|Home | About | Journals | Submit | Contact Us | Français|
This article identifies novel factors involved in cholesterol reduction by probiotic bacteria, which were identified using genetic and proteomic approaches. Approximately 600 Lactobacillus acidophilus A4 mutants were created by random mutagenesis. The cholesterol-reducing ability of each mutant was determined and verified using two different methods: the o-phthalaldehyde assay and gas chromatographic analysis (GC). Among screened mutants, strain BA9 showed a dramatically diminished ability to reduce cholesterol, as demonstrated by a 7.7% reduction rate, while the parent strain had a more than 50% reduction rate. The transposon insertion site was mapped using inverse PCR (I-PCR), and it was determined using bioinformatic methods that the deleted region contained the Streptococcus thermophilus catabolite control protein A gene (ccpA). In addition, we have shown using two-dimensional gel electrophoresis (2-DE) that several proteins, including a transcription regulator, FMN-binding protein, major facilitator superfamily permease, glycogen phosphorylase, the YknV protein, and fructose/tagatose bisphosphate aldolase, were strongly regulated by the ccpA gene. In addition, in vivo experiments investigating ccpA function were conducted with rats. Rats fed wild-type L. acidophilus A4 showed a greater than 20% reduction in total serum cholesterol, but rats fed BA9 mutant L. acidophilus showed only an approximately 10% reduction in cholesterol. These results provide important insights into the mechanism by which these lactic acid bacteria reduce cholesterol.
The ability to reduce serum cholesterol is a highly desirable attribute of probiotic cultures as a dietary component. Despite the fact that numerous studies have shown that Lactobacillus acidophilus can reduce serum cholesterol levels, neither the mechanisms of this process nor the molecules responsible for this activity are well understood. Cholesterol is an extremely important biological molecule that plays several vital roles, including serving as a precursor for synthesis of steroid hormones and bile acids, maintaining membrane structure, and insulating nerves (30); however, a high-fat diet, which is becoming increasingly more common, will lead to an accumulation of cholesterol in the human body. Excessively high levels of cholesterol have been shown to play a major role in human cardiovascular pathogenesis and cardiovascular disease, including coronary heart disease and stroke. The 2005 Dietary Guidelines for Americans Index ranked cardiovascular disease as the leading cause of death among adult Americans (10). High cholesterol has also has been associated with an increased risk of metabolic syndrome symptoms, including abdominal obesity (large waist circumference), hyperglycemia, hypertriacylglycerolemia, low-high-density lipoprotein (HDL) cholesterol, and hypertension, by as much as 3-fold (17). Since high serum cholesterol levels have been clearly shown to be a leading risk factor for human disease, more and more studies have focused on examining the effects of cholesterol on human health.
For the past several decades, there has been increasing interest in understanding the mechanisms of cholesterol reduction by probiotics. Fermented milk containing probiotics was first shown to exhibit hypocholesterolemic effects in humans in 1963 (32, 49). Since then, it has been repeatedly and consistently shown that some species of lactic acid bacteria, especially lactobacilli, can lower total cholesterol and low-density lipoprotein (LDL) cholesterol (2, 13, 14, 15, 47, 53, 54); however, the exact mechanisms by which these probiotic bacteria reduce total cholesterol levels are still unclear.
This study investigated factors involved in serum cholesterol reduction by L. acidophilus using a mutant that had decreased cholesterol reduction ability. Proteomic and genetic differences between the mutant and wild-type strains were examined. In addition, an in vivo study was conducted to examine differences in serum cholesterol reduction in rats that ate the mutant or wild-type bacteria. Until now, there have been few reports investigating cholesterol reduction by lactobacilli using proteomic and genetic analysis; therefore, the results of this study may provide vital insights into the mechanisms of probiotic cholesterol reduction.
L. acidophilus strain A4 (23), used in this study, was obtained from the Korean Collection for Type Cultures (KCTC, South Korea). The strain was cultured in de Man, Rogosa, and Sharpe (MRS) broth (Difco, Sparks, MD) at 37°C for 18 h. Stock cultures were stored at −80°C, using 50% glycerol as a cryoprotectant, and the strain was subcultured three times prior to use.
