Abstract

   The balanced crosstalk between miRNAs and autophagy is essential in
   hypertensive nephropathy. Hydrogen sulfide donors have been reported to
   attenuate renal injury, but the mechanism is unclear. We aimed to
   identify and verify the miRNAs and autophagy regulatory networks in
   hypertensive nephropathy treated with hydrogen sulfide donors through
   bioinformatics analysis and experimental verification. From the miRNA
   dataset, autophagy was considerably enriched in mice kidney after
   angiotensin II (AngII) and combined hydrogen sulfide treatment
   (H[2]S_AngII), among which there were 109 differentially expressed
   miRNAs (DEMs) and 21 hub ADEGs (autophagy-related differentially
   expressed genes) in the AngII group and 70 DEMs and 13 ADEGs in the
   H[2]S_AngII group. A miRNA–mRNA–transcription factors (TFs) autophagy
   regulatory network was then constructed and verified in human
   hypertensive nephropathy samples and podocyte models. In the network,
   two DEMs (miR-98-5p, miR-669b-5p), some hub ADEGs (KRAS, NRAS), and one
   TF (RUNX2) were altered, accompanied by a reduction in autophagy flux.
   However, significant recovery occurred after treatment with endogenous
   or exogenous H[2]S donors, as well as an overexpression of miR-98-5p
   and miR-669b-5p. The miR/RAS/RUNX2 autophagy network driven by H[2]S
   donors was related to hypertensive nephropathy. H[2]S donors or miRNAs
   increased autophagic flux and reduced renal cell injury, which could be
   a potentially effective medical therapy.

   Keywords: hypertensive nephropathy, hydrogen sulfide donors, miRNA,
   autophagy, bioinformatics analysis

1. Introduction

   Hypertension is recognized as one of the leading causes linked to the
   development of cardiovascular and renal disease [[32]1,[33]2], and it
   also promotes the progression of chronic kidney disease towards
   end-stage renal disease [[34]3]. Hypertensive nephropathy is mainly
   manifested as renal structural and functional damage caused by
   essential hypertension, among which renal inflammation is an important
   pathological feature [[35]4,[36]5]. In essential hypertension, the
   renin–angiotensin–aldosterone system (RAAS) is extensively activated,
   and the expression of its main effector angiotensin II (AngII) is
   upregulated. These subsequently lead to the dysfunction of renal
   endothelial cells, the release of various inflammatory factors, and the
   proliferation of mesangial cells, thereby inducing vascular remodeling,
   promoting renal inflammation and fibrosis, and ultimately leading to a
   decreased glomerular filtration rate, glomerular ischemia, sclerosis,
   and nephron atrophy [[37]6,[38]7]. Therefore, AngII is an important
   mediator of hypertensive nephropathy, and it is also the modeling
   condition chosen by many studies [[39]8]. In addition, many studies
   have shown that damage to and structural changes in podocytes are very
   important in renal injury [[40]9]. Here, we employed AngII-induced
   podocytes as a cell model. Autophagy is an important and highly
   conserved pathway, which acts as an essential role in maintaining the
   homeostasis of physiological and pathological processes, such as
   immunity, inflammation, aging and metabolism, etc. [[41]10].
   Accumulating evidence suggests that hypertensive nephropathy may be
   associated with the imbalance of autophagy regulation, and targeting
   the autophagy pathway may alleviate hypertension-induced kidney damage
   [[42]9,[43]11].

   Hydrogen sulfide (H[2]S) is an important physiological gas molecule,
   and it along with nitric oxide (NO) and carbon monoxide (CO) are
   considered as the three major gas signaling molecules in the body
   [[44]12]. Endogenous hydrogen sulfide is synthesized by
   cystathionine-γ-lyase (CSE), cystathionine-β-synthase (CBS), and
   3-mercaptopiruvate sulfurtransferase (MST) enzymes [[45]13]. Previous
   studies have shown that hydrogen sulfide is involved in the regulation
   of various physiological and pathological conditions, including the
   cardiovascular, renal, respiratory, central nervous, and digestive
   systems [[46]14]. In common kidney diseases, such as CKD and acute
   kidney injury, hydrogen sulfide production is reduced, and the
   expression of CBS, CSE, or MST is also downregulated to varying degrees
   [[47]15,[48]16,[49]17]. Treatment with hydrogen sulfide donors not only
   restores hydrogen sulfide levels but also ameliorates kidney function
   [[50]18]. Multiple studies have shown that hydrogen sulfide donors
   alleviate renal damage by regulating autophagy, oxidative stress,
   inflammation, and cell injury through various signaling pathways
   [[51]19]. However, few studies have focused on the epigenetic
   regulation of hydrogen sulfide donors in the treatment of hypertensive
   nephropathy, especially on microRNA regulation.

   MicroRNAs (miRNAs, miRs) are a group of single-stranded non-coding RNAs
   with a length of 19–22 nucleotides, which are important regulators of
   gene post-transcriptional expression [[52]20]. By specifically binding
   to the 3′UTR sequences of mRNA, miRNAs lead to the degradation or
   post-transcriptional inhibition of target genes and then participate in
   the regulation of different biological processes [[53]21]. The
   dysregulation of miRNAs is important in the occurrence and development
   of various kidney diseases [[54]22]. In hypertensive nephropathy, many
   miRNAs, such as miR-200a, miR-200b, miR-141, miR-429, miR-205 and
   miR-192, have increased expression [[55]23]. Studies have shown that
   angiotensin-converting enzyme inhibitors downregulate the expression of
   miR-324-3p in glomeruli and tubules, thereby reducing renal fibrosis
   [[56]24]. However, few studies have addressed miRNAs regulating
   autophagy pathways in hypertensive nephropathy. In addition, a growing
   number of studies have shown that hydrogen sulfide can also regulate
   miRNAs to improve cardiac and renal dysfunction [[57]25,[58]26]. Since
   the role of miRNAs in hypertensive nephropathy is emerging, the
   functional miRNAs and corresponding key mRNAs that regulate the
   autophagy pathway in hypertensive nephropathy and hydrogen sulfide
   donor therapy are explored to construct the miRNA-mRNA regulatory
   network, which is helpful for elucidating the molecular mechanism of
   hydrogen sulfide donors in the treatment of hypertensive nephropathy.

   Herein, we screened out differentially expressed miRNAs (DEMs) after
   AngII or a combination of AngII and hydrogen sulfide treatment using
   the Gene Expression Omnibus (GEO) database. Then, their target mRNAs
   were predicted by the validation and prediction databases, the top 5%
   hub mRNAs were identified using the protein–protein interaction (PPI)
   network and CytoHubba, and potential autophagy-related differentially
   expressed genes (ADEGs) were identified by the autophagy-related gene
   set. Subsequently, functional annotation and pathway enrichment
   analysis were conducted and a miRNA-ADEGs-TF (transcription factor)
   regulatory network was constructed and validated to better understand
   the underlying mechanism of the hydrogen sulfide treatment of
   hypertensive nephropathy.

2. Materials and Methods

2.1. Data Acquisition

   In order to obtain genetic analysis data for hypertensive nephropathy
   and hydrogen sulfide treatment, we used “hypertension”, “kidney”, and
   “hydrogen sulfide” as keywords to search for genome-wide expression
   studies in the GEO database ([59]https://www.ncbi.nlm.nih.gov/geo/,
   accessed on 1 October 2022). The non-coding RNA expression profiling
   dataset [60]GSE92857 was screened out and downloaded using the R
   package “GEO query” (v 2.62.2). The [61]GSE92857 dataset, containing 12
   kidney tissues each from normal control (Ctrl), AngII-treated (AngII),
   hydrogen-sulfide-treated, and a combination of AngII and hydrogen
   sulfide (AngII_H[2]S)-treated mice, was compiled on the [62]GPL19117
   ([miRNA-4] Affymetrix Multispecies miRNA-4 Array) platform [[63]26].
   This dataset was normalized with the RMA (robust multiarray analysis)
   algorithm using the Partek Genomics Suite software (Partek v6.6, St.
   Louis, MO, USA), and the independent [64]GSE37455 and [65]GSE37460
   datasets containing human hypertensive nephropathy samples were
   utilized for validation. The workflow chart of the methodology used in
   this study is shown in [66]Figure 1.

