J. Dairy Sci.

96:28162825
http://dx.doi.org/10.3168/jds.2012-6371
American Dairy Science Association, 2013.

Characterization of Lactobacillus plantarum Lp27 isolated from Tibetan kefir

grains: A potential probiotic bacterium with cholesterol-lowering effects
Ying Huang, Fei Wu, Xiaojun Wang, Yujie Sui, Longfei Yang, and Jinfeng Wang1
Central Research Laboratory, Second Hospital of Jilin University, Changchun 130041, Peoples Republic of China

ABSTRACT in the small intestine contribute to the regulation of

plasma cholesterol concentrations (Wilson and Rudel,
Lactobacillus plantarum Lp27 was isolated from 1994). Moreover, reducing the intestinal absorption
Tibetan kefir grains. The Lp27 isolate survived a 3-h of dietary and biliary cholesterol can decrease plasma
incubation at pH 2.0 and grew normally in 0.3% ox- cholesterol concentrations (Temel et al., 2007, 2010).
gall. In addition, the Lp27 isolate exhibited an adhe- The beneficial effects of lactic acid bacteria (LAB)
sion ratio of 9.5 2.5% with Caco-2 cells. Antibiotic in the intestinal tract and the potential of particular
susceptibility tests indicated that the Lp27 isolate was strains as probiotics have been reported in numerous
sensitive to gentamicin, tetracycline, erythromycin, studies (Franz et al., 2010). Specifically, the potential
and chloramphenicol, and was resistant to vancomycin of probiotic strains to lower the cholesterol level is gain-
with a minimum inhibitory value of 23 g/mL. The ing increased attention (Liong and Shah 2005, 2006;
Lp27 isolate inhibited cholesterol absorption through Ooi and Liong, 2010). However, the mechanism of the
downregulation of Niemann-Pick C1-like 1 (NPC1L1) hypocholesterolemic effects of probiotics has not been
expression in Caco-2 cells. The Lp27 isolate was fed fully elucidated.
to hypercholesterolemic rats at a dose of 109 cfu/d for Niemann-Pick C1-like 1 (NPC1L1) has been identi-
4 wk. The Lp27 feeding significantly lowered serum fied as a critical protein for intestinal cholesterol absorp-
total cholesterol, low-density lipoprotein cholesterol, tion; NPC1L1 is highly expressed in the small intestine,
and triglycerides concentrations, but no change was ob- especially at the surface of enterocytes (Altmann et al.,
served in the serum high-density lipoprotein cholesterol 2004). Mice deficient in NPC1L1 show dramatic reduc-
concentrations. In addition, liver total cholesterol and tions in dietary cholesterol absorption (Altmann et al.,
triglycerides were decreased in the Lp27-fed group. The 2004) and are completely resistant to both diet-induced
expression of NPC1L1 in the duodenum and jejunum hypercholesterolemia and apolipoprotein E deficiency-
was significantly decreased following Lp27 feeding. induced atherosclerosis (Davis et al., 2004, 2007). In
These results indicate that Lp27 might be an effective rats, the expression level of NPC1L1 along the length
cholesterol-lowering probiotic and a possible mechanism of the small intestine has been correlated with the
for the cholesterol-reducing effects of probiotics. efficiency of cholesterol absorption, with the highest
Key words: Lactobacillus plantarum Lp27, Tibetan expression found in the proximal intestine (duodenum
kefir grains, cholesterol reduction, Niemann-Pick C1- and jejunum) and the lowest expression found in the
like 1 distal intestine (ileum; Altmann et al., 2004).
INTRODUCTION Tibetan kefir grain (TK) is a natural starter for
fermented milk used in Tibet, China, which contains
Hypercholesterolemia and elevated concentrations of LAB, acetic acid bacteria, and yeasts (Farnworth
dietary cholesterol are both associated with an increased and Mainville, 2003). Tibetan kefir can be considered
risk of cardiovascular disease. A reduction as small as a probiotic resource because it can provide a variety
1% in serum cholesterol concentrations has been shown of health benefits in addition to its nutritional sta-
to decrease the risk of coronary heart disease in human tus (Urdaneta et al., 2007). Many studies that have
subjects by 2 to 3% (Manson et al., 1992). Endogenous- investigated the biological activities of TK identified
ly synthesized cholesterol, the absorption of dietary antimicrobial, immunomodulating, anti-inflammatory,
cholesterol, and the reabsorption of biliary cholesterol and hypocholesterolemic activities (Diniz et al., 2003;
Vinderola et al., 2005; Liu et al., 2006; Silva et al.,
2009). Various LAB present in TK have been isolated
and include Lactobacillus acidophilus (Angulo et al.,
Received November 12, 2012.
Accepted January 25, 2013.
1993), Lactobacillus plantarum (Garrote et al., 2001),
1
Corresponding author: [email protected] and Lactobacillus kefiranofaciens (Chen et al., 2012).
2816
 PROBIOTIC CHARACTERISTICS OF LACTOBACILLUS PLANTARUM Lp27 2817

