Lactobacillus plantarum 06CC2 reduces hepatic cholesterol levels and modulates bile acid deconjugation in Balb/c mice fed a high‐cholesterol diet

Abstract Previous study suggested that dietary intake of Lactobacillus plantarum 06CC2 (LP06CC2) isolated from Mongolian dairy products showed various health beneficial effects. Here, the effect of LP06CC2 on the cholesterol metabolism in mice fed a cholesterol‐loaded diet was evaluated. Cholesterol and LP06CC2 were incorporated into the AIN93G‐based diet to evaluate the effect on cholesterol metabolism in Balb/c mice. Serum and liver cholesterol levels were significantly increased in mice fed a cholesterol‐loaded diet whereas the LP06CC2 ingestion suppressed the increase of liver cholesterol. LP06CC2 suppressed the increase of the hepatic damage indices. The increase of the cecal content and fecal butyrate were observed in mice fed LP06CC2. The analysis of bile acids clearly showed that LP06CC2 increased their deconjugation indicating the decrease of bile acid absorption. The protein expression of hepatic Cyp7A1 was also suppressed by LP06CC2 in mice fed cholesterol. Finally, in vitro studies showed that LP06CC2 had the most potent ability to deconjugate bile acids using glycocholate among the tested probiotic lactic acid bacteria isolated from Mongolian dairy products. Taken together, LP06CC2 is a promising microorganism for the reduction of the cholesterol pool via modulation of bile acid deconjugation.

serum and liver lipid profiles indicating the putative usefulness of the LAB application for the prevention of lipid disorders and cardiovascular disease (Kurajo et al., 2018). Notably, the cholesterol-lowering effect was produced by various LAB through different molecular mechanisms. Therefore, LAB administration is expected to be a promising way to prevent or improve the cholesterol metabolism-related disorders. Among LAB, certain strains of Lactobacillus plantarum are shown to be useful for the reduction of the risk associated with cholesterol disorder in mice (Qu et al., 2020;Wang et al., 2019) and humans (Bukowska et al., 1998;Fuentes et al., 2013;Higashikawa et al., 2010). Several molecular mechanisms of this effect have been proposed as follows; assimilation of cholesterol (Dora & Glenn, 2002;Lim et al., 2017;Michael et al., 2016), promotion of bile acid or cholesterol excretion (Li et al., 2014;Qu et al., 2020;Zhai et al., 2019), and manipulation of hepatic cholesterol metabolism (Kim et al., 2014;Qu et al., 2020) In the context of these findings, it is of note that recent systematic study in L. plantarum showed ability to deconjugate bile acids (Prete et al., 2020). The apical sodium-dependent bile acid transporter (ASBT)/ileal bile acid transporter (IBAT)/solute carrier family 10 member 2 (SLC10A2) is highly expressed in the small intestine, and most of the conjugated bile acids secreted into the gastrointestinal lumen are absorbed through these transporters (Aguiar Vallim et al., 2013). About 95% of the bile acids secreted into the gastrointestinal lumen are recycled by the enterohepatic circulation, and some bile acids enter the lower gastrointestinal tract. Bile salt hydrolase of the intestinal bacteria removes amino acids from the conjugated bile acids and they escape uptake by ASBT/SLC10A2, reducing the bile acid pool in the body (Pavlović et al., 2012). Our previous study showed that L. plantarum 06CC2 (LP06CC2) isolated from Mongolian dairy products was able to alleviate influenza infection (Takeda et al., 2011) and stimulate the induction of helper type-1 T cells (Takeda et al., 2013). In contrast, the effect of LP06CC2 administration on the cholesterol metabolism has not been evaluated. The aim of this study was to evaluate the cholesterol-lowering effect of LP06CC2 on mice fed a cholesterol-loaded diet.

| Preparation of LP06CC2
Lactobacillus plantarum 06CC2 strain (LP06CC2), a potential probiotic from Mongolian dairy products, tolerates bile and gastric acids and adheres to Caco-2 cells. LP06CC2 was cultured at 37°C for 18 hr in de Man, Rogosa, and Sharpe (MRS) broth (Merck). Microorganisms were harvested by centrifugation at 1,500 g for 5 min, washed twice with phosphate-buffered saline, and lyophilized. The lyophilized LP06CC2 was stored at −30°C until the preparation of experimental diets.

