A total of 60 g of raw herbal pieces, including Salvia sinica Migo (25 g), Achyranthes bidentata Blume (10 g), Salvia coccinea Linn (15 g) and Caulis Spatholobi (10 g), was used. The herbal ingredients are shown in Table 1. Each herb was purchased from Dongfang Hospital affiliated with Beijing University of Traditional Chinese Medicine. The above herbs were boiled in a 10-fold volume of water at 100°C for 1.0 h. After filtration, the first extraction was boiled in an 8-fold volume of water for 0.5 h. Finally, both filtrates were mixed and concentrated to a volume of 60 ml containing 1 g/ml raw herbs. Valsartan capsules (Beijing Novartis Pharmaceutical Co., Ltd., batch number X2,375) were provided by Novartis (Bale, Switzerland).

TLYS decoction extract was combined with methanol and double distilled water (1:1, v/v) at 1:20, sonicated for 30 min and filtered through a 0.22 microns filtration membrane. Quality control of TLYS was performed using a UHPLC System (Dionex Ultimate 3,000, Thermo Corporation, United States) coupled with a mass spectrometer (LTQ-Oribitrap XL, Thermo Scientific). The chromatographic column was an Acquity UPLC C18 column (2.1 mm × 100 mm, 1.7 µm). The chromatographic conditions were as follows: 0.1% formic acid water (A) and methanol (B) were used as the mobile phase. The gradient elution conditions were as follows: 0–3 min, 5%–5% B; 3–45 min, 5%–75% B; 45–45.1 min, 75%–5% B; 45.1–50 min, 5% B. The column temperature was 30°C. The flow rate was 0.3 ml· min-1^, and the injection volume was 2 μL. A mass spectrometer equipped with an electrospray ionization source was used for both positive and negative ion mode with the mass range of 120–1800 m/z. The ionization voltages were 3500 V (positive mode) and 3000 V (negative mode), the capillary temperature was 320°C, the sheath gas flow rate was 35 arb and the auxiliary gas flow rate was 10 arb. XCMS software was used to import mass spectra. Peak integration, peak extraction, peak alignment, peak identification and retention time correction were carried out. Material identification of peaks was conducted with information collected from the databases and literature.

Male Sprague-Dawley (SD) rats (n = 15 per group, 180–200 g, 7–8 weeks of age) were provided by Beijing Vital River Laboratory Animal Technology Co., Ltd. (certificate number: SCXK (Beijing) 2016–0,006). The animal operation in this study was carried out according to the “Guiding Principles in the Use and Care of Animals” published by the US National Institutes of Health (NIH Publishing, No. 85–23, revised in 1996). This procedure was completed under the supervision of the Laboratory Animal Ethics Committee of Dongfang Hospital affiliated with Beijing University of Chinese Medicine (permit no. 202004). All animals were kept in a clean room at 22 ± 2°C and had free access to water and food.

As previously described, the UUO model was established in SD rats by ligation of the left ureter and sacrifice 14 days later (Masaki et al., 2003). In short, the rats were anesthetized by intraperitoneal injection of pentobarbital (50 mg/kg body weight); the left ureter was exposed via a midline incision and was ligated at two points with 4–0 silk sutures. The sham group underwent identical surgical procedures except for ligation of the ureter. The rats were randomly divided into four groups as follows: sham group, UUO group, TLYS group, and valsartan group. In the sham group and the UUO group, an equal volume of physiological saline was administered. The adult daily dosage for TLYS was 60 g. The daily dosage of TLYS in rats was calculated to be 7.8 g/kg by a correction factor equal to the human-rat body surface area ratio (6.3) (Xuan et al., 2021). The valsartan group was given 30 mg/kg/d of valsartan intragastrically.

The Scr and BUN levels were measured with a creatinine assay kit (C011-1, Nanjing Jiancheng Bioengineering Institute, Nanjing, China) and BUN assay kit according to the manufacturer’s instructions (C013-1, Nanjing Jiancheng Bioengineering Institute, Nanjing, China).

Six kidneys from each group were immediately fixed with 10% formalin, dehydrated, embedded in paraffin, and sectioned to a thickness of 5 µm. These sections were then stained with hematoxylin and eosin (H&E) and Masson’s trichrome. The kidney injury score was determined based on tubular atrophy and degeneration, renal papillary necrosis, interstitial inflammation, and fibrous hyperplasia as previously reported (Debelle et al., 2002).

