Cardiovascular effects of fermented milk containing angiotensin-converting enzyme inhibitors evaluated in permanently catheterized, spontaneously hypertensive rats.

Abstract

In this study, two strains of Lactobacillus helveticus were used to produce fermented milk rich in angiotensin-converting enzyme (ACE) inhibitors. In vitro tests revealed that the two milks contained competitive inhibitors of ACE in amounts comparable to what has been obtained in previously reported studies. The two milks were administered by gavage to spontaneously hypertensive rats that had had a permanent aortic catheter inserted through the left arteria carotis, and mean arterial blood pressure and heart rate were monitored from 4 to 8 h after administration. Unfermented milk and milk fermented with a lactococcal strain that does not produce inhibitors were used as controls. Highly significant blood pressure effects were observed; i.e., milk fermented with the two strains of L. helveticus gave a more pronounced drop in blood pressure than the controls. Significant differences in heart rate effects were detected with one of the strains.

Full text

Page 1

APPLIED AND ENVIRONMENTAL MICROBIOLOGY, July 2002, p. 3566–3569
0099-2240/02/$04.00?0DOI: 10.1128/AEM.68.7.3566–3569.2002
Copyright © 2002, American Society for Microbiology. All Rights Reserved.
Vol. 68, No. 7
Cardiovascular Effects of Fermented Milk Containing Angiotensin-
Converting Enzyme Inhibitors Evaluated in Permanently
Catheterized, Spontaneously Hypertensive Rats
Anders Fuglsang,1* Dan Nilsson,2and Niels C. B. Nyborg3
Royal Danish School of Pharmacy, Institute of Pharmacology, 2100 Copenhagen Ø,1Novo Nordisk A/S,
2670 Ma ˚løv,2and Chr. Hansen A/S, 2970 Hørsholm,3Denmark
Received 5 December 2001/Accepted 19 April 2002
In this study, two strains of Lactobacillus helveticus were used to produce fermented milk rich in angiotensin-
converting enzyme (ACE) inhibitors. In vitro tests revealed that the two milks contained competitive inhibitors
of ACE in amounts comparable to what has been obtained in previously reported studies. The two milks were
administered by gavage to spontaneously hypertensive rats that had had a permanent aortic catheter inserted
through the left arteria carotis, and mean arterial blood pressure and heart rate were monitored from 4 to 8 h
after administration. Unfermented milk and milk fermented with a lactococcal strain that does not produce
inhibitors were used as controls. Highly significant blood pressure effects were observed; i.e., milk fermented
with the two strains of L. helveticus gave a more pronounced drop in blood pressure than the controls.
Significant differences in heart rate effects were detected with one of the strains.
Angiotensin-converting enzyme (EC 3.4.15.1) (ACE) is
known to play a role in the regulation of blood pressure in
animals, including humans. The enzyme converts angiotensin I
into angiotensin II, the former being an inert peptide and the
latter being a pressor agent. The enzyme is also responsible for
the breakdown of bradykinin, which is a dilatory peptide. The
enzyme is thus an obvious drug target in treatment of certain
cardiovascular diseases, including hypertension. Lactic acid
bacteria are known to produce inhibitors of the enzyme in
various amounts during fermentation (5, 10, 13). The inhibitors
are formed by the bacterial proteinases when the lactic acid
bacteria hydrolyze milk proteins, mainly casein, into peptides,
which can be used as nitrogen sources necessary for growth. All
of the ACE inhibitors known to date that are formed in milk
during fermentation are peptides that act as competitive inhib-
itors, such as Ile-Pro-Pro and Val-Pro-Pro (11). In some cases
it has been demonstrated that fermented milk rich in inhibitory
substances can lower systolic blood pressure in spontaneously
hypertensive rats (SHR) after oral administration (11, 13, 14).
Since the inhibitors formed are peptides, which are formed
by degradation of caseins, proteolysis plays an important role.
In theory, both specificity and overall proteolytic activity may
have a significant role in the formation. The fact that the
proteolytic systems of lactic acid bacteria are quite unspecific,
i.e., that they cleave, for instance, caseins at a high number of
places (for a review, see reference 8), combined with the fact
that ACE can be inhibited by many different peptidic struc-
tures (for instance, Val-Pro-Pro [11], Ile-Tyr [15], and Lys-Val-
Leu-Pro-Val-Pro [9], all of which inhibit ACE in the micro-
molar range), led to the investigation of a possible correlation
between crude proteolysis (measured as free amino groups)
and ACE inhibition. It was shown that, in general, the higher
the number of free amino groups formed during fermentation,
the higher the chance that the same fermented substance will
be able to inhibit ACE significantly, even though this rule is not
always strictly obeyed. Among the strains showing a relatively
high amount of free amino group formation, two strains capa-
ble of producing milk with high activity in vitro against ACE
were identified and were shown to be able to decrease the in
vivo activity of circulatory ACE in rats, measured as a de-
creased ability to convert angiotensin I to angiotensin II after
oral administration of the fermented milk (3). This study did
therefore not involve measurements of blood pressure per se;
rather, it quantified pressure increases in response to injected
angiotensin I. Since the ACE inhibitors in modern use against
hypertension are able to decrease this pressure response to
angiotensin I injection, it seems relevant to proceed with a
more specific study on the effect of the fermented substances
on blood pressure.
The aim of this study was to use permanently catheterized
rats to investigate if the fermented milks, i.e., those that are
able to inhibit circulatory ACE in vivo as mentioned before,
are also able to decrease blood pressure in SHR measured
through a permanent aortic catheter. This technique, in com-
bination with a data acquisition system, also offers the advan-
tage of being able to extract heart rates from the pulsatile
signal. Since the mean arterial pressure (MAP) is roughly given
by the product of stroke volume, peripheral resistance, and
heart rate, the latter is an important parameter in the regula-
tion of blood pressure.
MATERIALS AND METHODS
Bacteria and fermentations. Lactobacillus helveticus CHCC637 and L. helveti-
cus CHCC641 were propagated in heat-treated 9.5% reconstituted skim milk,
and fermentations were carried out at 37°C overnight, using a 1% inoculum.
Lactococcus lactis CHCC1448 was propagated similarly but at 30°C. The three
strains were obtained from the Chr. Hansen Culture Collection (Chr. Hansen
* Corresponding author. Mailing address: Royal Danish School of
Pharmacy, Institute of Pharmacology, Universitetsparken 2, 2100
Copenhagen Ø, Denmark. Phone 45 35306000. Fax: 45 35306020.
E-mail: anfu@dfh.dk.
3566

