Journal Logo

Original Basic Science—General

Dual MicroRNA Blockade Increases Expression of Antioxidant Protective Proteins: Implications for Ischemia-Reperfusion Injury

Tingle, Samuel J. MRes1,2; Sewpaul, Avinash FRCS1,2; Bates, Lucy BSc1,3; Thompson, Emily R. MRCS1,2; Shuttleworth, Victoria PhD1; Figueiredo, Rodrigo MRCS1,2; Ibrahim, Ibrahim K. MCh1,2; Ali, Simi PhD1; Wilson, Colin PhD1,2; Sheerin, Neil S. PhD1

Author Information
doi: 10.1097/TP.0000000000003215



The deceased donor organ pool has been successfully expanded over recent years with the increased use of organs from expanded criteria donors (ECDs).1 While this increases the number of transplants which take place, organs from ECD are more sensitive to the ischemia-reperfusion injury (IRI) which is inevitable in transplantation.2 In transplanted kidneys, IRI manifests clinically as delayed graft function, which is associated with both acute rejection and chronic graft loss.3 Ameliorating IRI could improve outcomes in ECD grafts and further expand the donor pool.

MicroRNAs (miRNAs) are a type of small noncoding RNA, which are highly evolutionarily conserved and work by binding to target messenger RNAs (mRNAs) causing breakdown and translational repression.4 Each miRNA is capable of targeting many different genes, affecting a network of proteins, thereby mediating many intrinsically linked cellular processes.4 While many current therapeutics target only a single protein within a complex network, targeting miRNA biology may allow coordinated effects on networks of proteins. While initial studies identified miRNAs which were essential for normal cellular function, it has since become clear that overexpression of certain miRNAs is implicated in the pathogenesis of many diseases.5 The overexpression of detrimental miRNAs in pathological states results in decreased expression of protective genes.

The role of miRNAs in kidney transplantation and acute kidney injury has been thoroughly reviewed, and several detrimental miRNAs have been suggested.6,7 We selected target miRNAs to investigate further based on the following criteria: high expression in ischemic renal tissue, targeting of genes known to be protective in renal IRI, have previous studies showing that miRNA blockade ameliorates IRI and have shared mRNA targets such that dual blockade may have synergistic effects. miR-24-3p and miR-145-5p both fulfill these selection criteria.

These miRNAs both target antioxidant genes superoxide dismutase 2 (SOD2) and heme oxygenase 1 (HMOX1). SOD2 is highly expressed in the kidney, and its antioxidant protective role in IRI is well established.8,9 HMOX1 is an inducible antioxidant enzyme, which has been shown to ameliorate renal IRI in animal models when its expression is increased by either viral transduction or hemin treatment.10,11

miR-24-3p is enriched in endothelial cells and promotes apoptosis.12 Blockade of miR-24-3p has been shown to be therapeutic in mouse models of both cardiac and renal IRI.13,14 Lorenzen et al13 demonstrated that miR-24-3p is upregulated in renal tissue following IRI. In vivo treatment of mice with anti-miR-24-3p antisense oligonucleotide (ASO) before IRI led to a significant decrease in kidney injury markers neutrophil gelatinase-associated lipocalin (NGAL) and kidney injury marker 1 (KIM-1) and improved kidney function and survival. The interaction between miR-24-3p and HMOX1 has been luciferase reporter validated.15 MiR-24-3p is also predicted to have 3 independent seed sites on the 3′ untranslated region (UTR) of SOD2, but this interaction remains unstudied.16

miR-145-5p is upregulated following cerebral ischemia in rat brain tissue and in human serum following ischemic stroke.17,18 ASO blockade of miR-145-5p increases expression of its target SOD2, decreases infarct volume, and improves neurological outcome in rats following cerebral IRI.17,19 The role of miR-145-5p in renal IRI remains unknown, and no previous studies have investigated miR-145-5p in combination with miR-24-3p. miR-145-5p is predicted to target the 3′UTR of SOD2 and the 5′UTR of HMOX1.16

In conclusion, miR-24-3p and miR-145-5p are highly expressed in ischemic kidneys, target protective antioxidant genes, and have been shown to have a detrimental role in IRI. We hypothesized that expression of these miRs would increase following ischemia, therefore, targets of these miRNAs will decrease following hypoxia, and that anti-miR ASO therapy could ameliorate this decrease. As miRNAs have an inhibitory action on the genes which they target, and many miRNAs have shared mRNA targets, we hypothesized that dual blockade of miRNAs would lead to synergistic upregulation of shared target genes.