Five-micron thick paraffin-embedded kidney sections were deparaffinized, followed by antigen retrieval in ethylenediaminetetraacetic acid (1 mM). The samples were blocked with 0.3% H₂O₂ in methanol and 5% BSA. Kidney sections were incubated with α-SMA (1:200, 14395-1-AP, Proteintech, United States) and TGF-β1 (1:200, ab92486, Abcam, United States) primary antibodies overnight at 4°C, followed by horseradish peroxidase (HRP)-conjugated secondary antibodies (PV9001, Beijing Zhongshan Jinqiao Biotechnology Co., Ltd., Beijing). The reaction was visualized with DAB staining using a Leica Aperio Versa 8 system (Leica, Wetzlar, Germany). The cumulative optical density of the area of interest analysis was calculated using ImageJ software.

Total RNA of rat kidney cortical tissue (three samples per group) was extracted using TRIzol (Invitrogen, Carlsbad, CA, United States). After total RNA extraction, eukaryotic mRNA was enriched with oligo (dT) beads, and rRNA-enriched prokaryotic mRNA was removed with a Ribo-Zerotm magnetic kit (Epicentre). Then, fragmentation buffer was used to fragment the enriched mRNA fragments into short fragments, which were reverse-transcribed into cDNA with random primers. The second strand of cDNA was synthesized by DNA polymerase I, RNase H, dNTP and buffer. The cDNA fragment was then purified with a QIAquick PCR extraction kit, the ends were repaired, poly (A) was added and the fragments were attached to the Illumina sequencing adapter. The ligation products were detected by agarose gel electrophoresis, PCR amplification and Illumina HiSeqTM 2,500 sequencing.

First, the raw data were filtered, and the clean data obtained after filtering were compared to the reference genome of the species. Second, the expression level of each gene was calculated according to the comparison results. On this basis, differential expression analysis, enrichment analysis and clustering analysis of the samples were further performed. Finally, we used DESeq for gene expression analysis and screening of differentially expressed genes as follows: multiple expression differences | log2FoldChange |>1, significance p-value<0.05. The Pearson correlation coefficient between all samples was calculated using the function cor, and hierarchical clustering was performed using the hclust function in the stats package in R software. Then, GO function and KEGG pathway enrichment analyses were performed on the differentially expressed genes (DEGs).

Mitochondria were extracted from renal tissue using a mitochondrial extraction kit (C3606, Beyotime, China). Briefly, fresh kidneys were harvested, cut into small pieces, and washed thrice with precooled PBS. After digestion with trypsin, the renal tissues were homogenized in mitochondrial isolation reagent using a Dounce tissue grinder on ice. The homogenate was then centrifuged at 600 g for 10 min, and the supernatant was centrifuged at 1,500 g for 15 min to isolate the mitochondria. The change in MMP was measured by the JC-1 fluorescent probe, and the JC-1 red/green fluorescence intensity ratio was used to represent MMP. Fresh isolated mitochondria were incubated with 10 μg/ml JC-1 (at 37°C for 20 min), and the fluorescence intensity was measured by a Synergy H1 fluorometer microplate reader (BioTek, United States).

Renal tissues were homogenized followed by ultrasonic disruption to obtain renal tissue homogenates and then centrifuged at 3,000 rpm for 15 min at 4°C. Dihydroethidium (DHE) fluorescence was used to detect the level of ROS in rat renal tissues. Detection of the malondialdehyde (MDA) level as well as the superoxide dismutase (SOD) and glutathione peroxidase (GSH-Px) activities was performed according to the instructions provided by Shanghai Biyuntian Biotechnology Co., Ltd. (Shanghai, China).

Frozen sections were used to assess colocalization of Pink1 and TOM20. Kidney sections were blocked with 5% bovine serum albumin for 30 min at room temperature, followed by incubation overnight at 4°C with primary antibodies against Pink1 (1:200, SC517353, Santa Cruz, United States) and TOM20 (1:200, 11802-1-AP, Proteintech, United States). After the samples were washed with PBS, an Alexa Fluor 488-conjugated goat anti-mouse secondary antibody (1:300) and Alexa Fluor 594-conjugated goat anti-rabbit secondary antibody (1:300) were added for 1 h at room temperature. Finally, the slides were stained with DAPI solution for 10 min and captured by a laser scanning confocal fluorescence microscope (Olympus FV 1000, Japan). Ten nonoverlapping high-power fields (40X) were randomly captured in each specimen and analyzed by ImageJ software.