| Mice and diet
Studies were conducted using 5-week-old male Balb/c mice purchased from Japan SLC, Inc. and maintained at 22°C in a humidity-controlled room with a 12-hr light-dark cycle. All mice were acclimatized for 1-week and assigned to four groups (n = 6 each).
Mice were fed the AIN-93G-based normal diet (ND), a high-cholesterol diet (HCD) including 0.50% cholesterol and 0.25% sodium cholate, ND with the addition of LP06CC2 (5 wt%), or HCD with the addition of LP06CC2 (5%) for 3 weeks. Detailed compositions are shown in Table 1. Diets were vacuum-packed and stored at 4°C.
The mice had free access to the corresponding diet and water. The body weights and food intake were recorded every other day. The feces were collected from each mouse every other day and pre-
The membrane was washed three times and then treated with the secondary antibodies anti-rabbit IgG-HRP or anti-mouse IgG-HRP (Cell Signaling Technology) for 1 hr at 25°C. Signals were visualized using the ECL Western blotting substrate (Bio-Rad). Subsequently, the band intensities were determined using the chemiluminescent imaging system, LAS-4000 (Fujifilm). The intensity of the bands was normalized using a corresponding β-actin band as an internal control. The band intensities of Cyp7A1 were normalized with those of β-actin used as an internal control.

| LC/MS analysis of bile acids
Bile acids in feces were measured according to the method of Hagio et al. (2009). Feces were freeze-dried and ground thoroughly. One milliliter of ethanol was added to 100 mg of the ground samples to extract bile acids. Nordeoxycholic acid (NDCA, 25 nmol) was added as an internal standard to each sample. The samples were subjected to sonication and then heated at 60°C for 30 min in a water bath.
The samples were cooled through immersion in cold running water, heated in boiling water for 3 min, and centrifuged at 1,600 g for 10 min at 15°C. The supernatants were then collected, and the pellets were washed thrice with ethanol and centrifuged at 11,200 g for 1 min to collect supernatants. The pooled extracts were evaporated and resolved into 1 ml of methanol followed by purification using Ultrafree-MC-HV centrifugal filter units (Millipore) for LC/ MS analysis. Liquid chromatography (LC) separation was conducted using the ACQUITY UPLC H-class system (Waters) equipped with an ACQUITY UPLC HSS T3 column (1.7 μm, 100 mm × 2.1 mm; Waters) and maintained at 40°C. Solvent A was water containing 0.1% formic acid and solvent B was acetonitrile containing 0.1% formic acid.
Elution was conducted with a linearly increasing concentration gradient of acetonitrile at 0.4 ml/ml. The acetonitrile concentration was increased from 25% to 27% for 4 min, from 27% to 35% for 2 min, from 35% to 45% for 9 min, from 45% to 70% for 4 min, and from 70% to 100% for 0.5 min. The m/z values for each bile acid are shown in Table S1.

| LAB deconjugation of bile acids
Ten strains of LABs isolated from Mongolian dairy products (2.0 × 10 5 cfu) were seeded and cultured in MRS broth containing 0.2% glycocholate at 37°C for 20 hr. At the end of culture, the pH was adjusted to 7.0 with 5N NaOH. Then, 1 ml of the supernatant was collected and mixed with 2 ml of ethyl acetate. The solution was vigorously mixed and centrifuged at 2,300 g for 20 min at 4°C.
The ethyl acetate layer was collected and dried up under a nitrogen stream, then 0.25 ml of 0.01 N NaOH, 0.25 ml of 1% furfural, and 1.5 ml of 16 N H 2 SO 4 were added, and the reaction proceeded at 65°C for 13 min. Finally, 1.25 ml of acetic acid was added and the concentration of free bile acids was determined measuring the optical density at 660 nm.

| Statistical analyses
All data are represented with the mean ± SE. All statistical analyses were performed using 4-steps Excel Statistics (OMS). Post hoc tests were performed after a two-way ANOVA. When significant interaction (p < .05) was detected, the Tukey-Kramer test was conducted to evaluate the significant differences among dietary groups.

