Adeno-associated virus–based gene therapy treats inflammatory kidney disease in mice

Guochao Wu,1,2 Shuya Liu,1,2 Julia Hagenstein,1,2 Malik Alawi,3 Felicitas E. Hengel,1,2 Melanie Schaper,1,2 Nuray Akyüz,4 Zhouning Liao,1,2 Nicola Wanner,1,2 Nicola M. Tomas,1,2 Antonio Virgilio Failla,5 Judith Dierlamm,4 Jakob Körbelin,4 Shun Lu,1,2 and Tobias B. Huber1,2

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Address correspondence to: Shuya Liu or Tobias B. Huber, III. Department of Medicine, University Medical Center Hamburg Eppendorf Martinistr, 52 20246 Hamburg, Germany. Phone: 49.0.40.7410.51911; Email: s.liu@uke.de (S. Liu). Phone: 49.0.40.7410.53908; Email: t.huber@uke.de (TBH).

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Wu, G. in: JCI | PubMed | Google Scholar |

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Liu, S. in: JCI | PubMed | Google Scholar |

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Hagenstein, J. in: JCI | PubMed | Google Scholar

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Alawi, M. in: JCI | PubMed | Google Scholar |

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Hengel, F. in: JCI | PubMed | Google Scholar

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Schaper, M. in: JCI | PubMed | Google Scholar

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Akyüz, N. in: JCI | PubMed | Google Scholar

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Liao, Z. in: JCI | PubMed | Google Scholar

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Wanner, N. in: JCI | PubMed | Google Scholar |

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Tomas, N. in: JCI | PubMed | Google Scholar |

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Failla, A. in: JCI | PubMed | Google Scholar

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Dierlamm, J. in: JCI | PubMed | Google Scholar

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Körbelin, J. in: JCI | PubMed | Google Scholar |

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Lu, S. in: JCI | PubMed | Google Scholar |

1III. Department of Medicine,

2Hamburg Center for Kidney Health (HCKH),

3Bioinformatics Core,

4Department of Oncology, Hematology and Bone Marrow Transplantation with Section Pneumology, and

5Microscopy Imaging Facility, University Medical Center Hamburg-Eppendorf, Hamburg, Germany.

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Find articles by Huber, T. in: JCI | PubMed | Google Scholar |

Authorship note: GW and S. Liu contributed equally to this work. JK, S. Lu, and TBH contributed equally to this work.

Published September 3, 2024 - More info

Published in Volume 134, Issue 17 on September 3, 2024
J Clin Invest. 2024;134(17):e174722. https://doi.org/10.1172/JCI174722.
© 2024 Wu et al. This work is licensed under the Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/. Published September 3, 2024 - Version history
Received: August 16, 2023; Accepted: July 10, 2024 View PDF

Abstract

Adeno-associated virus (AAV) is a promising in vivo gene delivery platform showing advantages in delivering therapeutic molecules to difficult or undruggable cells. However, natural AAV serotypes have insufficient transduction specificity and efficiency in kidney cells. Here, we developed an evolution-directed selection protocol for renal glomeruli and identified what we believe to be a new vector termed AAV2-GEC that specifically and efficiently targets the glomerular endothelial cells (GEC) after systemic administration and maintains robust GEC tropism in healthy and diseased rodents. AAV2-GEC–mediated delivery of IdeS, a bacterial antibody-cleaving proteinase, provided sustained clearance of kidney-bound antibodies and successfully treated antiglomerular basement membrane glomerulonephritis in mice. Taken together, this study showcases the potential of AAV as a gene delivery platform for challenging cell types. The development of AAV2-GEC and its successful application in the treatment of antibody-mediated kidney disease represents a significant step forward and opens up promising avenues for kidney medicine.

Introduction

Most kidney diseases are associated with the dysfunction of the glomerular filtration barrier (GFB), which comprises 3 layers, including the glomerular endothelial cells (GEC), the glomerular basement membrane (GBM) and podocytes (1). GEC are fenestrated endothelial cells (EC) covered by a negatively charged surface layer known as the glycocalyx. Podocytes are specialized epithelial cells with foot processes spanned by cell-cell junctions known as slit diaphragms. The GBM is formed by secreted products from both GEC and podocytes. The functionality of the GFB relies strictly on its structural integrity, and breakdown of the GFB leads to the loss of kidney filtration function (2). Current interventions for kidney disease mostly target complications or manifestations of the disease and have limited disease-modifying effects (3). In past years, multiple molecules with therapeutic potential for the GFB have been identified. To date, more than 80 gene mutations or variants have been found to cause GFB disorders (4), and an increasing number of clinical trials are ongoing for drugs targeting the GFB (5). Therefore, the GFB is an important target for novel kidney therapies.

