NDT Advance Access originally published online on July 5, 2006
Nephrology Dialysis Transplantation 2006 21(10):2745-2753; doi:10.1093/ndt/gfl327
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Targeting of interstitial cells using a simple gene-transfer strategy
1Department of Internal Medicine and Therapeutics and 2Department of Advanced Technology for Transplantation, Osaka University Graduate School of Medicine, Suita 565-0871, Japan
Correspondence and offprint requests to: Yoshitaka Isaka, MD, PhD, Department of Advanced Technology for Transplantation, Osaka University Graduate School of Medicine, Suita 565-0871, Japan. Email: isaka{at}att.med.osaka-u.ac.jp
 |
Abstract |
|---|
 Top Abstract IntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Background. Interstitial fibroblasts are central to the inflammatory response during the progression of tubulointerstitial fibrosis. We examined the efficiency of a new gene transfer method that targets interstitial cells by using parenchymal injection of DNA followed by electroporation.
Methods. Fluoresceinisothiocyanate-labelled oligodeoxynucleotides (FITC-ODNs) or expression vectors were directly injected into the cortex of the kidney, followed by electroporation.
Results. Transfection with FITC-ODNs or the EGFP expression vector resulted in efficient transfection in interstitial fibroblasts, but not in tubular epithelial cells or glomerular cells. Transfection efficiency was optimal after using a total of 150 µg of DNA in 1000 µl of PBS, combined with clamping of the renal vessels prior to electroporation. Gene expression peaked at 4 days after transfection and decreased by two orders of magnitude at 6 weeks post-transfection; however, expression recovered to near peak levels after parenchymal or intraperitoneal injection of FR901228, a histone deacetylase inhibitor.
Conclusion. We demonstrated that direct parenchymal injection of DNA combined with electroporation enables gene transfer into interstitial fibroblasts.
Keywords: electroporation; gene transfer; interstitial fibroblast; parenchymal
|
Introduction |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Tubulointerstitial inflammation and fibrosis are commonly associated with most human progressive renal diseases, and their severity can determine the impairment of renal function and predict long-term prognosis [1]. Interstitial fibrosis is characterized by accumulation of extracellular matrix proteins in the renal tubulointerstitial compartment. Interstitial fibroblasts play a crucial role in disease progression through proliferation, promotion of local fibrogenic processes and by contributing to synthesis of extracellular matrix. Therefore, gene transfer techniques that specifically target interstitial fibroblasts will be helpful for exploring the mechanism of interstitial fibrosis and for developing therapeutic applications.
Two strategies have been reported for the transfection of DNA into interstitial fibroblasts: (i) a hydrodynamics-based delivery technique in which DNA is injected via the renal vein [2] and (2) retrograde DNA injection via the ureter, followed by application of electric fields [3]. We have previously shown that retrograde infusion of DNA via the ureter allows DNA to enter into the interstitial area by passing between papilla epithelial cells, and this is followed by diffuse DNA distribution into the cortical interstitial spaces [4]. Increased pressure may enhance the transfection efficiency of intravenous- or ureteral-injected plasmid DNA. The elevated pressure is thought to contribute to the delivery and distribution of the injected DNA to the cells [5]. Electric fields increase gene transfer not only by membrane permeabilization but also by direct effects to promote migration and cellular uptake of the attached DNA [3].
The walls of the peritubular capillaries (PTC) consist of extremely thin fenestrated endothelium [6]. We speculate that DNA enters the interstitial space by diffusing through the PTC network, which is permeable to small molecules [6]. It was previously reported that intrarenal and parenchymal injection of DNA–lipofection complex did not produce efficient transduction [7]. In the present study, a single parenchymal DNA injection with enhanced pressure was used to transduce remote interstitial fibroblasts through diffusion to provide a novel clinical approach for human gene therapy.