Figure 1.

   [67]Figure 1
   [68]Open in a new tab

   The flowchart of screening strategy. Ctrl, control kidney at 28 days.
   AngII, mice treated with AngII via mini osmotic pump for 28 days.
   AngII_H[2]S, mice treated with AngII via mini osmotic pump and hydrogen
   sulfide donor via i.p. injection for 28 days. DEMs, differentially
   expressed miRNAs. ADEGs, potential autophagy differentially expressed
   genes. GO, Gene Ontology. KEGG, Kyoto Encyclopedia of Genes and
   Genomes. PPI, protein–protein interaction. TF, transcription factor.

2.2. Analysis of Differentially Expressed miRNAs

   The normalized data matrix of the miRNA profiling dataset was acquired
   and represented as boxplot and density plot. The probes in the dataset
   were then annotated with the annotation file from the [69]GPL19117
   platform, and only probes interrogating M. Muculus transcripts were
   included. We screened out DEMs comparing the AngII and Ctrl groups and
   the AngII_H[2]S and AngII groups using the “limma” package (v 3.50.3)
   of the R software (v 4.1.2) [[70]27]. MiRNAs with a p value less than
   0.05 and an absolute fold change (FC) greater than 1.5 were considered
   as DEMs. Box line, density, and volcano plots of DEMs were generated
   using the “ggplot2” package (v 3.3.6) [[71]28], and the circular
   heatmap of the DEMs was illustrate through the TBtools software (v
   1.098769).

2.3. Prediction and Identification of Hub Target Genes

   The multiMiR package is a database of miRNAs and target genes, disease,
   and drug interactions, including 8 prediction databases
   (DIANA-microT-CDS, ElMMo, MicroCosm, miRanda, miRDB, PicTar, PITA, and
   TargetScan), 3 experimentally validated databases (miRecords,
   miRTarBase, and TarBase), and 3 databases containing
   drug-/disease-related information (miR2Disease, Pharmaco-miR, and
   PhenomiR) [[72]29]. The predicted and validated target genes of the
   DEMs were selected by multiMiR package (v 2.3) analysis, and the
   intersection of the two sets was selected as the target genes of DEMs.
   The PPI network of these intersecting target genes was then constructed
   using the Cytoscape software (v 3.9.1), followed by the Cytohubba (v
   0.1) plug-in’s Degree algorithm to take the top 5% genes as hub target
   genes [[73]30,[74]31].

2.4. Selection of Autophagy-Related Genes

   The autophagy-related gene set in Mus musculus was collected from the
   entries mmu04140 and mmu04136 in the Kyoto Encyclopedia of Genes and
   Genomes (KEGG) pathway database
   ([75]https://www.genome.jp/entry/mmu04140,
   https://[76]www.genome.jp/entry/mmu04136, accessed on 1 November 2022).
   The autophagy-related gene set is provided in [77]Table S1. In
   addition, we identified potential ADEGs via the Venn diagram package (v
   1.7.3).

2.5. Functional and Pathway Enrichment Analysis of ADEGs

   To evaluate the biological functions of the DEMs targeting ADEGs, Gene
   Ontology (GO) enrichment analysis and KEGG pathway analysis were
   performed using the “ClusterProfiler” package (v 4.2.2) in the R
   software. Among them, the species was set to Mus musculus, and the
   Benjamini–Hochberg method was used to adjust the p value. Finally, the
   enrichments or pathways with adjusted p values of less than 0.05 were
   screened out for analysis and plot display.

2.6. PPI Network and Recognition of Hub ADEGs and Core Modules

   The PPI network of ADEGs was visualized using the STRING database
   ([78]https://string-db.org/, v 11.5, accessed on 8 November 2022), with
   the interaction score setting at a high confidence (0.700) [[79]32].
   Afterward, the analysis data of the PPI network was imported into the
   Cytoscape software (v 3.9.1), and the CytoHubba plug-in (v 0.1) was
   used to assign values to each gene through the topological network
   algorithm, and then the top 10 genes of the “Degree” attribute were
   regarded as hub ADEGs [[80]31]. In addition, the Molecular Complex
   Detection (MCODE, v 2.0.2) plug-in was utilized to analyze the core
   modules [[81]33].

2.7. Construction of miRNA-mRNA-TF Regulatory Network

   Based on the results of the screened DEMs and targeted ADEGs, an
   autophagy-related miRNA-mRNA regulatory network was constructed using
   the Cytoscape software (v 3.9.1). Transcription factors (TFs) and
   miRNAs are two important regulators of gene expression, often
   controlling gene regulatory networks in a dependent or independent
   manner [[82]34,[83]35]. In order to further study the molecular
   mechanism of hydrogen sulfide regulation of autophagy, we used the
   iRegulon plug-in (v 1.3) to predict potential transcription factors
   associated with ADEGs [[84]36]. Ultimately, transcription factors
   enriched in the top 5 fractions, miRNAs co-regulated by AngII and
   hydrogen sulfide, and corresponding ADEGs were selected to construct an
   autophagy-related miRNA-mRNA-TF regulatory network.

2.8. Drugs and Reagents

   S-Propargyl-cysteine (SPRC), a substrate donor for H[2]S synthesis, was
   synthesized and purified as described in our previous study
   [[85]37,[86]38]. NaHS, interferon-γ, DL-Propargylglycine (PAG), and
   angiotensin II (AngII) were purchased form Sigma-Aldrich (St. Louis,
   MO, USA). Losartan was obtained from MCE (Shanghai, China). RPMI 1640
   medium, fetal bovine serum (FBS), penicillin/streptomycin solution, and
   0.25% trypsin were purchased from Gibco (New York, NY, USA). RNAiso
   Plus and RNAiso for small RNA reagents were obtained from TaKaRa
   (Shiga, Japan). A Hifair III 1st Strand cDNA Synthesis Kit (gDNA
   digester plus) and Hieff miRNA Universal qPCR SYBR Master Mix were
   obtained from Yeasen(Shanghai, China). ABScript III RT Master Mix for
   qPCR and 2X Universal SYBR Green Fast qPCR Mix were obtained from
   ABclonal (Wuhan, China). The miR-98-5p mimic (micrON mmu-miR-98-5p
   mimic), miR-669b-5p (micrON mmu-miR-669b-5p mimic), miR-NC (micrON
   mimic NC), and riboFECT™ CP were obtained from RiboBio (Guangzhou,
   China).

2.9. Cell Culture

   Conditionally immortalized mouse podocyte MPC5 cells were purchased
   from ATCC (USA). The MPC5 cells were cultured and differentiated as
   previously described [[87]39]. The MPC5 cells were first cultured at 33
   °C in RPMI 1640 medium containing 10% FBS, 1% penicillin/streptomycin,
   and 30 U/mL recombinant mouse interferon-γ. Then, the MPC5 cells were
   placed at 37 °C and cultured in complete RPMI 1640 medium without
   interferon for 10–14 days, and further experiments were performed after
   differentiation and maturity. When the MPC5 cells reached 80%
   confluence, they were digested with 0.25% trypsin and passaged into new
   culture dishes at a ratio of 1:3 to continue culturing.