The plasmid pMarA expresses Himar1 under the transcriptional control of the Bacillus subtilis housekeeping sigma factor (σA) for transposase expression. This plasmid also contains a kanamycin resistance transposable element but not the transposase and a temperature-sensitive replicon (28). The plasmid was obtained from the Bacillus Genetic Stock Center at the Department of Biochemistry, Ohio State University, 484 West 12th Ave., Columbus, OH 43210 (www.bgsc.org).
Electrocompetent cells for L. acidophilus were prepared as described by Kim et al. (22), and the random mutagenesis was performed as described by Le Breton et al. (28) with minor modification. Briefly, 2 μl pMarA plasmid (200 ng) was added to 50 μl ice-cold competent cell suspension (ca. 108 CFU/ml) in a disposable cuvette (Cuvette Plus; inner-electrode gap, 1.0 cm; Genetronics, San Diego, CA) and held on ice for 5 min. The mixture was then subjected to electroporation using a BTX 830 instrument (Genetronics) with the following conditions: 1.25-kV pulse strength (V/cm), 240-μs pulse length, 20 pulse number, and 500-ms pulse interval. After the pulse, the cell suspension was diluted to a volume of 1 ml in MRS broth and incubated at 37°C for 3 h. At the end of this expression period, cells were plated on MRS agar plates containing kanamycin (500 μl/ml) and incubated for 48 h. Transformed colonies were then assessed for plasmid-associated properties (28), i.e., kanamycin resistance (Kanr) and erythromycin resistance (Ermr) at the permissive temperature for plasmid replication (37°C) and Kanr and Erms at the restrictive temperature (42°C). Final isolates were then plated on MRS agar containing kanamycin and incubated at 42°C. The insertional colonies were then picked up randomly and used to establish mutant libraries, which were stored in MRS and 50% glycerol as a cryoprotectant. The mutant libraries were stored at −80°C, and strains were subcultured three times prior to use.
A quantitative assay to assess cholesterol removal was conducted as described by Kimoto et al. (24) with minor modifications. Freshly prepared MRS-THIO broth (MRS broth containing 0.2% sodium thioglycolate) (Sigma) was supplemented with 0.2% sodium taurocholate (Sigma) as a bile salt. The sodium thioglycolate functioned as an oxygen scavenger (5, 53). Fresh, filter-sterilized cholesterol solution (10 mg/ml in ethanol) was added to the broth to a final concentration of 100 g/ml. The broth was then inoculated with 1% culture (ca. 109 CFU/ml) and incubated for 20 h at 37°C. After incubation, cells were removed by centrifugation at 5,400 × g for 7 min. Then, total cholesterol in the spent broth was measured using the colorimetric method reported by Rudel and Morris (44) with minor modifications. Uninoculated sterile broth was also analyzed as a negative control.
Gas chromatographic analysis (GC) was used in addition to the colorimetric assay to more precisely measure cholesterol. The AOAC official method 994.10, the Cholesterol in Foods method (3), was used in this study. An Agilent (Wilmington, DE) HP 5890 GC-flame ionization detector (FID) system equipped with a FID and a 30-m, 0.25-mm (inside diameter) fused silica capillary column (HP5; 0.25-mm film thickness) were used for GC. GC operating conditions were as follows: injection port temperature, 250°C; detector temperature, 300°C; initial temperature, 160°C; hold for 2 min; increase of 20°C/min to 230°C; hold for 3 min; increase of 40°C/min to 255°C; hold for 25 min; flow rate, helium column, ca. 2 ml/min; injection mode, splitless. After setting the GC operating conditions, the sample was introduced into the injection port of the GC.
