects. Furthermore, Thompson et al.  have
studied the impact of cultured buttermilk consumption on
men and women (n Z 34).a
Effect of buttermilk consumption on serum lipid concentration and surrogate markers of cholesterol homeostasis in
Apolipoprotein B100, mg/dL
Apolipoprotein B48, mg/dL
C-reactive protein, mg/Le
Lathosterol/C, 102mmol/mmol Ce
Phytosterols/C, 102mmol/mmol Ce,f
Campesterol/C, 102mmol/mmol C
b-sitosterol/C, 102mmol/mmol Ce
Phytosterols, mg/Le, f
aC, cholesterol; PCSK9, protein convertase subtilisin kexin-9.
bValues are means ? SD.
cValues are expressed as percentage of change compared to placebo.
dp values from the main effect of diet in the MIXED model (SAS Institut. Cary. NC). P values were obtained using the PROC MIXED
procedures. The model was adjusted for sex.
eAnalysis were performed on log-transformed values.
fCampesterol þ b-sitosterol.
Cholesterol-lowering activity of buttermilk1259
lipid levels in a small group of healthy subjects and re-
ported no noticeable lipids-lowering effect. However,
cultured buttermilk represents the product of cow’s milk
fermentation and thus, does not possess the composition
specificities of fresh buttermilk (i.e. MFGM polar lipids).
Our data have shown that the cholesterol-lowering
property of buttermilk in humans appeared to be more
important among individuals with higher serum LDL-C con-
centrations. The impact of LDL-C status on diet-induced
LDL-C changes has been discussed previously . The LDL-
C response to buttermilk consumption (?0.24 mmol/L for
the “high” LDL-C group at screening) was close in magni-
tude to the cholesterol-lowering effectiveness of plant
sterol therapy, generally associated with an LDL-C reduc-
tion of 0.27e0.35 mmol/L . However, the effect of
buttermilk may differ from the effect of plant sterol ther-
apy by its ability to reduce serum TG levels as well. Indeed,
plant sterols inhibit intestinal cholesterol absorption with
no apparent effect on TG . In the present study, sup-
plementation with buttermilk reduced plasma concentra-
tions of apoB-48 by approximately 13% but this change was
not statistically significant. The absence of a correlation
between the reduction in plasma TG and in LDL-C with
buttermilk suggests that different mechanisms may be at
play. However, the TG-lowering effect associated with
buttermilk consumption and its underlying mechanisms
definitely deserves further research.
Although not statistically significant when considering
the whole study group, plasma phytosterols concentrations
were significantly reduced in subjects with high LDL-C
levels values at screening, i.e. in subjects showing a sig-
nificant LDL-C reduction after buttermilk consumption.
Multiple regression analyses have also shown that variations
in plasma b-sitosterol concentrations were the only multi-
variate predictor of the buttermilk-induced changes in
serum LDL-C and total-C levels. This suggests that butter-
milk consumption may reduce serum cholesterol primarily
through inhibition of intestinal cholesterol absorption.
APOE*3Leiden mice, when fed with a Western-type diet
supplemented with SL, showed significant reductions in
plasma cholesterol and TG levels . Supplementing the
western diet with SL also impaired fatty acid (FA) absorp-
tion, probably through ionic interactions between SL and FA
. Authors have not discussed the possibility that SL-
binding properties may also be the underlying mechanism
leading to impaired cholesterol and TG absorption in the
intestinal tract. SM and phosphatidylcholine, which repre-
sent respectively 25% and 35% of the bovine MFGM lipids
 have a high binding affinity with cholesterol, thereby
inhibiting intestinal cholesterol absorption in rats .