   For miRNAs mimic transfection, MPC5 cells were seeded into a 6-well
   plate containing complete culture medium the day before transfection,
   allowing the cell density to be about 40% at the time of transfection.
   Then, according to the manufacturer’s instructions, the transfection
   mixture was prepared and added dropwise to the cells in complete medium
   without penicillin/streptomycin, and the culture was continued in a 37
   °C incubator for 48 h for subsequent AngII treatment and other
   operations.

   For the H[2]S donor treatment, MPC5 cells were treated in the following
   groupings: Ctrl group (vehicle), AngII group (1 μM AngII treatment),
   NaHS or SPRC group (pre-treated with 50 μM NaHS or 50 μM SPRC for 1 h
   before exposure to 1 μM AngII), and SPRC + PAG group (pre-incubated
   with 2 mM PAG for 30 min and then pre-treated with 50 μM SPRC for 1 h
   before exposure to 1 μM AngII). For quantitative PCR and WB, AngII was
   stimulated for 24 or 72 h, respectively. NaHS, SPRC, and PAG were
   freshly prepared before each experiment.

2.10. Quantitative PCR

   The total RNA and miRNA were extracted using RNAiso Plus and RNAiso for
   small RNA reagents according to the manufacturer’s protocol. For the
   mRNA targets, total RNA was reversed into cDNA using ABScript III RT
   Master Mix. Subsequent quantitative PCR (qPCR) experiments were
   performed in a Bio-Rad CFX96 PCR instrument using 2X Universal SYBR
   Green Fast qPCR Mix. For miRNAs, complementary DNA (cDNA) was
   synthesized using specific stem–loop primers for miR-98-5p and
   miR-669b-5p through a Hifair III 1st Strand cDNA Synthesis Kit. Then,
   qPCR analyses were run in triplicate to calculate miRNAs expression
   using Hieff miRNA Universal qPCR SYBR Master Mix. U6 snRNA and Gapdh
   were used as a reference for miRNAs and mRNAs. All the primers used are
   shown in [88]Table S2.

2.11. Western Blot

   Western blot (WB) analysis was performed as previously reported
   [[89]40]. Briefly, proteins in MPC5 cells were lysed and extracted with
   RIPA buffer (Yeasen, China) and then boiled in 5× loading buffer
   (Yeasen, China). Subsequently, the protein concentration in each sample
   was determined using a BCA kit (Yeasen, China). Samples with the same
   amount of protein were subjected to SDS-PAGE (sodium dodecyl
   sulfate–polyacrylamide gel electrophoresis) and transferred to
   polyvinylidene fluoride (PVDF, Millipore, Burlington, MA, USA)
   membranes. The membranes were then blocked with 5% nonfat dry milk
   (Cell Signaling, Danvers, MA, USA) and incubated with the following
   primary antibodies: anti-PTEN, anti-KRAS, anti-NRAS, anti-RUNX2,
   anti-CSE, anti-MST (1:1000, Abclonal, Wuhan, China), anti-MAPK1
   (Erk1/2), anti-LC3B (1:1000, Cell Signaling, Danvers, MA, USA),
   anti-P62 (SQSTM1, 1:5000), anti-nephrin, anti-podocin, anti-CBS
   (1:1000, Proteintech, Rosemont, IL, USA), and anti-GAPDH (1:10,000,
   ABclonal, Wuhan, China). After washing with 1X TBST (Tris-buffered
   saline with tween 20), secondary HRP goat anti-rabbit or anti-mouse
   antibodies (1:10,000, Abclonal, Wuhan, China) combined with immobilon
   Western chemiluminescent HRP substrate (Millipore) were used to
   visualize bands under a ChemiDoc XRS+ system (BIO-RAD, Hercules, CA,
   USA). The grey value of each band was calculated using the Image J
   software (v1.54f, NIH, Bethesda, MA, USA), and the relative expression
   of the target protein was determined using the internal reference
   protein GADPH.

2.12. H[2]S Content Assay

   H[2]S content was measured as per the manufacturer’s instructions
   (H[2]S content assay kit, Solarbio, Beijing, China). In brief, a
   certain number of cells were collected using the extraction solution,
   and they were then subjected to ultrasound and centrifugation. The
   reaction reagents were added and incubated at room temperature for 10
   min. Subsequently, the absorbance of the solution at 665 nm was
   detected using a microplate reader (Varioskan Flash, Thermo Scientific,
   Waltham, MA, USA), and the corresponding H[2]S concentration was
   obtained by substituting it into the standard curve of the standard
   solution concentration.

   For fluorescence detection, a WSP-1 fluorescent probe (MCE, Shanghai,
   China) for H[2]S was used. After stimulation, 10 μM WSP-1 was added and
   incubated at 37 degrees for 60 min. After a quick wash with PBS twice,
   10 μg/mL Hoechst 33342 staining solution (Yeasen, Shanghai, China) was
   added and incubated for 10 min. Finally, observations and recordings
   were made under an Olympus CKX53 inverted fluorescence microscope
   (Tokyo, Japan).

2.13. Statistics

   In addition to the above-mentioned R package, the GraphPad Prism 9 (v.
   9.0.0, Boston, MA, USA) software was used for graphing and statistical
   analysis. For pairwise comparisons between more than two group sets,
   one-way ANOVA was performed, and Tukey’s multiple comparison post hoc
   test was utilized to correct the p value. Statistical significance was
   considered when the resulting p value was less than 0.05. All
   experiments were repeated at least three times independently, and data
   are shown as mean ± SD.

3. Results

3.1. MiRNAs Were Dysregulated in Mouse Kidney after AngII Induction and H[2]S
Donor Treatment

   A dataset of miRNA expression in kidneys from AngII- and/or
   H[2]S-treated mice was downloaded from the GEO database. The
   distribution trend in the boxplot of this dataset was basically a
   straight line ([90]Figure S1A), and the density plots of the values in
   the different groups were basically coincident ([91]Figure S1B). The
   results showed that the data quality was good and had good
   repeatability.

   We revealed a total of 109 DEMs between the AngII and Ctrl groups, with
   74 miRNAs upregulated and 35 miRNAs downregulated according to the
   principle of absolute FC > 1.5 and p value < 0.05 ([92]Figure 2A,C).
   After the AngII and H[2]S donor treatment, a total of 70 DEMs were
   screened out in comparison with the AngII alone group, with 33 miRNAs
   upregulated and 37 miRNAs downregulated ([93]Figure 2B,D). The
   hierarchical clustering analysis and volcano plots were used to assess
   the variation between the different groups ([94]Figure 2). All the DEMs
   are listed in [95]Supplementary Tables S3 and S4. These results showed
   the presence of abundant altered miRNAs in mouse kidneys after
   treatment with AngII alone or in combination with hydrogen sulfide
   donors.

Figure 2.

   [96]Figure 2
   [97]Open in a new tab

   Screening of DEMs in AngII- or H[2]S-treated mouse kidney. Clustered
   heatmap of DEMs between AngII and Ctrl groups (A) as well as comparison
   of AngII and H[2]S donor treatments (AngII_H[2]S) and AngII alone group
   (B). Green label represents Ctrl group, fuchsia represents AngII group,
   and khaki represents AngII_H[2]S group. The miRNA expression values
   were processed by log base 2 and are represented from blue to red in
   ascending order. (C) The volcano plot of DEMs between AngII and Ctrl
   groups. (D) The volcano plot of DEMs between AngII_H[2]S and AngII
   groups. Each of the top five miRNAs with the largest difference in
   forward and reverse changes are marked. Down means downregulated
   miRNAs, and these are shown as blue dots. Up, presented as red dots,
   means upregulated miRNAs. The gray dots show the not significantly
   changed miRNAs. The criteria were absolute FC > 1.5 and p value < 0.05.