In the present study, we identified and characterized reaction was performed in a total volume of 50 L that
a new strain of Lactobacillus from TK and evaluated its included 0.5 M of each primer, 1.25 units of Taq DNA
probiotic properties, including acid and bile salt toler- polymerase, 20 ng of DNA template, and 200 M of
ance, Caco-2 cell binding, and antibiotic susceptibility each deoxynucleotide triphosphates and PCR buffer
properties, in addition to hypocholesterolemic activi- containing Tris-HCl, KCl, and MgCl2. The amplifica-
ties. The in vitro uptake of cholesterol and the in vivo tion conditions were as follows: initial denaturation at
effects of the Lactobacillus strain on serum cholesterol 94C for 5 min, followed by 35 cycles of denaturation
levels in rats were determined. The mechanisms by at 95C for 30 s, annealing for 30 s at 55C, and a final
which the strain removed cholesterol in vitro and in extension of 10 min at 72C. The PCR-amplified DNA
vivo were also analyzed. Our goal is to develop a new fragments were resolved by agarose gel electrophoresis
LAB that can serve as a probiotic to lower cholesterol and stained with ethidium bromide (0.5 mg/mL) fol-
levels in humans. lowed by visualization under UV light.
Genus-specific PCR assays were performed using
the forward primer LbLMA-1 (5c-CTCAAAACTA-
MATERIALS AND METHODS
AAGTTTC-3c) and the reverse primer R-161 (5c-CTT-
TK GTACACACCGCCCGTTCA-3c; Dubernet et al.,
2002). Sequencing of the 16S ribosomal RNA genes was
The TK was collected from Tibet, China, and was performed to validate the phenotypic characterization
evaluated. The grains were cultured in sterile 10% re- of the Lactobacillus strains. The 16S ribosomal DNA
constituted skim milk at 20C for 20 h. Tibetan kefir of the selected strains were amplified by PCR using
grains were then filtered and stored at 4C. 2 universal primers: 27f and 1492r (Weisburg et al.,
1991). The PCR products obtained were cloned, se-
Isolation of LAB and Cultivation from TK quenced, and then compared using the NCBI BLAST
sequence database (http://www.ncbi.nlm.nih.gov/nuc-
The TK was recovered from the mother culture after core/JQ236617.1) to identify the species of each strain.
reaching the fermentative end-point. Ten grams of TK Sequencing was performed by Bioasia Co. (Shanghai,
was suspended in 90 g of sterile saline buffer (0.85%) China).
and homogenized with a Stomacher (Laboratory
Blender Stomacher 400, Seward, UK) for 20 min. Serial Acid Tolerance
dilutions of the suspended samples were used for mi-
crobial enumerations and isolation on de Man, Rogosa, The MRS broth was used to simulate the acidic condi-
Sharpe (MRS) agar (Difco, Detroit, MI) under both tions of the gut and was adjusted to different pH values
aerobic and anaerobic conditions and on LM17 agar (1.0, 2.0, and 3.0) with 1.0 N HCl; broth was adjusted
[M17 agar (Difco) with 0.5% (wt/wt) lactose (Sigma, to neutral pH (7.0) to serve as a control. Tubes of broth
St. Louis, MO)] under aerobiosis. In addition, 200 g/ adjusted to the different pH values were inoculated
ml of cycloheximide (Sigma) was added to both the (at 109 cfu/mL) with cultures of lactobacilli grown
MRS and LM17 agars to inhibit the growth of yeasts. overnight and incubated at 37C for 24 to 48 h. One
The plates were incubated at 30C for 7 d. After 7 d, milliliter of culture was immediately taken from each
yellowish colonies typical of lactobacilli were selected tube (0 h), and 10-fold serial dilutions were prepared in
for morphological examination under a microscope. 0.1% peptone water; pour plating was performed using
Pour plating was also performed with dilutions of 1:107 MRS agar. Similarly, 1 mL of culture was taken from
and 1:108, and the submerged colonies were selected for each tube after intervals of 1, 2, 3, and 4 h, followed by
morphological examination using Gram staining. The plating. The plates were incubated at 37C for 24 to 48
putative Lactobacillus isolates were further subjected to h and the colony-forming units were counted.
catalase testing and molecular typing methods. Stan-
dard bacterial cultures, namely Lb. acidophilus ATCC Bile Salt Tolerance
4356 and Lb. plantarum ATCC 8014, were obtained
from ATCC (Rockville, MD). The MRS broth supplemented with 0.2 and 0.3% (wt/
vol) oxgall was used to mimic the approximate levels
Identification of Lactobacillus Isolates by PCR of bile salts in the intestinal tract; MRS broth without
bile salt was used as a control. A fresh overnight culture
Genomic DNA was extracted from cultures grown at of Lp27 was inoculated at a final concentration of 1%
37C for 16 to 18 h in MRS broth using the protocol (vol/vol) into each media, incubated at 37C for 18 h,
described by Pospiech and Neumann (1995). A PCR and monitored for growth hourly by spectrophotometry
Journal of Dairy Science Vol. 96 No. 5, 2013
2818 HUANG ET AL.

