The Role of miR-217-5p in the Puromycin Aminonucleoside-Induced Morphological Change of Podocytes

Podocytes, alternatively called glomerular epithelial cells, are terminally differentiated cells that wrap around glomerular capillaries and function as a part of the glomerular filtration barrier in the kidney. Therefore, podocyte injury with morphological alteration and detachment from glomerular capillaries leads to severe proteinuria and subsequent renal failure through glomerulosclerosis. Previous RNA sequencing analysis of primary rat podocytes exposed to puromycin aminonucleoside (PAN), a well-known experimental model of injured podocytes, identified several transcripts as being aberrantly expressed. However, how the expression of these transcripts is regulated remains unclear. MicroRNAs (miRNAs) are small noncoding RNAs that posttranscriptionally inhibit the expression of their target transcripts. In this study, using small RNA sequencing analysis, miR-217-5p was identified as the most upregulated transcript in PAN-treated rat podocytes. MiR-217-5p overexpression in E11 podocyte cells led to shrunken cells with abnormal actin cytoskeletons. Consistent with these changes in cell morphology, gene ontology (GO) enrichment analysis showed that interactive GO terms related to cell morphogenesis were enriched with the predicted targets of miR-217-5p. Of the predicted targets highly downregulated by PAN, Myosin 1d (Myo1d) is a nonmuscle myosin predicted to be involved in actin filament organization and thought to play a role in podocyte morphogenesis and injury. We demonstrated that miR-217-5p targets Myo1d by luciferase assays, qRT–PCR, and Western blotting. Furthermore, we showed that miR-217-5p was present in urine from PAN- but not saline-administrated rats. Taken together, our data suggest that miR-217-5p may serve as a therapeutic target and a biomarker for podocyte injury.


Introduction
Approximately 10% of the world population is affected by chronic kidney disease (CKD), defined as abnormalities in kidney structure and/or function that are present for at least 3 months, irrespective of cause [1]. Millions of patients with CKD die every year due to inadequate treatment [1]. CKD shows few symptoms in initial stages and is often detected in advanced stages when symptoms become more obvious [2,3]. For this reason, CKD is often classified as a "silent" disease. When kidney functions are practically lost, this is expressed as 'renal failure' [4,5]. Unfortunately, no therapeutic methods to restore the function of chronically impaired kidneys have been established. During late stages of CKD, patients with CKD require renal replacement therapies including hemodialysis and kidney transplants for survival. Although hemodialysis is generally applied in cases of renal failure, patients receiving dialysis have significantly decreased life expectancy compared to the normal population [6]. Hemodialysis also reduces patient quality of life in physiological, mental, and/or social aspect(s) [7]. Furthermore, the high cost of dialysis places an economic burden on patients and is suggested to exert a negative influence on health economics [7]. While kidney transplantation can be a therapy option for end-stage CKD patients [8,9], the outcome is controversial among health care professionals [10,11]. Furthermore, patients usually need to wait a long time for a compatible donor and may die while waiting for a kidney [12]. Finally, for a transplant to be considered successful, the donated kidney must be accepted by the patient's immune system [13].
In the kidney, glomeruli are found in the cortex and serve as a hemofiltration device [14]. Podocytes, alternatively called glomerular epithelial cells, are highly differentiated cells that cover glomerular capillaries. Podocytes, together with endothelial cells and glomerular basal membranes, form a unit that functions as a glomerular filtration barrier in the kidney [15,16]. Podocytes have a characteristic morphology that contains many foot processes [14]. Between the foot processes, membranous structures called slit membranes form, and these membranes play an essential role as a final filtration barrier to prevent leakage of proteins into the urine during glomerular filtration [14][15][16]. The structure of the filtration barrier can be disrupted by podocyte damage, which occasionally causes podocyte detachment from the glomeruli; in these cases, severe proteinuria leading to renal failure may occur [16,17]. However, the mechanisms that underlie the structural disruption of podocyte remain unclear.
MicroRNAs (miRNAs) are small noncoding RNAs of 19-25 nucleotides in length that play a crucial role in posttranscriptional gene regulation by repressing the translation of or degrading target messenger RNAs (mRNA) [18,19]. Notably, mice with podocyte-specific deficiency in Dicer, a nuclease essential for miRNA biosynthesis, have been shown to develop severe proteinuria and die within weeks, indicating that miRNAs are indispensable for podocyte function [20][21][22]. Previous studies have also demonstrated that many miRNAs play essential roles in nephropathy-related podocyte injury [23,24]. However, the mechanisms underlying miRNA-mediated podocyte injury are not fully understood. Here, we performed an integrative analysis of miRNA and mRNA expression profiles associated with puromycin nucleoside (PAN)-induced podocyte injury. Our analysis showed that miR-217-5p upregulation in PAN-treated podocytes can lead to morphological alteration. Furthermore, we demonstrated that miR-217-5p is present in urine from PAN-but not saline-administrated rats, suggesting that miR-217-5p may serve as a therapeutic target and biomarker for podocyte injury.