Mapping of transposon insertion sites was performed as described by Le Breton et al. (28) with minor modifications. Chromosomal DNA was digested with TaqI and then circularized in a ligation reaction using T4 DNA ligase buffer (Promega) at a DNA concentration of 5 ng/μl. The ligation product was purified using a DNA purification kit (Promega) and resuspended in Tris-EDTA (TE) buffer at 10 ng/μl. Inverse PCRs were performed on aliquots of the ligation mixtures using the primers oIPCR1 (5′-GCTTGTAAATTCTATCATAATTG-3′) and oIPCR2 (5′-AGGGAATCATTTGAAGGTTGG-3′) (28). The reactions consisted of 30 cycles with an annealing temperature of 40°C (30 s) and an extension temperature of 72°C (30 s). PCR products were purified and concentrated during gel extraction.
The PCR products were then sequenced using the pGEM-T Easy vector system (Promega) as follows: the plasmid vector pGEM-T Easy was ligated with the amplified DNA fragments obtained above. To ensure that the DNA fragments were ligated into the vector, transformants were transferred into competent cells as a plasmid. After incubation, several colonies that appeared to be transformed were selected. Plasmid DNA from several colonies was then extracted from competent cells using the Wizard Plus Minipreps DNA purification kit (Promega), and the plasmid, including the insertion site, was sequenced using a pUC/M13 forward primer (24-mer). PCR and gene manipulation were conducted using the protocols described by Sambrook et al. (46). DNA sequencing and sequence analysis of the flanking regions were then conducted. All sequence analyses and DNA homology searches were conducted using the NCBI database (http://www.ncbi.nlm.nih.gov/).
Bacterial cells were grown for 20 h at 37°C in 100 ml MRS-THIO medium containing 0.2% taurocholate with or without cholesterol solution (100 g/ml) until they reached an optical density at 595 nm (OD595) of 1.0 (23). Each cell culture was centrifuged (4°C, 8,000 × g, 30 min) and washed three times with phosphate buffered-saline (PBS) buffer. Then, cells were resuspended in 5 ml lysis buffer (40 mM Tris-HCl, 1% Triton X-100, 1 mM EDTA, 1 mM MgSO4·7H2O; pH 8.0) containing 1 mM phenylmethylsulfonyl fluoride (PMSF) and lysed by sonication (Hielscher GmbH, Teltow, Germany) on ice 30 times for 30 s at an 80% pulse duration. Following sonication, the total protein was extracted twice in phenol, as described previously (4), and protein concentrations were measured by the Bradford method (43).
To perform the first-dimension electrophoresis, each sample was mixed with rehydration solution (7 M urea, 2% 3-[(3-cholamidopropyl)-dimethylammonio]-1-propanesulfonate [CHAPS], 50 mM dithiothreitol [DTT], 10% isopropanol, 5% glycerol, 0.5% ampholytes at a pH of 3 to 10, and a trace amount of bromophenol blue), resulting in a final protein volume of 500 μg in a total volume of 300 μl. The samples were then rehydrated on ReadyStrip IPG strips (17 cm, pH 3 to 10, nonlinear; Bio-Rad) and covered with mineral oil for 15 h at 20°C. Isoelectric focusing (IEF) was conducted using a Protean IEF cell (Bio-Rad) according to the manufacturer's instructions using the following conditions: 50 V for 20 min, 250 V for 30 min, 500 V for 30 min, 1,000 V for 1 h, 2,000 V for 1 h, 8,000 V for 2 h, 8,000 V for 30,000 V·h, and 500 V for 20 min. After IEF separation, the strips were incubated for 10 min in equilibration buffer I (6 M urea, 2% sodium dodecyl sulfate [SDS], 20% glycerol, 130 mM DTT, and 0.375 M Tris-HCl, pH 8.8) and then for an additional 15 min in equilibration buffer II (6 M urea, 2% SDS, 20% glycerol, 135 mM idoacetamide, and 0.375 M Tris-HCl, pH 8.8). After equilibration, the strips were transferred to 12.5% SDS-polyacrylamide gels (20 by 22 cm) for second-dimension electrophoresis. Separation was achieved using a Protean II xi system (Bio-Rad) at 10 mA per gel for 1 h and thereafter at 20 mA per gel at 4°C. Protein spots were visualized by blue-silver staining (6). The 6 stained gels were then scanned with a densitometric scanner (800 by 1,600 dpi, Uta 2100XL; Umax, Techville, Inc., Dallas, TX), and the spot images were analyzed with PDQuest software (Bio-Rad) according to the manufacturer's instructions. Only changes in spot intensity greater than 3.0-fold were considered to be different and were selected for further analysis.
