Background

We developed an electroporation-mediated gene transfer method targeting glomerular mesangial cells. Injecting DNA solution via renal artery followed by electric pulses using tweezers-type electrodes could result in efficient transfection in mesangial cells. Therefore, this gene transfer system opened a feasible strategy to manipulate the function of several cytokines and growth factors in mesangial cells. Recently, a new generation of catalytic nucleic acid composed of DNA, named DNA enzyme (DNAzyme), has been developed.

Method

We generated a DNAzyme (TGFDE) targeting transforming growth factor-1 (TGF-1), and examined the therapeutic effect of TGFDE in vitro and in vivo.

Results

In cultured rat mesangial cells, treatment with TGFDE blocked TGF-1 mRNA expression, and thereby suppressed type I collagen mRNA expression. Next, we introduced TGFDE or scrambled DNAzyme (TGFSCR) into anti-Thy-1 model of nephritic rats by electroporation 3 days after disease induction. Northern blot analysis and immunohistochemical staining demonstrated that glomerular message and protein expression of TGF-1, -smooth muscle actin (-SMA), and type I collagen were suppressed in TGFDE-transfected nephritic rats compared with untreated nephritic rats and TGFSCR-transfected rats on day 7. Consequently, we observed significant reduction in glomerular matrix score in TGFDE-transfected nephritic rats.

Conclusion

Inhibition of TGF-1 expression by electroporation-mediated DNAzyme transfer might be useful for the therapy of glomerulonephritis.

METHODS

Design of DNAzyme

Sequences of DNAzyme (TGFDE) for rat TGF-1 used in the present study were 5'-CGAGGGCGGCAGGCTAGCTACAACGAGGGGGAGGCT-3' (3' thymidine inverted is italic, catalytic domain is underlined) (Bex, Tokyo, Japan). TGFDE has nine and eleven nucleotide arms flanking the 15 nucleotide catalytic domain (underlined), which was designed to target the translational site adenine-uracil-guanine (AUG) in rat TGF-1 mRNA. For resistance to 3'-to-5' exonuclease digestion, the 3' terminus of the molecule was capped with an inverted 3'-3'-linked thymidine (italics). The nucleotide sequence in each arm of TGFDE was scrambled without altering the catalytic domain (TGFSCR). TGFSCR sequences were 5'-GGGGGAGGCGGCTAGCTACAACGACGAGGGCGGCAT-3' (3' thymidine inverted is italic, catalytic domain is underlined).

Effects of TGFDE in cultured mesangial cells

To examine the effects of TGFDE on TGF-1 and type I collagen expression on mesangial cells, subconfluent rat mesangial cells⁵ were growth arrested with 0.4% fetal calf serum (FCS). Quiescent subconfluent cells were treated with adding TGFDE or TGFSCR (1 and 5 mol/L) for 24 hours. Treated cells were then stimulated with 20% FCS for 8 hours and TGF-1 and type I collagen mRNA expression was determined by Northern blot analysis. TGFDE blocked FCS-induced TGF-1 and type I collagen mRNA expression on rat mesangial cells, while TGFSCR had no effect Figure 1.

Figure 1.

The effect of DNAzyme in cultured rat mesangial cells. Northern blot analysis demonstrated DNAzyme, TGFDE, blocked fetal calf serum (FCS)-induced transforming growth factor-1 (TGF-1) and type I collagen expression, while scrambled oligonucleoside, TGFSCR, had no effect. GAPDH is glyceraldehyde-3-phosphate dehydrogenase.

Full figure and legend (111K)

Experimental design in anti-Thy-1 nephritis

To determine the therapeutic effect of TGFDE on mesangial proliferative glomerulonephritis, we transferred TGFDE into nephritic rats by electroporation in vivo². All procedures were handled in a humane fashion in accordance with the guidelines of the Animal Committee of Osaka University. Six-week-old male Sprague-Dawley rats were anesthetized by intraperitoneal injection of pentobarbital (50 mg/kg) and anti-Thy-1 model of glomerulonephritis was induced by an intravenous injection of anti-Thy-1 monoclonal antibody, 1-22-3 (5 mg/kg)⁶. On day 3, TGFDE or TGFSCR (200 g) was transferred into nephritic rats (four rats in each group).

On day 7, treated left kidneys and untreated contralateral right kidneys were perfused with cold autoclaved phosphate-buffered saline (PBS), and samples of tissues for light microscopy were fixed with 4% paraformaldehyde overnight and dehydrated through a graded ethanol series and embedded in paraffin. Histologic sections (2 m) of the kidneys were stained with periodic acid-Schiff (PAS) reagent. Tissues for -smooth muscle actin (-SMA) immunostaining were fixed in methyl Carnoy's solution. For glomerular RNA preparations, glomeruli were isolated from the pooled remaining renal tissue by a standard sieving method. The experiments were repeated three times.

RESULTS

Effect of TGFDE on TGF--SMA and collagen I expression in vivo

To examine the effects of TGFDE in vivo, we transferred TGFDE or TGFSCR into nephritic rats on day 3, and examined glomerular mRNA levels on day 7. Northern blot analysis demonstrated that TGF-1 mRNA levels were markedly increased in untreated and TGFSCR-treated nephritic kidneys. In contrast, TGFDE transfection reduced the levels of TGF-1 mRNA. In addition, glomerular mRNA levels of -SMA and type I collagen were increased in untreated and TGFSCR-treated kidneys, while TGFDE transfection inhibited the -SMA and type I collagen expression Figure 2. Laser densitometric analysis revealed that TGF-1 and type I collagen mRNA in glomeruli from nephritic rats treated with TGFDE was reduced to 47% and 34% of those in untreated disease control, respectively.