Cell Culture and Induction of Podocyte Injury
Podocytes in primary culture were isolated from male Wistar rats at ages 7-8 weeks (SLC, Hamamatsu, Japan) and cultured as described previously [25,26]. The mouse podocyte cell line E11 was purchased from Cell Lines Service GmbH (Eppelheim, Germany) and cultured as described previously [27]. To cause cell injury, primary rat podocytes and E11 cells were treated with various concentrations of PAN (Wako, Osaka, Japan) for 24 h or 48 h.

Comprehensive Small RNA Sequencing
One µg of total RNA isolated from the rat podocytes in primary culture was subjected to small RNA sequencing to comprehensively analyze miRNA expression. Small RNA libraries were constructed using the TruSeq Small RNA Library Preparation Kits (Illumina, San Diego, CA, USA) and analyzed on a HiSeq-3000 sequencer (Illumina) at the Genome Information Research Centre of Osaka University. Raw reads obtained from the sequencing analysis were trimmed using Cutadapt v1.9.2 and subjected to miRNA-derived read counting and annotation using CLC Genomics Workbench 11.0.1 (Qiagen, Venlo, The Netherlands).
The reads obtained in this analysis and related metadata were deposited in the DNA Data Bank of Japan Sequence Read Archive (DRA) under the accession number DRA013173.

Real-Time Reverse-Transcription PCR for miRNA and mRNA
Quantitative RT-PCR (qRT-PCR) analysis using TaqMan ® MicroRNA Assays (Thermo-Fisher Scientific, Waltham, MA, USA) was conducted to validate the results of miRNA-seq analysis. Because expression levels of target miRNAs miR-217-5p were low, PCR preamplification with 12 cycles was conducted using TaqMan ® PreAmp Master Mix (Thermo-Fisher Scientific) before performing qRT-PCR. The expression of U6 small nuclear RNA (snRNA) was analyzed as an endogenous internal control to normalize expression levels of miRNAs. Monitoring the amplification of PCR products was conducted on a DICE Thermal Cycler Real-Time System (Takara-bio, Kusatsu, Japan).
Thunderbird SYBR qPCR Mix (Toyobo, Osaka, Japan) was used for the quantitative analysis of mRNA expression levels. GAPDH mRNA or 18S rRNA was used as an endogenous internal control to normalize mRNAs' expression levels (the primers sequences are listed in Table S1). Monitoring of amplification of PCR products was conducted on a DICE Thermal Cycler Real-Time System (Takara-bio). Dissociation curve analyses were also performed to verify specific amplification.

Cell Viability Assay
E11 cells were seeded in 96-well plates at 2 × 10 3 cells/well, cultured overnight, and treated with PAN (Wako, Osaka, Japan) for 48 or 96 h. After changing to fresh medium, the viability of podocytes was evaluated using Cell Counting Reagent SF (Nacalai Tesque, Kyoto, Japan) according to the manufacturer's instructions.

