Donor Simvastatin Treatment Prevents Ischemia-Reperfusion and Acute Kidney Injury by Preserving Microvascular Barrier Function

Authors

  • R. Tuuminen,

    Corresponding author
    1. Cardiac Surgery, Heart and Lung Center, Transplantation Laboratory, Haartman Institute, University of Helsinki, Helsinki University Central Hospital, Helsinki, Finland
    2. Institute of Biomedicine, Anatomy, University of Eastern Finland, Kuopio, Finland
    Search for more papers by this author
  • A. I. Nykänen,

    1. Cardiac Surgery, Heart and Lung Center, Transplantation Laboratory, Haartman Institute, University of Helsinki, Helsinki University Central Hospital, Helsinki, Finland
    Search for more papers by this author
  • P. Saharinen,

    1. Department of Virology, Molecular Cancer Biology Program, Research Programs Unit, Haartman Institute, University of Helsinki, Helsinki, Finland
    Search for more papers by this author
  • P. Gautam,

    1. Department of Virology, Molecular Cancer Biology Program, Research Programs Unit, Haartman Institute, University of Helsinki, Helsinki, Finland
    Search for more papers by this author
  • M. A. I. Keränen,

    1. Cardiac Surgery, Heart and Lung Center, Transplantation Laboratory, Haartman Institute, University of Helsinki, Helsinki University Central Hospital, Helsinki, Finland
    Search for more papers by this author
  • R. Arnaudova,

    1. Cardiac Surgery, Heart and Lung Center, Transplantation Laboratory, Haartman Institute, University of Helsinki, Helsinki University Central Hospital, Helsinki, Finland
    Search for more papers by this author
  • E. Rouvinen,

    1. Cardiac Surgery, Heart and Lung Center, Transplantation Laboratory, Haartman Institute, University of Helsinki, Helsinki University Central Hospital, Helsinki, Finland
    Search for more papers by this author
  • H. Helin,

    1. Division of Pathology, Genetics, HUSLAB, Helsinki University Hospital, Helsinki, Finland
    Search for more papers by this author
  • R. Tammi,

    1. Institute of Biomedicine, Anatomy, University of Eastern Finland, Kuopio, Finland
    Search for more papers by this author
  • K. Rilla,

    1. Institute of Biomedicine, Anatomy, University of Eastern Finland, Kuopio, Finland
    Search for more papers by this author
  • R. Krebs,

    1. Cardiac Surgery, Heart and Lung Center, Transplantation Laboratory, Haartman Institute, University of Helsinki, Helsinki University Central Hospital, Helsinki, Finland
    Search for more papers by this author
  • K. B. Lemström

    1. Cardiac Surgery, Heart and Lung Center, Transplantation Laboratory, Haartman Institute, University of Helsinki, Helsinki University Central Hospital, Helsinki, Finland
    Search for more papers by this author

Abstract

Ischemia-reperfusion injury (IRI) after kidney transplantation may result in delayed graft function. We used rat renal artery clamping and transplantation models to investigate cholesterol-independent effects of clinically relevant single-dose peroral simvastatin treatment 2 h before renal ischemia on microvascular injury. The expression of HMG-CoA reductase was abundant in glomerular and peritubular microvasculature of normal kidneys. In renal artery clamping model with 30-min warm ischemia, simvastatin treatment prevented peritubular microvascular permeability and perfusion disturbances, glomerular barrier disruption, tubular dysfunction and acute kidney injury. In fully MHC-mismatched kidney allografts with 16-h cold and 1-h warm ischemia, donor simvastatin treatment increased the expression of flow-regulated transcription factor KLF2 and vasculoprotective eNOS and HO-1, and preserved glomerular and peritubular capillary barrier integrity during preservation. In vitro EC Weibel–Palade body exocytosis assays showed that simvastatin inhibited ischemia-induced release of vasoactive angiopoietin-2 and endothelin-1. After reperfusion, donor simvastatin treatment prevented microvascular permeability, danger-associated ligand hyaluronan induction, tubulointerstitial injury marker Kim-1 immunoreactivity and serum creatinine and NGAL levels, and activation of innate and adaptive immune responses. In conclusion, donor simvastatin treatment prevented renal microvascular dysfunction and IRI with beneficial effects on adaptive immune and early fibroproliferative responses. Further studies may determine potential benefits in clinical cadaveric kidney transplantation.

Abbreviations
Ab

antibody

AKI

acute kidney injury

CCL

CC chemokine ligand

cDNA

complementary DNA

CoA

coenzyme A

CsA

cyclosporine A

DGF

delayed graft function

EC

endothelial cell

ELISA

enzyme-linked immunosorbent assay

EM

electron microscopy

eNOS

endothelial NOS

FITC

fluorescein isothiocyanate

H&E

hematoxylin and eosin

HIF

hypoxia-inducible factor

iNOS

inducible NOS

IRI

ischemia-reperfusion injury

MMP

matrix metalloproteinase

N

total sample size

NF-κB

nuclear factor kappa B

PCR

polymerase chain reaction

PDGF

platelet-derived growth factor

PFA

paraformaldehyde

ROS

reactive oxygen species

rRNA

ribosomal RNA

RT

reverse transcriptase

s.c.

subcutaneously

SEM

standard error of the mean

TLR

toll-like receptor

TUNEL

terminal deoxynucleotidyl transferase-mediated dUTP nick end-labeling.

Introduction

Progress in immunosuppressive medication has extensively reduced the rate of acute rejection and improved the long-term survival of kidney transplant patients [1, 2]. However, ischemia-reperfusion injury (IRI) is one of the major clinical challenges in kidney transplantation and is characterized by a series of rapid alloantigen-independent events [3]. Prolongation of kidney cold preservation is linked with the development of delayed graft function (DGF) and chronic allograft dysfunction [4, 5]. At present, there are only a few treatment modalities to prevent IRI [6].