GFB-targeting therapies not only involve rare diseases caused by genetic defects in the GEC or podocytes, but also common diseases characterized by glomerulosclerosis or glomerulonephritis, which are initiated in these 2 cell types. However, targeting the GFB is challenging (5). Traditional drugs like small molecules and monoclonal antibodies show limited efficacy or undesired side effects due to the lack of cell- or tissue-targeting specificity (5). Encouragingly, recent innovations in gene therapy made it possible to deliver therapeutic genetic cargo to difficult or previously undruggable cells, which can substantially improve the therapeutic efficacy.

Adeno-associated virus (AAV) is regarded as a promising viral-based platform for in vivo gene delivery (6), which has been licensed by the FDA and European Medicines Agency (EMA) and proven to have benefits in many genetic diseases, such as hemophilia and spinal muscular atrophy (7). AAV has been successfully applied in vivo as a delivery tool for the treatment of genetic diseases affecting many organs, but such applications are not yet available in the kidney (8). Natural AAV serotypes show insufficient targeting specificity and transduction efficiency in kidney cells and thus do not meet the requirements as a delivery tool for kidney-targeting therapy (9, 10). As the kidney is a complex organ comprising a variety of different cell types and tissues, approaches to broaden the tropism of AAV and screenings for kidney-specific AAV vectors are essential (11).

In this study, we aimed to discover new vectors targeting the renal glomerulus. We developed a kidney-specific selection protocol based on a previously described methodology (12) and screened a random AAV2 display peptide library in vivo. By integrating the experimental and bioinformatics workflows, we identified what we believe to be a new vector, termed AAV2-GEC, which specifically and efficiently targeted the GEC after systemic administration. AAV2-GEC exhibited robust GEC tropism in healthy C57BL/6J, Balb/c, BTBR mice, Sprague Dawley (SD) rats, and disease models that cause GEC damage. It also exhibited increased transduction efficiency and specificity in human primary GEC compared with WT AAV2 (AAV2-WT). Further, the potential of AAV2-GEC for kidney-targeting therapy was evaluated by delivering a bacterial cysteine proteinase, the IgG-degrading enzyme of Streptococcus pyogenes (IdeS), to the GEC. IdeS, also known as imlifidase, is a medication for the desensitization of highly sensitized patients undergoing kidney transplantation (13) and is currently being tested in clinical trials for therapeutic cleavage of kidney-bound IgG in patients with circulating anti-GBM antibodies (14). We show that AAV2-GEC-IdeS transduction efficiently produced IdeS in GEC, which provided sustained clearance of IgG and successfully prevented the onset and progression of anti-GBM glomerulonephritis in mice.

Results

In vivo selection of a random peptide library enriched glomeruli-targeted capsids. To select kidney-specific AAV capsids, we used an AAV2-displayed random heptamer peptide library (15, 16) with a calculated plasmid diversity of 1.5 × 108 (17) and we established an in vivo screening protocol for kidneys based on a previous report (12) (Figure 1A).

Figure 1

In vivo selection of the AAV2 peptide display library identified capsids enriched in renal glomeruli. (A) Schematic overview of in vivo selection in the murine kidney. (B) Pie charts demonstrating the distribution of peptide variants in each selection round. The frequency of particular peptide inserts was determined by NGS. “Others” indicates the occurrence of peptide variants ranked below the “top 100 peptides” in the total pool. (C) Heatmap demonstrating the top 10 peptide variants enriched in the glomerulus ranked by C scores. The combined C score (by multiplying GS and E) described the peptide performance regarding specificity (GS score) and efficacy (E score) with an ideal value of 1. (D) Quantification of vector genome distribution by qPCR. The number of vector genomes was quantified and normalized to vector copy numbers per diploid genome (vg/dg). Values are mean + SD. Significance was determined by 1-way ANOVA with Dunnett’s test, ****P < 0.0001 in all comparisons (glomeruli versus other organs).