Reporter Plasmid Construction and Luciferase Assay
The 2.2-kbp 3 -UTR of mouse Myo1d mRNA was amplified by RT-PCR from the total RNA of E11 cells using PrimeScript II reverse transcriptase (Takara-bio) and KOD-Plus DNA polymerase (Toyobo). The primers used for RT-PCR amplification were: 5 -CT GCTGCACATCAGAGGCCT-3 (forward) and 5 -TTTGTCGACAAGATTTAATGCTTTA TTGC-3 (reverse). The PCR product encompassing the 3 -UTRs of Myo1d was then subcloned into the pmirGLO vector (Promega, Madison, USA) at the DraI and SalI restriction sites to construct the pmirGLO-Myo1d 3 -UTR luciferase reporter plasmid. DNA ligation was conducted at 16 • C for 30 min using DNA Ligation Kit Ver.1 (Takara-bio). A reporter plasmid having a mutation at the predicted miR-217-5p binding site in the 3 -UTR of Myo1d mRNA was constructed using pmirGLO-Myo1d 3 -UTR as a template. First, the part between the predicted miR-217-5p binding site and the end of the 3 -UTR was PCR-amplified using KOD-Plus DNA polymerase (Toyobo) and the following primers: 5 -ACCTGGAATTCGGGGTGTGACTGACCACAGTAACAGCAGAGGAGAGGACACAG TGATTGTATGCATGGAGTAGGGGTCTCTTGAGTTAATGAAGATATCGTTATGGTTTG-3 (forward) and 5 -TTTGTCGACAAGATTTAATGCTTTATTGC-3 (reverse). The generated amplicon was digested using EcoRI and SalI at both termini. The amplicon then substituted the corresponding region of pmirGLO-Myo1d 3 -UTR. For reporter assays, E11 cells were transfected with these plasmids together with synthetic miRNA mimics (Thermo-Fisher Scientific) at 20 nM using Lipofectamin2000 (Thermo-Fisher Scientific). Twenty-four hours after transfection, the cells were processed using the Dual-Luciferase Reporter Assay System (Promega). Luminescence was detected using a TD-20/20 luminometer (Promega).

Urine Processing
PAN administration in Wistar rats and urine collection from rats were performed as described previously [25]. Urine samples were collected for 12 h at day 5 following the administration of PAN or saline. Collection on day 5 was chosen because we previously showed that urinary protein concentration, an index to evaluate the progression of glomerulopathy, was markedly increased from day 5 onwards [25]. The collected urine was filtered through a 0.22-µm filter to remove debris, then subjected to RNA isolation using an RNAiso Blood reagent (Takara, Tokyo, Japan). In parallel, to monitor the progression of glomerulopathy, the protein concentration of the filtered urine was measured as described previously [25].

Statistical Analysis
Student's t-tests were used to assess the significance of differences among groups for comparisons between two groups, and one-way analysis of variance followed by Ryan's test was used for comparisons among three or more groups. In all analyses, p < 0.05 was considered to be statistically significant.

Validation of PAN-Induced miR-217-5p Expression in Primary Rat Podocytes
Our small RNA-seq analysis indicated that miR-217-5p is the most highly (5.7-fold) upregulated miRNA in PAN-treated podocytes; no other miRNAs were upregulated more than five times. We performed qRT-PCR analysis to verify PAN-induced miR-217-5p expression in primary rat podocytes. Analysis of qRT-PCR analysis revealed that miR-217-5p expression in rat podocytes was markedly increased following PAN treatment ( Figure 2).

