Abstract

   Globally, metabolic dysfunction-associated steatotic liver disease
   (MASLD), previously termed nonalcoholic fatty liver disease (NAFLD), is
   one of the most common liver disorders and is strongly associated with
   copper deficiency. To explore the potential effects and mechanisms of
   Lactiplantibacillus plantarum LPJZ-658, copper deficiency combined with
   a high-sugar diet-induced MASLD mouse model was utilized in this study.
   We fed 40-week-old (middle-aged) male C57BL/6 mice a copper-deficient
   and high-sugar diet for 16 weeks (CuDS), with supplementary LPJZ-658
   for the last 6 weeks (CuDS + LPJZ-658). In this study, we measured body
   weight, liver weight, and serum biochemical markers. Lipid
   accumulation, histology, lipidomics, and sphingolipid
   metabolism-related enzyme expression were investigated to analyze liver
   function. Untargeted metabolomics was used to analyze the serum and the
   composition and abundance of intestinal flora. In addition, the
   correlation between differential liver lipid profiles, serum
   metabolites, and gut flora at the genus level was measured. The results
   show that LPJZ-658 significantly improves abnormal liver function and
   hepatic steatosis. The lipidomics analyses and metabolic pathway
   analysis identified sphingolipid, retinol, and glycerophospholipid
   metabolism as the most relevant metabolic pathways that characterized
   liver lipid dysregulation in the CuDS group. Consistently, RT-qPCR
   analyses revealed that the enzymes catalyzing sphingolipid metabolism
   that were significantly upregulated in the CuDS group were
   downregulated by the LPJZ-658 treatment. In addition, the serum
   metabolomics results indicated that the linoleic acid, taurine and
   hypotaurine, and ascorbate and aldarate metabolism pathways were
   associated with CuDS-induced MASLD. Notably, we found that treatment
   with LPJZ-658 partially reversed the changes in the differential serum
   metabolites. Finally, LPJZ-658 effectively regulated intestinal flora
   abnormalities and was significantly correlated with differential
   hepatic lipid species and serum metabolites. In conclusion, we
   elucidated the function and potential mechanisms of LPJZ-658 in
   alleviating copper deficiency combined with sugar-induced middle-aged
   MASLD and hope this will provide possible treatment strategies for
   improving MASLD.

   Keywords: metabolic dysfunction-associated steatotic liver disease,
   copper deficiency, liver lipidomics, serum metabolomics, gut microbiota

1. Introduction

   Metabolic dysfunction-associated steatotic liver disease (MASLD;
   formerly non-alcohol-related fatty liver disease (NAFLD)) is the most
   common liver disease globally and causes the majority of liver-related
   morbidity and mortality [[42]1,[43]2]. The spectrum of MASLD ranges
   from simple steatosis and nonalcoholic steatohepatitis (NASH) with
   variable degrees of fibrosis to cirrhosis, which is a major risk factor
   for hepatocellular carcinoma [[44]3]. In most patients, MASLD is
   associated with metabolic comorbidities such as obesity, type 2
   diabetes mellitus (T2DM), or dyslipidemia, and it is often considered
   to be a hepatic manifestation of metabolic syndrome [[45]4]. MASLD has
   become a global health problem, with its prevalence increasing
   significantly over time; the overall prevalence of MASLD is estimated
   to be 32.4% worldwide [[46]5]. Generally, gender and age differences
   exist in MASLD patients. Specifically, the overall prevalence of MASLD
   and NASH is significantly higher in men than in women. More so than
   with gender, the prevalence of MASLD increases with age and is
   significantly higher among people aged 50 years and older than in
   people under 50 years [[47]5,[48]6]. Copper is an essential dietary
   trace element and is a structural and enzymatic co-factor for proteins
   involved in critical cellular processes, including iron metabolism,
   mitochondrial function, and redox homeostasis [[49]7,[50]8]. The liver
   is the central organ for copper homeostasis [[51]9,[52]10], and hepatic
   copper is a key modulator of lipid catabolism [[53]11]. Inadequate
   copper intake is another important nutritional problem worldwide. There
   is growing evidence that copper deficiency (CuD) is associated with the
   development of MASLD [[54]12,[55]13,[56]14]. Liver copper levels are
   lower in patients with MASLD compared with normal controls and
   individuals with other diseases [[57]4]. Copper deficiency exacerbates
   hepatic fibrosis in bile duct-ligated rats, possibly mediated through
   reduced antioxidant defense and iron overload [[58]15]. Notably, copper
   deficiency is relatively common in advanced cirrhosis and is associated
   with an increased risk of infection, a unique metabolic profile, and an
   increased risk of death before transplantation [[59]16]. In addition,
   sugar intake further promotes the development of MASLD. Patients with
   MASLD have been shown to consume excessive amounts of fructose
   [[60]17]. Thus, increased sugar intake and inadequate copper intake are
   two key risk factors for MASLD. It is notable that dietary
   fructose–copper interactions impair the copper status and exacerbate
   MASLD [[61]18,[62]19].