Screening for miRNAs of Interest

From September 2015 to October 2015, 6 kidneys from the national organ allocation scheme that were deemed unsuitable for transplantation were used in this study. Consent for the use of the organs for transplantation and research was obtained from the donor family by the Specialist Nurses in Organ Donation before organ retrieval. Ethical approval was granted for the study by the national research ethics commission in the United Kingdom, National Research Ethics System (15/SC/0161).

To assess miRNA expression in ischemic renal tissue, biopsies were taken from a cohort of 6 human kidneys, from 4 donors, following extended cold ischemic times (further information in Table 1). Biopsies were stored in RNAlater at −80°C. RNA was extracted as described below. After quality control (Agilent 2100 Bioanalyzer System), miRNA profiling was performed using the miRCURY LNA Universal RT miRNA polymerase chain reaction (PCR) Human panel I + II (Exiqon). This uses real-time PCR to provide a miRNA profile, capable of measuring expression of 752 individual miRNAs. For normalization of the data, we used the mean value of the assays detected in each sample (Global mean normalization), as this was found to be the most stable normalizer.

Demographic information on the 6 kidneys used to screen for potential microRNAs of interest

Prediction of miRNA Targets

Prediction of miRNA-mRNA targets was performed using miRWalk2.0. This allows the following prediction tools to be used simultaneously: miRWalk2.0, miRanda, TargetScan, and RNA22.16 miR-24-3p is predicted to have 3 independent target sites on the 3′UTR of SOD2 and to also target the 3′UTR of HMOX1. miR-145-5p is predicted to have target sites within the 3′UTR of SOD2 and the 5′UTR of HMOX1.

Modeling IRI

All in vitro experiments were performed with primary human umbilical vein endothelial cells (HUVECs) from ATCC (PCS-100-010), with passage <6. HUVECs were grown to confluence in a 37°C, 5% CO2 incubator using Vascular Cell Basal Medium (PCS-100-030) supplemented with rh FGF-basic (5 ng/mL), rh insulin (5 µg/mL), ascorbic acid (50 µg/mL), l-glutamine (10 mmol/L), rh EGF (5 ng/mL), and 5% fetal bovine serum.

Hypoxia and reoxygenation were used to model IRI in vitro. HUVECs were seeded into 6-well or 96-well plates with seeding densities of 1 × 105 and 1 × 104 cells per well, respectively. Cells were kept in the media described above for 1 day, before being placed in a Sanyo Hypoxic incubator set to 1% O2, 5% CO2, and 37°C, for 24 hours. Cells were then transferred to a standard normoxic incubator with 5% CO2 at 37°C for 6 hours of reoxygenation. Cells were 80%–90% confluent when entering hypoxia. Cellular response to hypoxia was confirmed by a significant upregulation of vascular endothelial growth factor expression (Figure S1A, SDC,

ASO Blockade

Anti-miRNA ASOs were purchased from Qiagen (miRCURY LNA miRNA Power Inhibitors); these have a phosphorothioate-modified backbone and incorporate locked nucleic acids. MiRCURY LNA Power Inhibitor Negative Control A was used as a control ASO (termed scramble ASO). The use of scrambled sequence ASO is the recommended negative control for ASO miRNA blockade experiments.20 Lipofectamine-based transfection was used; culture media on 6-well plates was changed to 2 mL Opti-MEM Medium (Gibco) supplemented with 3 µL of Lipofectamine RNAiMAX Transfection Reagent (Invitrogen) and ASO to a final concentration of 25–50 nM. For reactive oxygen species (ROS) assays (further details below), 96-well plates were required; well volume was 200 µL; and lipofectamine/ASO was diluted such that their concentration matched the 6-well plates. HUVECs were transfected for 8 hours, 1 day before hypoxia and reoxygenation as described above. A no transfection control was treated in an identical fashion but lacking both lipofectamine and ASO.

RNA and Protein Extraction

The Qiagen TissueLyser with a single 5 mm RNAse-free stainless steel bead was used to homogenize core biopsy samples. Total RNA was extracted from biopsy sample homogenate and 6-well plates using Qiagen miRNeasy Mini Kit according to the manufacturer’s instructions. The NanoDrop One Microvolume UV-Vis Spectrophotometer (Thermo Fisher Scientific) was used for quantification and quality assessment of extracted RNA. Protein was extracted from 6-well plates using standard radioimmunoprecipitation assay buffer as per the Abcam protocol.21

Quantification of miRNA and Gene Expression

Quantification of miRNA and mRNA was performed with reverse transcription quantitative PCR (RT-qPCR). Reverse transcription was performed in a T100 Bio-rad Thermal Cycler using TaqMan MicroRNA Reverse Transcription Kit for miRNA and Tetro cDNA Synthesis Kit (Bioline) for mRNA, with 1 µg of total RNA loaded per reaction. PCR was performed using TaqMan gene expression assay primers and SensiFAST Probe Hi-ROX Mix (Bioline) in a StepOnePlus Real-Time PCR System.