Page 2

A/S, Hørsholm, Denmark). All chemicals and reagents were from Sigma unless
otherwise stated.
Measurement of in vitro ACE inhibition. The method of Cushman and
Cheung (2) was used with some modification. Fermented milk was centrifuged at
4°C in an Eppendorf 5804R centrifuge (Eppendorf GmbH, Hamburg, Germany)
at 4,500 rpm. The whey supernatant was adjusted to pH 8.3 with 10 N NaOH and
was given 14,000 rpm in a bench centrifuge (157MP; Ole Dich, Hvidovre, Den-
mark) for 10 min. The supernatant was used in the assay where 80 ?l was mixed
with 200 ?l of substrate buffer (0.3 M NaCl, 0.1 M boric acid, and 5 mM
hippuryl-histidyl-leucine, pH 8.3) and 20 ?l of ACE solution (rabbit lung ACE;
0.1 U/ml). The mixture was kept at 37°C for 60 min, and 250 ?l of 1 N HCl was
added to stop the reaction. The hippuric acid formed was extracted with 1,700 ?l
of ethyl acetate. Fourteen hundred microliters of the organic phase was trans-
ferred to a heat block at 95°C for 15 min to evaporate all ethyl acetate. The
residues were then resuspended in 2 ml of water and were measured at 214 nm
with a Secomam Anthelie spectrophotometer (Secomam, Ales, France) using
quartz cuvettes. We determined 100% activity by using 80 ?l of substrate-free
buffer instead of sample, and 0% activity was determined using 20 ?l of water
instead of enzyme solution. Unfermented milk cannot readily be analyzed this
way but was acidified to pH 3.75 and processed as described above.
Measurement of peptide content. The whey peptide content was quantified
using a modified version of the method by Church et al. (1). One hundred
milliliters of reagent was prepared by mixing 50 ml of 0.1 M borax, 5 ml of 20%
(wt/vol) sodium dodecyl sulfate, 540 ?l of thiolactic acid (Aldrich Chemicals),
and 2.5 ml of o-phthaldialdehyde solution (50 mg/ml in 96% ethanol) and
adjusting the volume to 100 ml with water. Twenty-five microliters of test solu-
tion diluted two- or threefold was mixed with 1,000 ?l of reagent, and the
reaction mixture was measured at 340 nm, with the reagent as reference. Since
no obvious peptide standards exist for the peptide mixtures formed by lactic acid
bacteria, the peptide content was quantified using both casein tryptone (Difco
Laboratories, Detroit, Mich.) and casein peptone (Organotechnie, La Cour-
neuve, France) as standards. Both peptone and tryptone yielded linear standard
curves in the range of 0 to 2.5 mg/ml (r2? 0.999).
Permanent catheterization of rats. Twelve-week-old male SHR were obtained
from Møllegaard & Bomholtga ˚rd A/S, Ejby, Denmark, and had free access to tap
water and standard Altromin pellet food. Lights were on from 6 a.m. to 6 p.m.
The rats were allowed to acclimatize at least 7 days before being used in the
experiment. Under narcolept anesthesia with midazolam (Alpharma, Oslo, Nor-
way) and fluanisone/fentanyl (Hypnorm; Janssen, Beersen, Belgium), the rats
had tygon tubing (Copenhagen University, Institute of Pharmacology, Copenha-
gen, Denmark) inserted in the aorta through the left carotid artery. The distal
end of the tubing protruded from the neck. The rats were given 3 days of
postoperative pain treatment by subcutaneous meloxicam injections (Metacam
Vet; Boehringer Ingelheim GmbH, Ingelheim, Germany). The catheters were
flushed with saline at up to 48-h intervals in order to keep them free of clots and
were sealed with a fluid consisting of glucose (50%), heparin (500 U/ml; Leo
Pharmaceuticals, Ballerup, Denmark), and streptokinase (10,000 U/ml; Aventis-
Behring, Danderyd, Sweden).
On the day of the experiment, the rats were placed in a restrainer and the
external end of the catheter was flushed with saline and connected to a pressure
transducer (P23XL; SpectraMed, Bilthoven, The Netherlands) for recording of
MAP. The pulsatile signal was recorded using a DAQ801/DaqEz data acquisition