   Probiotics have a long history of use and have attracted public
   interest in the course of promoting human intestinal health. The intake
   of probiotics has been suggested to enhance overall health and
   immunity. Studies have shown that probiotics can inhibit the growth and
   reproduction of pathogenic bacteria, regulate glucose and lipid
   metabolism, modulate the intestinal microbiota, enhance epithelial
   barrier function, and reduce intestinal inflammation [[63]20]. In
   several studies involving humans and animals, probiotics have been
   reported to have the potential to ameliorate MASLD, as shown in the
   associated improvement in hepatic steatosis, inflammation, oxidative
   damage, and fat reduction [[64]21,[65]22,[66]23]. Notably, our previous
   studies also identified that Lactobacillus rhamnosus GG (LGG)
   ameliorates MASLD by modulating intestinal barrier function and lipid
   metabolism [[67]24,[68]25,[69]26,[70]27]. However, few studies have
   focused on the role and mechanism of probiotics in copper
   deficiency-induced fatty liver disease.

   Lactiplantibacillus plantarum LPJZ-658 is a probiotic strain newly
   isolated by our group [[71]28]. In our previous study, we used
   metabolomics and gut microbiota analyses and found that LPJZ-658
   mitigated the development of NASH by correcting the dysregulation of
   bile acid metabolism and modulating the gut microbiota [[72]29]; thus,
   its potential role in improving MASLD was clarified.

   In this study, we utilized serum metabolomics, liver lipidomics, and
   gut microbiome analysis to elucidate the potential role and mechanism
   of LPJZ-658 in ameliorating copper deficiency combined with
   sugar-induced MASLD in middle-aged mice.

2. Materials and Methods

2.1. Probiotic Strain

   LPJZ-658 is a conserved strain grown in our laboratory. LPJZ-658 was
   cultured anaerobically in MRS broth (Beijing Solarbio Science &
   Technology Co., Ltd., Beijing, China) at 37 °C for 18–20 h.

2.2. Animal Experimental Design and Sample Collection

   The 40-week-old male C57BL/6J mice in this study were randomly divided
   into three diet treatment groups: the CuA group, CuDS group, and CuDS +
   LPJZ-658 group. All the groups were fed ad libitum for 14 weeks; diets
   were based on a modified AIN93G formulation (Xiao Shu You Tai (Beijing)
   Biotechnology Co., Ltd., Beijing, China). The CuA group was fed a
   formula containing 17.8 mg/kg Cu and normal sterilized tap water. The
   CuDS group and CuDS + LPJZ-658 group were fed formulas containing a
   diet with less than 0.1 mg/kg Cu and sugar solution consisting of
   glucose (18.9 g/L) and fructose (23.1 g/L) (Sigma-Aldrich, St. Louis,
   MO, USA). In addition, in the last 6 weeks of the experiment, the CuDS
   + LPJZ-658 group was orally administered daily with 1 × 10^9 CFU/day
   per mouse. After all experiments, the mice were fasted for 12 h and
   then euthanized for serum, liver, and cecal-content collection. Mice
   serum samples were processed for lipid profiles and metabolomics
   analysis. Liver samples were processed for histology, lipid profiles,
   lipidomic analysis, and gene expression analysis. Cecal contents were
   collected for gut microbiota analysis. For the animal experiment, n =
   6–8 per group. Experiments were repeated three times.

2.3. Blood Lipid Profiles and ALT/AST Assays

   Whole blood was centrifuged to separate serum. Alanine Aminotransferase
   Assay Kit and Aspartate Aminotransferase Assay Kit (Nanjing Jiancheng
   Institute of Bioengineering, Nanjing, China) were used to measure ALT
   and AST activity in serum samples and were utilized according to the
   manufacturer’s instructions. Serum lipid profiles, including
   triglyceride (TG), total cholesterol (T-CHO), high-density lipoprotein
   (HDL), low-density lipoprotein (LDL), and nonesterified fatty acid
   (NEFA), are measured using their corresponding assay kits (Nanjing
   Jiancheng Institute of Bioengineering, Nanjing, China).

2.4. Histological Analysis

   Liver samples were collected, fixed with 4% paraformaldehyde overnight,
   embedded into paraffin wax blocks, cut into 5-µm-thick sections, and
   stained with hematoxylin and eosin (H&E) (Wuhan Servicebio Technology
   Co., Ltd., Wuhan, China).

2.5. Liver Lipid Profiles Assay

   After homogenization and centrifugation (2500 rpm, 15 min, 4 °C) of
   liver tissue in ice-cold sodium chloride buffer, the supernatant was
   subjected to the determination of hepatic TG and T-CHO content using
   commercially available kits (Nanjing Jiancheng Institute of
   Bioengineering, Nanjing, China).