RNU-48 was used as our reference gene for miRNA analysis and was stably expressed following hypoxia. Ribosomal protein P0 (RPLP0) was used as our mRNA reference gene as more commonly used genes, such as hypoxanthine-guanine phosphoribosyltransferase 1 (HPRT1), were not stably expressed following hypoxia (Figure S1B and C, SDC,

Western Blot

The Abcam general Western blot protocol was used to perform Western blots.21 Following protein extraction, lysate protein concentration was quantified using the Pierce BCA Protein Assay Kit, and 8 µg of total protein was loaded per well of the Western blot. This was run on a NuPAGE 4%–12% Bis-Tris Protein Gel, and semi-dry transfer was used to blot onto Trans-Blot Turbo 0.2 µm nitrocellulose membranes (Bio-Rad). After blocking with 5% BSA, relevant antibodies were diluted in 1% BSA and used to probe for proteins of interest.

Primary antibodies were used at the following dilutions: SOD2 (Abcam: Ab13533) at 1:4000, HMOX1 (Abcam: Ab52947) at 1:1000, and α-tubulin (Sigma: T6074) at 1:4000. Horseradish peroxidase-conjugated secondary antibodies and Pierce ECL Western Blotting Substrate (Thermo Scientific) were used to develop the membrane, which was imaged using the Licor Odyssey Fc imaging system. Western blot densitometry was performed using ImageJ (v1.49) software to allow semiquantification of results. Densitometry fold change was calculated, normalized for densitometry of the loading control which was performed on the same blot.

Quantifying ROS Production

Quantification of ROS production was performed using a 2′,7′-dichlorodihydrofluorescein diacetate (DCFDA) Cellular ROS Assay Kit (Abcam: ab113851), according to the Abcam protocol, as previously described.22 Briefly, cells were seeded, transfected, and exposed to hypoxia and reoxygenation as described above. Following this, cells were treated with DCFDA at a concentration of 25 µM for 45 minutes in a normoxic incubator. A fluorescence microplate reader (at 485 nm excitation and 535 nm emission wavelengths) was used to measure production of oxidized DCFDA; this allows quantification of ROS including hydrogen peroxide, peroxyl radical, and peroxynitrite.22

Statistical Analysis

Results from RT-qPCR were analyzed using the 2(−ΔΔCt) method to assess fold change in expression levels between samples. Statistical analysis was performed using GraphPad Prism 8.01. P values are the result standard 1-way ANOVAs with Bonferroni correction for multiple analyses. Graphs display mean values with error bars representing SD.


miRNA Expression In Vivo

One of our criteria for selecting potentially detrimental miRNAs for further study was high expression in human ischemic renal tissue. To assess this, a cohort of 6 human kidneys were biopsied following organ retrieval and transport. These kidneys had undergone a significant period of ischemia (mean cold ischemic time was 17 h and 14 ± 217 min). Further information about the donors can be found in Table 1. High-throughput miRNA PCR panel was used to screen for the presence of highly expressed miRNAs which also have potential detrimental roles in IRI. This revealed that in human kidneys which have undergone extended cold ischemia, miR-24-3p and miR-145-5p were highly expressed (Figure 1). In this small sample, with similar cold ischemia times, there was no significant correlation between length of cold ischemia and expression of either miR-24-3p or miR-145-5p.

High-throughput PCR panel measuring the expression of 752 miRNAs in ischemic human renal tissue, displaying mean −ΔCq values, with global mean normalization (n = 6). miRNAs expressed at a high level and investigated further in this study are highlighted. miRNA, microRNA; PCR, polymerase chain reaction.

miRNA Expression In Vitro

We then looked to investigate the expression of these miRNAs in an in vitro model. RT-qPCR was used to assess miR-24-3p and miR-145-5p expression in HUVECs following 24 hours of hypoxia and a subsequent 6 hours of reoxygenation in vitro. As hypothesized, expression of both miRNAs increased significantly in response to hypoxia and remained elevated following reoxygenation (Figure 2A, B). miR-24-3p demonstrated a 1.51-fold increase (P ≤ 0.001), and miR-145-5p demonstrated a 1.95-fold increase (P ≤ 0.001), in response to hypoxia. Levels remained elevated following reoxygenation, at 1.39 (P ≤ 0.01) and 1.57 (P ≤ 0.05), respectively, with no significant change in either miR between hypoxia and reoxygenation.