Furthermore, the slow rate of hydrolysis of SM in the
gastrointestinal tract and its incomplete digestion may
enhance its capacity to bind cholesterol throughout the
small intestine . This phenomenon may obstruct the
hydrolysis and processing of other lipids, including micelle
formation, which are essential for cholesterol absorption,
thus reducing intestinal uptake . In recent studies,
buttermilk has been associated with a significant reduction
in the micellar solubility of cholesterol in vitro, thereby
providing further support to the thesis that buttermilk
constituents may impair intestinal absorption of cholesterol
PSCK9 plays a determinant role in cholesterol meta-
bolism by regulating the intracellular degradation of LDL
receptors . Stable isotope studies in humans have shown
that plasma PCSK9 is a significant correlate of the LDL
fractional catabolic rate , and therefore can be used as
a surrogate of LDL clearance . Buttermilk consumption
did not lead to a significant change in plasma PCSK9 con-
centrations compared to placebo. However, the magnitude
of the variation in plasma PCSK9 with buttermilk was
positively correlated with changes in serum total-C and
LDL-C concentrations. Therefore, it is possible that part of
the LDL-C lowering effect of buttermilk could be attributed
to an increased clearance of cholesterol. However, it is also
possible that the reduction in LDL-C with buttermilk in it-
self may have downregulated PCSK9 expression. Indeed,
buttermilk consumption resulted in increased plasma lath-
osterol concentrations, suggestive of an up-regulated
endogenous cholesterol synthesis. These data are consis-
tent with studies in animals having shown an increased
expression in genes involved in hepatic cholesterol uptake
and in hepatic cholesterol synthesis in APOE*3Leiden mice
supplemented with SL . It is also possible that circu-
lating products resulting from the digestion of buttermilk
may directly affect cholesterol synthesis and clearance
pathways. However, we propose that the apparent increase
in endogenous cholesterol synthesis and possibly the
apparent increase in LDL clearance associated with the
Buttermilk expressed as % change from the placebo phase in
subjects with LDL-C< or >3.7 mmol/L, the median LDL-C at
screening. There are 17 subjects in each group.
Individual changes in LDL-C concentrations with
1260 V. Conway et al.
cholesterol response to buttermilk consumption may rather
reflect feedback mechanisms compensating for the reduc-
tion in the body’s cholesterol pool, which may be due in
part to the magnitude of the buttermilk-induced inhibition
of intestinal cholesterol absorption.
This study has strengths and limitations. One of its
strengths resides in the fine characterization of the
buttermilk and placebo fed to the subjects. This allowed us
to ensure that the cholesterol and TG-lowering effects
observed were attributed specifically to components of
buttermilk, primarily MFGM. The repeated measurements
at screening and at the end of both treatments also
increased our capacity to detect very small changes in
outcome measures. Cholesterol absorption, synthesis and
clearance were estimated using surrogate markers and
additional studies using stable isotopes should be used to
confirm the findings of the present study. Indeed, although
the validity of indirect markers of cholesterol homeostasis
have been established [9,10,29] these markers do not
provide a measure of absolute amount of cholesterol
absorbed and synthesized.
In summary, the reduction in total-C and LDL-C concen-
trations following buttermilk consumption may be more
important among subjects with high baseline LDL, and ap-
pears to be primarily attributable to inhibition of intestinal
unique phospholipid content of buttermilk may be respon-
sible for this effect. The apparent increase in estimated
cholesterol synthesis with buttermilk is believed to be a
serum cholesterol concentrations. Consumption of butter-
milk may also affect LDL clearance directly or indirectly and
buttermilk is of interest and also deserves further research.
From a clinical perspective, buttermilk maybe considered as
a natural, well-tolerated and low cost dietary product for
improving lipid profiles in low risk patients.
We are grateful to the nurses and the laboratory staff of the
Institute of Nutraceuticals and Functional Food for their
technical assistance and the expert care provided to the
participants. We also express our gratitude to the partici-
pants, without whom the study would not have been
(6) according to LDL-C values at screening. The P values for interaction were obtained from the MIXED model using each variable as
a continuous variable. The median screening value of LDL-C (3.7 mmol/L) was used to categorize individuals with high or low levels
and to present interactions graphically. *Values are significantly different from those of the placebo treatment within each LDL
group (P < 0.05). Screening LDL-C :: <3.7 mmol/L (n Z 17), ?: >3.7 mmol/L (n Z 17). Values are means ? SEM.