2.6. Statistical Analysis

   Statistical significance was assessed using GraphPad Prism 7.0 software
   (GraphPad Software Inc., La Jolla, CA, USA). Data are presented as the
   means ± SEM. Differences between multiple groups were analyzed using
   one-way ANOVA, and a p-value < 0.05 was defined as the level of
   significance. Significance is noted as * p < 0.05, ** p < 0.01, and ***
   p < 0.001. The correlation analysis between gut microbiota, serum
   metabolites, and liver lipidomics was investigated using Cytoscape
   V3.7.1 software.

   Additional methods are described in [73]Materials and Methods in
   Supplementary Materials.

3. Results

3.1. Effect of LPJZ-658 on Liver, Body Weight, and Blood Metabolites in MASLD
Mice

   To examine the effects of LPJZ-658 on hepatic injury, we used a copper
   deficiency diet and combined it with a sugar solution-induced mouse
   model for MASLD. As shown in [74]Figure 1A–C, there were no significant
   differences in the body weight, liver/body weight, and eWAT (epididymal
   white adipose tissue)-to-body weight ratios between any of the groups.
   The metabolic index of the mice is shown in [75]Figure 1D. In
   comparison to the CuA group, the serum levels of T-CHO, LDL, and NEFA
   were significantly elevated, and the HDL levels were significantly
   lower in both the CuDS group and the CuDS + LPJZ-658 group of mice.
   Unexpectedly, LPJZ-658 did not affect CuDS-induced serum dyslipidemia.
   We next evaluated liver injury by assessing the plasma levels of the
   hepatic enzymes AST and ALT. As shown in [76]Figure 1E, the CuDS and
   CuDS + LPJZ-658 treatment did not affect the plasma ALT level. However,
   the serum AST levels were significantly increased in the CuDS group and
   markedly decreased in the CuDS + LPJZ-658 group. These results
   suggested that the LPJZ-658 intervention did not significantly improve
   the levels of relevant lipids in the peripheral blood of the mice
   compared to the CuDS group. However, LPJZ-658 treatment might
   effectively prevent liver injury in Cu deficiency combined with
   sugar-induced MASLD in mice.

Figure 1.

   [77]Figure 1
   [78]Open in a new tab

   Effect of LPJZ-658 on body weight, serum lipid homeostasis, and liver
   injury in MASLD mice. (A) Body weight. (B) Liver/body weight ratio. (C)
   eWAT/body weight ratio. (D) Serum levels of TG, T-CHO, HDL, LDL and
   NEFA. (E) Serum levels of ALT and AST. eWAT: epididymal white adipose
   tissue, TG: triglyceride, T-CHO: total cholesterol, HDL: high-density
   lipoprotein, LDL: low-density lipoprotein, NEFA: nonesterified fatty
   acid, ALT: Alanine Aminotransferase, AST: Aspartate Aminotransferase. *
   p < 0.05, ** p < 0.01, and *** p < 0.001.

3.2. Effect of LPJZ-658 on Liver Steatosis in MASLD Mice

   The accumulation of fat in the liver is one of the characteristics of
   liver injury in MASLD. To evaluate the effect of LPJZ-658 on Cu
   deficiency combined with sugar-induced hepatic lipid accumulation, we
   subsequently conducted a histopathological examination of the liver.
   Regarding the H&E staining, the CuDS group demonstrated significantly
   enhanced formation of hepatic lipid droplets compared with the CuA
   group, and this decreased in the CuDS + LPJZ-658 group ([79]Figure 2A).
   As expected, the LPJZ-658 treatment decreased the CuDS group-induced
   intrahepatic TG levels ([80]Figure 2B). To further investigate whether
   liver lipid metabolism is associated with the biological activity of
   LPJZ-658, we analyzed differences in the lipid composition in mice
   liver using lipidomics methods.

Figure 2.

   [81]Figure 2
   [82]Open in a new tab

   Effect of LPJZ-658 on hepatic steatosis and lipidomics profile in MASLD
   mice. (A) Representative images of H&E staining of liver tissue
   sections (scale bars: 100 µm). (B) Hepatic TG and T-CHO levels. PCA
   score plot of the annotated metabolites ((C) ESI+ and (E) ESI−).
   OPLS-DA analysis of identified metabolites ((D) ESI+ and (F) ESI−).
   Volcano plots of identified metabolites ((G) ESI+ and (H) ESI−). Mass
   spectra of the LysoPC(18:1(9Z)/0:0) at RT 2.45 min m/z 566.3491, full
   scan mass spectrum (I), tandem mass spectrum (J). (K) The pathway
   analysis visualized by a bubble plot. (L) Heatmap and hierarchical
   cluster analysis of differential metabolites. (M) The mRNA levels of
   genes related to sphingolipid metabolism. TG: triglyceride, T-CHO:
   total cholesterol, PCA: principal component analysis, OPLS-DA:
   Orthogonal Partial Least Squares-Discriminant Analysis. * p < 0.05, **
   p < 0.01, and *** p < 0.001.

