The mechanism underlying sepsis-associated acute kidney injury (SAKI), which is an independent risk factor for sepsis-associated death, is unclear. A previous study indicates that during sepsis miR-21a-3p accumulates in renal tubular epithelial cells (TECs) as the mediator of inflammation and mediates TEC malfunction by manipulating its metabolism. However, the specific mechanism responsible for the accumulation of miR-21a-3p in TECs during sepsis is unrevealed. In this study, a cecal ligation and puncture- (CLP-) induced sepsis rat model and rat TEC line were used to elucidate the mechanism. Firstly, miR-21a-3p and Ago2 levels were found out to increase in both plasma and TECs during sepsis, and the increase of intracellular Ago2 and miR-21a-3p could be mitigated when Ago2 was either inactivated or downregulated in septic plasma. Moreover, membrane Nrp-1 expression of TECs was increased significantly during sepsis and Nrp-1 knockdown also mitigated the rise of both the intracellular Ago2 and miR-21a-3p levels in TECs incubated with septic plasma. Furthermore, it was revealed that Ago2 can be internalized by TECs mediated with Nrp-1 and this process had no effect on the intracellular content of miR-21a-3p. Both Ago2 and miR-21a-3p could bind to TECs derived Nrp-1 directly. Finally, it was determined that miR-21a-3p was internalized by TECs via Nrp-1 and Ago2 facilitated this process. Taken together, it can be concluded from our results that Ago2 binding miR-21a-3p from septic plasma can be actively internalized by TECs via Nrp-1 mediated cell internalization, and this mechanism is crucial for the rise of intracellular miR-21a-3p content of TECs during sepsis. These findings will improve our understanding of the mechanisms underlying SAKI and aid in developing novel therapeutic strategies.
Dysregulated host response to infection results in the development of sepsis, which is the leading cause of death in the intensive care unit (ICU) worldwide [
Sepsis is reported to promote renal tubular epithelial cell (TEC) injury. Furthermore, TECs undergo malfunctioning rather than suffering from cell death during sepsis [
Recent studies have shown a growing consensus that both immune and nonimmune cells can constantly release miRs into an extracellular environment [
In our present study, relevant experiments on the CLP sepsis rat model and rat renal tubular epithelial cell line were carried out in order to find out the role receptor-mediated cell internalization played in the accumulation of miR-21a-3p in TECs during sepsis and relevant mechanisms.
Specific pathogen-free Sprague-Dawley rats with an average bodyweight of 200–300 g were procured from the Hubei Institute of Experimental Animal. The relevant study protocols were approved by the Institutional Animal Use and Care Committee of Fujian Medical University Union Hospital. The rats were housed in a laminar flow room under controlled temperature and humidity conditions with an artificial 12 h light/dark cycle. All rats had free access to tap water and standard rat chow. The rats were divided into the control group and cecal ligation and puncture (CLP) group randomly. The specimens of the control group were harvested 12 hours after sham operation. For the CLP group, the specimens were harvested at 12 hours, 18 hours, and 24 hours after the surgery, respectively, for further inspection and study.
CLP was performed as described before. In short, the rat was anesthetized with isoflurane and dissected to expose the cecum. The cecum was punctured in a through-and-through way at the midpoint between the end tip of the cecum and the ligation, which was located at the midpoint of the cecum. A small amount of feces were excluded before closing the abdominal cavity, and prewarmed normal saline (37°C, 5 ml/100 g) was injected subcutaneously for resuscitation. For the sham operation of the control group, the cecum was located and was returned into the abdominal before the abdominal cavity was closed.
TECs from the kidney of the rats were isolated as described before. Briefly, the cortex of the kidney was isolated carefully and ground on 80 and 100 mesh screens. The residue on the 100 mesh screen was digested with 0.2% trypsin (Hyclone). After being washed with PBS for 3 times, the TECs were sorted with magnetic cell sorting technique with FITC anti-cytokeratin 18 antibody (Abcam) and anti-FITC microbeads (Miltenyi Biotec).