at 620 nm. The time required to increase the absor- Bacteria, Cell Lines, and Coculture
bance at 620 nm by 0.3 units was measured in MRS
broth with and without oxgall. The difference in time The reference strains Lactobacillus plantarum ATCC
(h) between the 2 culture media was considered to be 8014 and Lactobacillus acidophilus ATCC 4356 were
the lag time during which the bile salt tolerance effect obtained from ATCC. For all experiments, Lp27 was
was expressed. precultured for 6 to 8 h at 37C on a rotating platform
(225 rpm). Precultures were then added to prewarmed
MRS medium (dilution 1:20). The cultures were cul-
Caco-2 Cell Adhesion Assay tivated overnight before use. One OD600nm (optical
density at 600 nm) of bacteria culture was equivalent
The adhesion capability of the isolates was assayed
to 1 108 Lp27/mL. The required amount of Lp27 was
according to the method described by Jacobsen et al.
resuspended in an appropriate volume of the respective
(1999). Initially, 1 105 Caco-2 cells were seeded into
prewarmed medium without antibiotics. The coculture
each well of a 6-well tissue culture plate. Dulbeccos
was prepared by washing Caco-2 cells with warm PBS.
modified Eagles minimal essential medium (DMEM)
The cells were incubated with Lp27 or with medium
supplemented with 10% (vol/vol) heat-inactivated (30
alone. The experiment was terminated by thoroughly
min, 56C) fetal bovine serum, 100 U/mL of penicillin,
washing the plates with ice-cold PBS.
and 100 mg/mL of streptomycin was used for cultur-
ing. The medium was replaced with fresh medium
every other day and an adhesion assay was performed Cholesterol Uptake Assay
after 20 d postconfluency. The cells were then washed Caco-2 cells were seeded onto 6-well plates at a den-
twice with 3 mL phosphate-buffered saline (pH 7.4). sity of 1 105 cells/well were incubated with Lp27,
Two milliliters of DMEM without serum or antibiot- Lb. plantarum ATCC 8014 or Lb. acidophilus ATCC
ics was added to each well and incubated at 37C for 4356 for 6 h. For comparison, nontreated Caco-2 cells
30 min. Approximately 109 cfu/mL of bacterial culture were also included. One hour before harvesting the
was suspended in 1 mL of DMEM (without serum or cells, 0.15 mL of a micellar solution containing 6.6 mM
antibiotics) and added to the different wells. The plate sodium taurocholate, 74,000 Bq (2 mCi) [3H] choles-
was incubated at 37C for 2 h in a 5% CO2, 95% air terol, 1 mM oleic acid, 0.5 mM monoolein, 0.1 mM
atmosphere. The monolayer was washed with sterile unlabeled cholesterol, and 0.6 mM phosphatidylcholine
PBS and the cells were detached by trypsinization. One was added to the Caco-2 cells. At the end of the incuba-
milliliter of 0.25% trypsin-EDTA solution (Sigma) was tion, the unincorporated, radiolabelled cholesterol was
added to each well of a 6-well plate, the plate was then removed by washing the cells 4 times with 1.5 mL of
incubated for 15 min at room temperature. The cell cold DMEM. The cellular lipids were extracted with 1.5
suspension was plated onto MRS agar by serial dilution mL of hexane:isopropyl alcohol:water (3:2:0.1, vol/vol/
to determine the number of adherent bacterial cells. vol). The radioactivity in the cellular lipid extract was
The plate was incubated for 24 to 48 h at 37C and the estimated by counting with a Packard liquid scintilla-
colonies were counted. The number of bacterial cells tion counter.
initially added to each well of the 6-well plates were also
counted by serial dilution and plating onto MRS agar. Animal Feeding and Grouping
The results of the adhesion assay were expressed as the
adhesion percentage (i.e., the ratio between adherent Animal experiments were conducted in compliance
bacteria and added bacteria per well). Caco-2 cells of with the Guide for Care and Use of Laboratory Ani-
the same passage were used for 3 independent experi- mals from the National Institutes of Health (Yin et
ments (n = 3) with 2 replicates in each experiment. al., 2010). Twenty male Sprague-Dawley rats (4 wk of
age) were obtained from the National Animal Breeding
Antibiotic Susceptibility Testing and Research Center, Beijing, China. The rats were
fed a commercial chow (Kangqiao Inc., Beijing, China),
The MIC of gentamicin, tetracycline, erythromycin, which included 32% protein, 5% fat, 2% fiber, 1.8%
chloramphenicol, and vancomycin (Sigma) was deter- Ca, 1.2% P, and 59% N-free extract for 1 wk. Rats
mined for Lp27 using the procedure as previously de- were individually housed in metal cages at a controlled
scribed (Rossetti et al., 2009). According to FEEDAP temperature (23 2C) and humidity (55 6%) under
(2005), MIC breakpoint values for gentamicin, tetracy- a 12-h light-dark cycle. After this adaptation period,
cline, erythromycin, chloramphenicol, and vancomycin rats were divided into 2 groups consisting of 10 rats
were 8, 8, 4, 4, and 4 g/mL, respectively. each; the initial average BW were similar between the

Journal of Dairy Science Vol. 96 No. 5, 2013

 PROBIOTIC CHARACTERISTICS OF LACTOBACILLUS PLANTARUM Lp27 2819

2 groups. Each group was assigned one of the following and the protein concentration was determined using
diet regimens: (1) high-cholesterol diet (HC diet group) the Bio-Rad Protein Assay Kit (Hercules, CA).
or (2) high-cholesterol diet + Lb. plantarum Lp27 (HC
Lp27 group). The high-cholesterol diet contained 1% Real-Time Quantitative PCR
(wt/wt) cholesterol, 10% lard, 5% sucrose, 0.3% sodium
cholate, 0.2% propylthiouracil, and 78.5% commercial The RNA was prepared using a Qiagen RNeasy Mini
chow (Aoboxing Biotech Co. Ltd., Beijing, China). Kit (Germantown, MD) according to the manufacturers
Each day of the study period, the HC Lp27 group re- protocol, and cDNA was synthesized using the cDNA
ceived 2 mL (109 cfu/mL) of an Lb. plantarum Lp27 Synthesis Kit from Invitrogen (Carlsbad, CA), accord-
solution intragastrically. The HC diet group received an ing to the manufacturers protocol. Transcripts were
equivalent amount of normal saline. The rats were fed detected using SYBR green-based (Applied Biosystems,
for 4 wk, during which time BW and food intake were Foster City, CA) real-time PCR. Forward and reverse
recorded daily. After the feeding period, the rats were primers were mixed in equal proportions and used
fasted overnight and used for subsequent testing. at a final concentration of 0.2 mM. Human NPC1L1
(forward: 5c-TATGGTCGCCCGAAGCA-3c, reverse:
Assay for Serum Lipids 5c-TGCGGTTGTTCTGGAAATACTG-3c) and human
-actin (forward: 5c-CCTGGCACCCAGCACAAT-3c,
Blood samples were collected from the tail veins of reverse: 5c-GCCGATCCACACGGAGTACT-3c) tran-
the rats under diethyl ether anesthesia on d 0, 7, 14, 21, scripts were measured. Additionally, rat NPC1L1
and 28. Approximately 1 mL of blood was taken from (forward: 5c-AACAGCGAGAGGCTCACATT-3c;
each rat, transferred to sterile tubes and kept on ice reverse: 5c-AGTGGCGTTCATGCCTGCCT-3c) and
for 30 min. The tubes were then centrifuged at 2,000 rat -actin (forward: 5c-ATTGTGATGGACTCCG-
g for 20 min at 4C. Serum samples were analyzed to GAGA-3c; reverse: 5c-CAGCTCATAGCTCTTCTC-
determine serum total cholesterol (TC), high-density CA-3c) transcripts were measured. -Actin served as a
lipoprotein cholesterol (HDL-C), low-density lipopro- housekeeping gene to normalize the expression of target
tein cholesterol (LDL-C), and triglyceride (TG) con- genes. Relative expression was calculated according to
centrations using a commercial kit (Biosino Biotechnol- the 2Ct method as previously described (Livak and
ogy and Science, Beijing, China). Schmittgen, 2001). Each experiment was performed
with duplicate samples, and the mRNA levels of each
Assay for Liver TC and TG sample were determined in triplicate. Real-time PCR
was performed using the ABI 7500 System for data
After 4 wk on the appropriate diet, all of the rats acquisition, and the data were analyzed using the ABI
were euthanized. The livers were removed, rinsed with 7500 System Sequence Detection software (both from
normal saline solution, blotted dry with filter paper Applied Biosystems). The data are presented as mean
and weighed. Liver TC and TG levels were determined values with standard deviations.
according to the method of Chiu et al. (2006).
Statistical Analysis
Preparation of Samples for RNA Measurements
Data are expressed as means with standard devia-
After the 4-wk feeding period, rats were euthanized tions. The statistical significance of the difference be-
with diethyl ether. The small intestines were removed, tween 2 means was evaluated using a Students t-test.
flushed with ice-cold PBS, and cut into 3 sections of For analyzing multiple mean values, Dunnetts test was
equal length. The sections were slit lengthwise, and the used after an ANOVA. In these tests, values of P <
mucosas were gently scraped and frozen in liquid N2 0.05 were considered to be significant.
and stored at 80C. Total RNA was extracted from
the tissue samples using RNA STAT-60 (Tel-Test,
RESULTS
Friendswood, TX). Total protein was recovered from
the organic phase that remained after RNA isolation Screening and Identification
by precipitating with isopropanol, washing with 0.3 M of Lactobacillus Isolates
of guanidine hydrochloride in 95% ethanol, and resus-
pending the protein pellet in 1% SDS and 50 mM of A total of 38 typical Lactobacillus isolates recovered
Tris-HCl, pH 8.8 (Banerjee et al., 2003). The RNA con- from kefir grain samples on MRS plates were initially
centration was determined by absorbance at 260 nm, selected for screening. Typical colonies of these Lac-