Microvascular dysfunction is pivotal for IRI and postischemic acute kidney injury (AKI) [7]. Cytoskeletal rearrangement of microvascular endothelial cells and their surrounding pericytes may lead to increased microvascular permeability, leukocyte extravasation, thrombosis, vasoconstriction and compromised tissue perfusion [8, 9]. The molecular mechanisms behind microvascular dysfunction involve reduced expression of the flow-dependent transcription factor Krüppel-like factor (KLF)2 [10, 11] and its downstream vasculoprotective genes such as endothelial nitric oxide synthase (eNOS) [12-14] and heme oxygenase (HO)-1 [15-17]. In addition, oxygen deprivation promotes phosphorylation of intracellular myosin light chain (MLC)2 in the vascular wall [18] through Rho kinase (ROCK) activation [19]. This leads to endothelial junction disruption and vasoconstriction through stress-fiber formation and also through the release of vasoactive factors such as angiopoietin (Ang)-2 [20] and endothelin (ET)-1 [21] from Weibel–Palade bodies [22, 23].

In clinical practice, statins, HMG-CoA reductase inhibitors, reduce morbidity and mortality caused by cardiovascular diseases through their lipid lowering activity. In addition, statins have cholesterol-independent immunomodulatory [24-28] and vasculoprotective [29, 30] properties. These properties protect the kidney [16, 31] and other organs [32, 33] against experimental IRI. In population-based studies, statins abrogated postischemic AKI and improved the long-term outcome after major elective surgery [34, 35]. Here, we examined whether HMG-CoA reductase inhibition with a single-dose of simvastatin 2 h before the onset of ischemia will counteract IRI in a rat renal artery clamping model with 30-min warm ischemia and in fully MHC-mismatched kidney allografts with 16-h cold ischemia. We demonstrate that simvastatin treatment preserves postischemic microvascular barrier function, prevents the no-reflow phenomenon and enhances recovery from acute tubulointerstitial injury. Our study highlights that early microvascular protection is fundamental to prevent IRI and the subsequent activation of innate and adaptive immune responses and fibrotic pathways.

Concise Methods

Experimental design

We used rat renal artery clamping, kidney preservation and transplantation models to investigate the effects of clinically relevant single-dose peroral donor simvastatin treatment on renal IRI. In a unilateral kidney clamping model glomerular and peritubular proportions of microvascular barrier disruption were assessed by albuminuria (n = 6/group) and a modified Miles Assay (n = 6/group). The severity of tubular reabsorption disturbances was analyzed by urinary flow rate (n = 6/group). Postischemic perfusion defects and no-reflow were analyzed by laser Doppler monitoring (n = 6/group). In a bilateral kidney clamping model peritubular capillary flow was analyzed by perfusion of renal arteries with FITC-labeled Lycopersicon esculentum (Tomato) lectin (n = 9/group). Renal function and degree of tubular kidney injury were determined by serum creatinine and urea nitrogen and neutrophil gelatinase-associated lipocalin (NGAL) measurements (n = 9/group). In a kidney preservation model glomerular filtration and peritubular capillary tight junction defects on cellular level were estimated by high-resolution imaging by transmission electron microscopy (n = 6/group). In a kidney transplantation model, the posttransplant degree of tubular kidney injury was analyzed by serum NGAL levels (n = 6/group). Intragraft inflammation, cortical hyaluronan (HA) accumulation and activation of RhoA/ROCK and fibrotic pathways were analyzed by immunohistochemistry (n = 6/group) and by quantitative real-time PCR (n = 6/group). Hypoxia-induced Weibel–Palade body exocytosis was analyzed by ELISA for human angiopoietin-2 and endothelin-1 from the culture growth media of human dermal blood endothelial cells (BEC) (n = 6/group). The effect of simvastatin treated macrophages on the activation of allogenic T cells in a mixed leukocyte culture after 5 days was analyzed by ELISpot (n = 3/group). Please see Supplementary Methods for detailed information.

Renal artery clamping model

Pathogen-free, inbred male Dark Agouti (DA, RT1a) rats (Scanbur, Sollentuna, Sweden) received simvastatin (Merck Research Laboratories, Whitehouse Station, NJ) 5.0 mg/kg p.o. diluted with polyethylene glycol (PEG) to a concentration of 1.5 mg/ml. The control rats received PEG vehicle p.o. After 2 h of simvastatin treatment, the rats were anesthetized with inhalational isoflurane. A midline abdominal incision was made, and the right or both renal arteries were clamped for 30 min, depending on the study model (please see detailed methods at Supporting Information). After clamp removal, the kidney was inspected for recovering blood flow and the abdomen was closed. The rats were administered 1 mL of saline and 0.1 mL of buprenorphinum (Temgesic 0.3 mg/mL, Schering-Plough, Kenilworth, NJ) for postoperative maintenance of fluid balance and pain relief, respectively.

Kidney preservation model

Two hours before kidney removal, the DA rats received simvastatin 5.0 mg/kg p.o. The rats were anesthetized and a midline abdominal incision was performed. The right kidney was removed, subjected either to no cold ischemia and no warm ischemia, or to 16-h cold ischemia and no warm ischemia, or to 16-h cold ischemia and 1-h warm ischemia, and further analyzed.

Heterotopic kidney transplantation model

Kidney allografts were transplanted from fully major histocompatibility complex (MHC)-mismatched male Dark Agouti (DA, RT1av1) to male Wistar Furth (WF, RT1u) rats (Scanbur). Donor DA rats received simvastatin (Merck, Whitehouse Station, NJ) 5.0 mg/kg p.o. 2 h before kidney removal. The rats were anesthetized and the donor right kidney was perfused with 5 mL cold PBS containing 50 IU/mL heparin and removed with a segment of the aorta and vena cava. The kidney was preserved in +4°C PBS for 16 h, whereafter the donor allograft aorta and vena cava were anastomozed end-to-side into heterotopic position of the recipient aorta and vena cava. The warm ischemia time (operation time) was standardized to 1 h. The left native kidney of the recipient was removed during the transplantation. The rats were administered 1 mL of saline and 0.1 mL of buprenorphinum (Temgesic, Schering-Plough) as above. The allografts were harvested 5 min, 6 h or 5 days after the transplantation. To avoid acute rejection, the recipients with a 5-day follow-up were given cyclosporine A (CsA, Novartis, Basel, Switzerland) 1.5 mg/kg/d s.c. diluted with Intralipid (Fresenius Kabi, Bad Homburg, Germany).