3.2. The Target Genes of These Dysregulated miRNAs Were Mainly Concentrated
in the Autophagy Pathway

   In order to identify the potential functions of these DEMs, we used
   eight prediction databases and three validation databases in the
   multiMiR database to find the target genes of these DEMs. In total,
   5550 validated and 15,194 predicted target genes and 3143 validated and
   14,779 predicated target genes were found in the AngII vs. Ctrl and
   AngII_H[2]S vs. AngII groups, respectively. Next, overlapping genes
   among the predicted and validated genes were selected as candidate
   target genes by crossover analysis. A total of 4667 and 2828 candidate
   target genes were selected accordingly ([98]Figure 3A,B). To further
   screen hub target genes, candidate genes were imported into the STRING
   database and Cytoscape software, and the genes with the top 5% of the
   PPI scores were marked as hub target genes according to the Cytohubba
   plug-in. A total of 233 and 141 hub target genes were marked in the
   AngII vs. Ctrl and AngII_H[2]S vs. AngII groups, respectively, and
   these were displayed with the PPI network ([99]Figure 3C,D).
   Afterwards, these hub target genes were interacted with the
   autophagy-related gene set to obtain 21 and 13 potential ADEGs,
   respectively ([100]Figure 3E,F and [101]Supplementary Table S5).

Figure 3.

   [102]Figure 3
   [103]Open in a new tab

   Hub target genes and ADEGs were selected for DEMs associated with AngII
   alone or AngII plus H[2]S donor treatment. Venn diagrams showing
   predicted target genes (red) and validated target genes (sky blue)
   corresponding to DEMs in comparison between AngII vs. Ctrl (A) and
   AngII_H[2]S vs. AngII (B). PPI network displaying the interaction of
   hub target genes from comparison between AngII vs. Ctrl (C), and
   AngII_H[2]S vs. AngII (D). The interacting genes between hub target
   genes and autophagy-related gene set in AngII vs. Ctrl (E) and
   AngII_H[2]S vs. AngII (F) comparison.

   In an effort to gain insight into the functions and mechanisms of these
   ADEGs, the GO and KEGG databases were used to enrich GO terms,
   including biological process (BP), cellular component (CC), and
   molecular function (MF), as well as the KEGG pathway via the
   ClusterProfiler package. In the comparison between the AngII and Ctrl
   groups, the enrichment analysis of the top 30 GO terms is shown in
   [104]Figure 4A. In the biological process category, AngII mainly
   affected the phosphorylation and activation of proteins such as protein
   kinase B signaling, protein phosphorylation, and peptidyl-threonine
   phosphorylation the mouse kidney. In addition, in terms of cellular
   components, AngII mainly altered aspects such as the
   phosphatidylinositol 3-kinase complex and mitochondria, while the major
   molecular functions affected were protein binding and enzymatic
   activity ([105]Figure 4A). In contrast, after treatment with hydrogen
   sulfide donors, protein phosphorylation in biological processes was
   also enriched, in addition to the positive regulation of gene
   expression and the JNK cascade ([106]Figure 4B). In the cellular
   component and molecular function categories, hydrogen sulfide donor
   treatment also altered the phosphatidylinositol 3-kinase complex as
   well as protein binding and kinase activation ([107]Figure 4B).

Figure 4.

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

   Enrichment plots from GO and KEGG functional analysis of targeted
   ADEGs. Barplots of the enrichment analysis for each of the top 10
   biological processes (green), cellular component (orange-red), and
   molecular function (blue) from AngII vs. Ctrl (A) and AngII_H[2]S vs.
   AngII (B). In addition, the barplots are arranged from smallest to
   largest according to the negative logarithm of the adjusted p value.
   The KEGG pathway enrichment analysis results of AngII vs. Ctrl (C) and
   AngII_H[2]S vs. AngII (D) are shown as dotplots. Among them, the size
   of the dots represents the number of ADEGs hitting the corresponding
   pathway, and the adjusted p value of each pathway is displayed from red
   to blue in ascending order. CellP., cellular processes; EnvIP.,
   environmental information processing; HumaD., human diseases; OrgaS.,
   organismal systems.

   The enrichment results of the top 20 KEGG pathways of ADEGs from the
   two comparison groups are shown in [110]Figure 4C,D. In the KEGG
   enrichment analysis, AngII mainly regulated autophagy, the mTOR
   signaling pathway, and the FoxO signaling pathway in kidney
   ([111]Figure 4C). Similarly, the H[2]S donor treatment also modulated
   autophagy, the FoxO signaling pathway, and the ErbB signaling pathway
   ([112]Figure 4D). These enriched GO terms and KEGG signaling pathways
   may be the main mechanism involved in the H[2]S treatment of
   hypertensive renal injury caused by AngII.

3.3. The Hub ADEGs of These Dysregulated miRNAs Were Analyzed

   To further investigate the function of the targeted ADEGs at the
   protein level and reveal the core interactions affected by hydrogen
   sulfide donor treatment, we constructed PPI networks from the Ctrl vs.
   AngII and AngII vs. AngII_H[2]S groups using the STRING database
   ([113]Figure 5A,B). The PPI networks of Ctrl vs. AngII with 21 nodes
   and 196 interaction relationships and AngII vs. AngII_H[2]S with 13
   nodes and 94 edges are shown in [114]Figure 5A and B, respectively.
   Then, the top 10 hub ADEGs were determined based on the cytoHubba
   Degree algorithm using the Cytoscape software. After the AngII
   treatment, the hub ADEGs identified were Kras, Irs1, Igf1r, Pik3ca,
   Mapk1, Pten, Pik3r1, Rps6kb1, Pdpk1, and Prkcd. ([115]Figure 5C). As
   for the H[2]S donor treatment, Kras, Nras, Mapk1, Prkcd, Mapk8, Pten,
   Pik3ca, Mapk9, Prkcq, and Pik3r3 were found to be hub ADEGs
   ([116]Figure 5D). Meanwhile, the core modules in these PPI networks
   were analyzed using MCODE with the k-score set to 2.0. A sub-network
   consisting of 9 nodes and 32 edges was considered as a core cluster in
   Ctrl vs. AngII, and Pik3ca, Pten, Kras, and Igf1r occupied the center
   of the module ([117]Figure 5E). In addition, the core module of AngII
   vs. AngII_H[2]S contained 10 nodes and 27 edges, and its center was
   Kras ([118]Figure 5F). The hub ADEGs and core modules revealed by these
   results may be important mechanisms by which hydrogen sulfide donors
   and AngII regulate the renal autophagy network.

Figure 5.

   [119]Figure 5
   [120]Open in a new tab

   Protein–protein interaction networks and core modules of ADEGs
   constructed by STRING and Cytoscape. PPI analysis of targeted ADEGs
   from AngII vs. Ctrl (A) and AngII_H[2]S vs. AngII (B). The network
   illustration of top 10 hub ADEGs explored by CytoHubba in comparison
   between AngII vs. Ctrl (C) and AngII_H[2]S vs. AngII (D). Hub ADEGs are
   ranked by Degree value from dark red to orange red. Core modules
   analyzed by molecular complex detection network clustering analysis in
   AngII vs. Ctrl (E) and AngII_H[2]S vs. AngII (F).