   The positive and negative ion current plots show an acceptable peak
   shape and relatively uniform distribution under the assay conditions of
   this experiment ([83]Figure S1). A principal component analysis (PCA)
   was performed on the liver content samples of each group. As shown in
   [84]Figure 2C,D, the samples in the CuA group were significantly
   distinct from those of the CuDS group, indicating that copper
   deficiency combined with sugar-induced significant lipid metabolite
   changes in the liver. The CuDS + LPJZ-658 group tended to separate from
   the CuDS group and partially overlap with the CuA groups, indicating
   that the disordered hepatic lipid metabolism was partially reversed
   after the LPJZ-658 intervention. Therefore, this method is reliable,
   and the results are credible. The OPLS-DA analysis and volcano plots of
   the differential metabolites in the dual-ion mode of the CuA and CuDS
   groups are shown in [85]Figure 2E–H. According to the criteria of an FC
   of > 1.2 or < 0.83 and a VIP of > 1, 76 significantly different lipid
   species were identified in the CuDS vs. CuA groups ([86]Table S1).
   Taking the ion at RT 2.45 min m/z 566.3469 as an example, the MS
   spectrum in [87]Figure 2I showed that the ion at m/z 566.3469 belonged
   to [M-H]^−. As shown in [88]Figure 2J, the compound produced three
   fragment ions, including m/z 506.3239, 281.2506, and 242.0799. The
   compound was identified as LysoPC (18:1(9Z)/0:0) after searching the
   LIPID MAPS database. To clarify the potential metabolic pathways of the
   76 significantly differential lipid species identified by the CuDS/CuA
   comparison, a pathway-enrichment analysis was performed. As shown in
   [89]Figure 2K, the results indicate that these lipid species mainly
   affect the sphingolipid, retinol, and glycerophospholipid metabolism
   pathways. These differential lipid species were plotted as heat maps to
   provide a more comprehensive picture of the differences between the
   three groups ([90]Figure 2L). In addition, six significantly reversed
   differential lipid species between the CuDS + LPJZ-658/CuDS groups were
   compared ([91]Table S1). To validate the analysis of the lipidomic
   metabolites, we performed qRT-PCR to measure the alteration in the mRNA
   levels of the sphingolipid metabolic-related enzymes. [92]Figure 2M
   shows that sphk2, Cers2, Sgms1, Smpd1, Smpd2, Smpd3, Smpd4, and Spt1c2
   were significantly upregulated in the livers of the CuDS group mice.
   Notably, the expression levels of sphk2, Cers2, Cers4, Smpd1, Smpd2,
   and Spt1c2 were significantly downregulated upon LPJZ-658 treatment.
   These findings suggested that dietary marginal copper deficiency and
   fructose feeding interact to exacerbate hepatic fat accumulation, and
   treatment with LPJZ-658 might effectively prevent hepatic dyslipidemia
   features in Cu deficiency combined with sugar-induced MASLD in mice.

3.3. Effect of LPJZ-658 on Serum Metabolome Profile in MASLD Mice

   UHPLC-Q-Orbitrap/MS was used to evaluate the serum extracts, and
   [93]Figure S2 displays the base peak chromatograms (BPCs) of the
   various groups. The peaks differed significantly in terms of the peak
   intensity and retention time. A multivariate statistical analysis was
   necessary for the identification of the metabolites due to the large
   number of ions present in each chromatogram. Following the generation
   of the Progenesis QI data, datasets, including the sample information,
   RT m/z values, and peak intensities, were generated for statistical
   analysis, and PCA models were constructed to visualize the
   classification patterns. Scoring plots were created based on the data
   obtained by LC-MS from serum extracts in both the positive and negative
   ion modes ([94]Figure 3A,B). We can observe a clear difference between
   the CuA group and the CuDS group. After the treatment with LPJZ-658,
   the points in the CuDS + LPJZ-6518 group show separation from the CuDS
   group. However, there is still a partial crossover, suggesting that the
   LPJZ-658 treatment could partially restore the metabolic changes in the
   CuDS group. The OPLS-DA model-derived VIP > 1, p < 0.05 in one-way
   ANOVA, and FC > 1.2 or FC < 0.83 were used as the criteria for
   differentially expressed metabolites ([95]Figure 3C–F). We identified
   33 metabolites showing significant differences between the CuA and CuDS
   groups ([96]Table S2). The KEGG pathway-enrichment analysis showed that
   these metabolites affect multiple metabolic pathways ([97]Figure 3G),
   including linoleic acid metabolism, taurine and hypotaurine metabolism,
   ascorbate and aldarate metabolism, etc. In addition, as shown in
   [98]Figure 3H and [99]Table S2, the LPJZ-658 treatment significantly
   reversed seven serum metabolites compared to the CuDS group. These
   findings demonstrated that LPJZ-658 supplementation might mitigate the
   metabolic changes associated with Cu deficiency combined with
   sugar-induced MASLD in mice by adjusting these pathways.