Journal of Dairy Science Vol. 96 No. 5, 2013

2820 HUANG ET AL.

tobacillus isolates were randomly selected from MRS Antibiotic Susceptibility

plates based on their morphological characteristics as-
sessed by microscopic examination after Gram staining The Lp27 isolate was found to be sensitive to tetracy-
and from negative tests for the catalase reaction. For cline, erythromycin, chloramphenicol, and gentamicin,
12 of the isolates, a 250 bp fragment was amplified and resistant to vancomycin (MIC of 23 g/mL), ac-
using 16S rRNA-based PCR primers specific for the cording to the FEEDAP (2005) breakpoint (vancomy-
genus Lactobacillus. Two standard probiotic cultures, cin, 4 g/mL; data not shown).
Lb. acidophilus ATCC 4356 and Lb. plantarum ATCC
8014, were used as positive controls (data not shown) Lp27 Inhibits NPC1L1 Expression in Caco-2 Cells
in the PCR assays.
Adherence of the putative probiotic organism to the The colon carcinoma cell line Caco-2 was used to
gastrointestinal tract is crucial for the organism to have elucidate the effects of probiotics on NPC1L1 expres-
an extended residence time in the host. Therefore, the sion in colonocytes. Cells were stimulated for 6 h with
12 isolates were subjected to an in vitro adherence test Lp27, Lb. plantarum ATCC 8014, or Lb. acidophilus
using the human carcinoma cell line Caco-2, which has ATCC 4356 (Figure 2A). The Lp27 isolate significantly
been widely accepted as an in vitro model to assess the downregulated the expression of NPC1L1 mRNA; Lb.
adhesion of potential probiotics to the intestinal epithe- plantarum ATCC 8014 and Lb. acidophilus ATCC 4356
lium. On qualitative evaluation, the isolate Lp27 most
strongly adhered with Caco-2 cells and was selected
for further characterization. The adhesion ratio of the
Lp27 isolate to the Caco-2 cells was estimated to be 9.5
2.5%, which was comparable to the adhesion values
of 6.8 1.2 and 12.7 1.9% of Lb. plantarum ATCC
8014 and Lb. acidophilus ATCC 4356, respectively, with
Caco-2 cells. The 16S rDNA gene sequencing identified
the Lp27 as an Lb. plantarum species.

Acid Tolerance

The 4-h survival curve of the Lp27 isolate at pH

values of 7.0, 3.0, 2.0, and 1.0 is shown in Figure 1A.
The viability of Lp27 increased when it was incubated
at pH 7.0 and 3.0. As no viable count was observed
after the fourth hour of incubation at pH 2.0 and after
the second hour at pH 1.0, it was concluded that Lp27
could survive at pH 2.0 for 3 h and pH 1.0 for 1 h. At
pH 3.0, Lp27 remained viable for more than 4 h.

Bile Salt Tolerance

As shown in Figure 1B, the time required to increase

the A620 nm reading by 0.3 units was 4.0 h when Lp27
was incubated in MRS broth without oxgall, whereas
the times required for this increase were 5.8 and 6.2 h
when Lp27 was grown in MRS broth containing 0.2 and
0.3% oxgall, respectively. The lag phase in 0.2 and 0.3%
oxgall concentration was 1.8 and 2.2 h, respectively.
Furthermore, no significant difference was observed
between the lag times for the 2 oxgall concentrations. Figure 1. Survival curves of Lactobacillus plantarum Lp27 in me-
dia with different pH values (A) and growth curves in media with
Maximum absorbance for Lp27 was acquired after 10 different bile salt concentrations (B). (A) Survival curve at pH 7.0
h in MRS broth without oxgall, whereas it required 11 (control; ), pH 3.0 (), pH 2.0(), and pH 1.0 (). (B) Absorbance
and 14 h in MRS broth with 0.2 and 0.3% oxgall, re- at 620 nm in de Man, Rogosa, Sharpe broth without bile salt (control;
), with 0.2% oxgall (), and 0.3% oxgall (). Error bars represent
spectively. However, the maximum value for absorbance standard deviations and each data point is the average of 3 repeated
at 620 nm was nearly the same for all treatments. measurements from 2 independently replicated experiments (n = 3).