3.4. Construction of the miRNA-mRNA-TF Autophagy Regulatory Network for H[2]S
Donor and/or AngII Treatment

   Finally, the DEMs and corresponding potential hub ADEGs of the two
   comparison groups were summarized and visualized using Cytoscape. This
   miRNA-mRNA regulatory network consisted of two main factors, namely
   AngII-induced hypertensive nephropathy and hydrogen sulfide donor
   treatment, as well as 21 differentially expressed miRNAs and 23 key
   ADEGs ([121]Figure 6A). To further elucidate the functions of these
   miRNAs and highlight the key processes of the transcriptional control
   of key ADEGs, regulatory networks of common miRNA–mRNA–transcription
   factors were constructed, respectively. The top five transcription
   factors in the enrichment score were selected to map the regulatory
   network of miR-98-5p and miR-669b-5p. The regulatory network of
   miR-98-5p contained 12 target genes and 5 transcription factors, of
   which the target gene Hif1a was also one of the enriched transcription
   factors ([122]Figure 6B). For miR-669b-5p, the network contained eight
   target genes and the top five transcription factors ([123]Figure 6C).
   Runx2 and Zfp219 are common transcription factors of miR-98-5p and
   miR-669b-5p targeted ADEGs.

Figure 6.

   [124]Figure 6
   [125]Open in a new tab

   Regulatory networks of the DEMs, target hub ADEGs, and transcription
   factors. (A) The miRNA-mRNA autophagy regulatory network in
   AngII-induced hypertensive nephropathy and hydrogen sulfide donor drug
   therapy. Octagons, rectangles, and ovals represent diseases/drugs,
   DEMs, and target hub ADEGs, respectively. The red fonts represent
   miRNAs and mRNAs co-regulated by AngII and H[2]S donors. The
   miRNA-mRNA-TF autophagy regulatory network of miR-98-5p (B) and
   miR-669b-5p (C). Green octagons represent transcription factors, blue
   ovals indicate target genes regulated by miRNAs, and purple ovals show
   target genes regulated by both miRNAs and transcription factors.

3.5. In AngII-Treated MPC5 Cells, the Hydrogen Sulfide Pathway Is
Dysregulated

   Furthermore, we used mouse podocyte MPC5 cells to evaluate the
   therapeutic molecular mechanism of H[2]S donors in hypertensive
   nephropathy. AngII stimulation of MPC5 cells was used to mimic the
   pathological state of hypertensive nephropathy. AngII significantly
   reduced the podocyte-specific markers of nephrin and podocin
   ([126]Figure 7A). In addition, we used NaHS and SPRC, a substrate donor
   for H[2]S synthesis, as endogenous and exogenous H[2]S generators
   [[127]41,[128]42]. NaHS and SPRC alleviated AngII-induced podocyte
   injury ([129]Figure 7A). Similar to hypertensive renal injury in vivo,
   AngII significantly increased the transcription levels of intracellular
   Tnf-α, il-1β, and il-6, while NaHS and SPRC alleviated this
   inflammatory state ([130]Figure 7B). This indicated that AngII puts
   cells in a state of damage and inflammation. Moreover, we also examined
   changes in H[2]S levels and synthases in this cellular state. The H[2]S
   content and fluorescent probe intensity in the MPC5 cells were
   significantly reduced under AngII stimulation, while SPRC partially
   restored the H[2]S level ([131]Figure 7C,D). Furthermore, we monitored
   the changes in H[2]S synthase CBS, CSE, and MST. In the MPC5 cells
   stimulated by AngII, CSE was significantly downregulated, while the
   other two enzymes, CBS and MST, did not change ([132]Figure 7E,F). This
   implicates the dysregulation of H[2]S pathway in AngII-treated MPC5
   cells.

Figure 7.

   [133]Figure 7
   [134]Open in a new tab

   H[2]S pathway in AngII-treated MPC5 cells. (A) Western blot analysis of
   nephrin, podocin, and GAPDH in MPC5 cells under different stimuli. n =
   3. (B) Normalized mRNA expression of Tnf-α, il-1β, and il-6. n = 3. (C)
   H[2]S content using spectrophotometer; n = 4. (D) H[2]S levels using
   fluorescent probe WSP-1. Bar = 50 μm; n = 5. (E) Western blot analysis
   of CBS, CSE, MST, and GAPDH in MPC5 cells under indicated stimuli; n =
   3. (F) Normalized mRNA expression of Cbs, Cse, and Mst; n = 4. # p <
   0.05, ## p < 0.01, ### p < 0.001 vs. AngII group, * p < 0.05, ** p <
   0.01, *** p < 0.001 vs. Ctrl group.

3.6. Changes in miR-98/669b, Related ADEGs, and Transcription Factors Were
Observed in the Autophagy Regulatory Network

   To validate the reliability of the regulatory network, the co-regulated
   DEMs, ADEGs, and TFs were further verified in human hypertensive
   nephropathy renal samples and AngII-treated MPC5 cells. Co-regulated
   gene expression data were extracted from datasets [135]GSE37455 and
   [136]GSE37460 as previously described. The expression of KRAS, MAPK1,
   NRAS, and PTEN was significantly upregulated in the tubulointerstitial
   region of patients with hypertensive nephropathy compared with healthy
   living donors ([137]Figure 8A, left panel). Similarly, the expression
   of MAPK1, NRAS, and PTEN was also significantly upregulated in the
   glomeruli of patients with hypertensive nephropathy ([138]Figure 8A,
   right panel). Next, we detected co-regulated miRNAs and mRNAs by AngII
   and H[2]S donors in MPC5 cells using qPCR. In addition, PAG, an
   inhibitor of the CSE enzyme, was used to block SPRC for verification
   [[139]37,[140]43]. Similar to the results of the miRNA dataset, the
   AngII treatment decreased the expression levels of miR-98-5p and
   miR-669b-5p, but both endogenous and exogenous H[2]S donors reversed
   these changes in MPC5 cells ([141]Figure 8B). To confirm the
   reliability of AngII’s regulation of these miRNAs, losartan was used
   here to bind to angiotensin receptors to block the effect of AngII. As
   shown in [142]Figure 8C, losartan almost restored the expression of
   miR-98-5p and miR-669b-5p to the control level.

Figure 8.

   [143]Figure 8
   [144]Open in a new tab

   Validation of co-regulated DEMs, ADEGs, and TFs in hypertensive
   nephropathy clinical samples and AngII-treated MPC5 cells. (A)
   Normalized mRNA expression of KRAS, MAPK1, NRAS, and PTEN in the
   tubulointerstitial region (left) and glomeruli (right) of human renal
   biopsies. Data were extracted from [145]GSE37455 and [146]GSE37460. # p
   < 0.05, ## p < 0.01, ### p < 0.001 vs. healthy living donor group using
   two-sided unpaired t-test; n = 21 and 20 (left) or 27 and 15 (right) in
   the healthy and hypertensive group. (B,C) RNA expression of miR-98-5p
   and miR-669b-5p were measured in MPC5 cells by qPCR. (D) The mRNA
   levels of Kras, Mapk1, Nras, Pten, Pik3ca, Pik3r3, Mapk9, and Runx2 in
   MPC5 cells were calculated using qPCR. For (B,D), the MPC5 cells were
   pre-incubated with or without 2mM PAG for 30 min and then pre-treated
   with NaHS (50 μM) or SPRC (50 μM) for 1 h before exposure to AngII (1
   μM) treatment. NaHS, SPRC, and SPRC + PAG groups were all treated with
   AngII. # p < 0.05, ## p < 0.01, ### p < 0.001 vs. Ctrl group, * p <
   0.05, ** p < 0.01, *** p < 0.001 vs. AngII group, and @@ p < 0.01, @@@
   p < 0.001 vs. SPRC group. For (C), MPC5 cells were pre-stimulated with
   or without losartan (300 μM) for 12 h and then co-incubated with or
   without AngII (1 μM). ### p < 0.001 vs. Ctrl group, ** p < 0.01 vs.
   AngII group. One-way ANOVA coupled with Tukey’s multiple comparison
   post hoc test was used. Data are shown as mean ± SD, n = 3. ns, not
   significant.

