Effective and Safe Gene Delivery to the Mouse Kidney via Slow Retrograde Renal Pelvis Injection of Adeno-Associated Virus Vectors

Summary

This report details the localized injection method of AAV vectors via the renal pelvis for effective and safe gene transfer into the mouse kidney.

Abstract

The development of effective in vivo tools and methods for gene delivery to the kidney is crucial for advancing basic kidney research and gene therapy for kidney diseases. In addition, growing awareness of monogenic kidney diseases, driven by advanced genetic testing, underscores the potential of gene therapy to treat and even cure difficult-to-treat genetic kidney diseases. In this regard, adeno-associated virus (AAV) vectors have garnered increasing attention as a robust platform for in vivo gene delivery; however, the most effective and safest method for AAV vector-mediated gene delivery to each therapeutically relevant cell type in the kidney has not yet been fully established. Here, the method of slow retrograde renal pelvis (RP) injection of AAV vectors is detailed, and its high potential to transduce mouse kidneys is demonstrated when used with appropriately selected AAV capsids. Moreover, this method is shown to be effective not only with the standard cesium chloride ultracentrifugation-purified AAV (CsCl AAV) vectors but also with centrifugally ultrafiltered AAV (CU AAV) vector minipreps, which can be prepared rapidly and simply using techniques that do not require specialized AAV vector expertise. As an example, this study demonstrates slow RP injection of both CsCl and CU AAV-KP3 vectors results in robust transduction of proximal tubules in the mouse kidney with no apparent tissue damage, unlike previously reported hydrodynamic approaches that inevitably lead to tissue damage. Thus, this study highlights that slow retrograde RP injection of select AAV capsid-derived vectors is an effective and safe method for gene delivery to the kidney. This method will be applicable to a broad spectrum of kidney research, including gene therapy studies. The use of CU AAV vector minipreps will significantly increase throughput and improve the time and cost efficiency needed to generate preliminary data and obtain crucial insights.

Introduction

Effective and safe methods for gene delivery to the kidney in small and large experimental animals provide crucial opportunities to advance basic kidney research and explore innovative therapeutic approaches. Establishing such methods is particularly relevant to gene therapy for various kidney diseases. Recent advancements in clinical genetic testing have revealed a higher prevalence of monogenic renal disorders than previously estimated, affecting approximately 30% of patients with chronic kidney diseases (CKDs)¹. CKD is characterized by progressive renal damage, ultimately leading to kidney failure with no truly effective disease-modifying therapies. Hemodialysis and kidney transplantation are currently available treatments; however, they have limitations that significantly impair the quality of life for patients with CKD². Thus, gene therapy, a clinically proven powerful approach to treating genetic diseases, has garnered significant attention in recent years as a novel avenue for the effective treatment of genetic kidney diseases leading to CKD^3,4,5,6. This approach can directly address the root causes of the diseases and potentially offer a cure.

Despite the steady advancements in the field, gene delivery to the kidney remains a significant challenge in both research and therapeutic settings. Efficiently targeting kidney cells with therapeutic genes is difficult due to the organ's complex structure and unique physiological barriers, including the glomerular filtration barrier, which is composed of glomerular endothelial cells, the glomerular basement membrane, and podocyte foot processes. Various strategies have been explored for kidney-targeted gene delivery, including direct renal cortex (RC) injection^7,8, renal artery (RA) injection^9,10,11, renal vein (RV) injection^12,13,14, renal pelvis (RP) injection^{15,16,17,18,19,20,21,22,23,24,25} and intravenous (IV) injection^{19,26,27,28,29,30} of viral or non-viral gene delivery vectors, as well as ultrasound and microbubble-mediated nucleic acid delivery^31,32. These approaches demonstrated variable transduction efficiencies or functional outcomes, often yielding inconsistent results that are difficult to reconcile (reviewed in references^3,33,34,35). Gene delivery to the kidney is still underdeveloped, making direct comparisons between vector systems and methodologies challenging. However, localized delivery methods, which facilitate the administration of highly concentrated agents directly to the kidney, have generally achieved better results than systemic IV delivery. In addition, adeno-associated virus (AAV) vectors, with their robust safety profiles and proven track record of clinical translation^36,37, have been the most studied and are regarded as the leading choice for renal gene transfer.

Common AAV serotypes and engineered capsids have demonstrated effective transduction of mesangial and mesenchymal cells in the kidney following IV injection in mice^28,30,38. Notably, recent studies have shown that IV administration of AAV9 can effectively transduce podocytes in mouse models of nephrotic syndrome and Alport syndrome, although this effect is not observed in healthy kidney^38,39, and that IV administration of an engineered AAV2 capsid can exclusively target glomerular endothelial cells in both healthy and inflamed kidneys in mice⁴⁰. Historically, renal tubules have been challenging to transduce via the IV route. However, it has recently been discovered that RP injection of select AAV capsids, such as AAV-KP1⁴¹, can achieve transduction of proximal tubules at levels more than an order of magnitude higher than those achieved with the benchmark AAV9 capsid³⁸. Although further improvements are needed, these advancements open new avenues for basic kidney research and gene therapy, offering opportunities that were previously unattainable.