Journal of Dairy Science Vol. 96 No. 5, 2013

 PROBIOTIC CHARACTERISTICS OF LACTOBACILLUS PLANTARUM Lp27 2821

were also able to downregulate NPC1L1 mRNA. Fur-

ther characterization of NPC1L1 expression revealed
that Lp27 downregulated NPC1L1 mRNA in a dose-
and time-dependent manner (Figure 2B-C).

Lp27 Decreases Micellar Cholesterol Uptake

To address whether Lp27 interferes with the uptake

of a micellar cholesterol solution, Caco-2 cells were
incubated with Lp27, Lb. plantarum ATCC 8014, or
Lb. acidophilus ATCC 4356 and were compared with
an untreated control. One hour before harvesting the
cells, 0.15 mL of a micellar solution was added to the
medium, and the amount of cholesterol taken up by
the cells was then measured (Figure 3). The uptake of
cholesterol by cells incubated with Lp27 was approxi-
mately 1.5-fold lower than the uptake of cholesterol
by the untreated control. The cells incubated with Lb.
plantarum ATCC 8014 or Lb. acidophilus ATCC 4356
also exhibited decreased uptake of micellar cholesterol
compared with the nontreated control.

BW and Food Intake

All rats appeared to be healthy throughout the feed-

ing period. No significant differences were observed in
BW gain, total food intake, or food intake efficiency (P
> 0.05) between the 2 groups of rats (Table 1).

Blood Lipid Analysis

Table 2 shows the effects of dietary cholesterol and

Lp27 on TC and TG levels in rats. The TC, LDL-
C, and TG concentrations observed in the HC Lp27
group were significantly decreased compared with those
observed in the HC diet group. However, the HDL-
C concentrations did not show a significant difference
between the groups.
Figure 2. Probiotics can downregulate Niemann-Pick C1-like 1
Liver Lipid Analysis (NPC1L1) expression in Caco-2 cells. (A) The expression of NPC1L1,
as determined by real-time PCR, in Caco-2 cells cocultured with
Lp27, Lactobacillus plantarum ATCC 8014 (Lp.8014; 107 per mL)
Table 3 shows the data for weight and lipid content of and Lactobacillus acidophilus ATCC 4356 (La.4356; 107 per mL) for
the rat livers. Average liver weight was not significantly 6 h was compared with the untreated (NT) control. The data are
different between the 2 groups. However, liver TC and presented as the means and standard deviations. An asterisk repre-
sents standard error of P < 0.01 compared with the NT control (by
TG concentrations were significantly lower in the HC Dunnetts test for multiple comparisons). (B) Expression analysis of
Lp27 group than in the HC diet group (P < 0.05). NPC1L1 in Caco-2 cells after 6 h of stimulation with Lb. plantarum
Lp27 at different concentrations. The data are presented as the means
and SD. An asterisk represents a standard error of P < 0.01 compared
Lp27 Inhibits NPC1L1 Expression with the NT control (by a Dunnetts test for multiple comparisons).
in the Small Intestine (C) Time course of the effects of Lb. plantarum Lp27 (107 per mL) on
NPC1L1 mRNA expression in Caco-2 cells. The data are presented as
the means and SD. An asterisk represents a standard error of P < 0.01
To address the mechanism underlying Lp27-mediated compared with the NT control (by Students t test).
inhibition of cholesterol absorption, mRNA levels of
NPC1L1 in the small intestines were assayed (Figure
4). The level of NPC1L1 mRNA varied in different seg-
Journal of Dairy Science Vol. 96 No. 5, 2013
2822 HUANG ET AL.

Table 1. Body weight gain, total food intake, and food intake efficiency
of rats fed a high-cholesterol (HC) diet for 4 wk1

Item HC diet2 HC Lp273

Body weight gain (g) 120.04 2.89a 118.52 3.12a
Total food intake (g) 589.76 21.04a 597.37 17.21a
Food efficiency4 (%) 20.35 0.37a 19.51 0.42a
a
Mean values within a row with different superscripted letters differ
significantly (P < 0.05).
1
The data are shown as the mean SD.
2
HC diet group = high-cholesterol diet.
3
HC Lp27 group = high-cholesterol diet + Lactobacillus plantarum
Lp27.
Figure 3. Lactobacillus plantarum Lp27 inhibits cholesterol uptake 4
in Caco-2 cells. Caco-2 cells were cocultured with Lb. plantarum ATCC Food efficiency (%) = (BW gain/food intake) 100.
8014 (Lp.8014) and Lactobacillus acidophilus ATCC 4356 (La.4356) as
well as the untreated (NT) control. One hour before harvesting the
cells, the medium was supplemented with 0.15 mL of a micellar solu- for 2 h and even grew well in 0.3% oxgall; thus, Lp27
tion. At the end of the incubation, the cells were washed thoroughly,
and cellular lipids were extracted. The radioactivity in the cellular exhibited good acid and bile salt tolerance.
lipid extract was measured. Values shown are the means and standard One of the required properties of probiotic strains is
deviations of 3 independent experiments. Statistically significant dif- that they are safe for human consumption. A qualified
ferences versus NT were determined by Dunnetts test (* = P < 0.05).
presumption of safety status has been granted to several
Lactobacillus species; Lb. plantarum is included in the
ments of the rat intestine, with the peak expression list of taxonomic units within the qualified presump-
observed in the proximal jejunum. In the duodenal and tion of safety status, provided that the lack of acquired
jejunal segments, NPC1L1 mRNA levels detected in antibiotic resistance is systematically demonstrated
the HC Lp27 group were significantly lower than those (EFSA, 2007). The Lp27 strain was tested for suscep-
detected in the HC diet group. In contrast, no signifi- tibility to 5 antibiotics belonging to the clinically most
cant difference was observed in NPC1L1 mRNA levels relevant antibiotics. The antibiotic susceptibility tests
in the ileal segment. indicated that Lp27 was only resistant to vancomycin.
The results were as expected, as lactobacilli are known
DISCUSSION to be naturally resistant toward vancomycin and such
resistance is usually intrinsic, chromosomally encoded,
To survive passage through the stomach, probi- and not transmissible (Klein et al., 1998, 2000).
otic bacteria must be tolerant of acidic environments. High levels of serum cholesterol have been associated
Stresses to ingested microorganisms begin in the stom- with an increased risk of coronary heart disease. The use
ach, which has a pH between 1.5 and 3, and continue in of probiotic bacteria to reduce serum cholesterol levels
the upper intestine, which contains a 0.03 to 0.3% (wt/ has attracted much interest in recent years (Ebringer
vol) concentration of bile salts. For probiotic strains, et al., 2008). The results of the present study indicate
survival at pH 3 for 2 h and in a bile concentration of that Lp27 was able to reduce cholesterol absorption by
0.3% (wt/vol) for 12 h is considered optimal (Usman inhibiting NPC1L1 gene expression in Caco-2 cells. Our
and Hosono, 1999). The Lp27 isolate survived at pH 3 previous in vitro study indicated that Lb. acidophilus