|
Material and Methods |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Parenchymal injection of India Ink
To examine whole kidney distribution of a solution injected directly into rat renal parenchyma, we used 700 µl of buffered India ink as a marker. Six-week old male Sprague–Dawley rats weighing 150 g (purchased from Japan SLC, Japan) were anesthetized by intraperitoneal administration of pentobarbital (50 mg/kg) and were handled in accordance with the guidelines of the Animal Committee of Osaka University. A midline incision was made to expose the left kidney. The abdominal aorta at the proximal site of the main branch of the celiac artery and renal vessels was clamped with two clips, which were released at the end of injection. A 10% India ink solution (700 µl in PBS) was carefully injected through a 30-gauge needle (Terumo Corporation, Tokyo, Japan) into the lower part of the left kidney. The tip of the needle was manually placed to reach the corticomedullary area. Following injection, the left kidney was perfused with 4% paraformaldehyde from the abdominal aorta and was harvested. For microscopic analysis, kidney sections were stained with hematoxylin-eosin after sufficient decolorization with 100% ethanol.
Parenchymal transfection
Parenchymal injection of 30 µg fluoresceinisothiocynate- labelled oligodeoxynucleotides (FITC-ODNs) in 700 µl PBS was followed by gene transfer through electroporation: tweezer-type electrodes were used to apply six consecutive 100 ms square electric pulses, each of which was 75 V, at 900 ms intervals [4]. The latter three pulses were administered with the counter current. After a 10 min interval, the kidney was perfused with PBS and harvested, and a rapid-frozen specimen was examined by fluorescence microscopy.
We also applied 200 µg of enhanced green fluorescent protein (EGFP) expression vector (pCAGGS-EGFP) in 700 µl PBS, driven by a cytomegalovirus enhancer and a chicken ß-actin promoter by using the same procedure [4]. Transgene expression was verified 4 days later.
To determine the effect of parenchymal transfection with or without electroporation, histological analysis was performed using Masson's trichrome staining.
Comparative luciferase assay
Optimal conditions for gene transfer by parenchymal injection were investigated by comparative analyses of several variables (delivery methods, interventions, injection volumes and DNA amounts) using a luciferase expression vector, pCAGGS-luciferase, as a reporter. Each variable group was comprised of more than three subjects. At 3 days after transfection, luciferase activity was examined using the Pica-gene luciferase assay kit (Toyo Ink MFG Co., Ltd., Tokyo, Japan). Luciferase activities in the transfected kidneys were normalized for protein concentration. All values are expressed as means ± SD. Statistical significance (defined as P < 0.01) was evaluated by one-way analysis of variance (ANOVA).
In an attempt to recover transgene expression, we administered FR901228 (0.3 mg/kg), a histone deacetylase inhibitor (kindly provided from Fujisawa Pharmaceutical Co. Ltd, Osaka, Japan), by parenchymal and intraperitoneal injection given at 6 weeks after parenchymal transfection. FR901228 was dissolved in dimethylsulfoxide and then diluted with PBS to its final concentration just before use.
Immunofluorescence
To identify the cells exposed to FITC-labeled ODN or pCAGGS-EGFP, transfected kidneys were stained with 4',6-diamidino-2-phenylindole, dihydrochloride (DAPI) and antibodies against rat laminin (a marker for tubular basement membrane; Monosan, Amuden, Netherlands) [3], rat RECA-1 (a marker for endothelial cells; Cosmo Bio, Tokyo, Japan), rat CD45 (a marker for leukocyte; Serotec, Ltd., Oxford, England) and ER-TR7 (a marker for fibroblasts; Biogenesis, New Fields, UK) [8].
|
Results |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Parenchymal injection with India Ink
To examine the distribution of a solution injected directly into rat renal parenchyma, we clamped the abdominal aorta proximal to the main branch of the celiac artery and renal vessels with two clips, and injected India ink into the lower portion of the left kidney. With the clamps in place, the injected ink spread throughout the whole kidney, whereas without the clamps, the ink spread via the renal vein and exhibited a limited and, sometimes, focal distribution. Macroscopic images demonstrated that the injected India ink spread homogeneously throughout the kidney (Figure 1A). Microscopic images indicated that the distribution of India ink was limited to the interstitial area but was not present in the tubular lumen or glomeruli (Figure 1B–D), suggesting that the India ink infiltrated along the PTC.
