SEARCH

SEARCH BY CITATION

Keywords:

  • Gene silencing;
  • ischemia-reperfusion injury;
  • liposomes;
  • liver;
  • siRNA;
  • TLR4

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

RNAi-based therapy is a promising strategy for the prevention of ischemia-reperfusion injury (IRI). However, systemic administration of small interfering RNA (siRNA) may cause globally nonspecific targeting of all tissues, which impedes clinical use. Here we report a hepatocyte-specific delivery system for the treatment of liver IRI, using galactose-conjugated liposome nanoparticles (Gal-LipoNP). Heptocyte-specific targeting was validated by selective in vivo delivery as observed by increased Gal-LipoNP accumulation and gene silencing in the liver. Gal-LipoNP TLR4 siRNA treatment resulted in a significant decrease of serum alanine transferase (ALT) and aspartate transaminase (AST) in a hepatic IRI model. Histopathology displayed an overall reduction of the injury area in the Gal-LipoNP TLR4 siRNA treated mice. Additionally, neutrophil accumulation and lipid peroxidase-mediated tissue injury, detected by MPO, MDA and ROS respectively, were attenuated after Gal-LipoNP TLR4 siRNA treatment. Moreover, therapeutic effects of Gal-LipoNP TLR4 siRNA were associated with suppression of the inflammatory cytokines IL-1 and TNF-α. Taken together, this study is the first demonstration of liver IRI treatment using liver-specific siRNA delivery.


Abbreviations: 
Lipo-NP

liposomal nanoparticles

Gal-Lipo-NP

galactose-conjugated Lipo-NP

IRI

ischemia-reperfusion injury

MDA

malondialdehyde

MPO

myeloperoxidase

ROS

reactive oxygen species

siRNA

small interfering RNA

TLR4

Toll-like receptor 4

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Liver ischemia-reperfusion injury (IRI) occurs during surgical procedures, which is unavoidable in liver transplantation (1). A clinically relevant consequence of IRI is diminished graft function and clinical outcomes associated with both acute and chronic rejection (2). Although the mechanisms of liver IRI remain to be fully elucidated, accumulating evidence suggests that pattern recognition receptors, such as toll-like receptors (TLRs), play a central role in liver IRI. TLRs “sense” signals associated with tissue injury such as damage-associated molecular patterns (DAMPs) (3). TLR-signaling pathways initiate innate immunity and inflammation and play critical roles in liver IRI (4). In support of this notion, recent studies have demonstrated that TLR4 is a critical mediator of inflammation and organ injury in IRI. The direct evidence of the role of TLR4 in IRI was shown in TLR4 deficient animals, in which reduced necrotic area and diminished inflammatory responses were observed in murine liver IRI models (5–7). Therefore, TLR4 has become a potential therapeutic target for liver IRI, However, there is not yet a clinically accepted therapy that will ameliorate or prevent cellular injury after liver ischemia based on gene-selective targeting.

The induction of RNA interference (RNAi) using small interfering RNA (siRNA) is a highly specific and effective method of gene silencing (8). We have previously demonstrated that silencing of complement and inflammatory genes using siRNA successfully prevented tissue injury in kidney and heart IRI models (9–15). Although RNAi-based therapeutics are exciting and promising, several hurdles exist, to prevent clinical application. One significant concern is nonspecific targeting, which may detrimentally cause global gene silencing in all cell types, as demonstrated by the administration of high doses of siRNA. Currently, viral vectors (16), hydrodynamic injection (17) and cationic liposomes (18) are the main methods for delivering siRNA in vivo. While viral vectors are associated with severe side effects, other methods require large volume and high injection speed, which are not clinically applicable (19). Moreover, these in vivo delivery methods are nontissue/cell specific. The objective of this study was to develop and optimize tissue/cell-specific targeting methods for the delivery of siRNA to hepatocytes and subsequent treatment of liver IRI. In recent years, cationic liposomes have become an attractive delivery system for siRNA clinical application (20) due to its low immunogenecity and ease of modification. However, cationic liposomes target cells in a nonspecific fashion. In order to endow liposomes with hepatocyte specificity, we modified the liposomes with galactose, which binds to the asialoglycoprotein receptor (ASPGR) that is expressed on the surface of hepatocytes (21). We hypothesized that galactose-conjugated liposome nanoparticles (Gal-LipoNP) may selectively deliver siRNA to hepatocytes.