Table 2. Serum total cholesterol (TC), low-density lipoprotein-cholesterol (LDL-C), and triglycerides (TG) levels in rats fed experimental diets1

Item2 Day 0 Day 7 Day 14 Day 21 Day 28

TC (mg/L)
HC-diet 986.18 14.5a 996.46 21.3a 1,089.32 31.2a 1,187.36 26.9a 1,269.18 31.53a
HC-Lp27 982.85 19.2a 972.57 15.4a 866.15 24.2b 832.85 19.2b 797.43 30.67b
LDL-C (mg/L)
HC-diet 278.21 18.20a 325.32 17.26a 493.46 12.48a 586.59 16.63a 687.24 21.32a
HC-Lp27 283.64 15.38a 278.47 13.52a 256.25 16.52b 245.37 15.58b 223.31 18.03b
TG (mg/L)
HC-diet 421.08 22.14a 445.92 12.89a 561.47 28.27a 675.62 37.43a 734.52 26.13a
HC-Lp27 428.63 12.58a 419.65 25.05a 403.62 21.23b 361.55 21.12b 339.52 24.07b
a,b
Mean values within the same column followed by different superscript letters are significantly different (P < 0.05).
1
The data are shown as the mean SD.
2
HC diet group = high-cholesterol diet; HC Lp27 group = high-cholesterol diet + Lactobacillus plantarum Lp27.

Journal of Dairy Science Vol. 96 No. 5, 2013

 PROBIOTIC CHARACTERISTICS OF LACTOBACILLUS PLANTARUM Lp27 2823
1
Table 3. Liver cholesterol and triglycerides (TG) levels in rats fed experimental diets

Treatment Liver Liver Liver TG

group weight (g) cholesterol (mg/g) (mg/g)
HC-diet2 9.08 0.18a 12.52 0.23a 36.28 0.27a
HC-Lp273 9.06 0.11a 8.25 0.12b 25.16 0.17b
a,b
Mean values within the same column followed by different superscript letters are significantly different (P
< 0.05).
1
The data are shown as the mean SD.
2
HC diet group = high-cholesterol diet.
3
HC Lp27 group = high-cholesterol diet + Lactobacillus plantarum Lp27.

4356 reduces cholesterol absorption by downregulating in rats and human subjects (Abd El-Gawad et al.,
the expression of NPC1L1 (Huang and Zheng, 2010). 2005; Wang et al., 2009). Significant reductions in TC
Substantial downregulation of NPC1L1 expression in and TG concentrations were also observed in the livers
Caco-2 cells by Lactobacillus rhamnosus BFE5264 and of rats in the Lp27-fed group. These findings demon-
Lb. plantarum has also been reported by Yoon et al. strated that the serum TC and TG levels in HC Lp27
(2013). These results indicate that downregulation of group were actually reduced rather than merely being
NPC1L1 might be an activity of most Lactobacillus redistributed from the blood to the liver.
strains and may provide a possible molecular mecha- Because cholesterol absorption primarily occurs in the
nism for the modulation of cholesterol concentrations. duodenum and proximal jejunum, with little absorp-
Moreover, Lp27 reduced the serum TC, LDL-C, and tion by the ileal segment of the intestine (Borgstrom,
TG concentrations of rats fed a high-cholesterol diet, 1960; Grundy, 1983), we investigated NPC1L1 mRNA
as reported earlier for other Lb. plantarum strains expression along the duodenum-ileum axis. The level
(Nguyen et al., 2007; Fazeli et al., 2010). No significant of NPC1L1 mRNA varied in the different segments of
differences in HDL-C concentrations were observed in rat intestine, with the peak expression detected in the
our study, which is in agreement with previous results proximal jejunum; this result is in agreement with a

Figure 4. Real-time PCR of Niemann-Pick C1-like 1 (NPC1L1) mRNA in the small intestines of rats fed Lactobacillus plantarum Lp27 was
compared with NPC1L1 mRNA levels in the small intestines of the control group [high-cholesterol (HC) diet group]. The data are presented as
the means and standard deviations. An asterisk represents mean values that differed significantly compared with the control group by Students
t test (P < 0.05).