Statistics

All data are mean ± SEM and analyzed by SPSS for Windows, version 15.0 (SPSS, Inc., Chicago, IL). For parametric comparison, Student's t-test between simvastatin treatment and the respective control groups was applied. For comparison in longitudinal studies, data was analyzed by repeated-measures ANOVA. p < 0.05 was regarded as statistically significant.

Results

HMG-CoA reductase is expressed in glomerular and peritubular microvasculature of normal rat kidneys

We investigated the expression of HMG-CoA reductase, a target molecule of statins with immunohistochemistry in normal DA rat kidneys. Immunoreactivity of HMG-CoA reductase was abundant in glomerular and peritubular microvascular structures (Figure 1), suggesting microvascular network as the direct therapeutic target of simvastatin in the kidney.

Figure 1.

HMG-CoA reductase is expressed in glomerular and peritubular vasculature of normal rat kidneys. Immunohistochemistry of normal DA rat kidneys for the localization of phosphorylated HMG-CoA reductase expression, a target molecule of statins. IgG controls are shown in insets. Scale bars = 50 μm.

Simvastatin treatment maintains tissue perfusion and microvascular stability in unilateral renal artery clamping model

First, we examined the clinical feasibility of single-dose simvastatin 5 mg/kg administration in the prevention of IRI. The dose of 5 mg/kg of simvastatin was based on our previous pharmacokinetic analyses in the rat [36]. The rats were treated either with simvastatin 5 mg/kg or PEG vehicle (p.o.). After 2 h, the right renal artery was clamped for 30 min. Unilateral kidney clamping model was used to appreciate the remote ischemia of contralateral kidney. Serial tissue Doppler analysis revealed that in vehicle-treated rats, cortical blood flow of the clamped right kidney was markedly diminished after reperfusion for at least 30 min, and had returned to normal levels at 6 h and at 3 days after reperfusion. In contrast, cortical blood flow of simvastatin-treated rats was restored within a few minutes after reperfusion (p > 0.001, Figure 2A). In a modified Miles assay, right renal artery clamping increased microvascular leakage of albumin-binding Evans blue dye 5 min after reperfusion, when compared to sham-operated controls (red dashed-line; Figure 2B; for localization see Figure 5B). Interestingly, IRI of the right kidney resulted in a clear, although more subtle, increase in microvascular permeability also in the nonclamped left kidney (Figure 2B). Simvastatin treatment significantly decreased microvascular leakage of albumin-binding Evans blue in the clamped right and the nonclamped left kidney, when compared to vehicle-treated appropriate controls (p < 0.001 and p < 0.05, Figure 2B).

Figure 2.

Simvastatin mediates direct vasculoprotective effects in rat right renal artery clamping model. The rats were given simvastatin 5 mg/kg p.o. 2 h before clamping the right renal artery to 30-min warm ischemia. (A) Serial laser Doppler analysis for dynamic microvascular blood flow measurements before clamping the renal artery and for 30 min after reperfusion. (B) Extravasated Evans blue dye measured by spectrophotometry 5 min after reperfusion. Twenty-four hours levels of (C) albuminuria and (D) urine flow rate (UFR) before clamping the right renal artery and for 3 days after reperfusion. Red dashed line represents sham-operated normal kidneys. N = 6 per group. Scale bars = 50 μm. *p < 0.05, **p < 0.01, ***p < 0.001.

To estimate the level of glomerular filtration barrier defects and tubular dysfunction, we analyzed the 24-h levels of albuminuria and urine flow rate, respectively, before clamping the right renal artery and for 3 days after reperfusion [37]. Simvastatin treatment significantly reduced the 24-h through albuminuria and urine flow rate after IRI, when compared to vehicle-treated rats (p < 0.01, Figure 2C, D).

Simvastatin treatment maintains peritubular capillary perfusion and prevents AKI in bilateral renal artery clamping model

We next used a bilateral kidney clamping model to validate the vasculoprotective effects of simvastatin, and to determine effects on kidney function parameters. The rats were treated either with simvastatin 5 mg/kg or vehicle (p.o.). After 2 h, both renal arteries were clamped for 30 min. Peritubular capillary perfusion 5 min after re-establishment of blood flow was determined by intravenous injection of endothelial cell binding FITC-labeled L. esculentum (Tomato) lectin. Simvastatin treatment prevented the loss of peritubular capillary perfusion (p < 0.05, Figure 3A), and maintained the kidney function measured as serum creatinine (p < 0.01, Figure 3B) and urea nitrogen (p < 0.01, Figure 3C) levels, when compared to vehicle-treated rats. Furthermore, ELISA analysis of NGAL serum levels, a biomarker of kidney injury, revealed that simvastatin treatment reduced tubulointerstitial injury at 3 days, when compared to vehicle-treated rats (p < 0.05, Figure 3D).

Figure 3.

Simvastatin improves perfusion and kidney function recovery after IRI in rat bilateral renal artery clamping model. The rats were given simvastatin 5 mg/kg p.o. 2 h before clamping both renal arteries to 30-min warm ischemia. (A) FITC-lectin perfused peritubular capillaries 5 min after re-establishment of blood flow. Serum analysis of (B) creatinine, (C) urea nitrogen and (D) neutrophil gelatinase-associated lipocalin (NGAL), a biomarker of tubular injury. Red dashed line represents sham-operated normal kidneys. N = 9 per group. Scale bars = 50 μm. *p < 0.05, **p < 0.01.