system (Quatech Inc., Ohio) running at a 50-Hz sampling rate. After recording
of their baseline MAP and heart rate, the rats were administered fermented milk
(10 ml/kg of body weight) by gavage and put back in the cage. Four hours after
the feeding they were put into the restrainer again and had their MAP contin-
uously measured for 4 h. As a negative control, unfermented milk was used along
with milk fermented by CHCC1448 (the latter being a strain that produces
fermented milk with very little ACE inhibition). This control more closely re-
sembles fermented milk and did not cause significant ACE inhibition in vivo in
a previous study.
Data analysis. BeePee v2 (a 32-bit Windows application programmed by
Anders Fuglsang) was used to extract data for the heart rate and MAP from the
data acquisition files. GraphPad Prism 2.01 (GraphPad Inc.) was used to perform
two-way analysis of variance on the blood pressure and heart rate data. This
program was also used to fit data for the inhibition in vitro to the Michaelis-
Menten equation and thus to obtain concentration-response curves and 50%
inhibitory concentrations (IC50s). A P of ?0.05 was considered statistically sig-
nificant.
RESULTS
Figure 1 shows the dose-response curves of in vitro ACE
inhibition of the four types of milk used in the study. The two
strains of L. helveticus clearly produce fermented milk with
ACE-inhibitory properties. The data fit with a maximal effect
of approximately 100% inhibition, which indicates that the
mode of inhibition is competitive. The ACE-inhibitory IC50s
(indicators of potency towards the enzyme) are listed in Table
1 along with data for optical density at 600 nm, peptide con-
tent, and pH. Note that the IC50given in microliters signifies
the volume that inhibits the enzyme by 50% under assay con-
ditions where the total volume is 300 ?l and is therefore a
measure of the concentration-dependent pharmacological
property called potency. The IC50given in milligrams per mil-
liliter tells how many milligrams of the peptide pool per mil-
liliter are needed to inhibit ACE by 50% and is therefore an
indicator of how specific the obtained peptide pool is against
ACE, regardless of whether or not the peptides are present in
large or small amounts (4, 7). Figure 2 shows the plot of ?MAP
versus time, where ?MAP is the initial MAP minus the MAP
at a specific time point. Figure 3 shows the change in heart rate
calculated the same way. Graphically, the curves for the test
groups display lower ?MAP than do the curves for their con-
trols. Both the blood pressures and heart rates are statistically
significantly different from those for the controls, as detailed in
FIG. 1. Dose-response curves (effect as function of the logarithm of
the dose in microliters) for the four types of milk used in this study.
The lines depict the best fit curves according to the Michaelis-Menten
equation. Note that unfermented milk had to be acidified with lactic
acid before construction of this kind of curve was possible. Data are
means of two replicates. Error bars represent minima and maxima.
White circles, L. helveticus CHCC637; black circles, L. helveticus
CHCC641; black triangles, L. lactis CHCC1448; and white triangles,
unfermented milk.
TABLE 1. Comparison of milk and bacterial strainsa
Strain or milk type OD600
pH
Peptide content
(mg/ml)b
IC50
(?l)
IC50
(mg/ml)c
L. helveticus CHCC637
L. helveticus CHCC641
L. lactis CHCC1448
Unfermented milk
6.0
5.0
3.1
1.0
3.3
3.4
4.3
6.7
4.90
4.73
2.28
0.27
3.53
3.41
1.65
0.21
9.8
16.5
?80
?80
0.16
0.26
ND
ND
aOD600, optical density at 600 nm; ND, not determined.
bThe left column is based on data obtained with casein peptone as the
standard; the right column is based on data obtained with casein tryptone as the
standard.
cCalculation based on the casein peptone standard.
VOL. 68, 2002 EFFECTS OF ACE3567