PAN-Induced Effects on Cell Viability, Cell Morphology, and miR-217-5p Expression in E11 Podocyte Cells
Considering the drastic induction of miR-217-5p expression in rat podocytes, we sought to determine whether the increased expression of miR-217-5p was involved in PAN-induced podocyte injury. However, primary rat podocytes are not suitable for this purpose because only a small number of podocytes can be isolated from a rat. Furthermore, it is generally difficult to efficiently transfect primary cells with DNA or RNA. Therefore, we evaluated whether E11, an immortalized cell line generated from mouse podocytes, could be used as an alternative in vitro model, since siRNA-mediated gene knockdown and vector-mediated gene overexpression have been previously conducted in this cell line [30]. First, we evaluated the effect of PAN on E11 cell viability. We found that PAN exerted a strong suppressive effect on E11 cell viability at >100 µg/mL ( Figure 3A). This concentration is higher than effective PAN concentrations reported previously for primary rat podocytes (>10 µg/mL) [25]. The PAN-resistant properties of mouse podocytes compared to rat podocytes has been attributed to a deficiency in the adenosine deamination pathway [31] and low expression of plasma membrane amine transporter [32]. We also found that PAN induced morphological changes in E11 cells. While E11 cells cultured in the absence of PAN largely exhibit a spread-out shape, cells appear shrunken when cultured with 100 µg/mL PAN ( Figure 3B). In addition, immunostaining with an antiβ-actin antibody revealed that actin filaments lining appeared disorganized or shortened in E11 cells treated with 100 µg/mL PAN ( Figure 3C).
Next, we evaluated whether miR-217-5p expression is upregulated in PAN-treated E11 cells. We found that miR-217-5p expression was upregulated in E11 cells treated with 100 and 300 µg/mL PAN ( Figure 3D). These data suggest that PAN exerts similar effects on cell viability and miR-217-5p expression in both primary rat podocytes and E11 cells, indicating that E11 cells can be used as a model to investigate the role of miR-217-5p in PAN-induced podocyte injury.

MiR-217-5p-Induced Effects on Cell Viability and Cell Morphology in E11 Cells
We next evaluated whether miR-217-5p overexpression in E11 cells affects cell viability and morphology. MiR-217-5p overexpression did not significantly affect cell viability of E11 cells after 48 h ( Figure 4A) and 96 h ( Figure S1). However, E11 cells with miR-217-5p overexpression in culture exhibited shrunken shapes ( Figure 4B). Furthermore, immunostaining with an anti-β-actin antibody revealed that actin filaments lining behind the cell membrane appeared disorganized or shortened in E11 cells with miR-217-5p overexpression ( Figure 4C). These data imply that miR-217-5p is possibly involved in cell morphogenesis through the regulation of actin filament formation.

Integrative Analysis of miRNA and mRNA Expression in PAN-Treated Podocytes
Although miRNAs are known to silence their target genes by attenuating translation and degrading mRNAs, previous studies have shown that mammalian miRNAs predominantly reduce target mRNA levels [33,34]. Therefore, expression levels of PANregulated miRNAs and their target mRNAs in podocytes are thought to be reciprocally regulated. In the present study, we selected PAN-regulated mRNAs as described in the Materials and Methods section. Through screening based on these criteria, 226 and 480 mR-NAs were identified as upregulated and downregulated mRNAs, respectively ( Figure S2, Tables S3 and S4). Next, the PAN-regulated mRNAs were screened to identify miR-217-5p targets satisfying the following criteria: (i) expression inversely correlated with miR-217-5p and (ii) in silico-predicted targets of miR-217-5p. After proceeding with these procedures, 12 mRNAs were identified as the predicted targets of miR-217-5p ( Figure S3 and Table 2).