Donor simvastatin treatment upregulates KLF-2, eNOS and HO-1 mRNA expression and maintains glomerular barrier integrity of the kidney during preservation

To investigate whether single-dose simvastatin treatment strategy has protective effects in kidney transplantation, the allografts were analyzed during preservation, IRI at 6 h, and acute rejection 5 days after transplantation. DA rat right kidneys were removed 2 h after treatment with either simvastatin 5 mg/kg or PEG vehicle (p.o.). The kidneys were analyzed immediately after removal from the donor, or subjected to 16-h cold ischemia at +4°C and no warm ischemia, or to 16-h cold ischemia at +4°C and 1-h warm ischemia at +4°C to correspond the cold ischemia time usually seen in clinical cadaveric renal transplantation, and warm ischemia occurring during the transplantation procedure. Simvastatin treatment significantly upregulated the mRNA expression of vasculoprotective flow-regulated transcription factor KLF2 (p < 0.01, Figure 4A), HO-1 (p < 0.05, Figure 4A) and eNOS (p < 0.001, Figure 4A), when compared to kidneys from vehicle-treated DA rats.

Figure 4.

Donor simvastatin treatment upregulates KLF2, its downstream HO-1 and eNOS gene expression in the kidney during cold and warm ischemia, maintains glomerular barrier integrity and inhibits hypoxia-induced Weibel–Palade body exocytosis. The DA rats were given simvastatin 5 mg/kg p.o. 2 h before the right kidney was removed and subjected either to no cold ischemia and no warm ischemia, or to 16-h cold ischemia at +4°C and no warm ischemia, or to 16-h cold ischemia at +4°C and 1-h warm ischemia at +25°C. (A) Real time RT-PCR analysis on relevant genes involved in vascular wall homeostasis. (B) Immunohistochemistry for phosphorylated intracellular MLC2 and membrane skeletal protein adducin as in situ indicators of RhoA/ROCK activation. (C, D) Transmission electron microscopy (TEM) to detect glomerular and peritubular capillary barrier defects. (C) Major detachment of glomerular endothelial cell layer and podocyte foot processes from the glomerular basement membrane and (D) loss of peritubular capillary interendothelial tight junctions were observed in nontreated kidneys subjected to 16-h cold and 1-h warm ischemia. (E) Exocytosis of Weibel–Palade body proteins angiopoietin-2 and endothelin-1 in human dermal blood endothelial cell culture after 24 h in either normoxic or hypoxic environment. (A) The results were normalized to 18S rRNA molecule numbers and are given as a ratio to mRNA in vehicle-treated kidneys subjected to no cold ischemia and no warm ischemia. Red dashed line represents (B–D) normal kidneys and (E) baseline of dermal blood endothelial cell culture. BEC, blood endothelial cell; e, endothelium; f, fenestra; g, gap; GBM, glomerular basement membrane; L, lumen; p, podocyte foot process; PCW, peritubular capillary wall; s, slit diaphragm; T, tubule; tj, tight junction; U, urinary space; V, vascular space. IgG controls are shown in insets. N = 6 per group. Scale bars = 50 μm (B) = 100 nm (C, D). *p < 0.05, **p < 0.01, ***p < 0.001.

As simvastatin inhibits RhoA/ROCK activation—an important regulator of vascular stability—we performed immunohistochemical staining for the phosphorylated forms of RhoA/ROCK downstream targets p-MLC2 and p-adducin [38, 39]. During preservation, donor simvastatin treatment significantly reduced microvascular immunoreactivity of p-MLC2 (p < 0.05, Figure 4B) and p-adducin (p < 0.01, Figure 4B) in the kidneys with 16-h cold and 1-h warm ischemia, when compared to the kidneys from vehicle-treated donors.

Ultrastructural changes of glomerular filtration barrier and peritubular capillary tight junction integrity were determined by transmission electron microscopy. In kidneys with 16-h cold and 1-h warm ischemia from vehicle-treated donors, preservation induced severe glomerular filtration and peritubular capillary barrier defects, detected as major detachment of glomerular endothelial cell layer and podocyte foot processes from the glomerular basement membrane (Figure 4C, middle photomicrograph) and loss of peritubular capillary interendothelial tight junctions (Figure 4D, middle photomicrograph). In kidneys with 16-h cold and 1-h warm ischemia from simvastatin-treated DA rats, the fenestrae of the endothelium did not reach over 100 nm in most of the glomerular units (p < 0.01, Figure 4C, right photomicrograph) and peritubular capillary interendothelial gaps were less frequent (p < 0.05, Figure 4D, right photomicrograph), suggesting preserved glomerular and peritubular capillary barrier.

Simvastatin treatment inhibits hypoxia-Induced exocytosis of angiopoietin-2 and endothelin-1 in endothelial cell culture

Next, we investigated whether simvastatin could prevent exocytosis of vasoactive factors from EC Weibel–Palade bodies and thereby mediate its vasoprotective effects on the microvasculature. Therefore, human dermal BEC cultures were exposed to normoxia or hypoxia (1% O2) for 24 h. ELISA was used to detect the presence of secreted angiopoietin-2 and endothelin-1 in growth media without or in the presence of 1.0 μM of simvastatin. In EC cultures, both normoxia and hypoxia induced a significant release of permeability factor angiopoietin-2 and vasoconstriction factor endothelin-1. Simvastatin inhibited the secretion of angiopoietin-2 (p < 0.001, Figure 4E) and endothelin-1 (p < 0.001, Figure 4E) to cell media both in normoxic and hypoxic environments, when compared to vehicle-treated cells.