The NRK52E cell line was cultured in Dulbecco’s modified Eagle’s medium (DMEM) (Hyclone) supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin solution at 37°C and 5% CO2. For specific protein expression knockdown groups and their negative controls, relevant siRNAs (Nrp-1 siRNA, siRNA NC (GenePharma)) were transfected to the cells cultured to 70-80% confluence using Lipofectamine™ 2000 according to the manufacturer’s instructions (Invitrogen) before further treatments. For different plasma incubations, firstly, the cells of different groups were cultured with the serum-free DMEM for 12 h; then, specific plasma was added accordingly and incubated with the cells for 12 hours. For the miR-21a-3p mimics and/or Ago2 treatments, all the cells with or without transfection were washed 3 times with PBS and cultured for 12 hours in a serum-containing medium; then, the cells were cultured in serum-free DMEM for 12 hours, after that the miR-21a-3p single strain mimics, biotin-labeled miR-21a-3p single strain mimics, FAM-labeled miR-21a-3p single strain mimics (GenePharma), and Ago2 (Sino Biological) were used to incubate the cells in serum-free DMEM accordingly for 12 hours before further inspections. For the Ago2 and miR-21a-3p costimulation, equimolar Ago2 and miR-21a-3p were premixed for 1 hour to form the Ago2/miR-21a-3p mimics complex before use.
Septic plasma was obtained from the sepsis rats euthanized 12 hours after the CLP operation. Ago2 monoclonal antibody (Novus) at the final concentration of 150 pg/ml was added to the septic plasma with or without further immunoprecipitation to form either Ago2 immunoprecipitation septic plasma or Ago2 monoclonal-treated septic plasma for further studies.
Plasma Ago2 levels were measured with commercial ELISA kits following the manufacturer’s instructions (Lifespan).
Real-time qPCR (RT-PCR) assay was carried out to quantitate the miR-21a-3p and Nrp-1 mRNA levels as described before. In short, total RNAs from either plasma or cells were collected and purified according to the manufacturer’s instructions (Invitrogen). First-Strand cDNA Synthesis Kit (TOYOBO) was used to synthesize cDNA while StepOne Real-Time PCR System (Invitrogen) was used to perform the relative real-time qPCR with Thunderbird SYBR qPCR Mix (TOYOBO). The U6 and GAPDH were used as internal control, respectively. The specific RNA levels were determined by the
The target protein expression levels were examined using Western Blotting as described previously. For TECs derived from the kidney of CLP rats, protease inhibitor containing RIPA buffer (Sigma) was used to extract the total protein. For samples derived from NRK52E in vitro, membrane proteins and cytoplasm proteins were firstly separated with Subcellular Protein Fractionation Kit for Cultured Cells (Thermo Scientific) in accordance with the manufacturer’s instructions; then, the proteins from either membrane or cytoplasm were prepared with protease inhibitor containing RIPA buffer (Sigma). The protein lysates were separated with SDS-PAGE and transferred to the PVDF films (Millipore). The PVDF films then were cut into pieces according to the molecular weights of the target proteins and incubated with the primary antibodies (Ago2 rabbit monoclonal antibody, 1 : 1000 (Novus); Nrp-1 rabbit monoclonal antibody, 1 : 1000 (Novus); and His-tag mouse monoclonal antibody, 1 : 1000 (MBL)) at 4°C for 12 h. Then, either AlexaFluor 680/790-labeled goat anti-rabbit IgG antibody (1 : 10000, LI-COR Biosciences) or AlexaFluor 680/790-labeled goat anti-mouse IgG antibody (1 : 10000, LI-COR Biosciences) was used as the second antibody, and the blots were visualized by LI-COR Odyssey Infrared Imaging System (LI-COR Biosciences). Quantity One was used to quantify the blots.
The immunofluorescent assay was carried out in the present study as described before. In brief, the cell climbing films were fixed in 4% paraformaldehyde with 0.1% Triton X-100 for 30 min at 4°C. Relevant primary antibodies (Ago2 rabbit monoclonal antibody, 1 : 200 (Novus); Nrp-1 rabbit monoclonal antibody, 1 : 200 (Novus)) and fluorescently labeled probe (LysoTracker Red DND-99 (YeasenBio)) were used to incubate with the films overnight at 4°C according to the manufacturers’ instructions. The films then were incubated with selected secondary antibodies (FITC-conjugated goat anti-rabbit IgG antibody and TRITC-conjugated goat anti-rabbit IgG antibody) specifically for 90 min in the dark. The nuclei were stained using 4
The flow cytometry assay was carried out to measure either cell membrane Nrp-1 level or cytoplasm Ago2 level. A Rat Neuropilin-1 PE-conjugated antibody (R&D) was used to determine the cell membrane Nrp-1 level by flow cytometry according to the manufacturer’s instruction. For the cytoplasm Ago2 measurement, Fixation/Permeabilization Solution Kit (BD) was used to permeabilized the cells accordingly; then, Ago2 mouse monoclonal antibody (Novus) and FITC-conjugated goat anti-mouse IgG antibody (Boster) were used to determine the cytoplasm Ago2 level with flow cytometry. The cytometry was executed with Beckman CytoFLEX FCM.