Page 3

Table 2. The reason why differences are shown rather than
actual values is that there was a large variability between the
individual blood pressures and heart rates among the rats.
DISCUSSION
The two strains of L. helveticus produce competitive inhibi-
tors (maximum inhibition according to the fit is approximately
100%), which is in accordance with earlier reports on the mode
of action for milk-derived inhibitors of ACE. The peptide
content, pH, and optical density resemble those previously
obtained with other bacteria (13).
The peptide content, though difficult to interpret directly
because of the lack of a standard that reflects the real distri-
bution of peptidic molecular weights of the peptides liberated
by the action of bacterial proteinases, was lowest in the two
controls as expected. Calculation of IC50in milligrams per
milliliter yields a number that is characteristic of the peptide
pool formed and is thus an indicator of the amount of peptide
pool necessary to inhibit ACE by 50% under the present assay
conditions. In theory, differences in the peptide pools might
therefore be detected by this value. Using nonparametric sta-
tistics could be a way of detecting differences in proteolytic
specificity among strains. This, however, would call for a higher
number of observations than present in this study. The data for
optical density, pH, and peptide content are similar to those in
previous reports on a strain of L. helveticus, designated CP790,
that had the ability to produce inhibitors (11). Milk fermented
with this strain proved highly hypotensive in a tail cuff study,
with a drop of up to about 30 mm Hg in systolic blood pressure
in SHR (11, 13). As blood pressure-lowering agents do not
change MAP and systolic pressure equally, the data presented
here cannot be directly compared to those from reports on
systolic effects. It is interesting, however, that we detect signif-
FIG. 2. The MAP effects of milk fermented with L. helveticus
CHCC641 (boxes, n ? 5) (a) and L. helveticus CHCC637 (boxes, n ?
6) (b) plus those of their controls, L. lactis CHCC1448 (triangles, n ?
5) and unfermented milk (circles, n ? 6). Data are represented as
values relative to the initial baseline blood pressure. The time scale is
relative to the time of feeding. Error bars represent standard errors of
the means.
FIG. 3. The heart rate effects of milk fermented with L. helveticus
CHCC641 (boxes, n ? 5) (a) and L. helveticus CHCC637 (boxes, n ?
6) (b) plus those of their controls, L. lactis CHCC1448 (triangles, n ?
5) and unfermented milk (circles, n ? 6). Data are represented as
values relative to the initial baseline blood pressure. The time scale is
relative to the time of feeding. Error bars represent standard errors of
the means. ?HR (bpm), change in heart rate (beats per minute).
TABLE 2. Comparison of two L. helveticus strainsa
Strain
Avg difference
(?MAP, mm Hg)b
Avg difference
(?HR, bpm)b
L. helveticus CHCC637
L. helveticus CHCC641
?12****
?11****
?14****
?13****
?8 (NS)
?28****
?22**
?59****
aThe average blood pressure and heart rate effects of the two types of ACE-
inhibiting rich milk are given. **, P ? 0.01; ****, P ? 0.0001; NS, not statistically
significant; and ?HR, bpm, change in heart rate, beats per minute.
bThe left column data are relative to those for the unfermented milk control;
the right column data are relative to those for the L. lactis CHCC1448 control.
3568 FUGLSANG ET AL.APPL. ENVIRON. MICROBIOL.