Donor simvastatin treatment reduces vascular permeability in kidney allografts after IRI

To study whether the 16-h cold and 1-h warm ischemia-induced endothelial barrier destabilization (Figure 4C, D) results in vascular leakage immediately after reperfusion, kidney transplantations were performed from fully MHC-mismatched male DA to male WF rats. Simvastatin 5 mg/kg or vehicle was given to the donors 2 h before kidney allograft removal and the allografts were preserved in PBS at +4°C for 16 h. Administration of intravenous Evans blue to the recipient at the time of reperfusion resulted in significant extravasation of the dye in kidney allografts from vehicle-treated donors 5 min after reperfusion (p < 0.001, Figure 5A), when compared to sham-operated normal kidney (red dashed line). Donor simvastatin treatment significantly decreased Evans blue extravasation in kidney allografts with 16-h cold ischemia time, when compared to vehicle-treated allografts (p < 0.001, Figure 5A).

Figure 5.

Donor simvastatin treatment decreases vascular leakage of kidney allografts. The DA rats were given simvastatin 5 mg/kg p.o. 2 h before kidney removal. The DA kidneys were subjected to 16-h cold preservation at +4°C and 1-h warm preservation at +25°C and transplanted to fully MHC-mismatched WF rat recipients. (A, B) Extravasated Evans blue dye measured by spectrophotometry 5 min after reperfusion. (B) Representative photomicrograph of the kidney allograft from a nontreated donor showing perfused vessels (green; endothelium binding FITC-conjugated Lycopersicon esculentum lectin) 5 min after reperfusion. Red dashed line indicates sham-operated normal kidneys. N = 6 per group. ***p < 0.001.

Donor simvastatin treatment significantly decreased infiltration of innate immune and alloimmune cells in kidney allografts after IRI

As donor simvastatin treatment resulted in rapid vasculoprotection immediately after IRI (Figure 5A), we next investigated the subsequent effects on innate and alloimmune responses, expression of danger-associated ligands, tubulointerstitial injury and activation of fibrotic cascades 6 h and 5 days after transplantation. In the 5-day study model, the recipients were also given CsA 1.5 mg/kg/day s.c. to avoid development of severe acute rejections.

In kidney allografts with 16-h cold and 1-h warm ischemia, donor simvastatin treatment significantly decreased the number of allograft-infiltrating myeloperoxidase+ neutrophils (p < 0.01, Figure 6A), ED1+ macrophages (p < 0.001, Figure 6B) and CD8+ T cells (p < 0.001, Figure 6D) at 6 h and the number of myeloperoxidase+ neutrophils (p < 0.001, Figure 6A), ED1+ macrophages (p < 0.001, Figure 6B), CD4+ T cells (p < 0.001, Figure 6C), CD8+ (p < 0.001, Figure 6D) and OX62+ dendritic cells (p < 0.05, Figure 6E) at 5 days. These results suggest that donor simvastatin treatment significantly reduced infiltration of innate immune cells at 6 h and alloreactive T cells at 5 days.

Figure 6.

Donor simvastatin treatment decreases leukocyte infiltration in kidney allografts. The density of allograft-infiltrating (A) myeloperoxidase+ neutrophils, (B) ED1+ macrophages, (C) CD4+ and (D) CD8+ T cells and (E) OX62+ dendritic cells 6 h and 5 days (representative photomicrographs) after kidney transplantation. IgG controls are shown in insets. N = 6 per group. Scale bars = 50 μm. *p < 0.05, **p < 0.01, ***p < 0.001.

Donor simvastatin treatment inhibits expression of danger-associated innate immunity ligand HA

As IRI induces synthesis and accumulation of renal cortical HA [40]—an endogenous danger signal that activates innate immunity and may also link graft injury to adaptive immunity activation—we studied the effect of donor simvastatin treatment on mRNA and protein expression of hyaluronan synthases (HAS1-3) and HA in kidney allografts after IRI. In kidney allografts with 16-h cold and 1-h warm ischemia, donor simvastatin treatment significantly reduced the HAS1 mRNA expression (p < 0.05, Table 1), the number of cortical cells immunoreactive for HAS1-3 expression (p < 0.05, Figure 7A) and the density of HA expression 6 h and 5 days after transplantation (p < 0.05, Figure 7B), when compared to allografts from vehicle-treated donors. Donor simvastatin treatment did not have any significant effect on the expression of the major HA receptor, CD44 (Figure 7C).