Figure 2.

Inhibition of transforming growth factor-1 (TGF-1), -smooth muscle actin (-SMA), and type I collagen mRNA in nephritic glomeruli by TGFDE. RNA was extracted from isolated glomeruli of normal kidney or untreated, TGFDE-treated and TGFSCR-treated nephritic kidney. Northern blot was probed for TGF-1, SMA, type I collagen, or glyceraldehyde-3-phosphate dehydrogenase (GAPDH).

Full figure and legend (133K)

TGF-1 Figure 3 and SMA Figure 4 protein expression was studied by immunohistochemistry using antihuman TGF-1 polyclonal antibody (Santa Cruz Biotechnology, Santa Cruz, CA, USA) and anti--SMA monoclonal antibody (Immunotech, Marseilles, France), respectively. Expression of TGF-1 was up-regulated in the glomeruli from nephritic rats, and this expression was significantly suppressed in those rats treated with TGFDE Figure 3d. In contrast, the level of TGF-1 expression was unchanged in TGFSCR-treated rats. We also observed -SMA expression in the glomeruli of untreated and TGFSCR-treated kidneys. However, immunostaining of -SMA was significantly weak in TGFDE-treated kidney Figure 4d.

Figure 3.

Inhibition of glomerular expression of transforming growth factor- (TGF-) by TGFDE. Representative photomicrographs show the immunohistochemical staining for TGF- in untreated (A), TGFDE-treated (B), and TGFSCR-treated (C) nephritic kidney (200). TGFDE (D) significantly suppressed glomerular TGF- expression (1.69 0.49, 0.81 0.47, and 1.73 0.63 in untreated, TGFDE, and TGFSCR, respectively). *P < 0.001.

Full figure and legend (300K)

Figure 4.

Inhibition of glomerular expression of -smooth muscle actin (-SMA) by TGFDE. Representative photomicrographs show the immunohistochemical staining for -SMA in untreated (A), TGFDE-treated (B), and TGFSCR-treated (C) nephritic kidney (200). TGFDE (D) significantly suppressed glomerular -SMA expression (2.17 0.72, 0.72 0.42, and 2.14 0.66 in untreated, TGFDE, and TGFSCR, respectively). *P < 0.001.

Full figure and legend (292K)

DISCUSSION

Effect on glomerular matrix accumulation

To determine the effect on the histological changes in nephritic kidneys, histologic analysis was performed using PAS staining. PAS staining showed marked ECM accumulation in untreated and TGFSCR-treated kidneys. In contrast, TGFDE transfection reduced ECM accumulation Figure 5. The degree of glomerular matrix accumulation was determined as the percentage of each glomerulus occupied by mesangial matrix. Glomerular matrix score was significantly reduced in TGFDE-treated kidneys compared with untreated and TGFSCR-treated kidneys. We further examined the effect of TGFDE on glomerular platelet-derived growth factor (PDGF) expression by reverse transcription-polymerase chain reaction (RT-PCR), and observed no differences between three groups (data not shown). In addition, glomerular cell number was not affected (79.5 9.7, 75.9 8.2, and 80.6 9.3 in untreated, TGFDE, and TGFSCR, respectively). TGF- is considered necessary for the glomerular remodeling. We examined the healing process of mesangiolysis and the later stage of the disease in the treated rats. TGFDE transfection did not interfere with the initial injury in the anti-Thy-1 model, because treated and untreated kidney had equivalent mesangiolysis. On the contrary, the healing process seemed accelerated in TGFDE-treated kidney probably due to the suppression of excessive ECM deposition.

Figure 5.

Inhibition of glomerular extracellular matrix (ECM) accumulation by TGFDE. Representative photomicrographs show periodic acid-Schiff (PAS) staining in untreated (A), TGFDE-treated (B), and TGFSCR-treated (C) nephritic kidney (200). Glomerular matrix score (D) was significantly reduced in TGFDE-treated kidneys compared with untreated and TGFSCR-treated kidneys (2.85 0.78, 1.97 0.72, and 2.68 0.73 in untreated, TGFDE, and TGFSCR, respectively). *P < 0.001.

Full figure and legend (315K)

In order to inhibit the gene expression by cleavage of the target mRNA, antisense oligonucleotides has been adopted⁷. We compared the effect of TGFDE with antisense oligonucleotides. Treatment with antisense oligonucleotide (200 g) for TGF-1 also reduced glomerular TGF-1 and type I collagen mRNA expression to 53% and 41% of untreatment, respectively. TGFDE was not less effective than antisense olignucleotides in treating nephritic disease was consistent with the previous report⁴. However, the effect of antisense may be transient, because RNase H bases antisense mechanism on the hydrolysis of RNA-DNA duplex⁸. In contrast, the 3' terminus of TGFDE was capped with an inverted 3'-3'-linked thymidine for resistance to 3'-to-5' exonuclease digestion, because phosphorothioate modified oligonucleotides may hamper the cellular function⁴.

We should establish the safety of this method prior to clinical application. However, we observed no damage in glomeruli and tubular epithelial cells when TGFDE was transferred into normal kidney on histologic examination. In addition, there was no difference between normal and TGFDE-treated rats in the serum lactate dehydrogenase (LDH) levels (117.5 26.7 and 129.3 29.2 in untreated and TGFDE, respectively), suggesting no major toxicity of electroporation-mediated TGFDE transfer.

CONCLUSION

Our study demonstrates that DNA injection via renal artery followed by electroporation could be a powerful therapeutic tool in vivo. The molecular intervention by DNAzyme for TGF- may be a promising strategy for treatment of glomerular diseases.

References

References