Specific pathogen-free Sprague–Dawley male rats (age, 7 weeks; weight, 180–200 g) were obtained from Shandong International Biotechnology Park Development (China) under license number SYXK (Lu) 20180030. The rats were maintained at 0°C–26°C and at a relative humidity of 40%–70%. Sterling MPPAE was obtained from New Era Health Industry (Group) Co., Ltd. Then, 0.3, 0.9, and 2.7 g/kg MPPAE/rat was adjusted for further analysis. The materials used in this study are shown in Table 1.

To induce kidney damage, Sprague–Dawley rats received intraperitoneal injections of Cd (2 mg/kg/d) for 7 days (Zhang et al., 2017), and rats in the negative control group (CK) received 0.9% saline solution via the same route injection. The rats then received different concentrations of MPPAE (0.3, 0.9, and 2.7 g/kg, respectively) or ultrapure water orally by gavage. The concentrations of MPPAE used in this study were determined based on previous reports and combined with preliminary experiments (Cong et al., 2015). Five groups were set for the following experiment: CK: negative control group; MG (model group): Cd alone; ELG (extremely low group): 0.3 g/kg; LG (low group): 0.9 g/kg MPPAE; HG (high group): 2.7 g/kg MPPAE. Indexes of malondialdehyde (MDA) and superoxide dismutase (SOD) in the kidney were investigated (n = 6 per group).

At 7 days post-treatment, the rats were euthanized and the kidneys were collected. Renal pelvis and papillae tissues were fixed in 4% paraformaldehyde and embedded in paraffin wax, and sections (3 μm thickness) were prepared. The sections were then stained with hematoxylin and eosin (HE) for the evaluation of histopathology under a microscope (×400 magnification).

Based on the in vivo experiments, transcriptomic and metabolomic analyses were conducted in the following groups: CK, MG, LG, and HG. For each group, total RNA was extracted from approximately 400 mg of kidney tissues using TRIzol, according to the manufacturer’s instructions. The quality of total RNA was evaluated using an Agilent 2100 Bioanalyzer. Subsequently, mRNAs were enriched using magnetic Oligo (dT) beads, followed by cDNA synthesis, and 250–300-bp cDNA fragments were selected for PCR amplification to construct the cDNA library. After quality control, the cDNA library was sequenced using the Illumina NovaSeq 6000 sequencing platform.

Raw sequencing data were filtered to remove reads that contained adapter sequences, N bases, and low-quality reads. The clean data were then mapped to the reference genome using HISAT2 software (Mortazavi et al., 2008). Novel transcripts were assembled using StringTie software (Pertea et al., 2015), and the transcripts were then annotated using the Pfam, SUPERFAMILY (El-Gebali et al., 2019), Gene Ontology (GO), and Kyoto Encyclopedia of Genes and Genomes (KEGG) databases (Chen et al., 2017).

The fragments per kilobase of transcript per million base pairs sequenced (FPKM) method, which is considered to reflect the influence of the sequencing depth and gene length of fragments, is commonly used to evaluate gene expression levels in transcriptomic data analysis (Bray et al., 2016). DESeq2 software was used to screen DEGs, defined as those with a fold change in expression >1, and a p-value < 0.5.

Kidney tissues (100 mg) were frozen in liquid nitrogen and ground before the addition of 500 μl of 80% methanol. The mixture was vortexed and placed on ice for 5 min before centrifugation at 15,000 × g for 20 min (4°C). Then, the methanol content in the supernatant was diluted to 53% and then centrifuged at 15,000 × g for 20 min (4°C). Finally, the supernatant was used for LC-MS analysis.

The chromatographic separation was performed using a Hypersil GOLD column (C18) at 40°C. The positive mode contained 0.1% formic acid as solvent A and methanol as solvent B. The negative mode consisted of 5 mM ammonium acetate as solvent A and methanol as solvent B. The mass spectrum scan ranged from 100 to 1,500 m/z, with other parameters set as follows: spray voltage: 3.5 kV, sheath gas flow rate: 35psi, auxiliary gas flow rate: 10 L/min, capillary temperature: 320°C, S-lens RF level: 60, auxiliary gas heater temperature: 350°C, polarity: positive and negative, and MS/MS: data-dependent scans.

All the raw data were processed using CD 3.1 search software, and the metabolites were identified and quantified through peak filtration, peak extraction, peak size quantitation, and finally, normalization. All the identified metabolites were annotated by the KEGG (https://www.genome.jp/kegg/pathway.html), HMDB (https://hmdb.ca/metabolites), and LIPID MAPS databases (http://www.lipidmaps.org/). The differences between metabolites detected in the two treatment groups were evaluated by t-tests, and those with variable importance in projection (VIP) > 1, p < 0.05, and fold change in levels >2 or <0.5 were regarded as significant DEMs.