Figure 3.

   [100]Figure 3
   [101]Open in a new tab

   Effect of LPJZ-658 on serum metabolites in MASLD mice. PCA score plots
   of the annotated metabolites ((A) ESI+ and (B) ESI−). OPLS-DA score
   plots of identified metabolites ((C) ESI+ and (D) ESI−). Volcano plots
   of identified metabolites ((E) ESI+ and (F) ESI−). (G) The pathway
   analysis visualized by a bubble plot. (H) Heatmap and hierarchical
   cluster analysis of differential metabolites between. PCA: principal
   component analysis, OPLS-DA: Orthogonal Partial Least
   Squares-Discriminant Analysis.

3.4. Effect of LPJZ-658 on Gut Microbiota Composition in MASLD Mice

   To elucidate the underlying mechanisms of LPJZ-658 in improving MASLD,
   we investigated its effects on the gut microbial composition. As shown
   in [102]Figure 4A, a Venn diagram of the OTUs was generated.
   Furthermore, the OTU rarefaction curves tended to be flat, suggesting a
   credible sequencing result ([103]Figure 4B). No significant differences
   were observed in the alpha diversity indices, including the Simpson,
   Chao 1, and Shannon indices, as presented in [104]Figure 4C. The
   principal coordinate analysis (PCoA) results are shown in [105]Figure
   4D, representing the beta diversity, reflecting the changes in the gut
   microbiota composition between each group. The important microorganisms
   for improving the gut microbiota of LPJZ-658 were determined by a
   linear discriminant analysis effect size (LEfSe) analysis to be sets of
   bacteria with linear discriminant analysis (LDA) values greater than 4,
   which were considered to be significantly different. The LDA score
   distribution map and cladogram are shown in [106]Figure 4E. As shown in
   [107]Figure S3, a total of 13 bacterial genera in the cecal-content
   samples were detected as being significantly different by the LEfSe, of
   which the CuA group accounted for only one (Dubosiella). Eight
   different genera, within two phyla (Proteobacteria and
   unidentified_Bacteria), three classes (Desulfovibrionia,
   Gammaproteobacteria, and Alphaproteobacteria), two orders
   (Desulfovibrionales and Pseudomonadales), and one family
   (Desulfovibrionaceae) were found in the CuDS group. In the CuDS +
   LPJZ-658 group, four bacterial taxa were found within one order
   (Erysipelotrichales), one family (Erysipelotrichaceae), one genus
   (Faecalibaculum), and one species (Faecalibaculum_rodentium).

Figure 4.

   [108]Figure 4
   [109]Open in a new tab

   Effect of LPJZ658 on the composition of gut microbiota: (A) Venn
   diagram. (B) Rarefaction curves of cecal contents samples. (C) Alpha
   diversity indices. (D) PCoA plots (Bray-Curtis distance). (E) The LEfSe
   analysis of differential microbial taxa: distribution histogram of LDA
   values (left); evolutionary branching diagram (right). The relative
   abundance of cecal contents bacterial phyla (F) and general. (H) The
   differential microbiota at the phylum (G) and genus (I) level. PCoA:
   principal coordinate analysis, LEfSe: linear discriminant analysis
   effect size, LDA: linear discriminant analysis. * p < 0.05, ** p <
   0.01.

   At the phylum level, Firmicutes, Bacteroidetes, and Verrucomicrobiota
   were the predominant phyla of microbiota in the cecal contents
   ([110]Figure 4F). As shown in [111]Figure 4G, the phylum-level analysis
   demonstrated that Cu deficiency combined with sugar-induced MASLD in
   the mice showed a significantly increased relative abundance of the
   Proteobacteria phylum, which was significantly reversed by the
   supplementation with LPJZ-658. In addition, a significant increase in
   the relative abundance of Firmicutes was observed in the CuDS +
   LPJZ-658 group compared to the CuDS group.