For western blot assays, renal tissues were lysed and homogenized in RIPA buffer supplemented with protease inhibitor cocktail l (C0001-1, Targetmol, China) and quantified with a BCA kit (P0013C, Beyotime, China). Protein sample extracts (30 mg/lane) were separated by SDS-PAGE and transferred onto a polyvinylidene difluoride membrane (PVDF). After the membranes were blocked with 5% BSA, they were incubated with the primary antibodies at 4°C overnight, followed by HRP-conjugated secondary antibody. Then, the membranes were incubated with HRP-conjugated secondary antibody (Boster Biological Technology Co., Ltd., China) at room temperature for 1 h. Films were scanned by a ChemiScope 6,000 system (Qinxiang, Shanghai, China). ImageJ software was used to measure the protein bands based on that of GAPDH.

Frozen sections were used to assess colocalization of Pink1 and TOM20. Kidney sections were blocked with 5% bovine serum albumin for 30 min at room temperature, followed by incubation overnight at 4°C with primary antibodies against Pink1 (1:200, SC517353, Santa Cruz, United States) and TOM20 (1:200, 11802-1-AP Proteintech, United States). After the samples were washed with PBS, an Alexa Fluor 488-conjugated goat anti-mouse secondary antibody (1:300) and Alexa Fluor 594-conjugated goat anti-rabbit secondary antibody (1:300) were added for 1 h at room temperature. Finally, the slides were stained with DAPI solution for 10 min and captured by a laser scanning confocal fluorescence microscope (Olympus FV 1000, Japan). Ten nonoverlapping high-power fields (40X) were randomly captured in each specimen and analyzed by ImageJ software.

GraphPad Prism software was used for statistical analysis. Quantitative data are expressed as the mean ± standard error of the mean (SEM). One-way ANOVA was used for all experimental data followed by Dunnett’s test. A p value <0.05 was considered significant.

To evaluate the major chemical components, we analyzed TLYS decoction using UHPLC-MS in positive and negative ion mode (Figure 1). Thirty-seven compounds (6 organic acids, 5 diterpene quinones, 4 sterones, 6 flavonoids, 10 phenolic acids and 6 other compounds) were detected at relatively high levels. The detailed information is shown in Table 2.

To study the effects of TLYS decoction on renal fibrosis, we adopted a rat model of UUO with 14-days TLYS treatment. We found that in the UUO group, the level of Scr was significantly higher than that in the sham group (66.79 ± 1.93 μmol/L vs 40.14 ± 1.83 μmol/L), and TLYS treatment partly decreased the levels to 57.25 ± 1.20 μmol/L (Figure 2A). Similarly, the rats with UUO showed significantly higher levels (4.36 ± 0.23 mmol/L) of BUN than the control rats (3.04 ± 0.11 mmol/L), while TLYS treatment significantly reduced the BUN levels to 3.52 ± 0.12 mmol/L (Figure 2B). As a positive control group, losartan (30 mg/kg) decreased serum concentration of BUN (p < 0.05) but not Scr. H&E staining and Masson’s trichrome staining were used to evaluate renal pathological injury (Figure 2C). H&E staining showed that the UUO group exhibited notable tubule atrophy and lumen dilation with diffuse interstitial inflammation. TLYS treatment attenuated kidney tubulointerstitial injury following UUO whereas the positive control, losartan, decreased the tubulointerstitial injury (Figure 2D). Masson’s staining showed substantial interstitial fibrosis in the UUO group; this fibrosis was significantly attenuated in the TLYS-treated rats (Figure 2E). These data suggested that treatment with TLYS could significantly mitigate renal injury.