|
Parenchymal transfection
Parenchymal injection of FITC-ODNs was followed by electroporation. FITC-ODNs were observed diffusely in the kidney (Figure 2A). Immunofluorescent analysis using both DAPI counter-stain and anti-laminin antibody showed that FITC-ODNs were localized in the nuclei of interstitial cells (Figure 2B). To identify the transfected cells, antibodies for CD45 (Figure 2C) and RECA-1 (Figure 2D) were used. Transfected interstitial cells (sky-blue, arrowhead) did not coincide with leukocytes (purple, arrow in Figure 2C) or endothelial cells (purple, arrow in Figure 2D), indicating that FITC-ODNs were introduced into interstitial fibroblasts.
|
We then transferred the EGFP expression vector (pCAGGS-EGFP) [4], and verified transgene expression 4 days later. In agreement with the results described in the preceding text, EGFP-positive cells were seen outside of the tubular basement membrane, suggesting that the EGFP-gene had been introduced into interstitial cells (Figure 3A and B). The shape of EGFP-positive cells was suggestive of fibroblasts. To identify the transfected cells, EGFP-positive cells were stained with ER-TR7 antibody, a marker for fibroblasts. EGFP-positive cells (Figure 3B) were merged with ER-TR-7-positive cells (Figure 3C), suggesting that the transfected cells were interstitial fibroblasts (Figure 3D). Thus, parenchymal injection of ODNs or naked plasmid DNA with electroporation may provide a practical gene-transfer technique that specifically targets interstitial fibroblasts.
|
Comparative luciferase assay
To evaluate the optimal conditions for gene transfer by parenchymal injection, we performed comparative analyses between several variables (delivery methods, interventions, injection volumes and DNA amounts) using the luciferase expression vector, pCAGGS-luciferase, as a reporter.
To compare delivery methods, pCAGGS-luciferase was injected via three different routes (parenchymal, ureteral [3], and renal vein injection [2]) while the renal vessels were clamped, followed by electroporation. Both parenchymal and ureteral injection resulted in higher luciferase activities than renal vein injection. There were no significant differences between parenchymal injection and ureteral injection (Figure 4A), which indicated a good efficiency of the parenchymal transfection method.
|
We then examined the effect of intervention (clamping of the renal vessels and/or electroporation) on transfection efficiency. Each intervention (clamping of renal vessels or electroporation) increased luciferase activity, and this effect was enhanced by applying both interventions (Figure 4B).
We assessed the optimal volume of parenchymal transfection using an injection volume gradient with a fixed amount of DNA. Parenchymal injection was performed by clamping the renal vessels combined with electroporation. Although we found a significant increase in luciferase activity between 700 µl and 1000 µl, higher volumes did not increase activity. We, therefore, decided to use 1000 µl as the standard injection volume (Figure 4C).
To estimate the optimal amount of DNA, we prepared a DNA gradient (5, 20, 50, 150 and 400 µg) while using a standard injection volume (1000 µl). Luciferase activity increased with the amount of DNA up to 150 µg. At 400 µg of DNA, however, there was no further significant increase in activity (Figure 4D).
Finally, we confirmed diffuse transgene expression throughout the kidney following parenchymal transfection of the lower portion of the kidney. Transfected kidneys were dissected into four parts (upper, middle and lower cortex and medulla) and luciferase activity was assessed in each part. We observed no significant differences in luciferase activity, which confirmed that a single parenchymal injection enables diffuse gene transfer to interstitial fibroblasts throughout the kidney (Figure 4E).
Time course of luciferase expression and FR901228 administration
To analyse transgene expression over time, we examined cortical luciferase activity for several days following parenchymal transfection with pCAGGS-luciferase. We used the most effective transfection conditions, which included 150 µg of DNA in a 1000 µl volume with clamping of the renal vessels followed by electroporation. Cortical luciferase activity, calculated from the average of upper, middle and lower cortex values, peaked at 4 days after transfection, and gradually decreased by two orders of magnitude within 6 weeks.