In this study, we report for the first time the use of liver-specific liposome-based siRNA delivery systems using Gal-LipoNP which efficiently silenced TLR4 in liver. Gene silencing of TLR4 resulted in the inhibition of TLR4 expression and reduction of liver IRI. Our results suggest that liver-specific Gal-LipoNP may have clinic potential.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Mice

CD1 mice were purchased from The Jackson Laboratory (Bar Harbor, ME, USA). The mice were maintained under specific pathogen-free conditions. Male mice between 6 and 8 weeks old were used. All experiments were performed in accordance with the Guide for the Care and Use on Animals Committee Guidelines.

TLR4 -siRNA design

Three target sequences of the TLR4 gene were selected. The oligonucleotides containing sequences specific for TLR4 (5′-GCATAGAGGTAGTTCCTAATA-3′, 5′-CACTCTTG ATTGCAGTTTCAA-3′ and 5′-GTTCCATTGCTTGGCGAA -3′) were synthesized and annealed. A pair of annealed DNA oligonucleotides was inserted into a pRNAT-U6.1/Neo siRNA expression vector that had been digested with BamHI and HindIII (Genescript, Piscataway, NJ, USA).

In vitrosilencing of theTLR4 gene on primary hepatocytes

Hepatocytes were isolated from the mice by an in situ collagenase perfusion technique previously described (22). Hepatocytes were then transfected with TLR4-siRNA using Lipofectamine 2000 (Invitrogen, Carlsbad, CA, USA). Scrambled (nonsense) siRNA were used as negative controls. The cells were plated into 24-well plates (1×105 cells per well) and transfected with 1 μg TLR4-siRNA or negative control siRNA. To activate the TLR4 pathway, we added LPS 10 μg/mL to the cells 24 h after transfection and incubated for another 24 h. Accordingly, we performed qPCR and Western blot to detect TLR4 expression.

Measurement of TNF-a and IL-1 stimulate by LPS

Posttransfection of TLR4-siRNA 24 h, hepatocytes were cultured in the presence of LPS(10 μg/mL), and culture supernatants were collected at 4h, 6h, 12h, 24h time points. The TNF-α and IL-1 levels were analyzed using standard commercial ELISA kits.

Preparation of liposomes

DSPE-PEG2000-Galactose lipid (Sigma-Aldrich Canada Ltd., Ontario, Canada) was prepared according to previously published methods (23). Liposomes containing DOTAP and cholesterol in a 1:1 molar ratio were prepared by a thin lipid film method (24). The lipids were dissolved in a glass tube containing chloroform, gently dried under nitrogen and further evaporated to dryness under vacuum. The multilamellar liposomes suspension was vortexed for 5 min and sonicated for 2 min using a bath sonicator above the Tc (50°C) and then subjected to extruded through polycarbonate membranes of pore size of 0.4, 0.2 and 0.1 mm while the temperature of the solution was kept at 50°C. The resulting liposomes were primarily unilamellar vesicles.

TLR4-siRNA solution was added to an equal volume of liposomes and protamine mixture at a ratio of 600 nmol DOTAP:30 μg protamine:50 μg siRNA to form liposome nanoparticles (LipoNP). LipoNPs were generated by incubating the mixture of LipoNP suspension with 10 mol% micelle solution of DSPE-PEG2000, Gal-DSPE-PEG2000, respectively, at 50°C for 10 min.

Particle diameter and zeta potential measured

Nontargeting and targeting LipoNP were freshly prepared and diluted with phosphate-buffered saline (PBS), and the mean particle diameter and surface charge (zeta potential) were measured by Zetasizer Nano Series (Malvern Instruments Ltd., UK), according to the manufacturer's protocols.

In vivodistribution of liposomes/siRNA complex

The Lipo-NP liposomes and Cy3-labeled siRNA complex solution was injected intravenously into mice (50 μg/mice) via tail vein. At 6 h following administration, the organs including heart, liver, spleen, lung and kidney were recovered and rinsed with saline, frozen and mounted for cryostat sectioning. Slides were viewed under a fluorescence microscope.