Since the kidney is a complex organ in terms of both anatomical structure and a large number of different cell types, it is important to monitor the selection kinetics and adjust the selection pressure during the process of in vivo screening. In the first 2 rounds of selection, the AAV library fragments were rescued from the genomic DNA of the whole kidney. From the third round of selection, we increased the selection pressure by rescuing the genomic DNA only from the isolated glomeruli instead of the whole kidney. Next generation sequencing (NGS) was performed after each round of selection to thoroughly analyze the enriched peptides from the target organ. After the fourth round of selection, we detected a dramatic enrichment. As expected, the number of peptide sequences decreased in each round of selection, whereas the percentage of the top 100 enriched peptides increased accordingly (Figure 1B and Table 1). The sequences of the top 10 enriched peptides in each selection round are listed in Table 2.

Table 1

Summary of the peptide diversity and enrichment

Table 2

Amino acid sequences of top 10 enriched peptides in each selection round

In the last round of selection, NGS was performed to analyze the peptides from off-target organs. We evaluated all enriched peptides by a rating system based on NGS data that reflects the relative frequency of a given peptide in the target tissue and its distribution in the target tissue compared with the off-target organs (see Methods) (16). In this study, the enrichment score (E score) reflected the changes in relative abundance from the third to fourth round of selection in glomeruli. The general tissue specificity score (GS score) reflected the relative abundance in glomeruli compared with multiple off-target organs. The combined score (C score) was determined by multiplying GS and E, reflecting the performance of a given peptide regarding both targeting specificity and efficacy. Thus, the enriched peptides were ranked by C scores. In the top 10 peptides (Figure 1C), QVLVYRE didn’t have the highest E score, but it outperformed other sequences with a better GS score, suggesting that it was not only highly abundant but also highly specific in glomeruli (Figure 1C). We further evaluated the targeting specificity and efficacy of AAV2-QVLVYRE by quantifying the distribution of the vector genome across all major organs. Quantitative PCR (qPCR) showed that AAV2-QVLVYRE was at least 10-fold more dominant in glomeruli compared with other organs, including the whole kidney (Figure 1D). Taken together, QVLVYRE was chosen as the most promising peptide targeting the renal glomerulus.

AAV2 vector displaying the QVLVYRE peptide specifically transduced the GEC. To evaluate the in vivo transduction profile of AAV2-QVLVYRE, we generated a self-complementary AAV reporter vector carrying the GFP gene driven by the constitutive CMV promoter and intravenously injected adult C57BL/6J mice with a dose of 5E12 vg/kg. Transgene expression was analyzed in different organs 2 weeks after the injection.

In the kidney, immunofluorescent staining (IF) showed that the GFP expression mediated by AAV2-QVLVYRE was restricted to the glomerulus and revealed excellent transduction efficiency in all renal glomeruli (Figure 2A). We further confirmed that AAV2-QVLVYRE specifically transduced the GEC, which was marked by CD31, but not podocytes marked by Wilms tumor protein (WT1), or mesangial cells marked by platelet-derived growth factor receptor β (PDGFRB) (Figure 2B). AAV2-QVLVYRE was hence termed AAV2-GEC.

Figure 2

AAV2-GEC specifically transduced the GEC. (A) Representative overview images of AAV2-GEC and AAV2-WT mediated GFP expression in kidneys from C57BL/6J. Original magnification x10. (B) AAV2-GEC mediated GFP expression was detected in the GEC marked by anti-CD31 antibody. Mesangial cells were marked by anti-PDGFRB antibody. Podocytes were marked by anti-WT1 antibody. Nuclei were counterstained with DAPI. Scale bars: 25 μm.