In an attempt to recover transgene expression, we administered FR901228, a histone deacetylase inhibitor, by parenchymal and intraperitoneal injection at 6 weeks after parenchymal transfection. Both the parenchymal and intraperitoneal injections of FR901228 resulted in recovery of luciferase activity to levels observed during initial transgene expression (Figure 5).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
Using a simple gene-transfer strategy, we demonstrated that interstitial cells can be targeted by a single parenchymal DNA injection in combination with electroporation. Direct injection of DNA into parenchyma provides several advantages for kidney-targeted gene transfer. For example, it provides a simple procedure with high transfection efficiency and reduced surgical invasiveness. Transfection efficiency was found to be similar to retrograde ureteral transfection [3], and was significantly better than that of retrograde renal vein injection [2]. When compared with other gene transfer methods using viral vectors or the HVJ-liposome method, direct injection of naked plasmid DNA allows easy and inexpensive sample preparation as well as increased safety by eliminating the risk of insertional mutagenesis, which has been associated with the use of viral vectors. In the present study, we performed midline incisions for parenchymal injection; however, a dorsolateral approach would further decrease the surgical invasiveness of this method.
We additionally demonstrated that parenchymal injection of naked plasmid DNA with electroporation targets interstitial fibroblasts. Although it is not clear how parenchymal transfection combined with electroporation resulted in specific transduction, we [3] and others [2] have previously reported similar findings. The expression of plasmid DNA requires cellular entry and transport into the nucleus. For the initial step, Budker et al. [5] hypothesized that plasmid DNA can be taken up by cells in vivo through a receptor-mediated process. For the latter step, there is evidence that plasmid DNA can transverse intact nuclear pores [9]. The number of cell membrane receptors is thought to partly determine differences in cellular transduction efficacy among renal cells, and receptors for plasmid DNA may be abundant in fibroblasts but scarce in endothelial or epithelial cells. Furthermore, the tubular basement membrane may preclude the introduction of plasmid DNA into tubular epithelial cells. It is our contention that the abundant attachment of DNA to the cell membrane explains specific transduction in interstitial fibroblasts. However, the precise mechanism of selective introduction into these cells is uncertain and remains to be elucidated.
Direct parenchymal injection resulted in global interstitial transduction. In a previous study, intrarenal and parenchymal injections of DNA-lipofectin complex did not cause transduction in mouse kidneys [7]. We speculate that differences in intrinsic hydrostatic renal pressure between the previous report and our findings may have affected transduction efficacy. We observed kidney expansion during the injection that included clamped renal vessels. The stress of a slight increase in intrinsic renal pressure may induce global distribution into interstitial spaces via the peritubular capillary network. Although India ink was observed in the interstitial area, interstitial cells did not show India ink staining. However, injection of India ink simply reflects how a solution distributes throughout the kidney, whereas electroporation helps in the entry of DNA across the plasma membrane.
Here, we employed parenchymal injection as a simple transfer method. The major factor in the clinical acceptability of electroporation-mediated gene delivery is its effect on the target tissue. Electroporation may damage the target tissue, and this depends on the electrical parameters. To be clinically acceptable for use in gene delivery, there should be no permanent damage to the kidney. The safety of electroporation has been circumstantially examined [10]. For example, electroporation of skin with 100 V pulses did not affect viability, did not cause erythema or oedema, and did not alter the scopic structure, which was comparable with control skin. Increased detachment was observed in the stratum corneum layers with increasing electroporation voltage (100–300 V). Microscopy revealed degeneration of the basal layer and breakdown of collagen fibers at 24 and 48 h following electroporation. However, skin viability was not affected at 0 and 24 h following electroporation. Degeneration of the basal layer and collagen fibre breakdown may not lead to cell death; therefore, mitochondrial dehydrogenase activity was not affected. In fact, microscopy studies showed that the affected basal layer and collagen fibres recovered to normal within 7 days.
Electroporation may influence the temperature of the targeted organ, and thereby induce the up-regulation of several genes, such as that of heat-shock proteins. It is known that factors such as temperature, pH and ionic strength modify the physical chemistry as well as permeability of the diffusing molecule. Although studies of temperature effects on electrical conductance and transport of macromolecules after electroporation showed that electroporation at mildly hyperthermic temperatures resulted in delivery of much higher quantities of macromolecules [11], temperature was maintained during the current experiment. In addition, electroporation-mediated gene-transfer studies showed no induction of heat-shock proteins (HSP27, HSP40, HSP 47 and HSP 90) after transfer [7].