   In addition, [112]Figure 4H displays the relative abundance of the
   microbiota in the mouse cecal contents at the genus level (top 35) and
   the remarkable differences in the microorganisms, as shown in
   [113]Figure 4I. The results showed that the relative abundances of
   Ruminococcus_torques_group were significantly increased in the CuDS
   group compared to the CuA group ([114]Figure 4I), while LPJZ-658
   supplementation significantly decreased them. In addition,
   Faecalibaculum was significantly elevated and Prevotellacea_UCG-001,
   [Eubacterium]_fissicatena_group was significantly decreased in the CuDS
   + LPJZ-658 group compared to the CuDS group. These results suggest that
   LPJZ-658 administration induces changes in the gut microbial
   composition, which may have beneficial effects on alleviating Cu
   deficiency with combined sugar-induced MASLD in mice.

3.5. Correlation Analysis of the Gut Microbiota–Serum Metabolites–Liver
Lipidomic Axis

   To determine whether the LPJZ-658-induced changes in the liver lipid
   metabolome and serum metabolome were associated with the effects of the
   gut microbiota, we selected the differential metabolites and
   genus-level colony abundance among the three groups for a Spearman
   correlation analysis. We found significant correlations between the
   LPJZ-658-induced change in lipid metabolomics, serum metabolomics
   metabolites, and genus-level gut microbiota abundance. As shown in
   [115]Figure 5A,B, the liver lipid metabolomics PC (18:0/18:1) was
   significantly positively correlated with the serum 2-Butoxyethanol, and
   the PC (18:0/22:4) was significantly positively correlated with the
   serum Pantothenic acid, Sphingosine 1-phosphate, and 2-Butoxyethanol.
   Furthermore, SM (d17:1/24:1) and PG (22:6/22:6) were significantly
   negatively correlated with the serum Pantothenic acid, Sphingosine
   1-phosphate, and Citrulline, and the gut abundance of
   [Ruminococcus]_torques_group. In addition, PG (22:6/20:3) was
   significantly positively correlated with the serum Pantothenic acid,
   Sphingosine 1-phosphate, Citrulline, and 2-Butoxyethano, and
   significantly negatively correlated with the gut abundance of
   Prevotellaceae_UCG-001. The serum metabolomics metabolite Citric acid
   showed negative correlations with the serum Sphingosine 1-phosphate and
   Citrulline and gut Prevotellaceae_UCG-001 abundance. The serum
   metabolomics metabolite Sphingosine 1-phosphate showed negative
   correlations with the serum 2-Butoxyethanol and gut
   Prevotellaceae_UCG-001 abundance. In addition, the serum Citrulline was
   positively correlated with the gut abundance of Faecalibaculum and
   negatively correlated with [Ruminococcus]_torques_group.

Figure 5.

   [116]Figure 5
   [117]Open in a new tab

   Correlation analysis. (A) Correlation network diagram between
   differential gut microbiota, serum metabolites, and liver lipidomic.
   Thicker lines indicate a higher correlation. The red line indicates a
   positive correlation, while the blue line indicates a negative
   correlation. (B) Correlation between differential gut microbiota, serum
   metabolites, and liver lipidomic. * p < 0.05, ** p < 0.01, and *** p <
   0.001.

4. Discussion

   MASLD is characterized by abnormal lipid metabolism, which results in
   lipid accumulation in hepatocytes [[118]2]. However, there has been a
   lack of consensus on the specific mechanism responsible for this until
   now. Copper is an important dietary trace element that has an essential
   role in mammalian physiology [[119]30]. Western-style diets are low in
   copper levels and high in fructose. In particular, given the central
   role of the liver in maintaining Cu metabolism, hepatic copper
   deficiency is strongly associated with metabolic diseases, including
   diabetes mellitus, metabolic syndrome, and MASLD
   [[120]14,[121]31,[122]32]. Studies of rodent models have found that low
   dietary copper increases lipid accumulation in the liver [[123]14] and
   induces dyslipidemia [[124]33,[125]34]. In recent years, many
   probiotics have been shown to prevent and treat MASLD
   [[126]35,[127]36,[128]37,[129]38]. LPJZ-658 is a strain of
   Lactiplantibacillus plantarum isolated by our group. Our practice has
   confirmed that LPJZ-658 improves lipid metabolism in late-laying laying
   hens and broiler chickens [[130]39]. It is important to note that we
   found that LPJZ-658 ameliorated hepatic steatosis, inflammation, and
   fibrosis in NASH model mice by mediating bile acid metabolism and the
   gut microbiota [[131]29], which suggests that LPJZ-658 has the
   potential to improve MASLD. In this study, copper deficiency combined
   with a high-sugar-diet-induced MASLD mouse model was used to explore
   the therapeutic mechanism of LPJZ-658 in MASLD-combined lipidomics,
   metabolomics, and gut microbiota strategies.