   Moreover, the RNA expression of miRNA hub target genes and TFs were
   further detected. In the MPC5 cells, Mapk1, Nras, Pten, Mapk9, and
   Runx2 were upregulated by AngII and downregulated by the H[2]S donors.
   In addition to the changes in the above genes, Pik3ca and Pik3r3 were
   downregulated by AngII, which could be reversed by H[2]S donor
   treatment ([147]Figure 8D). For the expression of Kras, AngII treatment
   increased its expression, with no statistical significance. However,
   both endogenous and exogenous hydrogen sulfide caused it to decrease.
   In addition, the regulatory role of SPRC on DEMs, ADEGs, and TFs was
   partially inhibited by PAG ([148]Figure 8B,D).

   Furthermore, miRNAs mimics were used to overexpress miR-98-5p and
   miR-669b-5p to find their genuine targets in the presence of AngII. The
   miR-98-5p mimic significantly increased the expression of miR-98-5p by
   about 6-fold, and the miR-669b-5p mimic increased the expression of
   miR-669b-5p by 5-fold, indicating successful overexpression
   ([149]Figure 9A). In the case of co-treatment with AngII, the miR-98-5p
   mimic significantly reduced AngII-induced elevations in Kras, Mapk1,
   Nras, Pten, Mapk9, and Runx2, and the miR-669b-5p mimic decreased the
   expression of Nras, Pik3ca, Pik3r3, and Runx2 changed by AngII
   ([150]Figure 9B).

Figure 9.

   [151]Figure 9
   [152]Open in a new tab

   Overexpression of miR-98 and miR-669b. (A) RNA expression of miR-98-5p
   and miR-669b-5p were measured in MPC5 cells by qPCR. (B) The mRNA
   levels of Kras, Mapk1, Nras, Pten, Pik3ca, Pik3r3, Mapk9, and Runx2 in
   MPC5 were calculated using qPCR. MPC5 cells were transfected with
   miR-NC, miR-98-5p mimic, and miR-669b-5p mimic. Forty-eight hours after
   infection, cells were incubated with or without AngII (1 μM) to detect
   RNA expression. Both miR-98-5p and miR-669b-5p mimic groups were
   incubated with AngII. # p < 0.05, ## p < 0.01, ### p < 0.001 vs. miR-NC
   group, * p < 0.05, ** p < 0.01, *** p < 0.001 vs. AngII group. One-way
   ANOVA coupled with Tukey’s multiple-comparison post hoc test was used.
   Data are shown as mean ± SD, n = 3.

3.7. H[2]S Donors and miR-98/669b Overexpression Reversed Target Gene
Changes, Restored Autophagy Flux, and Reduced Cell Damage

   To further investigate the protein-level changes in this regulatory
   network, we investigated the protein expression of these co-regulated
   DEMs and TFs in MPC5 cells. As shown in [153]Figure 10A, AngII
   stimulation significantly increased the expression of autophagy
   substrate P62 (SQSTM1) in the MPC5 cells, which implies a decrease in
   autophagic flux [[154]44]. However, both the exogenous and endogenous
   hydrogen sulfide donors NaHS and SPRC restored the P62 expression to
   some extent. At the same time, similar to the gene-level changes, AngII
   significantly increased the levels of NRAS and RUNX2 in the MPC5 cells,
   and these changes were reversed by NaHS and SPRC. Furthermore, the
   alterations in KRAS, NRAS, RUNX2, and P62 induced by SPRC were
   partially offset by PAG ([155]Figure 10A). In addition, we found small
   changes in PTEN and MAPK1, but most of them were not statistically
   significant. Subsequently, miRNA mimics were used to overexpress
   miR-98-5p and miR-669b-5p in the presence or absence of AngII to detect
   changes in autophagy markers and kidney injury markers. The AngII
   stimulation significantly increased P62 and decreased LC3B expression,
   implying a decrease in autophagy flux. However, the overexpression of
   miR-98-5p and miR-669b-5p significantly restored the expression of P62
   and LC3B and improved intracellular autophagy ([156]Figure 10B). In
   terms of kidney injury indicators, AngII reduced the expression of the
   podocyte markers nephrin and podocin, but the overexpression of
   miR-98-5p and miR-669b-5p restored the AngII-induced reduction in these
   two proteins ([157]Figure 10C). Finally, WB analysis was conducted to
   check which hub target genes and TFs were altered in response to
   miR-98-5p and miR-669b-5p overexpression as well as AngII stimulation.
   As shown in [158]Figure 10D, the miR-98-5p mimic reduced the
   AngII-induced increases in KRAS, NRAS, and RUNX2. Meanwhile, the
   miR-669b-5p mimic partially restored the AngII-induced changes in NRAS
   and RUNX2.

Figure 10.

   [159]Figure 10
   [160]Open in a new tab

   Validation of co-regulated DEMs, ADEGs, and TFs in protein levels. (A)
   Western blot analysis of PTEN, KRAS, NRAS, MAPK1, RUNX2, P62, and GAPDH
   protein expression levels in MPC5 cells under different stimuli. Cells
   were pre-incubated with or without PAG and then pre-treated with NaHS
   or SPRC for 1 h before exposure to AngII treatment. NaHS, SPRC, and
   SPRC + PAG groups were all treated with AngII. Expression levels were
   calculated by densitometric analysis and normalized to GAPDH. Protein
   quantification is shown in bar charts. # p < 0.05, ## p < 0.01, ### p <
   0.001 vs. Ctrl group, * p < 0.05, ** p < 0.01, *** p < 0.001 vs. AngII
   group, and @@ p < 0.01, @@@ p < 0.001 vs. SPRC group. For (B–D), MPC5
   cells were transfected with miR-NC, miR-98-5p mimic, and miR-669b-5p
   mimic and then incubated with AngII to detect the expression of
   autophagy markers P62 and LC3B (B), renal injury markers nephrin and
   podocin (C), and hub targets such as KRAS, NRAS, and MAPK1 as well as
   the TF RUNX2 (D) by Western blot analysis. Both miR-98-5p and
   miR-669b-5p mimic groups were incubated with AngII. With GAPDH as a
   normalized reference, protein quantification is shown in bar charts.
   Dark green, red, blue, and orange represent miR-NC, AngII, miR-98-5p,
   and miR-669b-5p mimic groups, respectively. # p < 0.05, ## p < 0.01,
   ### p < 0.001 vs. miR-NC group, * p < 0.05, ** p < 0.01, *** p < 0.001
   vs. AngII group. One-way ANOVA coupled with Tukey’s multiple comparison
   post hoc test was used. Data are shown as mean ± SD, n = 3.

   Collectively, in the hypertensive nephropathy samples and AngII-induced
   podocyte models, the renal injury markers were increased, autophagic
   flux was decreased, and differential miRNAs (miR-98-5p and
   miR-669b-5p), hub target genes (KRAS and NRAS), and transcription
   factor (RUNX2) were dysregulated. The endogenous and exogenous H[2]S
   donors restored the expression of miR-98-5p and miR-669b-5p as well as
   other hub genes in the regulatory network. The overexpression of
   miR-98-5p and miR-669b-5p increased autophagic flux, reduced markers of
   kidney injury, and also restored the abnormal expression of KRAS, NRAS,
   and RUNX2. This may be the mechanism by which H[2]S donors alleviate
   hypertensive nephropathy.