Protein spots were enzymatically digested in gel as described by Shevchenko et al. (50), using modified porcine trypsin (Promega). Gel pieces were washed with 50% acetonitrile to remove SDS, salt, and stain. The washed and dehydrated spots were then vacuum dried to remove solvent, rehydrated with trypsin solution (8 to 10 ng/μl) in 50 mM ammonium bicarbonate (pH 8.7), and incubated for 8 to 10 h at 37°C. Derivatization reactions were carried out as described by Wang et al. (58). Reagent solution was prepared by dissolving in 20 mM NaHCO3 (pH 9.5). The sulfonation reaction was carried out in a 0.6-ml tube by mixing 8.5 μl of reagent solution with 0.5 μl of tryptic digest. After incubation for 5 to 30 min at 55°C, the reaction was terminated by adding 1 μl of 5% trifluoroacetic acid (TFA). The sample was then loaded into a micropipette tip (C18 ZipTip; Millipore, Bedford, MA), washed three times with 10 μl 0.1% TFA, and eluted with 1 μl 50% acetonitrile-0.1% TFA. For protein identification, the fragment masses obtained from chemically assisted fragmentation-matrix-assisted laser desorption/ionization (CAF-MALDI) were identified using Sonar in the Ettan MALDI-time of flight (TOF) software or a protein identification search engine, PepFrag. Amino acid sequences of the peptides were obtained by determining the distances between consecutive peaks (y ions) in the PSD spectrum. The amino acid sequence was used to identify the proteins through a homology search using ProteinInfo (Proteometrics) or through a BLAST search using the ExPASy molecular biology server (www.expasy.ch).
Forty male Sprague-Dawley rats (NTacSam:SD; Samtako Bio, South Korea) were obtained at 10 weeks of age, with an initial average weight of 225 g. The animals were individually housed in mesh-bottom stainless steel cages in a temperature-controlled room (22 ± 2°C), with 12-h light-12-h dark lighting cycles. Rats had ad libitum access to water. A total of four treatment groups (n = 7) were fed the following experimental diets: normal diet (ND group), high-cholesterol diet (HCD group), high-cholesterol diet plus L. acidophilus A4 (wild-type group), and high-cholesterol diet plus BA9 mutant (mutant group). At the start of the experiment, average body weight did not differ among the four groups.
Freeze-dried L. acidophilus A4 and BA9 mutant cells were freshly prepared before rats were fed. The cultured strains were added to lactobacillus MRS broth (Difco, Sparks, MD) to a final culture concentration of 1% and incubated at 37°C for 18 h. Cultured cells were harvested by centrifugation at 2,000 × g for 20 min and washed three times with PBS. The cell pellets were suspended in 10% skim milk (used as a cryoprotectant). The suspended cells were frozen at −80°C overnight and then dried under vacuum for 48 h in a freeze-dryer. This process was performed each week. Body weight, food intake, and feeding efficiency did not differ among the groups over 6 weeks of observation.
Blood samples were collected by cardiac puncture after administration of ketamine-xylazine (90 mg per kg of body weight/10 mg per kg) each week. Approximately 1 ml of blood was taken from each rat, transferred to a 1.5-ml tube, and kept on ice for 30 min. Tubes were centrifuged at 2,000 × g for 20 min (4°C), serum was collected, and total cholesterol, high-density lipoprotein (HDL) cholesterol, low-density lipoprotein (LDL) cholesterol, and triglycerides were measured using commercially available kits (40).
Statistical analysis was performed using the SAS statistical software package (version 9.1; SAS Inc., Cary, NC). Analysis of variance (ANOVA) (α < 0.05), followed by Duncan's multiple-range (α = 0.05) test, was used to assess differences among groups. The level of significance was defined as an α value of <0.05.