Gene Ontologies Associated with the Predicted Targets of the PAN-Dysregulated miRNAs
GO enrichment analyses were performed using GOnet to predict biological processes associated with the predicted targets of miR-217-5p in podocytes (Table 2). Consistent with the mR-217-5p-induced abnormal morphology of E11 cells, interactive GO (biological process) terms including those related to cell morphogenesis were enriched with the predicted targets of miR-217-5p ( Figure S4). This study focused on Myosin 1d (Myo1d) as a predicted target because it is annotated to GO terms related to actin filament organization (Table 2), and previous studies have shown the involvement of nonmuscle myosins in podocyte morphogenesis and injury [35].  Immunohistochemical staining showed the cytosolic localization of Myo1d protein in E11 cells, consistent with the previously reported localization of Myo1d protein in the human bone osteosarcoma cell line U-2 OS ( Figure S5). We then examined whether Myo1d mRNA expression in podocytes was downregulated by PAN as shown in our microarray analysis. As expected, PAN downregulated Myo1d mRNA expression levels in primary rat podocytes ( Figure S6) and E11 cells (Figure 5), and PAN upregulated miR-217-5p expression in these cells ( Figure 3D). The reciprocal regulation of miR-217-5p and Myo1d mRNA expression raises the possibility that the miR-217-5p-Myo1d axis may be involved in PAN-induced podocyte injury. This possibility was supported by analyses of publicly available datasets of podocyte injury-related transcriptomes. The analysis of the dataset GSE124622, created through microarray analysis of immortalized human podocytes treated with PAN and Adriamycin (ADR) to cause podocyte injury [36], showed that Myo1d mRNA levels in the PAN-and ADR-treated podocytes decreased 0.41-fold (p = 7.54E-06) and 0.64-fold (p = 0.0364) compared to control podocytes, respectively (Table S5). Furthermore, analysis of the dataset GSE108629 analysis, created through a microarray-based study using a mouse model of immunotoxin-inducible podocyte injury, showed that Myo1d mRNA levels were reduced 0.61-fold (p = 0.00158) after podocyte injury [37]. These results indicate that a reduction in Myo1d mRNA expression in injured podocytes was observed in our rat model and in in vitro or in vivo models from other organisms.

Target Validation of miR-217-5p
To evaluate whether miR-217-5p targets Myo1d mRNA in podocytes through its 3 -UTR, we performed luciferase assays in E11 cells. Because the 3 -UTRs of mouse and rat Myo1d mRNAs are highly similar (84%) and have a predicted common miR-217-5p binding site at identical positions of their 3 -UTRs ( Figure 6A), we constructed reporter plasmid containing the mouse Myo1d 3 -UTR, designated as pmirGLO-Myo1d-3 -UTR ( Figure 6B). Cotransfection of miR-217-5p showed significantly downregulated luciferase activity in the reporter plasmid-transfected cells compared to the control ( Figure 6C). Alternatively, when E11 cells were transfected with the reporter plasmid with a mutation at the predicted miR-217-5p-binding site in the Myo1d 3 -UTR ( Figure 6B), the inhibitory effect of miR-217-5p on luciferase activity was weakened ( Figure 6C). However, the mutation only partly restored the luciferase activity attenuated by miR-217-5p. Therefore, there may be other miR-217-5p recognition site(s) in the 3 -UTR of Myo1d mRNA, which were not identified by the target prediction tool.
Next, we evaluated whether miR-217-5p overexpression in E11 cells affects Myo1d mRNA and protein expression levels through qRT-PCR and Western blotting, respectively.
As expected, expression levels of both Myo1d mRNA and Myo1d protein were reduced by overexpression of miR-217-5p compared to the control ( Figure 6D,E).

Detection of miR-217-5p in Urine from PAN-Administrated Rats
Because podocytes face the urinary space of the Bowman's capsule, it is conceivable that miRNAs in podocytes are secreted into glomerular filtrate. We evaluated the amount of miR-217-5p in urine from PAN-and saline-administrated rats ( Figure 7A). Consistent with the marked increase of miR-217-5p expression in PAN-treated rat podocytes (Figure 2), miR-217-5p was detected in cell-free urine from PAN-administrated rats but not in that from saline-administrated rats ( Figure 7B). By contrast, U6 snRNA, a small nuclear RNA ubiquitously expressed in many cell types including podocytes, was detected in urine at similar levels from both PAN-and saline-administrated rats ( Figure 7B).