The transduction properties of AAV2-GEC were compared with its parental AAV2-WT. No GFP expression was detected in the kidney of AAV2-WT–injected mice (Figure 2A). GFP expression was strong in the liver and heart and moderate in the spleen of AAV2-WT–injected mice, whereas it was far weaker in all of the same organs of AAV2-GEC–injected mice (Supplemental Figure 1A; supplemental material available online with this article; https://doi.org/10.1172/JCI174722DS1). Notably, GFP-positive cells were not colocalized with endothelial cell markers in the liver, heart, and spleen of AAV2-GEC–injected mice (Supplemental Figure 1B). AAV2-GEC–mediated GFP expression was analyzed over 360 days after intravenous injection (Supplemental Figure 2A) and the GFP signal intensity was quantified (Supplemental Figure 2B). Throughout the whole period, glomerular GFP peaked at day 14, was stable at high levels until day 120, and decreased from day 240. GFP signal was dominant in glomeruli, but from day 120 it was also detected in some EC in the tubular segment. Additionally, liver and spleen histology was analyzed (Supplemental Figure 3). No obvious histological lesions were observed over 360 days, indicating no tissue toxicity due to the injection of AAV2-GEC. Taken together, AAV2-GEC mediates specific and prolonged GFP expression in the GEC for at least 120 days upon intravenous injection.

AAV2-GEC maintained robust tropism in GFB-damaged mice. GEC are highly differentiated EC characterized by their unique fenestrae and surface layer glycocalyx, which are essential for glomerular filtration (18). The differentiation and permeability of GEC are also regulated by podocytes (5). Under disease conditions, the breakdown of the GFB due to the injury of GEC or podocytes could lead to changes in the GEC-directed tropism of our targeted AAV2. Therefore, we evaluated the AAV2-GEC tropism in mouse models with damaged GFB.

GEC injuries such as the loss of fenestrae and glycocalyx disruption are typically induced by hyperglycemia in diabetic kidney disease (DKD) (18). Previous studies report that BTBR mice expressing homozygous spontaneous obesity mutations (BTBRob/ob), a well-characterized DKD mouse model, exhibits early onset of hyperglycemia (19) and shows significant fenestrae changes in GEC (20). We thus used BTBRob/ob to evaluate the transduction profile of AAV2-GEC under the GEC injury condition. AAV2-GEC-GFP was intravenously injected in 16–18 week-old BTBRob/ob mice with a dose of 5 × 1012 vg/kg. Two weeks after injection, IF of kidney sections showed robust and efficient GFP expression in GEC (Figure 3, A and B) but not in podocytes or mesangial cells (Supplemental Figure 4A).

Figure 3

AAV2-GEC maintained robust tropism in BTBRob/ob and Nphs1ΔiPod mice. (A) Representative overview images of AAV2-GEC–mediated GFP expression in kidneys from BTBR ob/ob mice. Original magnification x10. (B) GFP expression was detected in the GEC marked by anti-CD31 antibody. (C) Representative overview images of AAV2-GEC mediated GFP expression in kidneys from Nphs1ΔiPod mice. Original magnification x10. (D) GFP expression was detected in the GEC marked by anti-CD31 antibody. Nuclei were counterstained with DAPI. Scale bars (B and D): 25 μm.

In podocytes, Nephrin is one of the essential slit diaphragm proteins. Loss of Nephrin at adult age results in podocyte injury and GFB leakage (21). We evaluated the transduction profile of AAV2-GEC in Nphs1ΔiPod mice, which have induced Nephrin-deficiency in podocytes after doxycycline administration (21). AAV2-GEC-GFP was intravenously injected in Nphs1ΔiPod mice 12 weeks after knock-out induction with a dose of 5 × 1012 vg/kg. IF was performed on the kidney sections after 2 weeks, showing robust and efficient GFP expression in GEC (Figure 3, C and D) but not in podocytes or mesangial cells (Supplemental Figure 4B).

To compare the transduction efficiency of AAV2-GEC in healthy and diseased states, AAV2-GEC-GFP was injected in C57BL/6J, BTBRob/ob, BTBR WT, Nphs1ΔiPod and noninduced Nphs1ΔiPod mice (hereafter referred to as Nphs1ctrl). GEC were isolated for transcriptome analysis. There were no significant differences in GFP expression in BTBRob/ob, BTBR WT, Nphs1ΔiPod, and Nphs1ctrl compared with C57BL/6J mice, and also no significant differences between BTBR WT and BTBRob/ob and between Nphs1ΔiPod and Nphs1ctrl mice (Supplemental Figure 5A). To investigate the effect of AAV transduction on GEC function, the gene expression levels in AAV transduced versus nontransduced GEC were compared. We identified 68 differentially expressed genes in AAV-transduced versus nontransduced cells (DEGs) (21 up and 47 downregulated) in BTBRob/ob mice and 24 DEGs (20 up and 4 downregulated) in Nphs1ΔiPod mice, whereas no significant DEGs were identified in C57BL/6J mice (Supplemental Table 1). The most significant DEGs are shown in a heatmap (Supplemental Figure 5B). Gene set enrichment analysis (GSEA) predicted significant gene ontology (GO) in AAV transduced GEC in BTBRob/ob and Nphs1ΔiPod mice (Supplemental Table 1), showing that RNA binding and ribosome biogenesis were affected by AAV transduction or transgene (GFP) expression (Supplemental Figure 5C).