| Growth parameters
LP06CC2 and cholesterol did not affect the body weight (Table 2).
Food intake and liver weight were higher in the HCD groups.
LP06CC2 did not affect the weight of the empty cecum. The weight of the cecum content is shown in Table 4. The weight of the epididymal fat was significantly lower in the LP06CC2 fed mice. The weight of the renal fat was lower in HCD groups; however, LP06CC2 had no apparent effect.

| Lipid parameters
Serum and liver lipid profiles were analyzed and are shown in Serum AST and ALT activities were measured as indices of hepatic injury. ALT is more specific to hepatic injury and cholesterol feeding significantly but moderately increased its serum activity. LP06CC2 slightly but not significantly suppressed the increase of ALT activity in mice fed HCD.

| Fecal parameters
Dietary cholesterol did not affect the fecal weight whereas LP06CC2 significantly increased it irrespective of the dietary condition (Table 4). In addition, the two-way ANOVA analysis showed the significant increase of cecal content by cholesterol and LP06CC2 whereas no interaction was detected. Cecum pH was slightly increased by LP06CC2.

| Bile acid analysis
Cecal and fecal bile acid compositions were analyzed using LC/MS to evaluate the effect of LP06CC2 on the bile acid metabolism. First, HCD increased the cecal total bile acids and the fecal bile acid excretion.
Under cholesterol-loaded condition, LP06CC2 increased the cecal total bile acid whereas it decreased the fecal bile acid excretion (Figure 1a,c).
To evaluate the effect on bile acid deconjugation, the taurine-conjugated bile acid content was calculated and is shown in Figure 1b (cecum) and 1D (feces). Cholic acid, α-muricholic acid, β-muricholic acid,  (Figure 1e). Cyp7A1 protein expression was slightly but not significantly decreased by LP06CC2 in mice fed HCD.

| Bile acid deconjugation in vitro
We isolated several LABs from Mongolian dairy products and selected some of them that were estimated to have probiotic properties. Here, we evaluated their ability to deconjugate bile acids in vitro (

| Fecal short-chain fatty acids analysis
Fecal SCFA levels were measured through LC/MS. LA and PA levels were significantly modulated by HCD (Figure 2). Although LP06CC2 did not regulate fecal LA, AA, and PA levels, the nBA level was increased by LP06CC2. The increase of nBA by LP06CC2 was notably observed in mice fed the ND + L diet compared with mice fed the ND diet and moderately observed in mice fed cholesterol-loaded diets.
iBA and iVA were not detected in any sample and nVA was detected

F I G U R E 1
Effect of Lactobacillus plantarum 06CC2 on the excretion, deconjugation, and biosynthesis of bile acids in mice fed a highcholesterol diet. (a) Cecal total bile acids, (b) Cecal conjugated bile acid level shown as taurine-conjugated fatty acids, (c) Daily fecal bile acid excretion, (d) Fecal conjugated bile acid level shown as taurine-conjugated fatty acids, (e) Hepatic Cyp7A1 expression shown as relative to β-actin. Data are means ± SE for six mice. Post hoc test was performed after a two-way ANOVA. When a significant interaction was detected, the Tukey-Kramer test was conducted to evaluate the significant differences among dietary groups. NS, not significant (p > .1) in 4/6 samples in ND + L group (0.07-0.17 mmol/g dry feces). The Pearson's correlation coefficient analysis revealed a strong and significant positive correlation between fecal LA and nBA, and between AA and nBA.