Parenchymal insertion is currently employed clinically for renal biopsies, indicating that our gene transfer technique may be adapted to humans. Instead of collecting kidney specimens, therapeutic genes can be delivered directly into the kidney in a non-invasive manner. Although the clinical safety of this method requires verification, we observed no histological damage in the glomeruli or the tubular epithelial cells and found no cellular infiltration, except for the scar at the injection site (Figure 6A). In addition, there were no differences in peak serum lactate dehydrogenase levels between normal and parenchymal-injected rats (224.3 ± 78.5 and 235.3 ± 93.5) at 12 h after parenchymal transfection, suggesting a lack of major toxicity following electroporation-mediated parenchymal transfer. To determine the effects of parenchymal transfection without (Figure 6D and F) or with (Figure 6B, C, E and G) electroporation, histological analysis was performed using Masson's trichrome staining at 8 days after transfection. The tubular cells attached to the electrode (Figure 6B) were moderately damaged, but there were no histological changes in the area apart from the cells in direct contact with electrodes (Figure 6C). The light microscopic structure of kidney specimens treated with parenchymal transfection with/without electroporation was comparable with control specimens at 8 days after transfection. We did not observe an increase in infiltrating cells in the electroporation-treated kidneys. In addition, electoroporation induced no harmful effects on glomeruli, while gene-transferred animals might show a slight tubular dilation when compared with untreated controls. These observations suggest that inflammatory responses resulting from parenchymal transfection are unlikely to cause long-term histological changes. The inner medulla is a possible niche for renal stem cells because it is located in the tubulointerstitium/inner papillae [12]. It is difficult to determine how these cells are affected by high-pressure injection of plasmids and electroporation, because good markers for stem cells are not available. It is clear that a less invasive technique is desirable. Although we did not observe histological damage or immunological reactions, except at the injection site and electrode-attached surface, parenchymal injection or electroporation may cause histological or immunological effects in large animals or humans. Further studies are needed to test the usefulness and safety of these techniques for application of in vivo electroporation in humans.
|
In this article, parenchymal or intraperitoneal injection of FR901228 resulted in recovery of luciferase activity to near initial levels of transgene expression; cortical luciferase activity had gradually decreased by two orders of magnitude after 6 weeks. Direct parenchymal injection of DNA combined with electroporation is a simple and powerful tool for targeting interstitial cells; however, the decrease in gene expression is a problem that would restrict the utility of this technique in clinical medicine. Recent reports examing chromatin remodelling revealed that histone acetylase and deacetylase regulate the level of transcription by modifying chromatin [13]. It has been shown that silencing of transgene expression occurs, even though transgenes are retained in the cells. Various studies have demonstrated that a decrease in transcription rather than elimination of the transfected genes is responsible for this effect [14]. Although the molecular mechanism remains unknown, the histone deacetylase inhibitor FR901228, has been shown to specifically enhance exogenous transgene expression in vivo [15]. Our data suggest that the decrease in transgene expression is associated with histone deacetylation, which was recovered by treatment with the histone deacetylase inhibitor FR901228. These findings strongly suggest that the intracellular balance between histone acetylation and deacetylation, which is normally important for cellular gene expression, is also relevant for parenchymal transfection. In addition, the inhibition of histone deacetylase and consequent transgene acetylation are both responsible for increased transduced gene expression. However, in light of recent findings showing that the terminal half-life of FR901228 from a phase I trial in adults was 486 min and that the clearance was
200 ml/min/m2 [16] and that activation of transgene expression may be transient, which could be followed by a gradual decrease. Therefore, repeated injection of FR901228 may be necessary for long-term transgene expression. In conclusion, we demonstrated that direct parenchymal injection of DNA combined with electroporation enables gene transfer into interstitial cells. This new strategy represents a simple and powerful tool for gene transfer that targets interstitial cells, and may provide a promising therapeutic tool for gene therapy. However, we were not able to establish the safety of this electroporation technique, and further experiments will be necessary to address the safety of electroporation in large animals prior to use in clinical applications. Finally, FR901228 may furnish a useful tool for overcoming the current limitation of gene therapy using plasmid DNA.