In the present article, slow retrograde RP injection of AAV vectors is detailed as an effective and safe method for gene delivery to the mouse kidney³⁸. As a representative example of this technique, AAV-KP3 was chosen as an exemplary capsid based on the discovery that it is one of the six capsids (i.e., AAV-KP1, AAV-KP2, AAV-KP3⁴¹, AAV-DJ⁴², AAV2G9⁴³, and AAV2.7m8⁴⁴) that showed enhanced renal transduction compared to AAV9 following RP injection in mice³⁸. Notably, slow retrograde RP-injected kidney tissues showed no apparent tissue damage in contrast to those treated with hydrodynamic approaches previously reported^23,24. Moreover, AAV miniprep vectors⁴⁵ crudely purified by quick and simple centrifugal ultrafiltration (i.e., CU AAV vectors) achieved renal transduction levels approximately 70% of those attainable by AAV vector preparations purified by the standard cesium chloride (CsCl) gradient ultracentrifugation (i.e., CsCl AAV vectors) when administered at the same vector genome titer. Thus, the method and observations presented here provide a valuable and practical resource for researchers seeking to enhance gene delivery to the kidney in mice and potentially larger animals.

Protocol

All animal experiments described here were approved by the Institutional Animal Care and Use Committee at Oregon Health & Science University (OHSU) and were performed in accordance with the guidelines for animal care at OHSU. In this study, eight-week-old C57BL/6J male mice were used. The reagents and the equipment used in this study are listed in the Table of Materials.

1. AAV vector production by an adenovirus-free plasmid transfection

1. Culture human embryonic kidney (HEK) 293 cells in Dulbecco's Modified Eagle Medium (DMEM)-high glucose (4.5 g/L) supplemented with 10% fetal bovine serum (FBS), 1% Penicillin and Streptomycin Mix, and 1 mM L-glutamine, in a 37 °C incubator with 5% carbon dioxide (CO₂).
2. On Day -2 (48 h prior to transfection), plate 4 x 10⁶ HEK 293 cells per T225 flask in 40 mL of the culture medium as described in step 1.1. This reaches ~95% confluency on Day 0.
   NOTE: The culture scale can be adjusted based on the application. For an AAV miniprep (step 2), prepare an appropriate number of flasks based on the target titer and vector concentration, considering that the purification process typically results in an approximate 50% loss of material. To achieve >6 x 10¹² vg/mL of AAV9 and AAV-KP3 vectors for RP injection in mice at a dose of 3 x 10¹¹ vg/mouse in a 50 µL vector solution, at least 2 flasks are recommended. For larger-scale AAV vector production, such as that for CsCl density gradient-purified AAV vectors (CsCl AAVs), 25 or more flasks are typically used.
3. Day 0: Plasmid DNA transfection
   1. Preparation for transfection
      1. Make sure that cells have reached ~95% confluency.
      2. Warm up DMEM supplemented with 1% Penicillin and Streptomycin Mix and 1 mM L-glutamine but without FBS (i.e., serum-free medium) in a 37 °C water bath.
      3. Allow the linear polyethylenimine (PEI) solution to reach room temperature.
   2. Preparation of plasmid DNA mixture
      1. Mix the plasmids (15 µg each of pAAV-CAG-tdTomato, pRepCap, and pHelper) in 1 mL of phosphate-buffered saline (PBS) without CaCl₂ or MgCl₂ in sterile 1.5 mL microcentrifuge tubes. The total amount of DNA is 45 µg/flask.
         NOTE: Plasmid DNA solution volumes can be disregarded if they are minimal. However, volume adjustment is advised if the total solution volume is ≥225 µL. The sequences of the plasmids used in this study are available in Supplementary File 1.
   3. PEI transfection
      1. Add 90 µL of PEI solution (1 mg/mL) to the 1 mL of PBS and mix with the PBS-plasmid DNA mix (prepared as above) in a 50 mL polypropylene tube. The final volume is approximately 2 mL (5% volume of the culture medium).
      2. Vortex the tubes for 5 s and briefly spin the tubes with a centrifuge to collect the liquid to the bottom of the tubes.
   4. Keep at room temperature for 15 min to form plasmid DNA-PEI complexes.
   5. Once the 15-min incubation is complete, add the 40 mL of prewarmed serum-free medium prepared in step 1.3.1.2 to the plasmid DNA-PEI complexes. The volume is approximately 42 mL.
   6. Replace the culture medium of the T225 flask with plasmid DNA-PEI complexes containing serum-free medium described in step 1.3.4.
   7. Maintain the cells in this transfection medium until the harvest on Day 5 (no medium change is required).
   8. On Day 2, observe cells transfected with the plasmid expressing fluorescent proteins (e.g., pAAV-CAG-tdTomato) under an inverted fluorescence microscope to assess plasmid transfection efficiency.
      NOTE: The conditions described above typically result in transfection efficiencies exceeding 70%. Some morphological changes (e.g., cell elongation) may be observed, likely due to the serum-free environment.
   9. On Days 3 to 5, continue to culture the transfected cells in a 37 °C incubator with 5% CO₂.
   10. On Day 5, collect both cells and virus-containing medium into a 50 mL polypropylene tube.
   11. Store the samples at -80 °C until use.