Discussion
This study performed miRNA-seq analysis of primary culture rat podocytes treated with or without PAN. The miRNA-seq dataset was combined with RNA-seq datasets of PAN-treated rat podocytes, which were acquired previously from the same RNA samples [25], and subjected to integrative bioinformatical analysis of miRNA and mRNA expression. MiR-217-5p was identified as the miRNA most strongly upregulated by PAN, with Myo1d as its possible target. We propose that the miR-217-5p-Myo1d axis may be associated with PAN-induced podocyte injury.
Many miRNAs have been previously associated with podocyte injury [23,24]. This study identified miR-217-5p as having the most highly upregulated expression (5.73-fold) following PAN-induced podocyte injury. The second most highly upregulated miRNA was miR-216a-5p (4.64-fold). These results are consistent with the fact that miR-217, -216a, and -216b are generated from a common host gene designated as MIR217HG, which spans an approximately 40-kbp region in the rat, mouse, and human genomes, allowing simultaneous regulation at a transcriptional level. Furthermore, the expression of MIR217HG-derived miRNAs is simultaneously dysregulated in different tissues under pathological conditions, such as in acute pancreatic injury [38][39][40][41][42][43], pancreatic cancer [44,45], and gastric cancers [46]; this may also be the case with podocyte injury.
Of the five dysregulated miRNAs, only miR-217-5p has been associated with podocyte injury in previous studies. Sun et al. reported that the expression of miR-217-5p is upregulated in podocytes stimulated with high glucose, and miR-217-5p is involved in high glucose-induced podocyte injury by attenuating the PTEN expression as its direct target [47]. These data are consistent with our findings that miR-217-5p expression is upregulated in response to podocyte injury. However, Jin et al. showed that reduced expression of the long noncoding RNA XIST prevents podocyte apoptosis in membranous nephropathy through the miR-217-TLR4 axis [48]. This report contrasts with our finding because claiming that miR-217-5p exerts a protective effect on podocyte injury. The inconsistency may be attributed to different pathological conditions between studies. The PAN nephropathy model exhibits pathological conditions similar to a minimal change disease rather than membranous nephropathy. However, whether miR-217-5p expression is differently regulated in various nephropathy models requires further investigation.
Although dysregulation of miR-217-5p expression to podocyte injury has been reported previously, the mechanism underlying the relationship is not understood. The GO enrichment analysis of our RNA-seq dataset showed that the predicted targets of miR-217-5p were closely associated with cell morphogenesis. Consistent with this analysis, E11 cells with miR-217-5p overexpression exhibited shrunken cell shapes with shortened or disorganized actin cytoskeletons. Previously, Solanki et al. demonstrated that the apoptosis-related p53 pathway is involved in actin cytoskeleton damage in PAN-or ADR-treated podocytes [36]. However, in our study, the p53 pathway was not associated with predicted targets of miR-217-5p in PAN-treated podocytes. Furthermore, unlike PAN, miR-217-5p overexpression did not affect E11 cell viability. Therefore, it is possible that PAN induces podocyte cell death through a mechanism independent of miR-217-5p. Of note, pathway analysis using DAVID (the Database for Annotation, Visualization and Integrated Discovery) revealed that mRNAs identified to be upregulated by PAN in our mRNA-seq analysis are most significantly associated with the p53 signaling pathway (Table S6). Therefore, we suggest that PAN may also induce morphological changes of podocytes as a result of p53-mediated apoptosis activated by the upregulated mRNAs. On the other hand, the mRNAs identified to be downregulated by PAN were not significantly associated with the p53 signaling pathway. Instead, these mRNAs were strongly associated with cell morphology-related pathways such as the "regulation of actin cytoskeleton" and "focal adhesion" pathways (Table S7), which were not associated with the upregulated mRNAs. It is possible that these pathways are partly associated with mRNAs negatively regulated by the PAN-induced miRNAs including miR-217-5p.