Immunoprecipitations were performed according to the manufacturers’ instructions. In short, the samples were incubated with the anti-Ago2 mouse monoclonal antibody (Novus) at 4°C overnight with shaking. Then, the mixtures were loaded with 40
The Magna RIP™ RNA-Binding Protein Immunoprecipitation Kit (Millipore) was used to perform RIP assay according to the instructions. Ago2 rabbit monoclonal antibody (Novus) was used as the primary antibody, SNRNP70 rabbit polyclonal antibody (Invitrogen) was used as positive control while normal rabbit IgG was used as a negative control (ProteinTech). The immunoprecipitated RNA was isolated, and RT-PCR was used to analyze the enrichment of miR-21a-3p.
miR-21a-3p and related controls were biotinylated by GeneParma. For TECs, the biotinylated oligonucleotides were incubated with TECs for 12 hours and the cells were lysed. For recombinant Nrp-1 (Abcam), the equimolar biotinylated oligonucleotides were added into the recombinant Nrp-1 PBS solution with or without recombinant Ago2 (SinoBiological) at the equimolar concentration. Subsequently, the samples were incubated with streptavidin-coated magnetic beads and biotin-coupled RNA complexes were pulled down. The relevant proteins were analyzed with immunoblotting.
Quantitative data were presented as the
To identify whether miR-21a-3p increased in both plasma and TECs, the CLP sepsis rat model was used in the present study. It can be found out with RT-PCR that both the miR-21a-3p levels in plasma and TECs increased significantly 12 h after the CLP procedure (Figure
miR-21a-3p and Ago2 levels of both TECs and plasma during sepsis. (a) Quantitative analysis of RT-PCR results of miR-21a-3p of either plasma or TECs at different time points. (b) Quantitative analysis of ELISA for Ago2 concentrations of plasma at different groups. (c) Representative RIP results of Ago2 binding miR-21a-3p in plasma of different groups. (d) Representative Western Blot results of Ago2 in TECs of the control group and CLP groups at different time points (
To further investigate the effects of septic plasma on the miR-21a-3p level in TECs and the role that plasma contained Ago2 plays in this process, we firstly obtained septic plasma from CLP rats at 12 h after the operation. After that, septic plasma was treated with either Ago2 monoclonal antibody (12 h+Ab) or Ago2 immunoprecipitation (12 h+IP) to regulate the activity and concentration of plasma Ago2 and Ago2 binding miR-21a-3p. It can be found out from the results that Ago2 immunoprecipitation significantly downregulated the concentrations of both plasma Ago2 and Ago2 binding miR-21a-3p. Ago2 monoclonal antibody did not affect the concentration of either Ago2 or Ago2 binding miR-21a-3p, but the monoclonal antibody would affect the biological activity of Ago2 in the plasma (Figure
Ago2 and miR-21a-3p levels in TECs stimulated with differently treated plasma. (a) Quantitative analysis of the results from ELIA for Ago2 and RT-PCR for miR-21a-3p of plasma with or without treatments. (b) Quantitative analysis of RT-PCR results of miR-21a-3p of TECs treated with different plasma. (c) Representative results of flow cytometry for Ago2 in TECs treated with different plasma. (d) Quantitative analysis of the results of flow cytometry for Ago2 in TECs treated with different plasma. (e) Representative Western Blot results of Ago2 in TECs treated with different plasma (
Nrp-1, a cell membrane receptor, is involved in the Ago2-mediated internalization of miRs. To find out whether miR-21a-3p could be internalized by TECs via the Ago2 pathway, it was important to verify the expression and distribution of Nrp-1 during sepsis. In vivo, our results of Western Blot and RT-PCR indicated that both the Nrp-1 expression and Nrp-1 mRNA transcription increased significantly in TECs during sepsis (Figures