RT-qPCR fold change in expression of miR-24-3p and miR-145-5p, normalized to RNU-48, in human umbilical vein endothelial cells following hypoxia and reoxygenation (n = 3). A, fold change in miR-24-3p (P = 0.0002). B, fold change in miR-145-5p (P = 0.00 08). ns (P > 0.05), *P ≤ 0.05, **P ≤ 0.01, **P ≤ 0.001. Representative of triplicate experiments. ns, nonsignificant; RT-qPCR, reverse transcription quantitative polymerase chain reaction.

Expression of miR-24-3p and miR-145-5p Targets in Hypoxia and Reoxygenation

If the increased expression of miR-24-3p and miR-145-5p is having a biological effect, a decrease in expression of the genes which they target would be expected. Expression of antioxidant enzymes, HMOX1 and SOD2, both decreased significantly following hypoxia and failed to recover following 6 hours of reoxygenation (Figure 3), consistent with the persistently high levels of miR-24-3p and miR-145-5p. At the mRNA level, SOD2 and HMOX1 showed 1.99- and 6.05-fold decreases in expression following 24 hours of hypoxia (P ≤ 0.0001 and P ≤ 0.0001; Figure 3A, B). There was no significant increase toward normal expression following 6 hours of reoxygenation.

Expression changes in miR-24-3p and miR-145-5p targets following hypoxia and reoxygenation. A and B, RT-qPCR fold change in SOD2 and HMOX1 mRNA, normalized to RPLP0, in human umbilical vein endothelial cells following hypoxia and reoxygenation (n = 3). A, Fold change in SOD2 mRNA (P < 0.0001). B, Fold change in HMOX1 mRNA (P < 0.0001). C, D, and E, Protein changes following hypoxia and reoxygenation (n = 3). Western blot densitometry displaying fold change in SOD2 protein (P = 0.009) and HMOX1 protein (P = 0.010), normalized to loading control α-tubulin, is shown in C and D, respectively. E, Western blot for SOD2 and HMOX1 with α-tubulin as a loading control (representative of n = 3). ns: (P > 0.05), *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001. HMOX1, heme oxygenase 1; mRNA, messenger RNA; ns, nonsignificant; RPLP0, ribosomal protein P0; RT-qPCR, reverse transcription quantitative polymerase chain reaction; SOD2, superoxide dismutase 2.

These results were mirrored when looking at a protein level; densitometry of SOD2 and HMOX1 Western blots revealed a 2.98- and 2.76-fold decrease following hypoxia (P ≤ 0.05 and P ≤ 0.05; Figure 3C, D). Again, there was no significant increase toward the baseline value following 6 hours of reoxygenation. A representative Western blot is displayed in Figure 3E.

Dual Blockade of miR-24-3p and miR-145-5p

We next wanted to investigate whether blocking miR-24-3p and miR-145-5p could increase the levels of antioxidant genes following hypoxia and reoxygenation, to assess whether dual blockade is a potential therapeutic strategy. ASOs which bind to, and therefore inhibit, miR-24-3p and miR-145-5p were delivered to HUVECs before hypoxia and reoxygenation (Figure 4A). HUVECs receiving ASO with a scrambled sequence were used as negative controls, as is recommended in the literature.20 Normoxic controls were also performed and can be found in Figure S2 (SDC, Blockade of the individual miRNAs in isolation as well as dual blockade of both miRNAs was performed.

Transfection with anti-miR-24-3p and anti-miR-145-5p ASO before hypoxia and reoxygenation. A, ASO treatment timeline. B and C, RT-qPCR fold change in SOD2 and HMOX1 mRNA, normalized to RPLP0. B, Fold change in SOD2 (P = 0.013). C, Fold change in HMOX1 mRNA (P = 0.0006). D, E, and F, Protein changes with ASO transfection and subsequent hypoxia and reoxygenation (n = 3). Western blot densitometry displaying fold change in SOD2 protein (P = 0.004) and HMOX1 protein (P = 0.003), normalized to loading control α-tubulin, is shown in D and E, respectively. F, Western blot for SOD2 and HMOX1 with α-tubulin as a loading control (representative of n = 3). G, Relative ROS production (measured using DCFDA fluorescence assay and displayed as fluorescence intensity) is lower after hypoxia and reoxygenation in cells transfected with both anti-miR-24-3p and anti-miR-145-5p ASO (n = 4 with triplicate repeats). ns: (P > 0.05), *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001. ASO, antisense oligonucleotide; DCFDA, 2′,7′-dichlorodihydrofluorescein diacetate; HMOX1, heme oxygenase 1; mRNA, messenger RNA; ns, nonsignificant; ROS, reactive oxygen species; RPLP0, ribosomal protein P0; RT-qPCR, reverse transcription quantitative polymerase chain reaction; SOD2, superoxide dismutase 2.