Previously we found that L. acidophilus A4 had the highest cholesterol-lowering activity out of 400 selected probiotic strains tested (23). To explore the effects of genes on cholesterol reduction activity in L. acidophilus A4, we used random mutagenesis. The cholesterol-reducing ability of approximately 600 L. acidophilus A4 random mutants was assessed and verified using the o-phthalaldehyde assay (24), and these results were confirmed by GC (3). In the colorimetric assay, L. acidophilus A4 and most of its 600 mutants were found to reduce cholesterol in the media by approximately 50%; however, one colony, BA9, reduced the amount of cholesterol in the media by only 7.7% (Fig. (Fig.1A).1A). This difference was confirmed by GC using a flame ionization detector (FID), which showed greatly attenuated cholesterol reduction in the BA9 mutant (Fig. (Fig.1B).1B). From these two analyses, it was clear that the BA9 mutant had an impaired ability to reduce cholesterol in the medium. There were no differences in biomasses or growth rates in the wild-type and BA9 strains (data not shown). Other probiotic characteristics of the BA9 mutant strain were also assessed. Acid tolerance, bile acid tolerance, and adhesion in HT-29 cells were examined. Significantly, these additional experiments showed that the other probiotic properties of the BA9 mutant strain were not significantly different from those of wild-type L. acidophilus A4 (data not shown). Therefore, with the exception of cholesterol reduction, the probiotic properties of wild-type L. acidophilus A4 appeared to be conserved in the BA9 mutant.
To identify the genes responsible for cholesterol reduction in L. acidophilus A4, we mapped the transposon insertion sites of the BA9 mutant, which had reduced cholesterol-lowering activity. The insertion sites were sequenced by inverse PCR, and we concluded that the BA9 mutant had two flanking regions. Based on sequence analysis, we found that one flanking region in the insertion was 99% homologous with the Streptococcus thermophilus catabolite control protein A (ccpA) gene. CcpA was previously shown to be expressed in a number of low-GC Gram-positive bacteria, including lactic acid bacteria (61). The ccpA gene is known to be important for carbon catabolism and its regulation and to be involved in a broad spectrum of novel regulatory processes in many Gram-positive bacteria, including lactic acid bacteria (26, 27, 34, 52). Also, it has been shown that in some lactic acid bacteria, ccpA not only activates carbon catabolism but also activates expression of FhyR (heme uptake regulator) regulatory protein (12), PepQ, which regulates nitrogen metabolism (31, 35), glycerol transport facilitator (GlpF), and glycerol kinase (GlpK) (7). Zomer et al. (61) constructed a Lactococcus lactis ccpA mutant and analyzed the wide range of COG (clusters of orthologous groups of proteins) categories that were significantly affected by deletion of ccpA. Interestingly, these included lipid metabolism, cell envelope biogenesis, outer membrane, intracellular trafficking and secretion, and other, unknown functions.
To explore the impact of the specific protein encoded by the ccpA gene, differences in protein expression profiles between the L. acidophilus A4 parent strain and the BA9 mutant were analyzed using 2-DE and CAF-MALDI. Protein expression was measured for mutant and wild-type L. acidophilus in the absence or presence of bile and/or cholesterol. It has previously been widely reported that bile is strongly correlated with cholesterol reduction by lactic acid bacteria (15, 24, 41, 55), but there were no previous reports on the mechanism of this association at the protein level.
A total of six gels were analyzed. Approximately 300 protein spots in each gel were visualized by Coomassie brilliant blue (CBB) G-250 staining (Fig. (Fig.2).2). Interestingly, 8 protein spots showing significantly different levels of protein displayed parallel responses to the presence of bile and/or cholesterol with bile in wild-type L. acidophilus A4 and the BA9 mutant. The addition of bile and/or cholesterol with bile to the medium increased the brightness of spots 1, 2, and 3 on both gels. However, spots 2 and 3 for the BA9 mutant were significantly less bright only in the presence of cholesterol with bile. Although the appearance of spots 4, 5, and 6 was not significantly affected by the addition of bile and/or cholesterol with bile, their expression in the BA9 mutant was downregulated in the presence of cholesterol with bile alone.