Conjoint analysis of the transcriptome and metabolome was performed to further reveal the post-transcriptional regulation of expressed genes. Correction analysis of the DEGs and DEMs was conducted based on Pearson coefficients. KEGG enrichment was evaluated to analyze the common pathways between DEGs and DEMs and clarify the main biochemical pathways and signal transduction pathways that link the DEGs and their corresponding DEMs.

Differences in the MDA and SOD levels in the kidneys between groups were analyzed using GraphPad Prism 7 software (Schneider et al., 2021). p < 0.05 was considered to indicate statistical significance.

The kidney MDA contents of the MG Cd control group were significantly higher than those in the negative control group, while the SOD levels were significantly lower, indicating that the Cd-induced kidney injury model was successfully constructed in rats. After the application of MPPAE, the decrease in kidney MDA contents and the increase in SOD contents in the LG and HG groups were exacerbated. Furthermore, MPPAE at 0.9 and 2.7 g/kg exerted significant protection against Cd-induced kidney injury in a dosage-dependent manner. Therefore, model rats were treated with MPPAE at 0.9 and 2.7 g/kg for subsequent histopathology evaluation and transcriptomic and metabolomic analyses (Figure 1).

After injection of Cd, the kidney exhibited marked changes characteristic of injury. In the MG group, the main lesion involved inflammatory cell infiltration of the renal papilla. Moreover, degeneration occurred in the cortex renal tubular epithelial cells with the frequency of 1/7 and in the renal tubular hyaline cast with the frequency of 3/7. Similar lesions were observed in the LG group. However, the frequency and severity of the lesions, especially the inflammatory reaction in the renal papilla, were significantly reduced compared to those in the MG group. In contrast, only one case of renal papilla mesenchymal inflammatory cell infiltration occurred in the HG group, and inflammatory cells were rare. Thus, our findings indicated that MPPAE protected against Cd-induced kidney injury in a dosage-dependent manner (Figure 2).

Transcriptomic sequencing yielded an average of 45,180,315 raw reads, from which an average of 6.68 G of clean data were filtered. Q20 in each group was higher than 97%, which indicated high-sequencing quality. In addition to the genes mapped to the reference genome, 2,031 novel transcripts were assembled. Finally, 34,914 genes were annotated in the Pfam database.

The GO classification showed that transcripts were enriched in three function categories of biological process, cellular component, and molecular function. Among these functions, cellular process (16,175 genes), cells (15,824 genes), and binding (13,404 genes) exhibit the highest proportion in each of these categories (Figure 3).

In comparison with the CK group, 3,205 DEGs (1,821 upregulated and 1,384 downregulated) were identified in the rats treated with Cd (Figure 4A), with the most significant changes observed in expression levels of genes encoding the L1 transposable element RBD-like domain, interleukin 1 receptor type 2, and cytochrome P450. In contrast, few DEGs (504 upregulated and 344 downregulated) were detected after treatment with the low concentration of MPPAE (Figure 4B), with the most significant changes observed in expression levels of genes encoding protein phosphatase 1, hypoxanthine phosphoribosyltransferase 1, and deoxyribonuclease 1-like 3. Compared with the MG group, 3,117 DEGs (1,503 upregulated and 1,614 downregulated) were identified in the rats treated with the high concentration of MPPAE (Figure 4C). The upregulated DEGs included several cytochrome P450-encoding genes and transporter-related genes. The number of common DEGs in both LG and HG groups compared with the MG group was 235, containing genes encoding cytochrome P450, zinc finger protein, ATP-binding cassette subfamily, etc.

The detected metabolites were annotated to three important databases. Metabolites from both positive and negative ion detection modes exhibited similar annotation information in the KEGG database. The dominant pathways were global and overview maps, amino acid metabolism, and lipid metabolism. In the HMDB database, the number of metabolites and annotated superclasses was different in the positive or negative ion detection modes. The dominant annotated superclasses in the positive detection mode were lipids and lipid-like molecules, organic acids and derivatives, and organoheterocyclic compounds. In the negative detection mode, the dominant annotated superclasses were lipids and lipid-like molecules, organic acids and derivatives, and organic oxygen compounds.