Single blockade of either miR-24-3p or miR-145-5p in isolation did not significantly change the expression of either SOD2 or HMOX1 mRNA when compared with negative control (Figure 4B, C). When looking at a protein level, single blockade of either miRNA in isolation did not significantly affect the expression of SOD2 (Figure 4D). On Western blot, single blockade of miR-24-3p or miR-145-5p in isolation appeared to increase the expression of HMOX1 protein; however, these changes failed to reach significance on densitometry fold changes of 3.79 (P = 0.260) and 4.33 (P = 0.176), respectively (Figure 4E, F).

Dual blockade increased expression of SOD2 and HMOX1 mRNA by a factor of 1.61 (P ≤ 0.05) and 2.62 (P ≤ 0.05), respectively (Figure 4B, C), when comparing to 50 nM scramble. The expression of HMOX1 mRNA was significantly higher using dual blockade compared with single blockade of either miR-24-3p or miR-145-5p in isolation (P ≤ 0.01 and P ≤ 0.05, respectively). Dual blockade significantly increased expression of SOD2 mRNA when compared with single blockade of miR-24-3p (≤0.05), but there was no significant difference in SOD2 mRNA expression between dual blockade and miR-145-5p blockade in isolation (P = 0.806).

The above changes were more pronounced at the protein level. Dual blockade increased expression of SOD2 and HMOX1 protein by a factor of 3.17 (P ≤ 0.05) and 6.97 (P ≤ 0.05), respectively, as measured by Western blot densitometry (Figure 4D, E), when comparing to 50 nM scramble. The increased level of SOD2 protein achieved with dual blockade was significantly higher than seen with single blockade of miR-24-3p or miR-145-5p in isolation (P ≤ 0.05 for both; Figure 4D).

To assess whether these changes had a functional role, a DCFDA assay was used to analyze cellular ROS production (Figure 4G). Dual blockade caused a significant reduction of ROS production compared with 50 nM scramble ASO negative control (P ≤ 0.05) following hypoxia and reoxygenation. There were significantly lower levels of ROS production with dual blockade compared with single blockade of either miR-24-3p (P ≤ 0.05) or miR-145-5p (P ≤ 0.05). The level of ROS production in the dual blockade-treated cells was not significantly different to cells which had not undergone hypoxia/reoxygenation (normoxia in Figure 4G).

Both of these miRNAs also have predicted and validated target sites on both sphingosine-1-phosphate receptor 1 (S1PR1; a promoter of vascular integrity with a role in IRI resolution) and H2A histone family member X (H2AFX; a protective gene regarding DNA damage and oxidative stress).13,16 Expression of both S1PR1 (P ≤ 0.05) and H2AFX (P = 0.144) was higher with dual blockade compared with 50 nM scramble ASO (Figure S1D and E, SDC, GAPDH, α-tubulin, and RPLPO are not known targets of either miRNA, and their expression did not change significantly with dual blockade.


miR-24-3p and miR-145-5p are 2 miRNAs with suggested detrimental roles in IRI, which are highly expressed in human kidneys following ischemia (Figure 1).13,14,17-19 An in vitro model was used to examine the effect of ASO blockade on these detrimental miRNAs and their targets. We show that both miRNAs are upregulated by hypoxia in human endothelial cells and that their expression remains elevated following 6 hours of reoxygenation. As expected, based on miRNA biology, the target antioxidant genes SOD2 and HMOX1 show reduced expression over the same time course.

Dual blockade of miR-24-3p and miR-145-5p with ASO caused a significant upregulation of SOD2 and HMOX1, both of which have been shown to be beneficial in IRI.9-11 This had a functional impact, ameliorating ROS production in cells following hypoxia and reoxygenation. The effects on target genes was more pronounced at the protein level than that at the mRNA level. This is to be expected given that miRNAs are known to cause translational repression; preventing this aspect should lead to increased protein expression, with minimal changes in mRNA expression.4

While we have focused on genes with importance in renal IRI, we theorize that the principle of dual blockade is more far reaching; if 2 highly expressed miRNAs share an mRNA target, then dual blockade seems to have a synergistic effect (Figure 5). This can be explained by ongoing mRNA breakdown and translational repression by 1 miRNA, even when the other miRNA is inhibited. This has relevance for any future research aiming to therapeutically target miRNAs.