Interestingly, pathway analysis using DAVID showed that the predicted targets of miR-217-5p in podocytes are significantly associated with the 'axon guidance' pathway. Previous studies have suggested there is common molecular machinery for the formation of podocyte foot processes and the axon guidance of neurons [49][50][51]. Taken together with the fact that the myosin I protein family was previously associated with both podocyte morphology [52] and axon guidance [53], we decided to focus on Myo1d, an actin-binding protein, as a possible target of miR-217-5p. As expected, Myo1d is a direct target of miR-217-5p; therefore, the miR-217-5p-Myo1d axis could be involved in podocyte injury and warrants further investigation as a research target for clinical applications.
Finally, we showed that miR-217-5p was present in urine from PAN-administrated rats but not in control rats. When podocytes are severely damaged, they occasionally become detached from the glomeruli and moved into urine [54]. Furthermore, urine is rich in exosomes, unilamellar small vesicles (50-100 nm in diameter) secreted from many cell types including podocytes [54]. Consistently, we detected small vesicles with varying diameter peaking at 100 nm in the PAN-and saline-administrated rats by dynamic light scattering (DLS) ( Figure S8). Previous work demonstrated that intracellular miRNAs are loaded into exosomes and secreted into the extracellular environment [55]; therefore, it is possible that miR-217-5p is present in not only urinary podocytes but also urinary exosomes from PAN-administrated rats. Taken together, miR-217-5p may be a promising biomarker of some kidney diseases involving podocyte injury.
In the present study, we could identify a novel miRNA-mRNA axis associated with podocyte injury. Nonetheless, this study has some limitations. First, it should be demonstrated that decreased Myo1d expression in E11 and rat primary podocytes leads to their morphological changes. Furthermore, the loss-of-function studies of miR-217-5p, e.g., the phenotypic analysis of miR-217-5p-knockout podocytes generated using a genome editing technique, is necessarily performed to support our findings. We hope that the results of such additional experiments will increase the clinical significance of our findings.
Supplementary Materials: The following information can be downloaded at https://www.mdpi. com/article/10.3390/ncrna8030043/s1. Figure S1. Effect of induced miR-217-5p expression on cell viability of E11 cells. Cell viability was evaluated at day 4 after transfection. n.s.: not significant compared to the NC-transfected cells. Figure S2. Selection of mRNAs dysregulated in PAN-treated podocytes. The numbers of upregulated (A) and downregulated (B) mRNAs selected based on the criteria are shown. Figure S3. Selection of mRNAs possibly targeted by miR-217-5p. Figure S4. Interactive GO terms enriched with the predicted targets of miR-217-5p in PAN-treated podocytes. Enriched GO terms (biological process) were filtered with a threshold of p-values greater than 8.35 × 10 −5 and analyzed for interconnection using the GOnet software. Of the 12 predicted targets, only Cris1 is not connected to any of the enriched GO terms. The more intense color of the GO term node indicates the more significant enrichment of the term (the smaller p-value). Figure S5. Myo1d mRNA levels in PAN-treated primary rat podocytes. Values were normalized relative to GAPDH mRNA expression. Primary rat podocytes were treated with 0-30 µg/mL PAN. The relative expression level of Myo1d mRNA at 0 µg/mL PAN was given an arbitrary value of 1. * p < 0.05 versus 0 mg/mL PAN. Data represent the means ± SD. Figure S6. Immunocytochemical detection of Myo1d protein in E11 cells. Myo1d proteins were detected in the cytosol of E11 cells. The cytosolic localization of Myo1d proteins have been also shown in the human bone osteosarcoma cell line U-2 OS (The Human Protein Atlas database; https://www.proteinatlas.org/ENSG00000176658-MYO1 D/subcellular#, accessed on 6 June 2022). Bar: 100 µm. Figure S7. Western blot images showing the detection of Myo1d protein in E11 cells with induced miR-217-5p expression. Arrowheads indicate the immunodetected bands of Myo1d protein. Figure S8. Size distribution of particles in urine from Wistar rats administrated with PAN (100 mg/kg) or saline. Whole particles in the urine were isolated by ultracentrifugation and analyzed by dynamic light scattering. Table S1. Primers designed for SYBR qRT-PCR. Table S2. Summary of small RNA sequencing results. Table S3. mRNAs upregulated in response to PAN-induced podocyte injury. Table S4. mRNAs downregulated in response to PAN-induced podocyte injury. Table S5. Decreased Myo1d mRNA expression in injured human podocytes based on the RNA sequencing datasets registered under the accession GSE124622 in Gene Expression Omnibus. Table S6. Pathways significantly associated with the mRNAs upregulated by PAN in podocytes. Table S7. Pathways significantly associated with the mRNAs downregulated by PAN in podocytes.