   The present study is similar to previous studies utilizing copper
   deficiency combined with fructose-induced MASLD [[132]13]; the lipid
   profiles were significantly disturbed in the CuDS group ([133]Figure
   1D). To our surprise, the treatment with LPJZ-658 did not improve this
   profile. We next evaluated liver injury by assessing the serum levels
   of hepatic enzymes; the AST levels were significantly increased in the
   CuDS group. Importantly, the elevated serum ALT levels were attenuated
   by the LPJZ-658 treatment ([134]Figure 1E). Prior work suggests that
   low Cu enhances MASLD severity [[135]14,[136]40], while metabolic
   disorders and their toxic lipid hoarding products are considered
   central to the development of MASLD [[137]41]. Here, we found that CuDS
   resulted in moderate hepatosteatosis through histology H&E staining and
   that LPJZ-658 supplementation alleviated hepatic steatosis ([138]Figure
   2A). Consistent with liver histology, the quantitative evaluations
   indicated that LPJZ-658 can also improve CuDS-induced hepatic steatosis
   by reducing the TG levels ([139]Figure 2B).

   Even though TGs are primarily associated with hepatic pathologies,
   other lipid fractions seem to be associated with the development and
   severity of MASLD. Therefore, in the present study, we investigated the
   effect of CuDS alone or after treatment with LPJZ-658 administration on
   hepatic steatosis, using untargeted lipidomic analyses to understand
   changes in the hepatic lipid profiles. The data analyses identified 76
   differential lipid species associated with the comparison between CuDS
   and CuA ([140]Table S1). These 76 differential lipid species are mainly
   involved in the sphingolipid, retinol, and glycerophospholipid
   metabolism pathways ([141]Figure 2K). Among them, sphingolipid
   metabolism attracted our attention. The sphingolipid and glycerolipid
   metabolic pathways share many common intermediates [[142]42,[143]43],
   and there is a strong correlation between the triglyceride and
   sphingolipid contents in the liver. The sphingolipid levels are
   involved in regulating major biological processes [[144]44,[145]45].
   The dysregulation of sphingolipid metabolism has been found to
   contribute to the development of MASLD. The sphingolipid family
   includes ceramides, sphingosines, glycosphingolipids, and
   sphingomyelins [[146]46]. As a critical bioactive lipid, ceramide is
   considered to be a metabolic hub as it lies at the crossroads of
   multiple metabolic pathways in mammals [[147]47]. A previous study
   found a significant increase in the ceramide content in the liver of
   rats fed with a high-fat diet (HFD) [[148]48]. Moreover, MASLD is
   associated with the increased expression of genes involved in the three
   different pathways leading to ceramide synthesis, including de novo
   synthesis, sphingomyelin hydrolysis, and the salvage pathway
   [[149]48,[150]49].

   Ceramides can be directly synthesized from sphingosines and acyl-CoAs
   expressed by six distinct genes (Cers1–6) through a salvage pathway.
   Specifically, CerS and Cers2, which synthesize long-chain ceramides,
   are the most prominent and are mainly expressed in the liver
   [[151]50,[152]51]. The enzymatic breakdown of sphingomyelin by
   sphingomyelinase (SMase, Smpd) is performed by the Cer-synthesizing
   enzyme that catalyzes the hydrolysis of sphingomyelin to Cer
   [[153]52,[154]53]. The deletion of Smpd improves MASLD progression by
   protecting against endoplasmic reticulum stress [[155]54]. There is
   further evidence that ceramides can also be generated from the
   degradation of complex sphingolipids, including glycerophospholipids
   and sphingomyelins. Next, ceramide can be converted by neutral
   ceramidase to a biologically active signaling molecule called
   Sphingosine (Sph), which can ultimately be phosphorylated by Sph kinase
   (SphK) to Sph-1-phosphate (S1P). S1P represents one of the final
   metabolites produced by sphingolipid metabolism and is a key
   biologically active sphingolipid involved in inflammation, cell
   survival, and migration [[156]55,[157]56]. By dissecting the
   dysregulation of sphingolipid metabolic enzymes at the mRNA level, we
   found that CuDS dramatically upregulated Sphk2, Cers2, Sgms1, Smpd1–4,
   and Spt1c2 in livers ([158]Figure 2M). Importantly, elevated mRNA
   levels of Sphk2, Cers2, Smpd1, Smpd2, and Spt1c2 were attenuated by
   LPJZ-658, suggesting the LPJZ-658 is important in alleviating
   CuDS-induced disorders of sphingomyelin metabolism.

   In MASLD, the phenotype is complex and dynamic as a result of multiple
   interactions between genetic and environmental factors. Metabolites can
   act as regulators of phenotypes. Metabolites present in blood provide
   rich information about individual physiological states. Many serum
   biomarkers have been reported to differentiate between MASLD and
   healthy individuals [[159]57]. Therefore, metabolomics may provide more
   insights and clues for the present study on the development and
   progression of copper deficiency combined with sugar-induced MASLD and
   the mechanism of action of LPJZ-658. In our previous studies, LPJZ-658
   has been found to affect multiple metabolic pathways, especially bile
   acid biosynthesis [[160]29].