Nrp-1 expression and distribution of TECs in different groups. (a) Representative Western Blot results of Nrp-1 in TECs of either control or sepsis (12 h: 12 hours after CLP) rat model). (b) Quantitative analysis of RT-PCR results of Nrp-1 mRNA transcription level in TECs of either control or sepsis (12 h: 12 hours after CLP) rat model). (c) Representative fluorescent results of Nrp-1 expression and distribution of TECs treated with different plasma. (d) Quantitative analysis of fluorescent results of Nrp-1 expression. (e) Quantitative analysis of RT-PCR results of Nrp-1 mRNA transcription level in TECs treated with different plasma. (f) Representative flow cytometry results of membrane Nrp-1 on TECs treated with different plasma. (g) Quantitative analysis of flow cytometry results of membrane Nrp-1 on TECs treated with different plasma. (h) Representative Western Blot results of membrane Nrp-1 on TECs treated with different plasma (
To further clarify the role of Nrp-1 in the elevation of intracellular Ago2 and miR-21a-3p in TECs stimulated by septic plasma, septic plasma was used to incubate with normal TECs (12 h group), siRNA NC-transfected TECs (12 h+SiRNA NC group) and Nrp-1 siRNA-transfected TECs (12 h+SiNRP-1 group) and relevant tests had been carried out. With the Western Blot of cytoplasm, it can be demonstrated that intracellular Ago2 level was increased significantly in the 12 h group than that in the control group. Moreover, Nrp-1 knockdown mitigated the increase of intracellular Ago2 of the TECs incubated with septic plasma remarkably. The intracellular level of Ago2 in 12 h+SiNrp-1 group was still remarkable higher than that of the control group. There was no significant difference in the content of intracellular Ago2 in the TECs of the 12 h groups and 12 h+SiRNA NC group (Figure
Ago2 and miR-21a-3p levels in TECs with or without Nrp-1 knockdown treated with septic plasma. (a) Representative Western Blot results of cytoplasm Ago2 of TECs with different treatments. (b) Quantitative analysis of RT-PCR results of miR-21a-3p of TECs with different treatments (
To further verify whether Nrp-1 expressed by TECs could mediate the extracellular Ago2 internalization, His-tagged recombinant Ago2 was used to stimulate both the normal TECs (Ago2 group) and TECs with Nrp-1 knockdown by Nrp-1 siRNA (Ago2+SiNrp-1 group), after that, relative measurements were carried out in the present study. With immunofluorescent measurement, it can be found out that the cytoplasm Ago2 content increased significantly in the Ago2 group but not that in the Ago2+SiNrp-1 group compared to the control (Figure
Exogenous Ago2 in TECs with different treatments. (a) Representative fluorescent results of cytoplasm Ago2 of TECs with different treatments. (b) Representative flow cytometry results of Ago2 in TECs with different treatments. (c) Quantitative analysis of flow cytometry results of Ago2 in TECs with different treatments. (d) Representative Western Blot results of cytoplasm Ago2 in TECs with different treatments. (e) Representative Western Blot results of cytoplasm His residue at 99 Kd in TECs with different treatments. (f) Representative results of immunoprecipitation of Ago2 binding His residue of TECs in the control and exogenous Ago2-treated groups. (g) Quantitative analysis of RT-PCR results of intracellular miR-21a-3p of TECs with different treatments (
To further verify whether TEC-derived Nrp-1 could mediate the internalization of Ago2 as well as the Ago2 binding miR-21a-3p in TECs, it is crucial to find out whether TEC-derived Nrp-1 could interact with Ago2 and miR-21a-3p directly. Firstly, immunoprecipitation was carried out with lysates from different groups of TECs, the results suggested that TEC-derived Nrp-1 did bind to Ago2 directly (Figure
Interaction of Nrp-1 with either Ago2 or miR-21a-3p. (a) Representative results of immunoprecipitation of the cell lysates of TECs with different treatments; the lysates were immunoprecipitated with Ago2 antibody and probed with Ago2 and Nrp-1 antibodies after electrophoresis. (b) Representative results of RNA pull-down assay for lysates of TECs treated with biotin-miR-21a-3p single strain mimics. Biotin-miR-21a-3p was precipitated; Ago2 and Nrp-1 were probed with immunoblotting. (c) Representative results of RNA pull-down assay for biotin-miR-21a-3p/Nrp-1 mixture. Biotin-miR-21a-3p was precipitated and Nrp-1 was probed with immunoblotting. (d) Representative results of RNA pull-down assay for biotin-miR-21a-3p/Ago2/Nrp-1 mixture. Biotin-miR-21a-3p was precipitated; Ago2 and Nrp-1 were probed with immunoblotting (Ctrl: control; SiNrp-1: Nrp-1 siRNA transfection; SiRNA NC: siRNA negative control transfection; NC: miR-21a-3p mimic negative control; Mimic: miR-21a-3p mimics).