Mouse warm hepatic IRI model and siRNA administration

A model of approximately 70% hepatic warm IRI was used (25). The mice were placed on a heating pad to maintain body temperature of 37°C during surgery. All structures in the portal triad to the left and median liver lobes were occluded with a microvascular clamp (Roboz Surgical Instrument, Gaithersburg, MD, USA) for 60 min. Blood was collected, and the liver tissue was recovered for analysis 6 h and 24 h after reperfusion. Sham animals underwent the same procedure, but without vascular occlusion.

A total of 50 μg of TLR4 siRNA was encapsulated by targeting and nontargeting LipoNP and diluted in 200 μL of PBS injected into mice at a normal pressure by tail vein. Other controls included the use of scramble siRNA, which were encapsulated by targeting liposome and sham groups.

Assessment of liver function

Blood samples were obtained from the inferior vena cava at 6 h and 24h postischemia. Serum levels of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) were measured by the core laboratory at the London Health Sciences Center to monitor liver function.

Histology and hepatic neutrophil infiltration

At 6 h postreperfusion, livers were dissected from mice and tissue slices were fixed in 10% formalin and processed for histology examination using standard techniques. Formalin tissue was embedded in paraffin, and 5-μm sections were stained with H&E. The histological severity of I/R injury was graded according to Suzuki's classification (26), in which sinusoidal congestion, hepatocyte necrosis and ballooning degeneration are graded from 0 to 4. The absence of necrosis or congestion/centrilobular ballooning is given a score of 0, whereas severe congestion/degeneration and >60% lobular necrosis are given a value of 4. Ten random fields were assessed for necrosis by standard morphologic criteria. These sections were examined in a blinded fashion by a pathologist.

The activity of myeloperoxidase (MPO), an enzyme specific to polymorphonuclear neutrophils, was used as an index of hepatic neutrophil accumulation (27).

Malondialdehyde assessment

Malondialdehyde (MDA) levels were measured using the thiobarbituric acid (TBA) method (28). The amount of lipid peroxides was quantified based on the production of MDA, which in combination with TBA forms a pink chromogen compound, and was measured at 532 nm wavelength.

Measurement of reactive oxygen species (ROS)

ROS production was measured using 2,7-dichlorodihydro-fluorescein diacetate (DCF-DA, Molecular Probes, Invitrogen). The DCF assay provides a global approach to evaluating the production of ROS (29). DCF-DA was dissolved in DMSO to a final concentration of 10 μM before use. Liver tissue samples (200 μg proteins) were incubated with 10 μL of DCF-DA for 30 min at 37°C in a 96-well plate. The fluorescence was measured with multilabel readers (Victor 3, Wallac).

Immunohistochemistry for TLR4, TNF-α and IL-1 deposition in liver tissue

The liver tissue was fixed in 4% paraformaldehyde, and endogenous peroxidase activity was eliminated with 3% H2O2 in methanol. Sections were incubated with primary antibody against TLR4, TNF-α (Abcam, Cambridge, MA, USA) and IL-1 (R&D Systems, Minneapolis, MN, USA). Then the sections were incubated with biotinylation secondary antibody for 45 min, followed by horseradish peroxidase-labeled streptoavidin at 37°C for 45 min. Slices were finally developed with diaminobenzidene (DAB) and counterstained with hematoxylin.

Quantitative PCR for gene expression of TLR4 and proinflammatory cytokines

Total RNA was extracted from kidneys and cells using Trizol Invitrogen, Carlsbad, CA, USA). Total RNA was reverse-transcribed using oligo-(dT) primer and reverse transcriptase (Invitrogen). Primers used for the amplification of murine TLR4, IL-1, TNF-α and GAPDH were as follows: TLR4, 5′-CACTGTTCTTCTCCTGCCTGAC-3′ (forward), 5′-CCTGGGGAAAAACTCTGGATAG-3′ (reverse);

TNF-α, 5′-CTCCCTCCAGAAAAGACACCAT-3′ (forward), 5′-.ATCACCCCGAAGTTCAGTAGACAG-3′ (reverse);

IL-1, 5′-CAGCCTTATTTCGGGAGTCTATTC-3′ (forward), 5′-TATCCCTTTGTTAACCCATCTGTA-3′ (reverse); and GAPDH, 5′-TGATGACATCAAGAAGGTGGTGAA-3′ (forward),

5′ -TGGGATGGAAATTGTGAGGGAGAT-3′ (reverse). The reaction conditions were 10 min at 95°C, 30 s at 95°C, 45 s at 58°C and 45 s at 72°C (40 cycles). Samples were normalized using the housekeeping gene GAPDH, and a comparative CT method was used for the analysis.