Conflict of interest statement. None declared.
|
References |
|---|
|
TopAbstractIntroductionMaterial and MethodsResultsDiscussionReferences |
|---|
- Risdon R, Sloper J, de Vardener H. (1968) Relationship between renal function and histologic changes founf in renal-biopsy spcimens from patients with persistent glomerulonephritis. Lancet 2:363–366.[ISI][Medline]
- Maruyama H, Higuchi N, Nishikawa Y, et al. (2002) Kidney-targeted naked DNA transfer by retrograde renal vein injection in rats. Hum Gene Ther 13:455–468.[CrossRef][ISI][Medline]
- Nakamura H, Isaka Y, Tsujie M, et al. (2002) Introduction of DNA enzyme for Egr-1 into tubulointerstitial fibroblasts by electroporation reduced interstitial alpha-smooth muscle actin expression and fibrosis in unilateral ureteral obstruction (UUO) rats. Gene Ther 9:495–502.[CrossRef][ISI][Medline]
- Tsujie M, Isaka Y, Ando Y, et al. (2000) Gene transfer targeting interstitial fibroblasts by the artificial viral envelope type hemagglutinating virus of Japan liposome method. Kidney Int 57:1973–1980.[CrossRef][ISI][Medline]
- Budker V, Budker T, Zhang G, et al. (2000) Hypothesis: naked plasmid DNA is taken up by cells in vivo by a receptor-mediated process. J Gene Med 2:76–88.[CrossRef][ISI][Medline]
- Lemley K and Kriz W. (1994) Structure and function of the renal vasculature with clinical and functional correlations. In Tisher C and Brenner B (Eds.). Renal Pathology(J.B. Lippincott, Philadelphia) pp. 981–1026.
- Grossin L, Cournil-Henrionnet C, Pinzano A, et al. (2006) Gene transfer with HSP 70 in rat chondrocytes confers cytoprotection in vitro and during experimental osteoarthritis. Faseb J 20:65–75.
[Abstract/Free Full Text] - Maxwell PH, Ferguson DJ, Nicholls LG, et al. (1997) The interstitial response to renal injury: fibroblast-like cells show phenotypic changes and have reduced potential for erythropoietin gene expression. Kidney Int 52:715–724.[ISI][Medline]
- Hagstrom JE, Ludtke JJ, Bassik MC, et al. (1997) Nuclear import of DNA in digitonin-permeabilized cells. J Cell Sci 110 [Pt 18]:2323–2331.
- Medi BM and Singh J. (2005) Skin targeted DNA vaccine delivery using electroporation in rabbits II. Safety. Int J Pharm 294:53–63.[CrossRef][ISI][Medline]
- Murthy SN, Sen A, Zhao YL, et al. (2004) Temperature influences the postelectroporation permeability state of the skin. J Pharm Sci 93:908–915.[CrossRef][ISI][Medline]
- Oliver JA, Maarouf O, Cheema FH, et al. (2004) The renal papilla is a niche for adult kidney stem cells. J Clin Invest 114:795–804.[CrossRef][ISI][Medline]
- Grunstein M. (1997) Histone acetylation in chromatin structure and transcription. Nature 389:349–352.[CrossRef][Medline]
- Palmer TD, Rosman GJ, Osborne WR, et al. (1991) Genetically modified skin fibroblasts persist long after transplantation but gradually inactivate introduced genes. Proc Natl Acad Sci USA 88:1330–1334.
[Abstract/Free Full Text] - Yamamoto S, Yamano T, Tanaka M, et al. (2003) A novel combination of suicide gene therapy and histone deacetylase inhibitor for treatment of malignant melanoma. Cancer Gene Ther 10:179–186.[CrossRef][ISI][Medline]
- Sandor V, Bakke S, Robey RW, et al. (2002) Phase I trial of the histone deacetylase inhibitor, depsipeptide (FR901228, NSC 630176), in patients with refractory neoplasms. Clin Cancer Res 8:718–728.
[Abstract/Free Full Text]
Accepted in revised form: 9. 5.06
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||