Theory of dual miRNA blockade synergism. miRNAs normally function by binding to target mRNA causing mRNA breakdown and translational repression, leading to decreased expression of the target protein. When 2 miRNAs which share a target gene are highly expressed, and single ASO blockade of one of the miRNAs is performed, the other miRNA continues to bind to target mRNA and repression of the target protein remains. Dual blockade prevents interaction of both highly expressed miRNAs, allowing translation of target mRNA into protein. ASO, antisense oligonucleotide; miRNA, microRNA; mRNA, messenger RNA.

A working model for the role of miR-24-3p and miR-145-5p in IRI can be found in Figure 6. Our findings that miR-24-3p and miR-145-5p are upregulated in hypoxia is supported by previous in vitro and in vivo work.13,14,17 In agreement with our data, other groups have found that HMOX1 is downregulated in hypoxia in human endothelial cells.23 Although not studied in human endothelial cells, SOD2 expression has been shown to decrease in response to hypoxia in alternative models.24 We showed that ASO miRNA blockade could increase the levels of HMOX1 and SOD2 proteins and ameliorate the surge in ROS that occurs with IRI.

Working model of the detrimental roles of miR-24-3p and miR-145-5p in human IRI. The present study demonstrated endothelial upregulation of miR-24-3p and miR-145-5p and downregulation of SOD2 and HMOX1 in response to hypoxia; these changes persist to at least 6 h postreoxygenation. This reduction in antioxidants may limit the defense against the surge in ROS which is inevitable with reperfusion. ASO blockade of miR-24-3p and miR-145-5p in combination resulted in a relative increase in levels of HMOX1 and SOD2 proteins following hypoxia and reoxygenation. This decreases ROS production in cells and may ameliorate ROS-mediated IRI. ASO, antisense oligonucleotide; HMOX1, heme oxygenase 1; IRI, ischemia-reperfusion injury; miRNA, microRNA; ROS, reactive oxygen species; SOD2, superoxide dismutase 2.

The predominantly in vitro nature of this study is a limitation, and results may not extrapolate to the in vivo setting. Therefore, future work to investigate the effects of dual blockade on functional outcomes in in vivo and ex vivo models is essential. In addition, high-throughput sequencing at the tissue level is essential to identify potential off target effects. On the basis of previous research, we predict that the augmented antioxidant defense and protection from ROS production which we have demonstrated will translate into improved outcomes. Previous in vivo work has demonstrated that blockade of miR-24-3p or miR-145-5p in isolation is able to ameliorate renal and cerebral IRI, respectively.13,17,1825 We propose that these therapeutic effects are likely at least in part due to HMOX1 and SOD2 upregulation. Upregulating HMOX1 in renal transplant recipients with heme arginate is the focus of an ongoing large randomized controlled trial (Heme Arginate in Transplantation Study; HOT2), based on promising preclinical work and a previous phase IIb study.26,27 Our proposed therapy has the potential benefit of upregulating additional protective genes in conjunction with HMOX1.

Several ASO therapies have been successfully translated into the clinic in recent years, with strong evidence supporting the efficacy and safety of ASOs which target mRNA.28-30 Unlike other DNA therapeutics, ASOs can be successfully delivered in vivo by gymnosis (naked delivery) without the need for potentially harmful transfection reagents. Furthermore, their effect is transient and causes no genomic alterations. More recently ASO targeting miRNA have been investigated. The success of Miravirsen, an ASO targeting miR-122 which significantly decreases HCV viral titer, provides in-man proof of concept that ASOs targeting miRNAs can be safe and effective, without significant off target toxicity.31,32

Delivery of miRNA-targeting ASOs using ex vivo normothermic machine perfusion (NMP) has been described. In this instance, ASO can be delivered directly to an organ before transplantation, avoiding systemic delivery. Goldaracena et al33 demonstrated successful delivery of Miravirsen to pig livers using ex vivo NMP. Using an orthotopic porcine liver transplant model, they showed that delivery of Miravirsen during NMP caused significant upregulation of miR-122 targets. These changes persisted for at least 72 hours following transplantation.33 Studies undertaken by our group have shown that anti-miR-24-3p ASO can be effectively delivered during NMP of human kidneys, without the need for transfection reagents, with significant increases in HMOX1 within 6 hours.15 Therefore, NMP may be an ideal platform for delivery of dual miRNA blockade therapies, and we will be investigating this in the future.

NMP of kidneys is currently performed at the recipient center (termed end ischemic).34,35 However, continuous NMP during transport will likely be accomplished in the future, as is the case for liver perfusion.36 Here, IRI would be partly ameliorated, and warm ischemia in the donor would become the major factor. It remains unknown whether strategies to further ameliorate IRI would be required, or successful, if continuous NMP from donor to recipient is established. Despite these concerns, continuous NMP would enable delivery of drugs, including ASO, while the organ is in transport without extending total preservation time, an improvement over end-ischemic treatment at the recipient center.