Journal of Dairy Science Vol. 96 No. 5, 2013

2824 HUANG ET AL.

study by Altmann et al. (2004). Moreover, NPC1L1 Davis, H. R. Jr., L. M. Hoos, G. Tetzloff, M. Maguire, L. J. Zhu, M. P.
Graziano, and S. W. Altmann. 2007. Deficiency of Niemann-Pick
mRNA levels in the duodenal and jejuna segments C1 like 1 prevents atherosclerosis in ApoE / mice. Arterio-
of rats in the Lp27-fed group were significantly lower scler. Thromb. Vasc. Biol. 27:841849.
than those in the duodenal and jejunal segments of the Davis, H. R. Jr., L. J. Zhu, L. M. Hoos, G. Tetzloff, M. Maguire, J.
Liu, X. Yao, S. P. Iyer, M. H. Lam, E. G. Lund, P. A. Detmers, M.
control group. The present results indicate that Lp27 P. Graziano, and S. W. Altmann. 2004. Niemann-Pick C1 Like 1
is able to reduce cholesterol absorption by inhibiting (NPC1L1) is the intestinal phytosterol and cholesterol transporter
NPC1L1 mRNA transcription in the small intestine, and a key modulator of whole-body cholesterol homeostasis. J.
Biol. Chem. 279:3358633592.
which is in agreement with our previous results (Huang Diniz, R. O., L. K. Garla, J. M. Schneedorf, and J. C. T. Carvalho.
et al., 2010). 2003. Study of anti-inflammatory activity of Tibetan mushroom, a
symbiotic culture of bacteria and fungi encapsulated into a poly-
saccharide matrix. Pharmacol. Res. 47:4952.
CONCLUSIONS Dubernet, S., N. Desmasures, and M. Gueguen. 2002. A PCR-based
method for identification of lactobacilli at the genus level. FEMS
In summary, Lb. plantarum Lp27 isolated from TK Microbiol. Lett. 214:271275.
exhibited efficient cholesterol-reducing ability in vitro Ebringer, L., M. Ferencik, and J. Krajcovic. 2008. Beneficial health
effects of milk and fermented dairy productsReview. Folia Mi-
and in vivo. The results of this study indicate that the crobiol. (Praha) 53:378394.
probiotic potential of the Lp27 strain for the control European Food Safety Authority. 2007. Introduction of a qualified
of cholesterol is at least partially mediated by the presumption of safety (QPS) approach for assessment of selected
microorganisms referred to EFSA. EFSA J. 587:116.
downregulation of NPC1L1. Furthermore, these results Farnworth, E. R., and I. Mainville. 2003. Kefir: A fermented milk
indicate a possible mechanism for the cholesterol- product. Pages 77111 in Handbook of Fermented Functional
reducing effects of probiotics. In future studies it will Foods. CRC Press, Boca Press, Boca Raton, FL.
Fazeli, H., J. Moshtaghian, M. Mirlohi, and M. Shirzadi. 2010. Reduc-
be necessary to test more animals, using varying doses tion in serum lipid parameters by incorporation of a native strain
of bacteria over long durations, to assess the long-term of Lactobacillus plantarum A7 in mice. Iran. J. Diabetes Lipid
probiotic potential of Lp27. Disord. 9:17.
FEEDAP. 2005. Opinion of the scientific panel on additives and prod-
ucts or substances used in animal feed on the updating of the cri-
ACKNOWLEDGMENTS teria used in the assessment of bacteria for resistance to antibiotics
of human or veterinary importance. EFSA J. 223:112.
Franz, C. M. A. P., G. Y. Cho, and W. H. Holzapfel. 2010. Probiotics:
This work was supported by the National Natu- Taxonomy and technological features. Pages 241245 in Probiotic
ral Science Foundation of China (Changchun; No. and Prebiotic Foods: Technology, Stability and Benefits to Human.
31100022) and the Program of Science and Technology N. P. Shah, A. G. da Cruz, and J. de Assis Fonseca Faria, ed. Nova
Science Publishers, Hauppauge, NY.
Development Plan of Jilin Province (Changchun; No. Garrote, G. L., A. G. Abraham, and G. L. de Antoni. 2001. Chemical
20110737). and microbiological characterization of kefir grains. J. Dairy Res.
68:639652.
Grundy, S. M. 1983. Absorption and metabolism of dietary cholesterol.
REFERENCES Annu. Rev. Nutr. 3:7196.
Huang, Y., J. Wang, Y. Cheng, and Y. Zheng. 2010. The hypocho-
Abd El-Gawad, I. A., E. M. El-Sayed, S. A. Hafez, H. M. El-Zeinia, lesterolaemic effects of Lactobacillus acidophilus American Type
and F. A. Salehb. 2005. The hypocholesterolaemic effect of milk culture collection 4356 in rats are mediated by the down-regulation
yoghurt and soy-yoghurt containing bifidobacteria in rats fed on a of Niemann-Pick C1-like 1. Br. J. Nutr. 104:807812.
cholesterol-enriched diet. Int. Dairy J. 15:3744. Huang, Y., and Y. Zheng. 2010. The probiotic Lactobacillus acidophi-
Altmann, S. W., H. R. Jr. Davis, L. J. Zhu, X. Yao, L. M. Hoos, G. lus reduces cholesterol absorption through the down-regulation of
Tetzloff, S. P. Iyer, M. Maguire, A. Golovko, M. Zeng, L. Wang, Niemann-Pick C1-like 1 in Caco-2 cells. Br. J. Nutr. 103:473478.
N. Murgolo, and M. P. Graziano. 2004. Niemann-Pick C1 like 1 Jacobsen, C. N., V. Rosenfeldt Nielsen, A. E. Hayford, P. L. Moller,
protein is critical for intestinal cholesterol absorption. Science K. F. Michaelsen, A. Paerregaard, B. Sandstrom, M. Tvede, and
303:12011204. M. Jakobsen. 1999. Screening of probiotic activities of forty-seven
Angulo, L., E. Lopez, and C. Lema. 1993. Microflora present in kefir strains of Lactobacillus spp. by in vitro techniques and evaluation
grains of the Galician region (north-west of Spain). J. Dairy Res. of the colonization ability of five selected strains in humans. Appl.
60:263267. Environ. Microbiol. 65:49494956.
Banerjee, S., A. Smallwood, A. E. Chambers, and K. Nicolaides. 2003. Klein, G., C. Hallmann, I. A. Casas, J. Abad, J. Louwers, and G.
Quantitative recovery of immunoreactive proteins from clini- Reuter. 2000. Exclusion of vanA, vanB and vanC type glycopep-
cal samples following RNA and DNA isolation. Biotechniques tides resistance in strains of Lactobacillus reuteri and Lactobacillus
35:450452., 454, 456. rhamnosus used as probiotics by polymerase chain reaction and
Borgstrom, B. 1960. Studies on intestinal cholesterol absorption in the hybridization methods. J. Appl. Microbiol. 89:815824.
human. J. Clin. Invest. 39:809815. Klein, G., A. Pack, C. Bonaparte, and G. Reuter. 1998. Taxonomy and
Chen, Y. P., P. J. Hsiao, W. S. Hong, T. Y. Dai, and M. J. Chen. 2012. physiology of probiotic lactic acid bacteria. Int. J. Food Microbiol.
Lactobacillus kefiranofaciens M1 isolated from milk kefir grains 41:103125.
ameliorates experimental colitis in vitro and in vivo. J. Dairy Sci. Liong, M. T., and N. P. Shah. 2005. Acid and bile tolerance and
95:6374. cholesterol removal ability of lactobacilli strains. J. Dairy Sci.
Chiu, C. H., T. Y. Lu, Y. Y. Tseng, and T. M. Pan. 2006. The ef- 88:5566.
fects of Lactobacillus-fermented milk on lipid metabolism in ham- Liong, M. T., and N. P. Shah. 2006. Effects of a Lactobacillus casei
sters fed on high-cholesterol diet. Appl. Microbiol. Biotechnol. symbiotic on serum lipoprotein, intestinal microflora, and organic
71:238245. acids in rats. J. Dairy Sci. 89:13901399.