These eight spots were identified by CAF-MALDI (Table (Table1).1). Of the eight spots, the following six proteins were identified: transcription regulator (spot 1), flavin mononucleotide (FMN)-binding protein (spot 2; 6-fold decreased), major facilitator superfamily permease (spot 3; 5.4-fold decreased), glycogen phosphorylase (spot 4; 3.4-fold decreased), the YknV protein (spot 5; 2.9-fold decreased), and fructose/tagatose bisphosphate aldolase (spot 6; 2.4-fold decreased). Unexpectedly, two of the spots could not be identified in this study.
Numerous studies have suggested that cholesterol either is incorporated into L. acidophilus or adheres to the bacterial cell membrane. Incorporated cholesterol would be less available for absorption from the intestine into the blood, and the assimilated cholesterol would be excreted with the fecal matter (13, 14). Noh et al. (36) arrived at this hypothesis by postulating that cholesterol, which can be incorporated into the cells of L. acidophilus, alters the cell membrane or cell wall of the organism. Consequently, we hypothesized that the membrane-associated proteins identified in our proteomic study might play an important role in this probiotic cholesterol reduction process.
FMN-binding protein (spot 2) is expressed in some bacterial strains, where it acts as an electron transport carrier, and it may be involved in redox systems (25). Flavin is expressed in relatively high concentrations in dairy products (16), and its effects on oxidation reduction are very sensitive to pH, favoring an acidic environment (33, 60). These effects of FMN-binding protein and flavin have been shown in several studies, and particularly in Lactobacillus strains (16). Generally, FMNs function as coenzymes for the flavoproteins of flavoenzymes and are essential factors for the metabolism of carbohydrates, amino acids, and lipids. In addition, FMN-complexed proteins act as intermediaries in the transfer of electrons in biological oxidation-reduction reactions. Interestingly, in some bacteria, flavin adenine dinucleotide (FAD) and FMN act as cholesterol oxidases (9, 33), and other studies have shown that cholesterol oxidases from numerous bacteria tightly bind the flavin prosthetic group (11, 20, 21, 37, 56). In addition, the major facilitator superfamily (MFS) protein (spot 3), which was downregulated in the BA9 mutant in the presence of cholesterol with bile, is the largest family of secondary transport carriers found in living organisms, including bacteria. Its function as a transporter is very diverse and includes the transport of sugars, drugs, metabolites, amino acids, anions, and other major facilitators (38, 45). The substrate translocation mechanisms of MFS are thought to be associated with the alternating accessibility of the substrate-binding site to the other surfaces of the membrane (1, 59).
Importantly, a range of functional categories may be affected by ccpA mutations in lactic acid bacteria, including lipid metabolism, inorganic ion transport and metabolism, cell envelope biogenesis, outer membrane and intracellular trafficking and secretion, and carbohydrate transport and metabolism (61). Recent studies have identified ccpA-regulated cell surface proteins in bacterial strains (18, 51). Interestingly, fructose-1,6-bisphosphate aldolase was shown to be downregulated in a ccpA-deleted strain and was thus classified as a ccpA-dependent cell wall-associated protein (18). Consistent with this, our proteomic study showed downregulation of fructose/tagatose bisphosphate aldolase (spot 6) in the ccpA-deleted BA9 mutant. Ling et al. (29) also showed that fructose-1,6-bisphosphate aldolase, a cytoplasmic glycolytic enzyme, is cell wall associated. In addition, downregulation of glycogen phosphorylase (spot 4), which was observed in the BA9 mutant in the presence of cholesterol with bile, was found to be associated with HPr in ccpA-dependent carbon catabolite regulation in certain bacteria (48, 52).