As our work has focused only on endothelial cells, the results must be extrapolated with caution. We selected endothelial cells as they are key mediators of renal IRI and show enrichment of miR-24-3p.12,37 Further investigation on renal epithelial cells is clearly required to assess the effects of dual blockade on the entire kidney. We plan to address this by implementing dual blockade in a large animal model, which will allow us to see the effect of dual blockade on a tissue level, as well as delineate the effect in different cell types.

The fact that we used HUVECs make our results more generalizable; the described combination of ASOs may be beneficial in the plethora of IRI-mediated pathologies which face surgeons. For example, dual blockade of miR-24-3p and miR-145-5p may represent a therapeutic option for conditions such as acute limb ischemia, acute mesenteric ischemia, acute tubular necrosis, and the IRI which accompanies cardiopulmonary bypass.


Ischemia and reperfusion are inevitable in kidney transplantation, but the injury they cause is variable and potentially modifiable. Dual blockade of miR-24-3p and miR-145-5p is able to synergistically upregulate protective antioxidant genes HMOX1 and SOD2 and decrease ROS production in vitro. Delivery of miR-24-3p and miR-145-5p ASOs to prevent IRI is a potential therapy worthy of further research.


1. Aubert O, Kamar N, Vernerey D, et al. Long term outcomes of transplantation using kidneys from expanded criteria donors: prospective, population based cohort study. BMJ. 2015; 351:h3557
2. Saat TC, van den Akker EK, IJzermans JN, et al. Improving the outcome of kidney transplantation by ameliorating renal ischemia reperfusion injury: lost in translation? J Transl Med. 2016; 14:20
3. NHS Blood and Transplant. Organ Donation and Transplantation: Activity Report 2017/18. Available at Accessed February 13, 2020
4. Lim LP, Lau NC, Garrett-Engele P, et al. Microarray analysis shows that some microRNAs downregulate large numbers of target mRNAs. Nature. 2005; 433:769–773
5. Tüfekci KU, Oner MG, Meuwissen RL, et al. The role of microRNAs in human diseases. Methods Mol Biol. 2014; 1107:33–50
6. Wilflingseder J, Reindl-Schwaighofer R, Sunzenauer J, et al. MicroRNAs in kidney transplantation. Nephrol Dial Transplant. 2015; 30:910–917
7. Guo C, Dong G, Liang X, et al. Epigenetic regulation in AKI and kidney repair: mechanisms and therapeutic implications. Nat Rev Nephrol. 2019; 15:220–239
8. Marklund SL. Extracellular superoxide dismutase and other superoxide dismutase isoenzymes in tissues from nine mammalian species. Biochem J. 1984; 222:649–655
9. Salvemini D, Cuzzocrea S. Superoxide, superoxide dismutase and ischemic injury. Curr Opin Investig Drugs. 2002; 3:886–895
10. Chen HH, Lu PJ, Chen BR, et al. Heme oxygenase-1 ameliorates kidney ischemia-reperfusion injury in mice through extracellular signal-regulated kinase ½-enhanced tubular epithelium proliferation. Biochim Biophys Acta. 2015; 185210 Pt A2195–2201
11. Wu YD, Wang GX, Wei JX, et al. Heme oxygenase-1 gene transfer protects rat kidney transplant from ischemia/reperfusion injury. Zhonghua Wai Ke Za Zhi. 2005; 43:1215–1218
12. Zhou Q, Gallagher R, Ufret-Vincenty R, et al. Regulation of angiogenesis and choroidal neovascularization by members of microRNA-23~27~24 clusters Proc Nat Acad Sci USA. 2011; 108:8287–8292
13. Lorenzen JM, Kaucsar T, Schauerte C, et al. MicroRNA-24 antagonism prevents renal ischemia reperfusion injury. J Am Soc Nephrol. 2014; 25:2717–2729
14. Fiedler J, Jazbutyte V, Kirchmaier BC, et al. MicroRNA-24 regulates vascularity after myocardial infarction. Circulation. 2011; 124:720–730
15. Sewpaul A, Figueiredo RS, Bates L, et al. Modifying renal gene expression by anti-sense oligonucleotide delivery during normothermic machine perfusion Transplantation. 2018; 102:S728
16. Dweep H, Gretz N. Mirwalk2.0: a comprehensive atlas of microRNA-target interactions. Nat Methods. 2015; 12:697