   In this study, we examined the changes in metabolites between the three
   groups using untargeted metabolomics. A clear separation between the
   CuA and CuDS groups was observed by both ESI+ and ESI− in the PCA and
   OPLS-DA analyses, suggesting that CuDS treatment has a significant
   effect on metabolism in mice. We found 33 differential metabolites
   between the CuA and CuDS groups that mainly affected metabolic
   pathways, including linoleic acid metabolism, taurine and hypotaurine
   metabolism, and ascorbate and aldarate metabolism, which is strongly
   associated with MASLD development [[161]58,[162]59,[163]60,[164]61].
   Furthermore, LPJZ-658 treatment reversed 6 of these 33 differential
   metabolites. In conclusion, these data demonstrated that
   supplementation with LPJZ-658 may improve MASLD by regulating multiple
   metabolic pathways.

   The gut microbiota is an important bridge to systemic metabolism. The
   gut microbial imbalance may lead to the dysfunction of the host’s
   machinery, which contributes to the pathogenesis of MASLD [[165]62]. To
   elucidate the mechanisms of LPJZ-658, we investigated the changes in
   the composition of the cecal-content microbiota. We analyzed the
   relative abundance of bacteria at the phylum and genus levels,
   respectively. At the phylum level, although the two dominant phyla,
   Bacteroidetes and Firmicute, did not change significantly in the CuDS
   group, we found that the abundance of Proteobacteria increased
   significantly, which is consistent with clinical findings [[166]63].
   The rise in the Proteobacteria phylum proportion was the most prominent
   alteration in the gut–liver axis-induced hepatic steatosis in
   high-fructose-induced MASLD [[167]64]. Remarkably, we found a marked
   reduction in the Proteobacteria abundance in the CuDS + LPJZ-658 group
   ([168]Figure 4G). Moreover, at the genus level, the CuDS group was
   characterized by a significantly higher relative abundance of
   Ruminococcus_torques_group compared with the CuA group. Noteworthily,
   LPJZ-658 supplementation significantly changed the gut microbiota
   composition to varying degrees. The CuDS + LPJZ-658 group displayed a
   significantly increased abundance of Faecalibaculum and a decreased
   relative abundance of Prevotellaceae_UCG-001,
   Ruminococcus_torques_group, and Eubacterium fissicatena group
   ([169]Figure 4I). In the gut of patients and animals with MASLD, the
   abundance of Ruminococcus_torques_group is increased [[170]65].
   Interestingly, multiple studies have found that the intestinal
   Ruminococcus_torques_group was crucial in lipid deposition and hepatic
   steatosis in poultry [[171]66,[172]67,[173]68]. In the present study,
   the reduction in Ruminococcus_torques_group abundance after the
   LPJZ-658 treatment may be involved in the regulation of hepatic lipid
   metabolism in MASLD mice. The genus Parabacteroides has been reported
   to be associated with obesity, insulin resistance, and inflammation in
   HFD-fed mice [[174]69]. Although some strains of Prevotellaceae seem to
   be inflammatory pathogens, they play a critical role in immune
   regulation [[175]70]. In the present study, we observed that the
   abundance of Prevotellaceae_UCG-001 decreased in the CuDS + LPJZ-658
   group, suggesting that LPJZ-658 promoted the abundance of microbes
   regulating immunity. Faecalibaculum may mediate these beneficial
   effects through carbohydrate and energy metabolism pathways in rodents
   [[176]71]. In our study, the LPJZ-658-induced increase in
   Faecalibaculum may favorably influence the function of energy
   metabolism and improve nutrient absorption and metabolic processes in
   the intestine. The Eubacterium fissicatena group is highly correlated
   with obesity and obesity-related metabolic disorders in hosts
   [[177]72]. Another notable finding indicated that the proportion of the
   Eubacterium fissicatena group was significantly higher in MASLD mice
   [[178]73,[179]74]. Although CuDS did not elevate the abundance of the
   Eubacterium fissicatena group in the present study, LPJZ-658
   significantly reduced its abundance. The above studies revealed the
   beneficial role of LPJZ-658 in regulating the intestinal flora.
   Therefore, the above studies revealed that LPJZ-658 may alleviate
   CuDS-induced hepatic steatosis through the modulation of the gut
   microbiota composition. More importantly, the results of the Pearson
   correlation showed a significant correlation between the differential
   liver lipid profiles, serum metabolites, and gut flora at the genus
   level ([180]Figure 5).

5. Conclusions

   In conclusion, this study has shown that copper deficiency combined
   with sugar causes liver steatosis and gut dysbiosis. Through a
   multi-omics analysis, we demonstrate that LPJZ-658 supplementation
   improves abnormalities of hepatic lipid metabolism significantly in
   middle-aged MASLD mice.

Acknowledgments