AAV2-GEC maintained robust tropism in Balb/c mice and SD rats. Since the transduction by AAV vectors may vary substantially between strains and species (22, 23), we evaluated AAV2-GEC tropism in adult Balb/c mice and SD rats. AAV2-GEC-GFP was intravenously injected in Balb/c mice with a dose of 5 × 1012 vg/kg. After 2 weeks, IF was performed on the kidney sections. We observed robust and efficient GFP expression in the GEC of the Balb/c mice (Figure 4, A and B). Similar tropism of AAV2-GEC was also observed in the SD rat at the dose of 5 × 1012 vg/kg, in which GEC were marked by endothelial cell–specific biomarker rat endothelial cell antigen 1 (RECA-1) (Figure 4, C and D). Of note is that the dose of 5 × 1012 vg/kg used here was below the low dose range (1.2 × 1013 vg/kg) for rats, as previously reported (24).
Figure 4
AAV2-GEC maintained robust tropism in Balb/c mice and SD rats. (A) Representative overview images of AAV2-GEC mediated GFP expression in kidneys from Balb/c mice. (B) GFP expression was detected in the GEC marked by anti-CD31 antibody. (C) Representative overview images of AAV2-GEC mediated GFP expression in kidneys from SD rats. (D) GFP expression was detected in the GEC marked by anti–RECA-1 antibody. Nuclei were counterstained with DAPI. Scale bars (B and D): 25 μm. Original magnification: x20 (A) and x10 (C).

AAV2-GEC exhibited enhanced transduction in human primary GEC. To validate the targeting efficacy of AAV2-GEC on human primary GEC, we applied AAV2-GEC or AAV2-WT luciferase reporter vectors. Reporter gene activity of AAV2-GEC was 2-fold higher than that of AAV2-WT (Supplemental Figure 6A). To evaluate cell type specificity of AAV2-GEC, transduction efficiency was also investigated in other human glomerular cell types, including immortalized podocytes and human primary mesangial cells. AAV2-GEC showed a 95% decrease in reporter gene activities compared with AAV2-WT in both glomerular cell types (Supplemental Figure 6, B and C). These data suggest that AAV2-GEC exhibited enhanced transduction in human GEC compared with AAV2-WT, but not in other human glomerular cell types.

AAV2-GEC delivery of IdeS successfully treated anti-GBM glomerulonephritis. To investigate the feasibility of using AAV2-GEC for in vivo delivery of therapeutic transgenes, we developed a treatment strategy for anti-GBM glomerulonephritis. AAV2-GEC vectors carrying secretory IdeS (see Methods) and GFP under the control of the CMV promoter were used as treatment and control vectors, respectively.

For the prophylactic interventions, the treatment and control vectors were intravenously injected in adult C57BL/6J male mice with a dose of 1 × 1013 vg/kg. Two weeks after AAV injection, all mice received 150 μl anti-GBM serum produced in sheep to induce glomerulonephritis (Figure 5A).
Figure 5
AAV2-GEC delivery of IdeS successfully treated anti-GBM glomerulonephritis. (A) Schematic of the prophylactic intervention protocol. C57BL/6J mice were divided into control (AAV2-GEC-GFP) and treatment (AAV2-GEC-IdeS) groups. n = 10 per each group. (B) UACR was measured at 0, 1, 3, and 7 days. (C) Representative images showing remaining sheep IgG Fc and the deposition of C1q in kidneys at 7 days. Scale bars: 100 μm. (D) Schematic of the therapeutic intervention protocol. C57BL/6J mice were divided into control (AAV2-GEC-GFP) and treatment (AAV2-GEC-IdeS) groups. n = 6 per each group. (E) UACR was measured at 0, 1, 4, 8, 15, and 22 days. (F) Representative images showing remaining sheep IgG Fc and the deposition of C1q in kidneys at 22 days. Nuclei were counterstained with DAPI. Values are mean ± SEM. Significance: 2-way ANOVA with repeated measures, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; only statistically significant comparisons are shown.