Statistical analysis

All results were presented as means ± SD. Statistical comparisons between groups were performed using Student's t-test. Statistical significance was determined as p < 0.05.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Preparation and characterization of Gal-LipoNP

In this study, we chose TLR4 as a targeting gene, given that recent evidence has implicated TLR4 as a central mediator of inflammation and organ injury after IRI (30). TLR4 siRNA was encapsulated within the aqueous interior of LipoNP that were conjugated to galactose, a ligand that binds the ASGPR which are specifically expressed on hepatocytes (21). TLR4 Gal-Liposomes were prepared by mixing neutral lipid cholesterol and DOTAP to form liposomes through the freeze/thawing technique. Subsequently, liposomes were repeatedly extruded through polycarbonate filter membranes with pore sizes of 400, 200, 100 nm to produce a liposome population of uniform size. Next, protamine-condensed TLR4-siRNA was mixed with cationic liposome to form the LipoNP. Finally, a postinsert technique (31) was used to construct Gal-LipoNP by modifying the cationic liposomes complex with galactose lipids.

The diameter and surface charge of the liposomes delivery systems are known to be the major factors influencing the liposomes’ transfection efficiency and biodistribution (32). The physicochemical characteristics of the LipoNP-siRNA complex are summarized in Table 1. The particle size was around 130 nm. The surface charge, of the LipoNP, represented by zeta potential in Table 1, was about 50 mV and manifested a slightly positive charge, which is known to increase transfection efficiency (33). The diameters of targeted LipoNP and nontargeted LipoNP were similar and no difference of surface charge was found between the targeted LipoNP and nontargeted LipoNP.

Table 1.  Particle size and zeta potential of liposomes and its complex
FormulationsParticle size (nm)Potential (mV)
LipoNP-TLR4137.43 ± 1.1050.77 ± 0.49
Gal-LipoNP-TLR4138.03 ± 1.1056.50 ± 1.08

Distribution of Gal-LipoNPin vivo

In order to test whether Gal-LipoNP can specifically target liver, we investigated the biodistribution of siRNA in vivo. The Gal-LipoNP Cy3-labeled siRNA complex was administered intravenously to mice. The distribution of Cy3-siRNA in the liver, spleen, kidney and lung was detected and compared with untreated mice. Figure 1 shows red fluorescence, which represents the distribution of Gal-LipoNP/Cy3-siRNA complex in different organs. The results showed that Cy3-siRNA was abundantly detected in the liver at 6 h. Indicative of selectivity, the fluorescence intensity was higher than that in the mice injected with nontargeted LipoNP/siRNA complex. In mice receiving nontargeted LipoNP, the majority of signal was detected in the spleen, with minor amounts in the liver. These results indicate that Gal-LipoNP can specifically target hepatocytes and efficiently deliver siRNA into the liver.

image

Figure 1. Distribution Gal-LipoNP carried siRNA in vivo. Gal-LipoNP was generated as described in the Materials and Methods section. Cy3-labled siRNA (50 μg) was incorporated in Gal-LipoNP. LipoNP with galactose (Gal-LipoNP) and nonconjugated (LipoNP) were intravenously injected into CD1 mice. Six hours after administration of LipoNP, the mice were killed and heart, liver, spleen, lung and kidney were collected. The Cy3-labled siRNA in organs was detected by fluorescence microscope.