4. Discussion

   Hypertensive nephropathy is one of the major causes of chronic kidney
   disease and end-stage renal disease [[161]45]. Recently, emerging
   evidence has shown that autophagy plays a crucial role in the
   maintenance of kidney function and kidney diseases [[162]46,[163]47].
   Targeting the autophagy pathway may have therapeutic benefits in the
   treatment of kidney diseases [[164]48]. In acute kidney injury,
   autophagy is induced in the proximal tubule as a protective mechanism
   [[165]47]. The deletion of key autophagy proteins increases P62 levels
   and oxidative stress, which impairs renal function [[166]49]. In
   diabetic nephropathy, the dysregulation of autophagy in podocytes,
   glomerular endothelial cells, and proximal tubules is thought to
   contribute to disease progression [[167]50]. Dysregulated autophagy has
   also been reported in many kidney diseases, such as tubulointerstitial
   fibrosis, focal segmental glomerulosclerosis, and polycystic kidney
   disease [[168]48,[169]51]. However, the role of autophagy in
   hypertensive nephropathy or kidney injury caused by hypertension has
   not yet been elucidated. Here, we have demonstrated that autophagy may
   be involved in AngII-induced hypertensive nephropathy. We screened out
   109 differentially expressed miRNAs and the corresponding 4667 target
   genes after AngII treatment. Then, 21 of the 233 hub genes, selected
   from the top 5% CytoHubba algorithm, were related to autophagy. If hub
   genes are not counted, the 57 of the 4667 target genes were also
   associated with autophagy. This indicates that autophagy-related genes
   accounted for a large proportion of the hub target genes of DEMs after
   the AngII treatment.

   Studies have shown that hydrogen sulfide play a therapeutic role by
   regulating the autophagy pathway [[170]52]. In renal disease, exogenous
   H[2]S treatment can protect the kidney from ROS-induced autophagy and
   renal injury by inhibiting the ROS-AMPK pathway [[171]53]. However,
   there are few reports on the relationship between hydrogen sulfide and
   autophagy in hypertensive nephropathy. Our study has demonstrated that
   hydrogen sulfide donor treatment may affect the autophagy pathway
   altered by AngII. We identified 70 DEMs and the corresponding 2828
   target genes in the AngII_H[2]S vs. AngII group, 44 of which were
   associated with autophagy, and 13 of the 141 key target genes were
   associated with autophagy. To our knowledge, this is the first
   bioinformatic report describing the regulation of autophagy pathways by
   hydrogen sulfide donors in hypertensive nephropathy.

   miRNA dysregulation is an important component of epigenetic regulation,
   and studies have identified a large number of miRNAs that play a role
   in hypertensive nephropathy [[172]54], and the role of hydrogen sulfide
   and microRNA crosstalk has also gradually emerged in kidney diseases
   [[173]25,[174]55]. In the present study, 12 miRNAs targeting key ADEGs
   were altered upon AngII treatment, and correspondingly, 11 miRNAs were
   altered upon co-treatment with AngII and hydrogen sulfide. Among them,
   miR-98-5p and miR-669b-5p were co-altered miRNAs, and AngII caused
   their decrease, while the hydrogen sulfide donors significantly
   increased the expression of both. Studies have shown that miR-98-5p is
   decreased in the kidney tissue of diabetic mice and in
   high-glucose-treated human podocytes (HPCs) and HK2 cells, and the
   overexpression of miR-98-5p inhibits apoptosis, proteinuria, and the
   renal fibrosis process [[175]56,[176]57]. In cardiovascular diseases,
   the overexpression of miR-98-5p can also protect cardiomyocytes from
   apoptosis [[177]58,[178]59]. These all indicate the protective effect
   of miR-98-5p. For miR-669b-5p, it was only found in mus musculus, and
   there are few reports on its function. But for its family, miR-669c was
   protective in a mouse model of ischemic stroke, and long-term miR-669a
   treatment attenuated chronic dilated cardiomyopathy in dystrophic mice
   [[179]60,[180]61]. These studies have suggested a possible protective
   role of the miR-669 family in cardiovascular diseases.

   This study focused on changes in miR-98-5p and miR-669b-5p, but the
   remaining DEMs may also play a role in the progression of hypertensive
   nephropathy. Among the other AngII-upregulated miRNAs, previous studies
   have shown that silencing microRNA-132 reduced renal fibrosis by
   selectively inhibiting myofibroblast proliferation [[181]62]. The
   miR-132 family also directly targets the pro-autophagy FoxO3
   transcription factor [[182]63]. miR-762 is significantly upregulated in
   septic acute kidney injury [[183]64]. For other miRNAs downregulated by
   AngII, the most studied is miR-204-5p. miR-204-5p is downregulated in
   renal biopsy specimens from patients with hypertensive nephrosclerosis,
   diabetic nephropathy, and in the kidneys of rats fed with a high-salt
   diet [[184]65,[185]66]. Furthermore, the knockdown or knockout of
   miR-204-5p results in substantial pathological changes in the kidneys
   as well as proteinuria [[186]66], and the protective role of miR-204 in
   chronic kidney disease is related to SHP2 and STAT3 [[187]67]. For the
   downregulated DEMs in the hydrogen sulfide group, miRNA-363-3p promotes
   apoptosis in response to heavy-metal-induced kidney injury by
   downregulating phosphoinositide 3-kinase expression [[188]68]. In
   H[2]S-upregulated DEMs, miR-129-5p attenuates LPS-induced podocyte
   injury and acute kidney injury, and this kind of protection is related
   to the regulation of the NF-κB signaling pathway [[189]69]. In short,
   these altered miRNAs may be the underlying mechanism behind
   AngII-induced hypertensive nephropathy and the therapeutic effect of
   hydrogen sulfide.

   A total of 21 and 13 ADEGs were screened out, and their potential
   biological functions were identified by GO and KEGG enrichment
   analysis. Interestingly, these ADEGs collectively affect protein
   phosphorylation, phosphatidylinositol 3-kinase complex, protein
   binding, and protein kinase activity. These aspects are all related to
   autophagy. Protein kinases are integral to autophagy. Both autophagy
   initiation and autophagy signaling utilize kinase mechanisms [[190]70].
   Among them, kinase-catalyzed phosphorylation is an important part of
   the most-studied post-translational modification of autophagy
   [[191]71]. For example, phosphatidylinositol 3-kinase (PI3K) interacts
   with various regulatory proteins to form multiple complexes that
   selectively participate in different stages of autophagy. A complex of
   PI3K and ATG14 is involved in the formation of autophagic vesicles
   [[192]72]. In addition, the ADEGs that affect these biological
   functions may be genes such as PTEN, AMPK1, KRAS, and NRAS. In the
   enrichment study of signaling pathways, autophagy and FoxO signal
   transduction are commonly enriched signaling pathways. This further
   illustrates that AngII and H[2]S donors affect the autophagy pathway in
   the kidney. The role of the FoxO transcription factor family in
   autophagy has been widely reported. For example, FoxO1 and FoxO3
   enhance autophagic flux by increasing the expression of autophagy
   genes, such as Ulk2, Becn1, Lc3b, Atg12, and GabarapL1 [[193]73].
   Furthermore, SNAI2 is transcriptionally activated by FoxO3 and
   interacts with FoxO3 to form a feed-forward regulatory loop to enhance
   the expression of autophagy genes [[194]74]. In addition to general
   autophagy, FoxO3 is also involved in the regulation of mitophagy
   [[195]75]. Among the other AngII-enriched signaling pathways, mTOR
   signal transduction is closely related to autophagy, and mTORC1 in the
   signaling pathway is also a promising pharmacological target for
   regulating autophagy [[196]76]. Future experiments will be necessary to
   explore the potential biological functions of these ADEGs regulated by
   AngII and hydrogen sulfide donors.