Journal of Dairy Science Vol. 96 No. 5, 2013

 PROBIOTIC CHARACTERISTICS OF LACTOBACILLUS PLANTARUM Lp27 2825
Liu, J. R., S. Y. Wang, M. J. Chen, H. L. Chen, P. Y. Yueh, and C. W. Niemann-Pick C1-like 1 regulates biliary cholesterol concentra-
Lin. 2006. Hypocholesterolaemic effects of milk-kefir and soyamilk- tions and is a target of ezetimibe. J. Clin. Invest. 117:19681978.
kefir in cholesterol-fed hamsters. Br. J. Nutr. 95:939946. Urdaneta, E., J. Barrenetxe, P. Aranguren, A. Irigoyen, F. Marzo, and
Livak, K. J., and T. D. Schmittgen. 2001. Analysis of relative gene F. C. Ibanez. 2007. Intestinal beneficial effects of kefir-supplement-
expression data using real-time quantitative PCR and the 2(-Delta ed diet in rats. Nutr. Res. 27:653658.
Delta C (T)) method. Methods 25:402408. Usman, , and A. Hosono. 1999. Bile tolerance, taurocholate deconjuga-
Manson, J. E., H. Tosteson, P. M. Ridker, S. Satterfield, P. Hebert, tion, and binding of cholesterol by Lactobacillus gasseri strains. J.
G. T. OConnor, J. E. Buring, and C. H. Hennekens. 1992. The Dairy Sci. 82:243248.
primary prevention of myocardial infarction. N. Engl. J. Med. Vinderola, C. G., J. Duarte, D. Thangavel, G. Perdigon, E. Farn-
326:14061416. worth, and C. Matar. 2005. Immunomodulating capacity of kefir.
Nguyen, T. D., J. H. Kang, and M. S. Lee. 2007. Characterization J. Dairy Res. 72:195202.
of Lactobacillus plantarum PH04, a potential probiotic bacterium Wang, Y., N. Xu, A. Xi, Z. Ahmed, B. Zhang, and X. Bai. 2009. Ef-
with cholesterol-lowering effects. Int. J. Food Microbiol. 113:358 fects of Lactobacillus plantarum MA2 isolated from Tibet kefir
361. on lipid metabolism and intestinal microflora of rats fed on high-
Ooi, L. G., and M. T. Liong. 2010. Cholesterol-lowering effects of cholesterol diet. Appl. Microbiol. Biotechnol. 84:341347.
probiotics and prebiotics: A review of in vivo and in vitro findings. Weisburg, W. G., S. M. Barns, D. A. Pelletier, and D. Lane. 1991. 16S
Int. J. Mol. Sci. 11:24992522. ribosomal DNA amplification for phylogenetic study. J. Bacteriol.
Pospiech, A., and B. Neumann. 1995. A versatile quick preparation 173:697703.
of genomic DNA from Gram-positive bacteria. Trends Genet. Wilson, M. D., and L. L. Rudel. 1994. Review of cholesterol absorp-
11:217218. tion with emphasis on dietary and biliary cholesterol. J. Lipid
Rossetti, L., D. Carminati, M. Zago, and G. Giraffa. 2009. A qualified Res. 35:943955.
presumption of safety approach for the safety assessment of Grana Yin, Y. N., Q. F. Yu, N. Fu, X. W. Liu, and F. G. Lu. 2010. Effects of
Padano whey starters. Int. J. Food Microbiol. 130:7073. four bifidobacteria on obesity in high-fat diet induced rats. World
Silva, K. R., S. A. Rodrigues, L. Xavier, and A. S. Lima. 2009. An- J. Gastroenterol. 16:33943401.
timicrobial activity of broth fermented with kefir grains. Appl. Yoon, H. S., J. H. Ju, H. N. Kim, H. J. Park, Y. Ji, J. E. Lee, H. K.
Biochem. Biotechnol. 152:316325. Shin, M. S. Do, and W. Holzapfel. 2013. Reduction in cholesterol
Temel, R. E., J. K. Sawyer, L. Yu, C. Lord, C. Degirolamo, A. McDan- absorption in Caco-2 cells through the down-regulation of Nie-
iel, S. Marshall, N. Wang, R. Shah, L. L. Rudel, and J. M. Brown. mann-Pick C1-like 1 by the putative probiotic strains Lactobacil-
2010. Biliary sterol secretion is not required for macrophage re- lus rhamnosus BFE5264 and Lactobacillus plantarum NR74 from
verse cholesterol transport. Cell Metab. 12:96102. fermented foods. Int. J. Food Sci. Nutr. 64:4452.
Temel, R. E., W. Tang, Y. Ma, L. L. Rudel, M. C. Willingham, Y.
A. Ioannou, J. P. Davies, L. M. Nilsson, and L. Yu. 2007. Hepatic

Journal of Dairy Science Vol. 96 No. 5, 2013