(A) Total-C, (B) LDL-C, (C) Phytosterol and (D) b-sitosterol response to buttermilk treatment compared with placebo
of changes in surrogate makers of cholesterol homeostasis
and the percentage of changes in total serum cholesterol
and LDL-C concentrations with buttermilk (n Z 34).a
Univariate correlations between the percentage
Surrogate marker (%6)%6 Total-C%6 LDL-C
aSpearman correlations were carried out between the change
(%D) in total cholesterol (C) and LDL-C and the change (%D) in
surrogate makers of cholesterol homeostasis, buttermilk vs.
Cholesterol-lowering activity of buttermilk1261
possible. We also thank YC in helping with fatty acids lab- Download full-text
oratory analyses. We are thankful to Pharmalab Inc in
helping us with the formulation and packaging of the ready-
to-used test products (buttermilk and placebo) consumed
in the present study.
The authors’ responsibilities were as followsdBL, PC, SG
and YP have designed and obtained funding for this study;
AC: coordinated the clinical study; VC: characterized and
designed the test products, performed statistical analyses
with the precious help of CR, analyzed the data and wrote
the present manuscript. VC was supported by the Alexander
Graham Bell Canada Graduate Scholarships from the Natu-
ral Sciences and Engineering Research Council of Canada.
BL is a member of and Scientific Advisory Board for the
Dairy farmers of Canada. BL, PC, YP and SG received a grant
from the Dairy Research Cluster for this research (Dairy
Farmers of Canada, Agriculture and Agri-Food Canada, Ca-
nadian Dairy Commission). Other authors report non con-
flict of interest in relation with this study.
 Clark LT. Treating dyslipidemia with statins: the risk-benefit
profile. Am Heart J 2003;145:387e96.
 Ward S, Lloyd Jones M, Pandor A, Holmes M, Ara R, Ryan A,
et al. A systematic review and economic evaluation of statins
for the prevention of coronary events. Health Technol Assess
(Winch, Eng) 2007;11:1e160 [iiieiv].
 Ros E. Intestinal absorption of triglyceride and cholesterol.
Dietary and pharmacological inhibition to reduce cardiovas-
cular risk. Atherosclerosis 2000;151:357e79.
 Jones PJH, AbuMweis SS. Phytosterols as functional food in-
gredients: linkages to cardiovascular disease and cancer. Curr
OpinClin Nutri Metab
 Third report of the national cholesterol education program
(NCEP) expert Panel on detection, evaluation, and treatment
of high blood cholesterol in adults (adult treatment panel III)
final report. Circulation December 17, 2002;106(25):3143.
 Dawson PA, Rudel LL. Intestinal cholesterol absorption. Curr
Opin Lipidol 1999;10:315e20.
 Conway V, Gauthier SF, Pouliot Y. Effect of cream pasteuri-
zation, microfiltration and enzymatic proteolysis on in vitro
cholesterol-lowering activity of buttermilk solids. Dairy Sci
 Seidah NG. PCSK9 as a therapeutic target of dyslipidemia.
Expert Opin Ther Targets 2009;13:19e28.
 Nissinen MJ, Gylling H, Miettinen TA. Responses of surrogate
markers of cholesterol absorption and synthesis to changes in
cholesterol metabolism during various amounts of fat and
cholesterol feeding among healthy men. Br J Nutr 2008;99:
 Simonen P, Gylling H, Miettinen TA. The validity of serum
squalene and non-cholesterol sterols as surrogate markers of
cholesterol synthesis and absorption in type 2 diabetes.
 Goulet J, Nadeau G, Lapointe A, Lamarche B, Lemieux S.
Validity and reproducibility of an interviewer-administered
food frequency questionnaire for healthy French-Canadian
men and women. Nutr J 2004;3:13.