In the LIPID MAPS database, there were significant differences in the annotation information between the positive or negative ion detection modes. In the positive ion detection mode, the dominant annotated classes were glycerophosphoethanolamines, steroids, and glycerophosphocholines. In the negative ion detection mode, the dominant annotated classes were fatty acids and conjugates, glycerophosphocholines, and eicosanoids.

Numerous metabolites were differentially expressed after the injection of Cd. In the negative ion detection mode, 55 metabolites were upregulated after Cd and 77 were downregulated compared with the CK group. The most downregulated metabolite was 2-[(2-amino-6-methylpyrimidin-4-yl)thio]-4,6-dimethylnicotinonitrile, although no annotated information was available in any of the three databases, followed by hydroquinone. The most upregulated metabolites were sodium cholate and chenodeoxycholic acid. In the positive ion detection mode, 37 metabolites were upregulated and 160 were downregulated after the injection of Cd. The most upregulated metabolites were O-phospho-L-tyrosine and PC (20:5e/18:2). The most downregulated metabolites were tropolone and 2-(carboxymethoxyl)-4-methoxybenzoic acid.

Only minor changes in metabolites were observed after the application of low-dosage MPPAE. In the positive ion detection mode, 28 metabolites were upregulated, while only seven metabolites were downregulated. L-ascorbate and LPC 16:2 were the most upregulated metabolites, while prolyl leucine and ureidoisobutyric acid were the most downregulated. In the negative ion detection mode, 21 metabolites were upregulated, while only six were downregulated. The most upregulated metabolite was ciliatine, which was detected at 41.5-fold higher levels in the low-dosage MPPAE group compared with the control group without MPPAE treatment. The most downregulated metabolite was 4-hydroxy-3- methoxyphenylglycol sulfate.

Differences were detected in the degree of regulation of DEMs between the high- and low-dosage MPPAE groups. In the negative ion detection mode, 35 metabolites were upregulated, and the most differentially expressed metabolite was ciliatine. The levels of ciliatine in the high- and low-dosage MPPAE groups were 195.5-fold higher and 41.5-fold higher, respectively, than those in the control group without MPPAE treatment. These data indicated that ciliatine plays a crucial role in the mechanism by which MPPAE mediated the detoxification of Cd. The metabolites Dl-3,4-dihydroxymandelic acid and glyphosate were upregulated by 151.3-fold and 134.4-fold, respectively. In addition, 11 metabolites were downregulated, with D-glucuronic acid identified as the most differentially expressed metabolite. In the positive ion detection mode, 28 metabolites were upregulated and seven were downregulated.

Venn diagrams were generated to compare the overlapping and unique DEMs among the different treatment groups detected in the positive and negative ion detection modes (Figure 5). The number of common DEMs in both LG and HG groups compared with the MG group was 8 under the negative ion detection mode and 10 under the positive ion detection mode. One DEM, namely, 4-hydroxyretinoic acid, was detected under both negative and positive ion detection modes. From Venn diagram analysis, one DEM ascorbic acid and one DEM N1-(1,3-thiazol-2-yl)-2-chlorobenzamide were detected under negative and positive ion detection modes, respectively, among all groups. Hierarchical clustering analysis was also conducted to compare the metabolic changes under different treatments detected in the positive and negative ion detection modes.

After the injection of Cd, 41 and 50 KEGG pathways were enriched in the positive and negative ion detection modes, respectively. In the positive ion detection mode, the most enriched pathways were metabolic pathways and tyrosine metabolism, while in the negative ion detection mode, the most enriched pathways were metabolic pathways and phenylalanine metabolism (Figure 6). After the application of low-dosage MPPAE, nine and 22 KEGG pathways were enriched in the positive and negative ion detection modes, respectively, with metabolic pathways found to be dominant (Figure 6). Similarly, after the application of high-dosage MPPAE, metabolic pathways were dominant in both the negative (24 pathways) and positive (14 pathways) ion detection modes (Figure 6).

Pearson correlation analysis of the DEGs and DEMs was conducted, with coefficients <0 representing a negative correlation, while correlation coefficients >0 represent a positive correlation (Figure 7).

KEGG enrichment analysis was conducted to identify the shared pathways associated with DEGs and DEMs, thereby clarifying their common biochemical pathways and signal transduction pathways. After the injection of Cd, 96 and 80 KEGG pathways were enriched in the positive and negative ion detection modes, respectively. After the application of low-dosage MPPAE, the enriched 16 KEGG pathways were enriched in the positive ion detection mode and 36 were enriched in the negative ion detection mode. In the low-dosage MPPAE treatment group, 26 KEGG pathways were enriched in the positive ion detection mode and 82 in the negative ion detection mode (Figure 8).