Since miR-21a-3p could interact with TEC-derived Nrp-1, it was important to clarify if miR-21a-3p could be internalized by TECs and what the roles of Nrp-1 and Ago2 play in this process. To answer the above questions, TECs were divided into several groups based on whether or not Nrp-1 knockdown was performed and either FAM-miR-21a-3p mimics or Ago2/FAM-miR-21a-3p mimics complex was used to incubate with the different groups of TECs. Firstly, confocal microscopy was carried out to measure the cytoplasm FAM-miR-21a-3p content, while lysosomes were marked with Lyso-tracker to visualize the cytoplasm. It can be found out from the immunofluorescence results that cytoplasm FAM optical intensities were both increased significantly when TECs were incubated with either FAM-miR-21a-3p or Ago2/FAM-miR-21a-3p complex. Moreover, FAM optical intensity in TECs incubated with Ago2/FAM-miR-21a-3p complex was remarkably higher than that in TECs incubated only with FAM-miR-21a-3p. Nrp-1 knockdown could significantly mitigate this phenomenon (Figures
miR-21a-3p in TECs with different treatments. (a) Representative fluorescent results of intracellular FAM-miR-21a-3p mimics in TECs with different treatments. (b) Quantitative analysis of FAM optical intensities of fluorescent results of intracellular FAM-miR-21a-3p mimics in TECs with different treatments. (c) Quantitative analysis of RT-PCR results of miR-21a-3p in TECs with different treatments. (
Sepsis is a global health concern as it causes millions of deaths each year. The prevalence of SAKI in patients admitted to ICU is 10–20%. Thus, sepsis is the most common cause of AKI in critically ill patients. Compared with that among patients without AKI, the mortality rate is 50% higher among patients with SAKI [
Using in vitro cell lines for further study can eliminate the interference of in vivo confounding factors and be more accurate. In rats, sepsis is induced at 12 h post-CLP operation [
There must be receptors expressed on the cell membrane of TECs which can mediate internalization so that miR-21a-3p could be internalized associated with Ago2. Neuropilin-1 (Nrp-1) is known to be the receptor that can bind various ligands via its extracellular part consisted of several domains [
Next, to further verify our hypothesis, it was important to find out whether Nrp-1 can mediate the internalization process of extracellular Ago2 and miR-21a-3p in TECs. First of all, to verify if exogenous Ago2 could be internalized by TECs via Nrp-1, His-tagged recombinant Ago2 was used to incubate with either normal TECs or Nrp-1 knockdown TECs and relative measurements were carried out in the present study. It can be demonstrated from the results that Nrp-1 was crucial for the increase of intercellular Ago2 in TECs incubated with His-tagged recombinant Ago2. However, these findings do not indicate if the increased Ago2 in the cells is due to the internalization of exogenous Ago2 or the active syncretization of the cells after being stimulated. His tag is an amino acid residue consisting of 6 histidines, and it is one of the most common tags used to facilitate the purification of recombinant proteins [
Previous studies have demonstrated that Nrp-1 was the receptor for receptor-mediated cell internalization of many exogenous molecules such as VEGF in neurons and tumor cells [
To the best of our knowledge, this is the first study to demonstrate that Ago2 binding miR-21a-3p from septic plasma can be actively internalized by TECs via Nrp-1-mediated cell internalization, which enhances the intracellular miR-21a-3p levels in the TECs during sepsis. Although further researches are needed to verify some unsolved questions such as (1) through which specific pathway that Ago2-miR-21a-3p enters the TECs under the mediation of Nrp-1, (2) what other cellular mechanisms can mediate this process; (3) does exosome miR-21a-3p also participate in the rise of intracellular miR-21a-3p in TECs during sepsis; (4) does the mechanism elucidated in this study also exist in other cell types; and (5) is the internalization mechanism of miR-21a-3p also applicable to other miRs in TECs. For miR-21a-3p can manipulate cell metabolisms of TECs during sepsis, which is one of the most important mechanisms that induced SAKI, our findings will provide a novel basis for understanding the mechanisms underlying SAKI and aid in developing new relative therapeutic strategies accordingly.
The internalization of plasmatic Ago2 binding miR-21a-3p mediated by membrane Nrp-1 is an important mechanism underlying miR-21a-3p accumulation in TECs during sepsis.
The data used to support the findings of this study are either included within the article or available from the corresponding author upon request.
The authors have declared that no competing interest exists.
Zhiqiang Zou and Qin Lin contributed equally to this work.
This work is supported by the National Natural Science Foundation of China (Grant No. 81701889) and the Natural Science Foundation of Fujian (Grant No. 13185044).