 Labonte ´ M-E`, Couture P, Paquin P, Chouinard Y, Lemieux S,
Lamarche B. Comparison of the impact of trans fatty acids
from ruminant and industrial sources on surrogate markers of
cholesterol homeostasis in healthy men. Mol Nutr Food Res
 Miettinen TA, Tilvis RS, Kesaniemi YA. Serum plant sterols and
cholesterol precursors reflect cholesterol absorption and
synthesis in volunteers of a randomly selected male popula-
tion. Am J Epidemiol 1990;131:20e31.
 Lohman TG, Roche AF, Martorell R. Anthropometric stan-
dardization reference manual. In: Lohman Timothy G,
Roche Alex F, Martorell Reynaldo, editors. Champaign, Illinois:
Human Kinetics Books; 1988.
 Motard-Be ´langerA,Charest
Chouinard Y, Lemieux S, et al. Study of the effect of trans
fatty acids from ruminants on blood lipids and other risk
factors for cardiovascular disease. Am J Clin Nutr 2008;87:
 Ohlsson L, Burling H, Duan RD, Nilsson A. Effects of a
sphingolipid-enriched dairy formulation on postprandial lipid
concentrations. Eur J Clin Nutr 2010;64:1344e9.
 Ohlsson L, Burling H, Nilsson A ˚. Long term effects on human
plasma lipoproteins of a formulation enriched in butter milk
polar lipid. Lipids Health Dis 2009;8:1e12.
 Thompson LU, Jenkins DJ, Amer MA, Reichert R, Jenkins A,
Kamulsky J. The effect of fermented and unfermented milks
on serum cholesterol. Am J Clin Nutr 1982;36:1106e11.
 Schaefer E,Lamon-Fava
Clevidence B, Judd J, et al. Individual variability in lipoprotein
cholesterol response to National Cholesterol Education Pro-
gram Step 2 diets. Am J Clin Nutr 1997;65:823e30.
 Gupta AK, Savopoulos CG, Ahuja J, Hatzitolios AI. Role of
phytosterols in lipid-lowering: current perspectives. QJM
 Duivenvoorden I, Voshol PJ, Rensen PC, van Duyvenvoorde W,
Romijn JA, Emeis JJ, et al. Dietary sphingolipids lower plasma
cholesterol and triacylglycerol and prevent liver steatosis in
APOE*3Leiden mice. Am J Clin Nutr 2006;84:312e21.
 Dewettinck K, Rombaut R, Thienpont N, Le TT, Messens K, Van
Camp J. Nutritional and technological aspects of milk fat
globule membrane material. Int Dairy J 2008;18:436e57.
 Nyberg L, Duan R-D, Nilsson A ˚. A mutual inhibitory effect on
absorption of sphingomyelin and cholesterol. J Nutr Biochem
 Eckhardt ERM, Wang DQ-H, Donovan JM, Carey MC. Dietary
sphingomyelin suppresses intestinal cholesterol absorption by
decreasing thermodynamic activity of cholesterol monomers.
 Noh SK, Koo SI. Egg sphingomyelin lowers the lymphatic ab-
sorption of cholesterol and alpha-tocopherol in rats. J Nutri
 Feng D, Ohlsson L, Ling W, Nilsson A ˚, Duan R-D. Generating
ceramide from sphingomyelin by alkaline sphingomyelinase in
cholesterol uptake in Caco-2 Cells. Dig Dis Sci 2010;55:
 Garmy N, Taı ¨eb N, Yahi N, Fantini J. Interaction of cholesterol
with sphingosine. J Lipid Res 2005;46:36e45.
 Chan DC, Lambert G, Barrett PHR, Rye K-A, Ooi EMM,
Watts GF. Plasma proprotein convertase subtilisin/kexin type
9: a marker of LDL apolipoprotein B-100 catabolism? Clin
 Matthan NR, Raeini-Sarjaz M, Lichtenstein AH, Ausman LM,
Jones PJH. Deuterium uptake and plasma cholesterol pre-
cursor levels correspond as methods for measurement of
endogenous cholesterol synthesis in hypercholesterolemic
women. Lipids 2000;35:1037e44.
A, GrenierG,Paquin P,
S, Ausman L,OrdovasJ,
1262V. Conway et al.