Page 4

icant effects on the heart rate. It is well known that drugs used
in the treatment of hypertension may affect heart rate; ?-ad-
renergic antagonist (e.g., propranolol) blocks the sympathetic
action of the ?-receptors on the heart and thus decreases the
heart rate directly. In contrast to this, calcium channel blockers
of the dihydropyridine type (e.g., felodipine) cause a compen-
satory stimulation of the sympathetic nervous system, thereby
raising the heart rate. As angiotensin II raises arterial tone,
thereby raising blood pressure, one could argue that ACE
inhibition could evoke a reflex-mediated increase in the heart
rate, but since angiotensin II activates the sympathetic nervous
system by presynaptic facilitation (12), one could also argue
that inhibition of ACE in vivo should decrease the heart rate.
Even though receptors for angiotensin II are present in both
cardiac atria and ventricles throughout the life cycle (6), the
high number of other endogenous regulatory and compensa-
tory pathways makes it difficult to predict what effects on the
heart rate ACE inhibitors should have, especially when we do
not know if the active substances affect other targets. The
present results therefore allow us only to conclude that the
antihypertensive effects observed with ACE-inhibitory fer-
mented milk may be mediated by a decreased heart rate but do
not absolutely prevent us from excluding dilatory effect in the
periphery. Regional flow studies may be a way of assessing
such effects.
From Table 2 it is also clear that the fermented milk pro-
duced by L. helveticus CHCC637 and L. helveticus CHCC641
does not yield identical probability values versus the two con-
trols. The reason for this could well be differences in the pool
or potency of the obtained sample, as Table 1 suggests. It is
also worth noting that the heart rate effect was insignificant
versus that of unfermented milk but significant versus that of
the L. lactis CHCC1448 product. The reason must be differ-
ences in the chemical composition of these two controls, such
as differences in pH or peptide content.
The data in this study do not answer some very fundamental
questions, such as when the effects are most pronounced or
how long the duration of action is. It is not within our present
permission to measure the animals outside the interval of 4 to
8 h after oral administration. During the 4 h in the restrainer
when MAP and the heart rate are being measured, the animals
do not have access to water or food. This situation is far from
real life, and so the data presented may not accurately reflect
the pharmacological potential of the samples. This problem
could be overcome in the future by using radiotelemetry.
All blood pressure curves (both tests and controls) seem to
decrease with time, which may be explained by the fact that the
animals lose body fluid during the 4 h. Another possibility is
that the rats gradually acclimatize to being in the experiment
over these 4 h. In that case, the initial blood pressures mea-
sured here do not reflect the factual baseline blood pressures,
and the significant figures could instead indicate an altered
ability to acclimatize during acute stress (as imposed by the
experimental procedure).
An obvious future objective is to evaluate the two types of
fermented milk in hypertensive animals using both tail cuffs
and radiotelemetry. Studies on the plasma half-lives of ACE-
inhibitory, milk-derived peptides are under way in order to
elucidate and characterize the true mechanism of action of