To monitor the GFB function, the urinary albumin-to-creatinine ratio (UACR) was measured during the progression of anti-GBM glomerulonephritis (Figure 5B). On day 1 after anti-GBM serum injection, albuminuria was detected in control mice, which was persistent from day 3 until day 7. In contrast, the onset of albuminuria was prevented in treated mice. Only a very mild increase in UACR was measured at day 1, which declined to baseline at day 3 and was maintained at the low level until day 7. These results suggest that delivery of the AAV2-GEC-IdeS efficiently prevented albuminuria in the progression of anti-GBM glomerulonephritis.

IdeS specifically cleaves IgG in the hinge region, yielding the Fab and Fc fragments (25). Kidney IF sections showed that the Fc fragments of sheep IgG were barely detectable in treated mice but predominantly deposited on the GBM of control mice (Figure 5C). The accumulation of sheep and mouse IgG on the GBM was also strongly reduced in treated mice (Supplemental Figure 7A), which then substantially reduced the deposition of complement C1q and C3 (Figure 5C and Supplemental Figure 7A).

Next, we evaluated the therapeutic potential of AAV2-GEC-IdeS by injecting the treatment and control vectors 1 day after the induction of anti-GBM glomerulonephritis (Figure 5D). While both control and treated mice experienced an initial peak in albuminuria on day 1, the subsequent course differed. Control mice displayed decreased UACR but albuminuria persisted until day 22, while treated mice exhibited a marked decrease in UACR, reverting to baseline levels from day 8. This differential response suggests a successful therapeutic effect of AAV2-GEC-IdeS in mitigating albuminuria (Figure 5E).

Kidney IF sections showed that the Fc fragments of sheep IgG, as well as the complement C1q and C3 were barely detectable in treated mice but predominantly deposited on the GBM of control mice (Figure 5F and Supplemental Figure 7B). The deposition of sheep and mouse IgG on the GBM was weaker in treated mice than in control mice (Supplemental Figure 7B). These results suggest that delivery of the AAV2-GEC-IdeS after anti-GBM glomerulonephritis also efficiently prevented albuminuria and the disease progression.

Additionally, serum liver function indexes (ALTL, ASTL, GGT-2, CHOL2, TRIGL, and BILD2) were normal in both treated and control mice (Supplemental Figure 8), indicating no liver toxicity caused by AAV2-GEC-IdeS injection.

The long-term expression of secretory IdeS was analyzed by monitoring its serum level for 240 days (Supplemental Figure 9). IdeS was fused with nanoluciferase (Nluc) and delivered to GEC by intravenous injection of AAV2-GEC-IdeS-Nluc with a dose of 5 × 1012 vg/kg. The concentration of circulating IdeS was maintained from day 3 until day 240, indicating a stable expression of IdeS by transduced GEC.

Taken together, these results suggest that AAV2-GEC-IdeS transduction efficiently produced IdeS in GEC, which provided sustained clearance of kidney-bound IgG and successfully prevented the progression of anti-GBM glomerulonephritis.
Discussion
Discovering new AAV vectors with cell-targeting properties plays an important role in developing new therapeutic approaches for kidney disease. AAV capsid engineering by directed evolution allows the generation of diverse capsid libraries, and the iterative selection of such libraries enables the identification of AAV vectors with desired tropisms. This strategy has been successfully applied to several organs and tissues (15, 16, 26–34), but less progress has been made in the kidney.

In this study, we tailored the selection process for the kidney to identify GFB-targeting AAV vectors. Anatomically, the kidney can be divided into the glomerular and tubular compartments, which exert filtration and reabsorption functions, respectively. The tubular compartment contains mainly epithelial cells that form different tubular segments, whereas the glomerular compartment contains cell types other than epithelial cells, including EC, mesangial cells, and podocytes, the latter being nontypical epithelial cells with contractile characteristics (

View original article

JOURNAL OF CLINICAL INVESTIGATION

Like

Share Bookmark

0 0 0 0 0 0 0

Comments (0)

Login or register to post your comment

More from this channel

Adeno-associated virus–based gene therapy treats inflammatory kidney disease in mice

Comments (0)