Given the protective effect of TLYS on renal fibrosis, we investigated the expression of α-SMA and TGF-β1 by immunohistochemistry. As shown in Figures 3A,B, the expression of α-SMA in the tubular interstitium was much higher in the UUO group than in the sham group. However, in the TLYS group, this increase was significantly suppressed. In addition, compared with the sham group, the UUO group showed a dramatic increase in TGF-β1 levels, while TLYS treatment significantly inhibited this abnormal increase in TGF-β1 (Figures 3A,C). Similarly, valsartan treatment significantly reduced the level of α-SMA and TGF-β1 (Figures 3A–C).

To explore the mechanisms of TLYS decoction, we performed RNA-Seq analysis. There were 1,541 DEGs with upregulated expression and 877 DEGs with downregulated expression in the UUO group relative to the sham group. Ninety-five DEGs had upregulated expression and 200 DEGs had downregulated expression in the TLYS group relative to the UUO group. Additionally, the differences were significant (|log2 (fold-change)| > 1 and p < 0.05, Figures 4A,B). Next, these overlapping DEGs were analyzed by GO analysis of biological processes, cellular components, and molecular functions. The number of DEGs with upregulated expression was significantly higher, whereas the number of DEGs with downregulated expression was lower in the UUO group than in the sham group (Figure 4C). After TLYS treatment, the number of DEGs with downregulated expression significantly increased and was greater than that of DEGs with upregulated expression (Figure 4D). These results further showed that TLYS has a comprehensive regulatory effect in the rats with UUO.

To screen out the most representative DEG group affected by TLYS, we performed a trend analysis of all selected DEGs. DEGs were divided into 8 categories (Figure 5A), of which profiles 2, 6 and 7 were significant (Figure 5B). We calculated the proportion of each DEG trend in the corresponding pathways, and signal transduction was commonly affected by the above three DEG trends (Figure 5C). Then, by KEGG enrichment analysis, three significant pathways were identified for TLYS treatment; the pathways were phagosomes (a membrane source in the earlier stages of autophagy), lysosome (an important component of autolysosome), and oxidative phosphorylation, which may participate in mitochondrial function (Figure 5D). Therefore, TLYS treatment might be associated with mitochondrial dysfunction and mitophagy in the rats with UUO.

A reduction in MMP suggests damage to mitochondrial function, which is indicated by a lower ratio of red-to-green fluorescence. The MMP was lower in the UUO group than in the sham group, but TLYS increased the MMP level (Figure 6A). Mitochondrial division and fusion are crucial for maintaining morphology and function. The expression of proteins related to mitochondrial fusion (Mfn1 and Mfn2) showed a significant reduction in the UUO group (p < 0.01), and this trend was reversed after TLYS treatment (p < 0.01) (Figures 6B–D). As shown in Figures 6B,E,F, the expression of the proteins related to mitochondrial fission (Drp1 and Mff) was upregulated in the UUO group compared with the sham group, and TLYS treatment significantly prevented this trend.

Oxidative stress induced by the accumulation of ROS causes serious damage to mitochondria; therefore, we investigated the effects of TLYS on oxidative stress. Our results demonstrated that the activities of SOD and GSH-PX were significantly lower and that the content of ROS and MDA was higher in the UUO group than in the sham group. Interestingly, TLYS treatment significantly enhanced the levels of SOD and GSH-PX and inhibited the increase in ROS and MDA (Figures 7A–D). The results indicated that TLYS could ameliorate oxidative stress in the rats with UUO.

Damaged mitochondria can be degraded by mitophagy, which is driven by Pink1/Parkin signaling. In our study, we assessed the protein expression of this pathway. Compared with the sham group, the UUO group showed significantly increased protein levels of Pink1 and Parkin in the kidney. However, TLYS treatment significantly decreased the expression of Pink1 and Parkin. Moreover, immunofluorescence confirmed and further revealed that Pink1 in the renal tubular epithelial cell was diminished in the rats with UUO compared with the sham rats (Figures 8A–C). The colocalization of mitochondria (marked by TOM20) and Pink1 was markedly decreased in the TLYS group compared with the UUO group (Figure 8D). These results suggested that the translocation of Pink1 from the cytoplasm to mitochondria machinery is inhibited, resulting in the accumulation of Pink1 at the outer mitochondrial membrane in the UUO group, and TLYS could reverse this change. These data indicated that TLYS inhibited Pink1/Parkin-mediated mitophagy in the rats with UUO.