antihypertensive peptides formed during milk fermentation.
ACKNOWLEDGMENTS
This work was supported by the Danish Heart Foundation, grant
number 01-1-2-19-22-893. The Danish Animal Experimentation In-
spectorate approved the use of permanently catheterized rats for this
study.
We thank Dennis Neal (Rpm-Psi Inc., Northridge, Calif.) for com-
petent help during the data acquisition setup. Special thanks go to
Henrik Holm-Kjar (Zealand Pharmaceuticals A/S, Glostrup, Den-
mark), who taught us the in vivo technique.
REFERENCES
1. Church, F. C., H. E. Swaisgood, D. H. Porter, and G. L. Catign. 1983.
Spectrophotometric assay using o-phthaldialdehyde for determination of
proteolysis in milk and isolated milk proteins. J. Dairy Sci. 66:1219–1227.
2. Cushman, D. W., and H. S. Cheung. 1971. Spectrophotometric assay of the
angiotensin converting enzyme of rabbit lung. Biochem. Pharmacol. 20:
1637–1648.
3. Fuglsang, A., F. P. Rattray, D. Nilsson, and N. C. B. Nyborg. Lactic acid
bacteria: inhibition of angiotensin converting enzyme in vitro and in vivo.
Antonie van Leeuwenhoek, in press.
4. Gabrielsson, J., and D. Wiener. 2000. Pharmacodynamic concepts, p. 175–
257. In Pharmacokinetic and pharmacodynamic data analysis: concepts and
applications, 3rd ed. Swedish Pharmaceutical Society, Stockholm, Sweden.
5. Gobbetti, M., P. Ferranti, E. Smacchi, F. Goffredi, and F. Addeo. 2000.
Production of angiotensin-I-converting-enzyme-inhibitory peptides in fer-
mented milks started by Lactobacillus delbrueckii subsp. bulgaricus SS1 and
Lactococcus lactis subsp. cremoris FT4. Appl. Environ. Microbiol. 66:3898–
3904.
6. Hunt, R. A., G. M. Ciuffo, J. M. Saavedra, and D. C. Tucker. 1995. Quan-
tification and localisation of angiotensin II receptors and angiotensin con-
verting enzyme in the developing rat heart. Cardiovasc. Res. 29:834–840.
7. Kenakin, T. 1997. Pharmacologic analysis of drug-receptor interaction, 3rd
ed., p. 1–43. Lippincott Williams & Wilkins, Philadelphia, Pa.
8. Kunji, E. R., I. Mierau, A. Hagting, B. Poolman, and W. N. Konings. 1996.
The proteolytic systems of lactic acid bacteria. Antonie Leeuwenhoek 70:
187–221.
9. Maeno, M., N. Yamamoto, and T. Takano. 1996. Identification of an anti-
hypertensive peptide from casein hydrolysate produced by a proteinase from
Lactobacillus helveticus CP790. J. Dairy Sci. 79:1316–1321.
10. Meisel, H., A. Goepfert, and S. Gu ¨nther. 1997. ACE-inhibitory activities in
milk products. Milchwissenschaft 52:307–311.
11. Nakamura, Y., N. Yamamoto, K. Sakai, A. Okubo, S. Yamazaki, and T.
Takano. 1995. Purification and characterization of angiotensin I-converting
enzyme inhibitors from sour milk. J. Dairy Sci. 78:777–783.
12. Tsuda, K., H. Shima, M. Kuchii, I. Nishio, and Y. Masuyama. 1987. Effects
of captopril on neurosecretion and vascular responsiveness in hypertension.
Clin. Exp. Hypertens. Part A Theory Pract. 9:375–379.
13. Yamamoto, N., A. Akino, and T. Takano. 1994. Antihypertensive effects of
different kinds of fermented milk in spontaneously hypertensive rats. Biosci.
Biotechnol. Biochem. 58:776–778.
14. Yamamoto, Y., N. Nakamura, K. Sakai, and T. Takano. 1995. Effect of sour
milk and peptides isolated from it that are inhibitors of angiotensin-I-con-
verting enzyme. J. Dairy Sci. 78:1253–1257.
15. Yokoyama, K., H. Chiba, and M. Yoshikawa. 1992. Peptide inhibitors for
angiotensin I-converting enzyme from thermolysin digest of dried bonito.
Biosci. Biotechnol. Biochem. 56:1541–1545.
VOL. 68, 2002EFFECTS OF ACE 3569