Download figure to PowerPoint

Knock down of TLR4 using Gal-LipoNP

As an indication of principle, we transfected TLR4 siRNA into primary hepatocytes in which TLR4 was detected. The TLR4 was significantly elevated after stimulation with LPS. The results showed that contrary to the scrambled control, the TLR4 siRNA was capable of abolishing LPS-induced upregulation of TLR4 at the transcription (Figure 2A) and protein (Figure 2B) levels. In addition, the secretion of proinflammatory cytokines TNF-a (Figure 2C) and IL-1 (Figure 2D) by hepatocytes was dramatically increased as early as 4 h after stimulation by LPS. These increased cytokines were effectively suppressed after the knockdown of TLR4 gene using siRNA.

image

Figure 2. Gene silencing of TLR4 in vitro and in vivo. (A and B) Gene silencing of TLR4 in vitro. Primary hepatocytes were prepared as described in the Materials and Methods section. Cells were cultured in vitro and transfected with a mixture of three different sequences of TLR4-specific siRNA using lipofectamine 2000. Twenty-four hours after gene silencing, LPS was added to the cells and incubated for another 24 h. The expression of TLR4 from the silenced and nonsilenced cells was determined using quantitative RT-PCR (A) and Western blot (B). (C and D) Inflammatory cytokines secreted by hepatocytes. Hepatocytes were cultured and transfected with TLR4 siRNA, as described above. The level of TNF-α (C) and IL-1 (D) in the culture supernatants was measured by ELISA. (*) Statistical significance when compared with liver cell+LPS and liver cell+LPS +TLR4 siRNA groups was denoted at p < 0.05. (E) Gene silencing of TLR4 in vivo. CD1 mice were administrated with 50 μg of Gal-LipoNP-scramble containing scrambled siRNA, Gal-LipoNP-TLR 4 containing TLR4 siRNA, and nonconjugated LipoNP that incorporated with TLR4 siRNA. Forty-eight hours after gene silencing, liver were clamped for IRI. After 6 h reperfusion, liver tissues were collected and total RNAs were extracted. Transcripts of TLR4 and GAPDH were determined qPCR. Relative quantity of TLR4 mRNA was expressed as mean ± SD. (F) TLR4 deposition in liver tissue detained by IHC staining. Tissues from the above experiments were also used to detect TLR4 using IHC, as described in the Materials and Methods section. (a) CD1 mice were untreated; (b) or treated with 50 μg of Gal-LipoNP that incorporated with scrambled siRNA; (c) nonconjugated LipoNP that incorporated with TLR4 siRNA; (d) Gal-LipoNP that incorporated with TLR4 siRNA. (*) Statisical significance when compared with Gal-LipoNP-scrambled and LipoNP-TLR4 groups was denoted at p < 0.05.

Download figure to PowerPoint

Our and other scholarly studies indicate that 50 μg siRNA can significantly knock down targeted gene expression (10,12,14,34). To test gene silencing efficacy of siRNA delivered by Gal-LipoNP in the liver, we intravenously injected 50 μg Gal-LipoNP TLR4 siRNA into mice followed by induction of IRI. TLR4 gene expression levels were determined by qPCR. As shown in Figure 2E, expression of the TLR4 gene was significantly elevated after IRI compared to sham-treated animals. After treatment with Gal-LipoNP TLR4 siRNA, the expression of TLR4 was knocked down to basal levels. While significant gene silencing was achieved in Gal-LipoNP treated mice, treatment with LipoNP TLR4 siRNA that was not conjugated with galactose caused only marginal suppression of TLR4. In a parallel experiment, the efficacy of gene silencing was confirmed by immunohistochemistry. TLR4-siRNA suppressing the expression of TLR4 at protein level was observed (Figure 2F).

Attenuation of liver IRI using Gal-LipoNP TLR4 siRNA

After validating gene silencing using Gal-LipoNP TLR4 siRNA, we evaluated the therapeutic potential in a model of hepatic injury. Gal-LipoNP was administered prior to IRI, and ALT level was measured. The level of ALT increased significantly 6 h and 24 h postreperfusion compared to sham controls (Figure 3A). Contrastingly, treatment with Gal-LipoNP TLR4 siRNA prior to inducing IRI effectively prevented the increase of ALT. Treatment of TLR4 siRNA with nontargeted LipoNP did not display the therapeutic effects against IRI.