Table 1. Relative mRNA levels after simvastatin treatment in kidney allografts at 6 h and 5 days
mRNA6 h5 days
No treatmentSimvastatinNo treatmentSimvastatin
  1. Donors were treated with simvastatin 5 mg/kg p.o. 2 h before graft harvest. Nontreated donors served as controls. Kidney allografts were collected 6 h or 5 days after reperfusion for quantitative real-time RT-PCR (N = 6/group). The results were normalized to 18S rRNA and are given as a ratio to mRNA in control allografts at 6 h.
  2. Data is given as mean ± SEM. Student t-test was used for two-group comparison. Bold represents statistically significant difference between the study qroups. *p ≤ 0.05, **p < 0.01, ***p < 0.001.
  3. Ang, angiopoietin; Arg, arginase; ET, endothelin; HAS, hyaluronic acid synthase; HIF, hypoxia inducible factor; iNOS, inducible nitric oxide synthase; MHC II, major histocompatibility complex class II; MMP, matrix metalloproteinase; NF-kB, nuclear factor kappaB; PDGF, platelet derived growth factor; TGF, transforming growth factor; TLR, toll-like receptor; VEGF, vascular endothelial growth factor.
Ang-11 ± 0.090.67±0.06*1.33 ± 0.121.35 ± 0.10
Ang-21 ± 0.130.56 ± 0.17**0.31 ± 0.070.28 ± 0.04
Arg-11 ± 0.310.18 ± 0.04*0.27 ± 0.050.37 ± 0.06
Bax/Bcl-21 ± 0.080.71 ± 0.08*0.79 ± 0.050.82 ± 0.06
CD-801 ± 0.110.53 ± 0.03*3.61 ± 0.762.46 ± 0.30
CD-831 ± 0.070.66 ± 0.10*1.40 ± 0.081.76 ± 0.15
CD-861 ± 0.130.62 ± 0.08*1.06 ± 0.061.06 ± 0.10
ET-11 ± 0.100.64 ± 0.07*0.37 ± 0.030.40 ± 0.04
HAS-11 ± 0.150.56 ± 0.11*0.09 ± 0.010.10 ± 0.02
HIF-1α1 ± 0.111.21 ± 0.261.83 ± 0.211.02 ± 0.07***
iNOS1 ± 0.310.65 ± 0.240.77 ± 0.130.46 ± 0.04*
MHC II1 ± 0.130.53 ± 0.098.15 ± 1.355.97 ± 0.48
MMP-91 ± 0.220.34 ± 0.05*0.91 ± 0.220.99 ± 0.13
NF-κB1 ± 0.141.07 ± 0.122.07 ± 0.211.33 ± 0.11*
PDGF-A1 ± 0.090.64 ± 0.11*0.56 ± 0.060.55 ± 0.04
T-best1 ± 0.070.73 ± 0.0915.88 ± 1.8412.88 ± 0.92
TGF-β11 ± 0.070.68 ± 0.10*4.55 ± 0.703.07 ± 0.22
Tie-21 ± 0.120.57 ± 0.070.89 ± 0.150.82 ± 0.08
TLR-21 ± 0.280.80 ± 0.252.28 ± 0.451.42 ± 0.07*
VEGF-A1 ± 0.761.29 ± 0.120.95 ± 0.120.56 ± 0.03*
VEGF-C1 ± 0.110.57 ± 0.07**0.61 ± 0.080.82 ± 0.04
VEGFR-31 ± 0.110.54 ± 0.09**0.48 ± 0.030.47 ± 0.04
18S rRNA1 ± 0.030.95 ± 0.051.02 ± 0.041.01 ± 0.01
Figure 7.

Donor simvastatin treatment inhibits the expression of danger-associated innate immunity ligand hyaluronan in kidney allografts. Cortical immunoreactivity of (A) hyaluronan synthases (HAS1-3), (B) hyaluronan (HA) and (C) the levels of CD44, a major hyaluronan receptor 6 h and 5 days (representative photomicrographs) after kidney transplantation. Insets show IgG controls. N = 6 per group. Scale bars = 50 μm. *p < 0.05, **p < 0.01, ***p < 0.001.

Donor simvastatin treatment reduces mRNA expression of vascular permeability factor Ang-2 and genes involved in macrophage dendritic cell activation in kidney allografts after IRI

Kidney allografts with 16-h cold and 1-h warm ischemia were analyzed by quantitative real-time RT-PCR to define the mRNA expression levels of genes involved in microvascular homeostasis and innate immune activation after IRI (Table 1 and Supplementary Table 1). At 6 h, donor simvastatin treatment significantly reduced the mRNA expression of vascular permeability factor Ang-2, its receptor Tie-2 and genes involved in macrophage/dendritic cell activation such as VEGF-C and VEGFR-3, CD80, CD83 and CD86 and MHC class II, when compared to allografts from vehicle-treated donors (p < 0.05, Table 1).

At 5 days, donor simvastatin treatment significantly downregulated the mRNA levels of transcription factor HIF-1α, its downstream iNOS and VEGF as well as innate immune response receptor TLR-2 and its downstream NF-κB in kidney allografts with 16-h cold and 1-h warm ischemia (p < 0.05, Table 1).

Simvastatin treatment of DA rats decreased macrophage TNF-a and iNOS mRNA expression but did not affect macrophage-Mediated priming of alloreactive T cells in vitro

Our results so far indicated that donor simvastatin treatment has direct vasculoprotective and anti-inflammatory effects at 6 h resulting in diminished alloimmune reaction at 5 days. As the effects of donor simvastatin treatment are carried along with the transplanted kidney, the possible cellular targets include passenger leukocytes. To simulate possible direct simvastatin effects on passenger leukocytes, macrophages of DA rats were isolated 2 h after treatment with either simvastatin 5 mg/kg or PEG vehicle (p.o.). Quantitative real-time PCR revealed that simvastatin treatment decreased TNF-α and iNOS that are implicated in macrophage polarization to M1 phenotype, but also decreased Arg-1 that is implicated in M2 macrophage phenotype (p < 0.05, Supplementary Figure 1A). Furthermore, IFN-γ ELISpot assay showed that culture of simvastatin-treated DA macrophages with WF splenocytes resulted in similar activation of allogenic T cells compared to PEG-treated macrophages (Supplementary Figure 1B).

Importantly, this simplified in vitro experimental approach does not exclude the possibility that passenger leukocytes or graft-infiltrating recipient antigen presenting cells may be indirectly activated by kidney allograft injury in vivo. Collectively, in vitro results together with the results obtained from allografts analyzed at 6 h after reperfusion suggest that donor simvastatin treatment may have prevented the activation of passenger leukocytes or dendritic cells in the kidney allograft through reduced IRI and thus possibly affecting their ability to migrate to secondary lymphoid organs for antigen presentation to alloreactive T cells.

Donor simvastatin treatment enhances recovery from acute tubulointerstitial injury

Semiquantitative assessment of tubular necrosis revealed severe tubular necrosis both in kidney allografts from vehicle- and simvastatin-treated donors 6 h after reperfusion (Figure 8A). Additionally, interstitial hemorrhage was observed in kidney allografts from vehicle-treated donors (Figure 8A). Donor simvastatin treatment reduced the pro-apoptotic BAX/Bcl-2 mRNA ratio (p < 0.05, Table 1) and the number of TUNEL-positive apoptotic kidney cells (p < 0.01, Figure 8B) at 6 h, and that of tubulointerstitial injury marker Kim-1 immunoreactive tubuli 6 h (p < 0.01, Figure 8C) and 5 days (p < 0.05, Figure 8C) after transplantation.