| D ISCUSS I ON
In  (Jones et al., 2008;Ridlon et al., 2014), and the increase of deconjugation by the ingestion of L. plantarum H6 strain is explained by the manipulation of this microflora (Qu et al., 2020). Further studies are needed to understand the detailed mechanism for the LP06CC2 direct or indirect manipulation of microflora. The composition of primary bile acids is somewhat different between mice and humans, and α,β-muricholic acid, which is rarely detected in humans, was detected in this and our previous studies (Kosakai et al., 2019).
On the other hand, the conjugated bile acids are reabsorbed in the small intestine and the free bile acids are metabolized to secondary bile acids in the lower gastrointestinal tract in mice and humans.
Therefore, the cholesterol-lowering effect of LP06CC2 through bile acid deconjugation is expected to be exerted in humans. On the other hand, because the composition of the entire gut microbiota and the degree LP06CC2 occupancy may affect this effect, further studies on the gut microbiota analysis and the most appropriate dosage are needed.
Although the reduction of taurine-conjugated bile acids was observed in the LP06CC2 ingestion, the total bile acid excretion was significantly reduced. In addition, the expression of Cyp7A1, a rate-limiting enzyme of the bile acid synthesis from cholesterol, was also suppressed. As observed in Cyp7A1-deficient mice, the fecal bile acid level was lower than that in wild-type mice (Erickson et al., 2003). Therefore, the expression of hepatic Cyp7A1 protein reflects the hepatic cholesterol level and the influx of bile acid from the enterohepatic circulation (Chiang, 2015). In contrast, several reports showed that LAB ingestion promotes hepatic Cyp7A1 protein expression together with an increase of the efflux as bile acids (Heo et al., 2018;Hu et al., 2013;Qu et al., 2020) It seems that such cholesterol-reducing mechanism did not occur in this study.
The increase of the cecal content is promoted through the intestinal fermentation of dietary fiber and is often accompanied with an increase of SCFAs such as AA, PA, and nBA (Berggren et al., 1995;Besten et al., 2013). Some of the health beneficial effects of probiot-

ics and dietary fibers may be explained via the increase of intestinal
SCFAs. Here, our data revealed that LP06CC2 specifically increased fecal nBA but no other SCFAs indicating stimulation of nBA bacteria such as Clostridium cluster IV, XIVa (Barcenilla et al., 2000;Pryde et al., 2002). Notably, LP06CC2 failed to increase fecal LA even though it is considered a probiotic strain. Recent studies have uncovered the importance of LA as the main substrate for nBA synthesis by butyrate-producing intestinal bacteria such as Clostridium XIVa (Bourriaud et al., 2005;Duncan et al., 2004). As observed in nBA showed a strong and significant correlation (r = .8206, p < .01, Figure S1). This finding indicates that the intestinal LA supply may be an important factor for the manipulation of BA production. Generally, SCFAs regulate cholesterol metabolism-related gene expression leading to the production of SREBP2, LDL receptor, and Cyp7A1, and results in the reduction of the cholesterol level (Zhao et al., 2017). This mechanism for the regulation of cholesterol levels may not be appropriate for LP06CC2 in that the increase of fecal AA and PA levels and hepatic Cyp7A1 expression was not observed. Notably, nBA may reduce cholesterol absorption via downregulation of the Niemann-Pick C1-Like 1 (NPC1L1) and upregulation of ABCG5 and G8 expression (Chen et al., 2018;Nguyen et al., 2019). . In addition, some reports reveal that LAB directly regulates NPC1L1 and results in the suppression of cholesterol absorption in vitro (Le & Yang, 2019;Lim et al., 2017).
NPC1L1 is a cholesterol transporter and regulates the whole body cholesterol pool and NPC1L1 knockout mice resistance to diet-induced hepatic cholesterol increase (Davis et al., 2004). Therefore, further studies are needed to evaluate the putative mechanism via regulation of the intestinal cholesterol transporter including NPC1L1.
Unexpectedly but intriguingly, serum TG and epididymal fat weight were significantly decreased by LP06CC2. We have not evaluated the detailed mechanism underlying these observations but it may be important for the application of LP06CC2 to prevent obesity and hyperlipidemia as shown in other strains of L. plantarum (Choi et al., 2020;Wu et al., 2015).

CO N FLI C T O F I NTE R E S T
We declare no conflict of interest associate with this manuscript.
Mikako Minesaki, Yuko Miyamoto, Asuka Iwakiri, Kenjiro Ogawa, and Kazuo Nishiyama contributed to the analysis and interpretation of data and assisted in the preparation of the manuscript. All authors equally contributed to this work and approved the final version of the manuscript.

O RCI D
Masao Yamasaki https://orcid.org/0000-0001-7758-9405 F I G U R E 2 Effect of Lactobacillus plantarum 06CC2 on the fecal short-chain fatty acid levels in mice fed a highcholesterol diet. Results are shown as μmol/g dry feces. Data are means ± SE for six mice. Post hoc test was performed after a two-way-ANOVA. When a significant interaction was detected, the Tukey-Kramer test was conducted to evaluate the significant differences among dietary groups. NS: not significant (p > .1)