image

Figure 3. Preventing liver IRI by Gal-LipoNP TLR4 siRNA. (A)ALT level after hepatic warm ischemia/reperfusion. CD1 mice were administrated with 50 μg of Gal-LipoNP containing scrambled siRNA, TLR4 siRNA and nonconjugated LipoNP that incorporated with TLR4 siRNA. Forty-eight hours after gene silencing, liver were clamped for IRI. The blood samples were collected for the test of ALT level 6 h and 24 h after reperfusion, respectively (*p<0.05, comparing with scrambled siRNA or LipoNP without Gal). (B) Liver histopathology. CD1 mice were untreated (a, e) or treated with 50 μg of Gal-LipoNP that incorporated with scrambled siRNA (b, f), nonconjugated LipoNP that incorporated with TLR4 siRNA (c, g) and Gal-LipoNP that incorporated with TLR4 siRNA (d, h). Forty-eight hours after gene silencing, livers were clamped for IRI. After 6 h reperfusion, the livers were stained with H&E (a, b, c, d) and MPO (e, f, g, h). (C) MDA levels. The mice were subjected to liver IRI and treated as described in (A).The MDA levels of liver tissue were measured by the TBA method. (D) ROS production. Mice were induced liver IRI and treated as described in (A). The ROS production of liver tissue was measured by DCF assay, as described in the Materials and Methods section. Mean±SD is shown in A, C and D; (*) Statistical significance when compared with Gal-LipoNP-scrambled and LipoNP-TLR4 groups was denoted at p < 0.05.

Download figure to PowerPoint

We then examined the severity of IRI by determining histopathology changes in livers on the basis of the Suzuki histological classification (26). In the livers of scrambled siRNA or LipoNP TLR4 siRNA treated mice, significant hepatocyte edema, severe sinusoidal congestion, cytoplasmic vacuolization and extensive hepatocellular necrosis were displayed (30–60%; Figure 3B-b and 3B-c; score 3.4 ± 0.6 and 3.2 ± 0.5, respectively). In contrast, animals treated with Gal-LipoNP TLR4 siRNA showed mild-to-moderate edema, sinusoidal congestion and cytoplasmic vacuolization (Figure 3B-d; score, 1.2 ± 0.6; p<0.05), whereas livers in the sham group showed no edema or necrosis (Figure 3B-a; score 0; p<0.01). Neutrophilic infiltration, an important feature of IRI-induced inflammation, was significantly higher in IRI liver, as determined by MPO assay (Figure 3B, lower panel). In contrast, mice pretreated with Gal-LipoNP TLR4 siRNA demonstrated significant attenuation of all pathological changes except for minimal hepatocyte ballooning and slightly dilated sinusoid. Remarkably, neutrophil infiltration as well as necrosis was almost entirely eliminated.

It has been demonstrated that TLR4 signaling cascade is critical for oxidative stress induced by ROS production in IRI (35). Accordingly, we determined the MDA content as an index of lipid peroxidation. Hepatic tissue MDA levels were significantly lower in mice treated with Gal-LipoNP TLR4 compared to mice treated with scrambled siRNA or nontargeted LipoNP (p < 0.05, Figure 3C). We also examined the ROS production in liver tissue using DCF assay (29). The ROS production was significantly decreased in the mice treated with Gal-LipoNP TLR compared to those treated with scrambled siRNA and nontargeted LipoNP (p<0.05, Figure 3D). Taken together, treatment with Gal-LipoNP TLR4 siRNA significantly attenuated IRI-induced liver injury.

Inhibition of inflammatory response by Gal-LipoNP TLR4 siRNA

To further investigate the molecular mechanisms of TLR4 activation responsible for the generation of proinflammatory cytokines involved in the recruitment of inflammatory cells, we measured the expression of IL-1 (Figure 4A and C) and TNF-α (Figure 4B and D), detected by qPCR (Figure 4A and B) and immunohistochemistry (Figure 4C and D), in livers after treatment with Gal-LipoNP TLR4 siRNA. IL-1 and TNF-α were found to increase as a result of IRI. In comparison to the sham-operated control, livers undergoing IRI had significantly higher expression of both genes in the scrambled and nontargeted group. Treatment with Gal-LipoNP TLR4 siRNA suppressed these cytokines in IRI livers (Figure 4A–D). These data imply that protective effects of Gal-LipoNP TLR4 siRNA are associated with downregulation of proinflammatory cytokines in IRI livers.