Figure 8.

Donor simvastatin treatment inhibits acute tubulointerstitial injury in kidney allografts. (A) Paraffin-embedded kidney samples were stained with hematoxylin and eosin for semiquantitative assessment of tubular necrosis. Left photomicrograph; severe tubular epithelial cell necrosis and interstitial hemorrhage, right photomicrograph; severe tubular epithelial cell necrosis (lower right part), well-preserved tubuli (upper left part) and two normal glomeruli. (B) TUNEL-positive apoptotic kidney cells and (C) tubulointerstitial injury marker Kim-1 immunoreactive tubuli 6 h and 5 days after kidney transplantation. (D) Serum creatinine and urea nitrogen levels and ELISA analysis of serum levels of kidney injury biomarker NGAL. Red dashed line represents normal rats. Representative photomicrographs at 6 h (A, B) and 5 days (C). Normal kidneys are shown in insets. N = 6 per group. Scale bars = 50 μm. *p < 0.05, **p < 0.01.

Next, we analyzed renal function with serum creatinine and urea nitrogen measurements. Donor simvastatin treatment reduced serum creatinine levels at 6 h (p < 0.01, Figure 8D). At 5 days serum creatinine and urea nitrogen returned to near normal levels both with vehicle and simvastatin treatment. ELISA analysis of serum levels of kidney injury biomarker NGAL revealed that 16-h cold and 1-h warm ischemia induced a significant tubulointerstitial injury in kidney allografts 6 h and 5 days after transplantation, when compared to normal rats (red dashed line; Figure 8D). Donor simvastatin treatment significantly decreased tubulointerstitial injury measures by serum NGAL levels from kidney allograft recipients at 5 days (p < 0.01, Figure 8D).

Donor simvastatin treatment decreased TGF-β1 signaling and initiation of profibrotic cascades in kidney allografts after IRI

As inflammation and TGF-β1 signaling regulate fibroproliferation, kidney allografts with 16-h cold and 1-h warm ischemia were stained with antibodies specific for TGF-β1, phosphorylated Smad2 (p-Smad2) which indicates activation of the TGF-β1 pathway, molecular marker fibroblast specific protein-1 (FSP1) for epithelial- and endothelial-to-mesenchymal transition, and fibroblast marker prolyl-4-hydroxylase. In allografts from vehicle-treated donors, clear TGF-β1 signaling and activation of the fibrotic cascade was observed (Figure 9A–D). Donor simvastatin treatment significantly reduced the TGF-β1 activation measured as p-Smad2+ cells (p < 0.01, Figure 9B), FSP1 immunoreactivity (p < 0.01, Figure 9C) and prolyl-4-hydroxylase+ fibroblast cell count (p < 0.01, Figure 9D) at 5 days, when compared to vehicle-treated donors. Moreover, quantitative real-time PCR analysis revealed that donor simvastatin treatment downregulated intragraft pro-fibrotic ET-1, PDGF-A and TGF-β1 and tissue remodeling MMP-9 mRNA (p < 0.05, Table 1).

Figure 9.

Donor simvastatin treatment decreases TGF-β1 signaling and fibroproliferative activity in kidney allografts 5 days after transplantation. (A) The immunoreactivity of TGF-β1 and (B) its activation pathway analyzed by phosphorylated Smad2 immunoreactivity. (C) The number of fibroblast specific protein-1 (FSP1) for tissue fibroblasts and for epithelial and endothelial cells converting into fibroblasts. (D) Prolyl-4-hydroxylase+ immunostaining for fibroblast density. IgG controls are shown in insets. N = 6 per group. Scale bars = 50 μm. **p < 0.01.

Discussion

Immunosuppressive therapies in solid organ transplantation target mainly T cell mediated adaptive immune responses in allograft recipients. However, adaptive immune responses are enhanced by processes that activate innate immunity such as brain death, preservation injury, IRI and infections. Unfortunately the therapeutic strategies against innate immune responses are currently limited. The aim of this study was to investigate whether donor simvastatin treatment regulates preservation injury, IRI, AKI and innate and adaptive immune responses in kidney allografts subjected to 16-h cold preservation. We found that HMG-CoA reductase—a therapeutic target for statins—was expressed in glomerular and peritubular vascular networks of native kidneys. Importantly, a clinically relevant donor simvastatin treatment protocol abolished kidney allograft IRI through direct vasculoprotective effects. Donor simvastatin treatment modified the expression of vasculoprotective genes, reduced the release of vasoactive peptides from endothelial storage granules, and inhibited microvascular RhoA/ROCK activation and endothelial barrier integrity, indicating that several mechanisms contributed to the vasculoprotective effects of simvastatin.

IRI is generally viewed as a biphasic event comprising the loss of oxygen supply and metabolic substrates during preservation, and the production of reactive oxygen species (ROS) during reintroduction of blood flow. These events reprogram the expression of vasoactive genes in a way that in turn may induce vascular dysfunction and contribute to the development of IRI. In addition to hypoxia and re-oxygenation, loss of pulsatile flow and vascular mechanical shear stress during allograft preservation have been shown to play a central role in IRI. In particular, downregulation of the shear stress-regulated transcription factor KLF-2 during flow-cessation may result in endothelial barrier disruption [41], and activation of proinflammatory [42] and prothrombotic pathways [11]. In accordance with previous reports on statins and KLF2 [43, 44], we found that donor simvastatin treatment upregulated the mRNA expression of KLF2, eNOS and HO-1 in the kidney during 16-h cold preservation. These observations suggest that at least part of the vasculoprotective effects of donor simvastatin treatment may be due to the induction of the shear stress-regulated KLF-2 transcription factor and eNOS [11, 44] and HO-1 [45].