The YknV protein (spot 5), which was also downregulated in the BA9 mutant in the presence of cholesterol with bile, was also found to be an important multipass membrane protein, especially in the ABC transport system (42, 57), but its function is not fully understood. Taken together, this evidence indicates that all of the downregulated proteins are related to the ccpA gene and/or carbohydrate catabolism. Based on these proteomic and genetic studies, the absent or downregulated factors in the BA9 mutant appear to be connected, and this connection may be related to the bacterial cell membrane. Therefore, it seems reasonable to conclude that the ccpA gene and the proteins identified in this study might affect cholesterol reduction in L. acidophilus A4 by cell membrane modulation. However, the results of this experiment still need to be verified by further targeted gene mutation experiments.
To further investigate the effects of the ccpA gene and specific proteins regulated by ccpA, experiments were conducted with rats to assess differences in cholesterol metabolism in animals fed L. acidophilus A4 or the BA9 mutant. The experimental animals appeared healthy and had glossy fur throughout the study. Body weight and feeding efficiency did not differ among the different dietary treatment groups (data not shown), with the exception of the high-cholesterol diet group, which had a slightly higher average body weight. Hypercholesterolemia in rats was confirmed by measurement of total cholesterol serum levels. Total serum cholesterol levels significantly increased after 14 days of feeding with the high-cholesterol diet (α < 0.05).
The differences between treatment groups during the 6-week study period are shown in Table Table2.2. Blood samples from rats fed L. acidophilus A4 (wild-type group) or the BA9 mutant (mutant group) were used to measure serum levels of total cholesterol, LDL cholesterol, and HDL cholesterol.
Total serum cholesterol (Table (Table2)2) was significantly different in the ND and HCD groups over 4 weeks. The important thing is that total serum cholesterol in the wild-type group, which was fed wild-type L. acidophilus A4, was reduced by approximately 25%. In contrast, total cholesterol in the mutant group, which was fed the BA9 mutant, was not reduced by the same amount as in the wild-type group. Moreover, there was a significant increase in cholesterol levels after 2 weeks in the mutant group, which was not observed in the wild-type group (α < 0.05). In addition, there was also a slight increase in LDL cholesterol in the HCD groups after a 2-week hypercholesterolemia induction period. After 2 more weeks, differences in LDL cholesterol were observed between the wild-type and mutant groups (α < 0.05) (Table (Table2).2). However, HDL cholesterol was only slightly increased in each treatment group over the 4 weeks of the experiment (Table (Table22).
These in vivo studies revealed that there was a reduction in total serum cholesterol in rats fed L. acidophilus A4 (wild-type group) (25%). In contrast, there was much less reduction in total serum cholesterol in rats fed the BA9 mutant. Moreover, there were minor differences in the HDL cholesterol and LDL cholesterol between these two treatment groups.
Over the past few decades, a considerable number of studies have been conducted on the effects of probiotics on serum cholesterol. It has been reported that probiotic bacteria and dairy products can modulate serum cholesterol levels in many animal models of induced hypercholesterolemia (8, 19, 39), but the mechanism responsible for this phenomenon is still not clear. The aim of these in vivo experiments was to establish that the microbial molecular factors identified in the in vitro experiments would affect serum cholesterol in vivo. We compared serum cholesterol levels in rats fed wild-type and BA9 mutant L. acidophilus and found that the BA9 mutant did not lower serum cholesterol levels as much as the wild-type strain. These results lead us to the conclusion that the modified genetic and proteomic factors identified in the in vitro studies may be responsible for the lowered cholesterol levels and/or modulated lipid metabolism observed in vivo. However, further studies will be needed to verify this conclusion.
To the best of our knowledge, this is the first report that ccpA, which encodes catabolite control protein A, plays an important role in cholesterol reduction by probiotic bacteria. In addition, very few studies have systematically examined the cholesterol-reducing properties of lactic acid bacteria using proteomic analysis. This study has demonstrated that specific genetic and proteomic mutations in a probiotic microorganism can alter serum cholesterol levels in an animal model.
This study was supported by the Technology Development Program for Agriculture and Forestry, Ministry for Food, Agriculture, Forestry and Fisheries, Republic of Korea (108141-3), and a Korea Science and Engineering Foundation (KOSEF) grant funded by the South Korean government (MEST) (20090083882).
Published ahead of print on 21 May 2010.