image

Figure 4. Inhibiting inflammatory cytokines by Gal-LipoNP TLR4 siRNA. CD1 mice were administered with 50 μg of Gal-LipoNP containing scrambled siRNA, TLR4 siRNA and nonconjugated LipoNP that incorporated with TLR4 siRNA. Forty-eight hours after gene silencing, the livers were clamped for IRI. After 6 h reperfusion, liver tissues were collected and total RNAs were extracted. The levels of IL-1 (A and C) and TNF-α (B and D) were measured by qPCR (A and B) and immunohistochemical staining (C and D). (a) CD1 mice were untreated; (b) or treated with 50 μg of Gal-LipoNP containing scrambled siRNA; (c) nonconjugated LipoNP that incorporated with TLR4 siRNA; (d) Gal-LipoNP containing TLR4 siRNA. Relative quantity of IL-1 and TNF-α mRNA was expressed as mean ± SD. (*) Statistical significance when compared with Gal-LipoNP-scrambled and LipoNP-TLR4 groups was denoted at p < 0.05.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Accumulating evidence has demonstrated that the TLR-signaling pathway is critical in IRI (30,35), although effective therapy of liver IRI by specific blocking a TLR4 pathway through hepatocyte-specific targeting method has not been reported to our knowledge. In this study, we successfully developed a liver-specific targeting method based on selective affinity of Gal-LipoNP to hepatocyes. We observed that gene silencing using Gal-LipoNP TLR4 siRNA can efficiently inhibit the expression of TLR4 in the liver induced by IRI, which resulted in the prevention of liver IRI.

There are two distinct phases of liver injury after warm IRI (36). The initial phase (<2 h after reperfusion) is characterized by oxidant stress, where production and release of reactive oxygen species (ROS) appear to directly result in hepatocellular injury. The late phase of liver injury occurs from 6 to 48 h after hepatic reperfusion and is an inflammatory disorder mediated by recruited neutrophils. At this point, most parameters of IRI (including ROS products, histology changes, neutrophil infiltration and cytokine release) become evident. In this study, we demonstrated that treatment of TLR4 siRNA at the very least protects liver IRI in the late phase of liver IRI.

Although the promise of RNAi for therapeutic application is exciting and promising, several factors impede its use as drugs in human patients. Currently high concentrations of siRNA are needed to induce therapeutic effects in vivo. While “flooding” the system with siRNA in the form of hydrodynamic administration is effective in animal models, this approach is not practical for clinical intervention (18). The ability to selectively deliver siRNA to specific cells would allow for lower concentrations of siRNA to be used, as well as possibly alleviating need for systemic delivery.

In order to achieve hepatocyte-targeted delivery of TLR4 siRNA, we conjugated galactose to the LipoNP, which specifically binds to the ASGPR that is expressed on the surface of liver cells with high affinity at a rapid internalization rate (21). ASGPR naturally binds and internalizes terminal galactose-bearing asialoglycoproteins. Lipids conjugated with galactose have been used to deliver the gene to liver cells (37,38). Therefore, Gal-LipoNP was chosen as a potentially viable candidate. Our data (Figure 1) demonstrated that Gal-LipoNP is able to specifically deliver siRNA into the liver tissue. Studies have shown that while a high dose of galactose (700 mg/kg) might cause liver injury, a low dose would not trigger liver inflammation (39). In our experiment, the dose of galactose per mouse is 100 nmol (2.713 × 10−3 mg), almost 6500 times lower than the reported toxic dose. No evident inflammation was observed in this study.

In addition to ability of the galactose-conjugated LipoNP to bind hepatocytes, this complex appears to be capable of delivering siRNA to hepatocytes and subsequently releasing it from endosomes into cytoplasm after internalization. In support of this notion, the analysis of biodistribution in our study showed that the Gal-LipoNP was capable of selectively delivering the fluorescence-labeled siRNA into the liver compared with nontarget LipoNP. These results suggest that the Gal-LipoNP enhanced hepatic cellular uptake of siRNA by directing the whole siRNA-nanopaticles exclusively to the liver and markedly reduced the distribution of siRNA to other organs.