In contrast to changes in the mRNA expression, the release of vasoactive peptides from EC storage granules—the Weibel–Palade bodies—serves as a rapid stress signal during IRI. Several IRI-related stimuli such as hypoxia, reoxygenation and inflammation result in exocytosis of vasoactive peptides such as von Willebrand factor, P-selectin, Ang2 and ET-1 from EC Weibel–Palade bodies [46, 47], factors related to adverse cardiovascular events [48, 49]. Although we found that donor simvastatin treatment had no effects on Ang2 or ET-1 mRNA expression in kidney allografts during preservation, it decreased the mRNA expression of Ang2 and ET-1 during IRI. In addition, in vitro assays showed that simvastatin inhibited EC release of Ang2 and ET-1. Similarly, a previous report shows that simvastatin inhibits von Willebrand factor release from EC [22]. Together, these results indicate that statins may mediate their vasculoprotective mechanisms by inhibition of the Weibel–Palade body exocytosis from endothelial cells.

Loss of endothelial barrier function has a key role in IRI and results in increased vascular permeability, tissue edema and inflammation. Here we show that simvastatin treatment abolished IRI-induced vascular permeability and albuminuria indicating preserved endothelial integrity in peritubular microvasculature and glomeruli, respectively. Furthermore, we found structural evidence of preserved glomerular endothelial integrity with diminished glomerular EC–EC gap formation in simvastatin-treated kidneys. As activation of the RhoA/ROCK pathway leads to formation of intracellular stress fibers, disruption of EC–EC junctions, cellular contractility and renders kidney medulla vulnerable to AKI during postischemic no re-flow [50-52], it is interesting that donor simvastatin treatment inhibited microvascular RhoA/ROCK activation in kidney allografts during preservation. This is in line with findings of the role of simvastatin as a RhoA/ROCK inhibitor. In addition, a recent report shows that specific ROCK inhibition counteracts IRI in the kidney through vasculoprotection further supporting the central role of RhoA/ROCK in IRI [53]. Although in our models early microvascular protection was proven, it is probable that simvastatin has effects not only on vascular cells but also on tubular epithelial cells [54] and—in renal artery clamping model—direct HMG-CoA reductase-independent effects on leukocytes [24].

Recent studies have highlighted the importance of danger-associated endogenous ligands as a link between innate and adaptive immune responses as well as fibrotic processes. Here, we found that IRI resulted in profound HA expression and that donor simvastatin treatment abolished HA synthesis and its accumulation in the cortex of kidney allografts. HA is synthesized extensively in acute cellular stress such as renal IRI [40], and is broken into low molecular weight (LMW) fragments. These fragments in turn act as endogenous danger ligands for TLR2 receptor inducing activation and maturation of dendritic cells that stimulate naïve T lymphocytes [55]. As we found that donor simvastatin treatment inhibited the infiltration of adaptive immunity cells at 5 days, it is possible that the diminished activation of innate immunity delayed adaptive immune responses. Furthermore, HA-CD44 interaction orchestrates the TGF-β1-dependent fibroproliferative phenotype [56, 57], as CD44 contains a TGF-βRI-binding site [58]. Similarly to the diminished adaptive immunity activation, donor simvastatin treatment reduced the expression of profibrotic genes, activation of TGF-β1 and allograft fibroblast density indicating that inhibition of IRI and danger-associated ligand expression may have sustained beneficial effects at later time points.

Clinical trials of statin treatment after kidney transplantation have shown that recipient statin treatment does not affect the incidence of acute rejections but it may reduce the risk of cardiovascular events [59]. In contrast to recipient statin treatment, donor statin treatment targets the transplanted organ and the very proximal events that may lead to early allograft injury and predispose the organ to the development of DGF [60]. Donor treatment with another HMG-CoA reductase inhibitor atorvastatin for 2 days has been shown to protect against IRI in a rat kidney transplantation model [61], although it failed to protect kidney allografts from rejection in a nonimmunosuppression brain death model that results in extensive allograft injury [62]. Importantly, our study shows that kidney IRI may be prevented with a single-dose donor simvastatin administered per os only 2 h before kidney ischemia. This fits to the time window to treat a cadaveric organ donor that is usually 6–8 h from the statement of brain death to organ procurement, and is also supported by our previous results on the pharmacokinetics after donor simvastatin treatment [36]. We have therefore initiated a randomized clinical trial to investigate the effect of donor simvastatin treatment as an adjunct therapy on short- and long-term results of kidney allografts. As experimental models cannot take into account all the variables involved in the clinical situation, this clinical study should reveal whether donor simvastatin treatment has beneficial effects for example on DGF.

In conclusion, we found that a clinically relevant donor simvastatin treatment protocol abolished IRI in kidney allografts, which is in conjunction with our previous results in cardiac allografts [36]. Therefore, the vasculoprotective effects of statins may not be restricted to specific organs—important information when considering donor simvastatin treatment and multiorgan donation. The mechanisms of statins may involve several vasculoprotective pathways such as inhibition of endogenous danger-associated ligands, inhibition of innate and adaptive immune responses and profibrotic cascades. These results highlight the profound vasculoprotective effects of statins, and implicate donor simvastatin treatment as a feasible strategy for organ protection. Ultimately, clinical trials are needed to determine the potential of donor simvastatin treatment in clinical kidney transplantation.

Acknowledgments

We gratefully acknowledge Ms. Jaana Komulainen, MLT, Ms. Eija Rahunen, MLT, and Mr. Kari Kotikumpu, MLT, for excellent technical assistance. This study was supported by grants from the Academy of Finland, the Sigrid Juselius Foundation, Helsinki University Central Hospital Research Funds, Finnish Life and Pension Insurance Companies, the Finnish Kidney Foundation, University of Helsinki, Research and Science Foundation of Farmos, Biomedicum Helsinki Foundation, The Finnish Medical Foundation, Sirpa and Markku Jalkanen Foundation, Emil Aaltonen's Foundation and Finnish Transplantation Society.

Disclosure

The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.

Ancillary