17. Dharap A, Bowen K, Place R, et al. Transient focal ischemia induces extensive temporal changes in rat cerebral MicroRNAome. J Cereb Blood Flow Metab. 2009; 29:675–687
18. Gan CS, Wang CW, Tan KS. Circulatory microRNA-145 expression is increased in cerebral ischemia. Genet Mol Res. 2012; 11:147–152
19. Xie X, Peng L, Zhu J, et al. miR-145-5p/Nurr1/TNF-α signaling-induced microglia activation regulates neuron injury of acute cerebral ischemic/reperfusion in rats. Front Mol Neurosci. 2017; 10:383
20. Horwich MD, Zamore PD. Design and delivery of antisense oligonucleotides to block microRNA function in cultured drosophila and human cells. Nat Protoc. 2008; 3:1537–1549
21. Abcam. General Western blot protocol: guidance for running an efficient and accurate experiment. 2010. Available at Accessed February 13, 2020
22. Degl’Innocenti D, Ramazzotti M, Sarchielli E, et al. Oxadiazon affects the expression and activity of aldehyde dehydrogenase and acylphosphatase in human striatal precursor cells: a possible role in neurotoxicity. Toxicology. 2019; 411:110–121
23. Nakayama M, Takahashi K, Kitamuro T, et al. Repression of heme oxygenase-1 by hypoxia in vascular endothelial cells. Biochem Biophys Res Commun. 2000; 271:665–671
24. Gao YH, Li CX, Shen SM, et al. Hypoxia-inducible factor 1α mediates the down-regulation of superoxide dismutase 2 in von Hippel-Lindau deficient renal clear cell carcinoma. Biochem Biophys Res Commun. 2013; 435:46–51
25. Pimstone NR, Engel P, Tenhunen R, et al. Inducible heme oxygenase in the kidney: a model for the homeostatic control of hemoglobin catabolism. J Clin Invest. 1971; 50:2042–2050
26. Marsonn L; Heme Arginate in Transplantation Study (HOT2). identifier: NCT03646344. Available at Accessed February 13, 2020
27. Thomas RA, Czopek A, Bellamy CO, et al. Hemin preconditioning upregulates heme oxygenase-1 in deceased donor renal transplant recipients: a randomized, controlled, phase IIB trial. Transplantation. 2016; 100:176–183
28. Corey DR. Nusinersen, an antisense oligonucleotide drug for spinal muscular atrophy. Nat Neurosci. 2017; 20:497–499
29. Mendell JR, Rodino-Klapac LR, Sahenk Z, et al.; Eteplirsen Study Group. Eteplirsen for the treatment of Duchenne muscular dystrophy. Ann Neurol. 2013; 74:637–647
30. Raal FJ, Santos RD, Blom DJ, et al. Mipomersen, an apolipoprotein B synthesis inhibitor, for lowering of LDL cholesterol concentrations in patients with homozygous familial hypercholesterolaemia: a randomised, double-blind, placebo-controlled trial. Lancet. 2010; 375:998–1006
31. van der Ree MH, van der Meer AJ, de Bruijne J, et al. Long-term safety and efficacy of microRNA-targeted therapy in chronic hepatitis C patients. Antiviral Res. 2014; 111:53–59
32. van der Ree MH, van der Meer AJ, van Nuenen AC, et al. Miravirsen dosing in chronic hepatitis C patients results in decreased microRNA-122 levels without affecting other microRNAs in plasma. Aliment Pharmacol Ther. 2016; 43:102–113
33. Goldaracena N, Spetzler VN, Echeverri J, et al. Inducing hepatitis C virus resistance after pig liver transplantation-a proof of concept of liver graft modification using warm ex vivo perfusion. Am J Transplant. 2017; 17:970–978
34. Nicholson ML, Hosgood SA. Renal transplantation after ex vivo normothermic perfusion: the first clinical study. Am J Transplant. 2013; 13:1246–1252
35. Hosgood SA, Saeb-Parsy K, Wilson C, et al. Protocol of a randomised controlled, open-label trial of ex vivo normothermic perfusion versus static cold storage in donation after circulatory death renal transplantation. BMJ Open. 2017; 7:e012237
36. Nasralla D, Coussios CC, Mergental H, et al.; Consortium for Organ Preservation in Europe. A randomized trial of normothermic preservation in liver transplantation. Nature. 2018; 557:50–56
37. Weight SC, Bell PR, Nicholson ML. Renal ischaemia–reperfusion injury. Br J Surg. 1996; 83:162–170

Supplemental Digital Content

Copyright © 2020 Wolters Kluwer Health, Inc. All rights reserved.