   After the PPI analysis, we validated co-regulated DEMs, ADEGs, and TFs
   using clinical patient databases as well as several cell experiments.
   In the kidney cell experiments, AngII stimulation significantly
   increased the expression of the autophagy substrate P62, implying a
   decrease in autophagic flux. Previous studies have shown that P62, a
   substrate of the autophagy–lysosomal degradation pathway, was observed
   to accumulate in the kidneys of STZ-induced diabetic mice, Wistar obese
   rat bodies, and kidney biopsies of patients with type 2 diabetes,
   suggesting that autophagy is impaired [[197]77]. Targeting the
   autophagy pathway to activate and restore autophagic activity may be
   renoprotective. For example, the endogenous calpain inhibitor
   calpastatin can maintain autophagy to alleviate AngII-mediated podocyte
   injury [[198]9]. In addition, AngII can also induce the activation of
   TLR4 to inhibit autophagy and induce apoptosis and kidney injury
   [[199]78]. At the same time, changes in KRAS and NRAS were also
   observed. On the one hand, the inhibition of KRAS increases autophagic
   flux, and the inhibition of KRAS/RAF/MEK/ERK signaling triggers
   autophagy [[200]79,[201]80]. On the other hand, for NRAS, there are not
   many corresponding reports. In the MPC5 cells, we found that both
   endogenous and exogenous hydrogen sulfide reversed AngII-induced KRAS
   and NRAS changes to some extent, which may partially restore autophagy
   in cells. Changes in PTEN and MAPK1 were found in the clinical samples
   and kidney cells. Published findings suggest that PTEN activates
   autophagy through the AKT/mTOR pathway [[202]81]. Furthermore,
   autophagic flux and ATG7 (autophagy-related 7) levels were increased by
   the activation of MAPK1/3 [[203]82]. We found that PTEN and MAPK1 were
   increased at the gene level, which may be related to the positive
   feedback regulation of autophagy in the early stage of disease. In
   cervical cancer cells, the overexpression of miR-98-5p significantly
   increased the expression of the PTEN protein, and silencing miR-98-5p
   had the opposite result [[204]83]. In the future, more experiments are
   needed to explore the effect of timing of AngII regulation on PTEN and
   autophagy.

   The synergistic effect of transcription factors and miRNAs has not yet
   been fully elucidated, and this study is also the first attempt to
   reveal the interactions between DEMs, hub target genes, and
   transcription factors in the treatment of AngII-induced hypertensive
   kidney injury with hydrogen sulfide donors. We identified and validated
   one transcription factor, Runx2, that co-regulates miR-98-5p and
   miR-669b-5p target hub genes. Runx2 is a well-known transcription
   factor for skeletal development, but its role in non-bone tissues is
   gradually being reported. Studies have shown that AngII can
   significantly increase the level of Runx2 in aortic tissue and human
   aortic smooth muscle cells, which is similar to our validation results
   [[205]84,[206]85]. In addition, the activation of Runx2 promotes aortic
   valve fibrosis as well as aortic fibrosis and stiffness in patients
   with type 2 diabetes [[207]86,[208]87]. In chronic kidney disease,
   Runx2 is deacetylated by SIRT6 to inhibit osteogenic trans
   differentiation of vascular smooth muscle cells [[209]88]. Runx2 is
   also regulated by miRNAs. The downregulation of miR-204 promoted
   hyperproliferation and calcified lesions of pulmonary artery smooth
   muscle cells by increasing the expression of the transcription factor
   Runx2 [[210]89]. Increased miR-204-5p also suppressed Runx2 expression,
   thereby inhibiting aortic valve calcification [[211]90]. Interestingly,
   studies have shown that Runx2 also promoted autophagy by enhancing the
   trafficking of LC3B vesicles [[212]91]. In this study, the
   transcription factor Runx2 was upregulated by AngII, and this
   upregulation was reversed by endogenous and exogenous hydrogen sulfide
   donors. These results suggest that the transcription factor Runx2 may
   be a good therapeutic target for hydrogen sulfide donors to treat the
   pathological process of renal fibrosis in hypertensive nephropathy. The
   detailed functions and underlying mechanistic pathways of Runx2 require
   further investigation. More in vitro and in vivo studies and
   validations are needed in the future to elucidate the role of these new
   markers, including DEMs, hub ADEGs, and TFs in hypertensive nephropathy
   and hydrogen sulfide therapy.

5. Conclusions

   By using bioinformatics analysis, 109 DEMs in mouse kidney tissue after
   treatment with AngII alone and 70 DEMs after treatment with a combined
   treatment of AngII and H[2]S donors were identified. Through the
   regulatory network and validation study, the H[2]S donor treatment
   increased autophagic flux, reduced renal cell injury, and restored the
   two identified and co-regulated miRNAs, miR-98-5p and miR-669b-5p,
   attenuated the dysregulation of ADEGs (KRAS, NRAS) and a key
   transcription factor (RUNX2). These findings help to define the
   pathophysiological mechanism of AngII-induced hypertensive nephropathy
   and shed new light on the therapeutic effects of hydrogen sulfide
   donors.

Supplementary Materials

   The following supporting information can be downloaded at:
   [213]https://www.mdpi.com/article/10.3390/antiox13080958/s1. Figure S1:
   Profiles of the non-coding RNA expression profile dataset
   [214]GSE92857, Table S1: Autophagy-related gene set in mouse, Table S2:
   Primer sequences, Table S3: Differentially expressed miRNAs between
   AngII and Ctrl group, Table S4: Differentially expressed miRNAs between
   AngII_H[2]S and AngII group, Table S5: DEMs and their target ADEGs in
   AngII vs Ctrl and AngII_H[2]S vs AngII comparison.
   [215]antioxidants-13-00958-s001.zip^ (1.1MB, zip)

Author Contributions

   Conceptualization, Q.Y., Y.M. and Y.-Z.Z.; data curation, Q.Y., M.R.,
   Y.M. and Y.-Z.Z.; formal analysis, Q.Y. and M.R.; funding acquisition,
   Y.-Z.Z.; investigation, Q.Y. and D.F.; methodology, Q.Y. and M.R.;
   project administration, Q.Y. and Y.-Z.Z.; resources, Q.Y. and Y.M.;
   software, Q.Y. and M.R.; supervision, Y.M. and Y.-Z.Z.; validation,
   Q.Y., M.R. and D.F.; visualization, Q.Y.; writing—original draft, Q.Y.
   and M.R.; writing—review and editing, Q.Y., Y.M. and Y.-Z.Z. All
   authors have read and agreed to the published version of the
   manuscript.

Institutional Review Board Statement

   Not applicable.

Informed Consent Statement

   Not applicable.

Data Availability Statement

   All data generated or analyzed during this study are included in this
   published article, it’s [216]supplementary information files, and
   publicly available repositories. The datasets [217]GSE92857,
   [218]GSE37455, and [219]GSE37460 were downloaded from the GEO database.

Conflicts of Interest

   The authors declare no conflicts of interest.

Funding Statement

   This work was funded by Macau Science and Technology Development fund
   (FDCT (0012/2021/AMJ, 003/2022/ALC, 0092/2022/A2, 0144/2022/A3)),
   Shenzhen-Hong Kong-Macao Science and Technology Fund (Category C:
   SGDX20220530111203020), and Shanghai Committee of Science and
   Technology (No. 23ZR1413200).

Footnotes

   Disclaimer/Publisher’s Note: The statements, opinions and data
   contained in all publications are solely those of the individual
   author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI
   and/or the editor(s) disclaim responsibility for any injury to people
   or property resulting from any ideas, methods, instructions or products
   referred to in the content.

References