TLR4 also acts as a receptor for several DAMPS (40) generated by injured tissue/cells such as HMGB1 (35), low molecular weight hyaluronic acid degradation products (41) and heat shock protein-70 (42,43). Thus IRI may well fit within a theoretical model in which ATP depletion and release of oxygen-free radicals due to oxygen deprivation causes an initial phase of cell death. Molecules released from these dead/injured cells act as endogenous TLR4 ligands, which subsequently activate the innate immune system, evoking inflammation and causing the IRI in liver. Although the role of TLR4 on IRI has been extensively studied, effective therapy through knocking down this molecule has not been developed for preventing liver IRI.

It has been reported that hepatocytes do express TLR4 (44–46) and are capable of responding to extremely low (nanogram) concentrations of LPS (44). Hepatocytes were initially viewed as by stander cells from an immune standpoint (47). Recent data, however, revealed that hepatocytes also play a critical role in inflammation (48). Hepatocytes are responsible for bacterial products and LPS-mediated desensitization (44), suggesting that hepatocytes are capable of responding directly to microbial products as part of an innate immune response. Accordingly, we hypothesize that endogenous ligands released from necrotic or injured hepatocytes are directly recognized by TLR4 during liver IRI, thereby initiating the TLR4 signaling cascade that causes inflammation and organ damage.

So far, TLR4 has been identified in rodents and transfers signals intracellularly via two pathways: toll-interleukin 1 receptor domain containing adaptor protein (TIRAP) bridges MyD88 for MyD88-dependent signaling and TRAM bridges TRIF for MyD88-independent signaling (49,50). While the MyD88-dependent pathway initiates the induction of gene transcription and production of proinflammatory cytokines and chemokines, such as IL-1, IL-6 and TNF-α (51–53), the MyD88-independent pathway results in the stimulation of interferon (IFN)-inducible genes and the production of IFN-β (51). Zhai et al. (54) demonstrated that TLR4 is actively involved in initiating the liver IR injury cascade leading to local induction of proinflammatory cytokines TNF-a in a murine live warm ischemia reperfusion model. In this study, the proinflammatory cytokine mRNA levels were markedly upregulated 6 h after reperfusion while Gal-LipoNP TLR4-siRNA treatment specifically significant decreased expression of IL-1 and TNF-α gene. Levels of these cytokines have been reported to correlate with levels of TLR4 activation and severity of liver-specific injury in several studies (55). Our results are congruous with these findings, demonstrating that gene silencing of TLR4 significantly suppressed inflammatory cytokines in IRI livers.

It has been demonstrated that TLR4 expression in the donor organ also plays a key role in the mechanism of hepatic IRI after liver transplantation, as local targeting of innate TLR4 signaling in the graft could be an efficient strategy to control the IRI-related processes (56). Therefore, post-transplant treatment with Gal-LipoNP TLR4-siRNA, which induces local gene silencing in liver, may be beneficial in protecting graft function in liver transplantation.

In summary, we report here a new system to deliver siRNA specifically to the liver using galactose-ligated LipoNP. Gal-LipoNP carried TLR4-siRNA can efficiently knock down TLR4 gene synthesis in liver. Treatment of Gal-LipoNP carried TLR4-siRNA attenuated liver IRI, protected liver function, decreased neutriphil infiltration and suppressed inflammatory cytokines in liver. Thus, Gal-LipoNP carried TLR4-siRNA could be a promising approach for preventing liver IRI.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

This study was supported by the Heart and Stroke Foundation of Canada. D.C. was supported by the CIHR Strategic Training Program in Cancer Research. W.-P.M. is an awardee of Department of Surgery Institute Scientist, University of Western Ontario. Dr. A.M.J. is supported by a Pfizer-CIHR Chair in Clinical Transplantation. N.J. was supported by the grants from the Major State Basic Research Development Program (973 Program) of China (No. 2009CB522404), National 11th Five-Year Science and Technology Plan major projects of China (No. 2008ZX10002-026). We thank Weihua Liu for her technical assistance for histopathology experiments.

Disclosure

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References

Disclosure: Commercial Organizations.

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

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

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgments
  8. Disclosure
  9. References