2. Centrifugal ultrafiltration (CU) purification of AAV vectors (AAV miniprep)

1. Thaw the frozen sample from step 1.3.11 in a 37 °C water bath.
2. Keep 1 mL of crude lysate aside for DNA dot blot assay (used as Fraction 1).
3. Add 80 µL of 1M magnesium chloride (MgCl₂, final concentration of 2 mM), 160 µL of 0.1M sodium hydroxide (NaOH, final concentration of 0.4 mM), and 16 µL of Serratia marcescens endonuclease (final concentration of 100 units/mL) into a 50 mL tube. Incubate at 37 °C for 1 h. Vortex the sample every 15 min during the incubation.
4. Collect the Serratia marcescens endonuclease-treated sample into a 250 mL centrifuge bottle and centrifuge at 10,000 x g at 4 °C for 30 min.
5. Transfer the supernatant to a 50 mL polypropylene tube using a 50 mL serological pipette.
6. Filter the supernatant to remove debris using a 0.45 µm PVDF syringe filter. Keep 1 mL of 0.45 µm-filtrated sample aside for DNA dot blot assay (used as Fraction 2).
7. Add 10 mL of molecular grade water to a 100 kDa molecular weight cut-off (MWCO) centrifugal ultrafiltration tube and centrifuge at 2,000 x g at 4 °C for 10 min. Remove the flow-through.
8. Add 15 mL of 5% sorbitol in PBS to the centrifugal ultrafiltration tube and centrifuge at 2,000 x g at 4 °C for 10 min. Remove the flow-through.
9. Add up to 15 mL of 0.45 µm-filtrated sample as described in step 2.6 and centrifuge at 2,000 x g at 4 °C for 15 min for concentration until the target volume of approximately 500 µL is achieved. Keep 1 mL of flow-through aside for DNA dot blot assay (used as Fraction 3).
   NOTE: Before each centrifugation, resuspend the retentate with a pipette to disrupt the concentration gradient. Do not touch the membrane with the pipette tip. If the target volume is not reached after 15-min centrifugation, extend the centrifugation as needed by adjusting the duration based on the remaining volume in the tube.
10. Repeat step 2.9 until the entire sample is filtered, reaching the target volume of approximately 500 µL. Keep 1 mL of the flow-through from the first and second iterations of step 2.10 aside for DNA dot blot assay (used as Fraction 4 and Fraction 5).
    NOTE: A single centrifugal ultrafiltration tube can be used to process samples obtained from two or more flasks.
11. Add 15 mL of 5% sorbitol in PBS and centrifuge at 2,000 x g at 4°C for 15 min until the target volume of approximately 500 µL is achieved. Keep 1 mL of flow-through aside for DNA dot blot assay (used as Fraction 6).
12. Repeat step 2.11 (used as Fraction 7).
    NOTE: If higher AAV titers are required for downstream experiments, the final target volume can be further reduced to 150 µL.
13. Transfer the ultrafiltered AAV vector to a 2 mL screw cap tube with an O-ring. Keep 10 µL of ultrafiltered AAV vector aside for DNA dot blot assay (used as Fraction 8).
14. Determine AAV titers by DNA dot blot assay as previously described⁴⁶.
    NOTE: CU-purified vectors (CU AAV) contain higher levels of empty capsid contamination and host cell-derived impurities compared to CsCl-purified vectors (CsCl AAV). These factors may impact transduction efficiency and immune responses, requiring careful consideration in data interpretation.

3. CsCl density-gradient purification of AAV vectors

1. Thaw the frozen sample from step 1.3.11 in a 37 °C water bath.
2. Centrifuge at 10,000 x g at 4 °C for 30 min to remove the cell debris.
3. Collect the supernatant and precipitate AAV vectors in the supernatant by adding Polyethylene Glycol 8000 (PEG, final concentration of 8%) and sodium chloride (final concentration of 500 mM).
4. Incubate on ice for 3 h.
5. Centrifuge at 10,000 x g at 4 °C for 30 min to obtain the pellets containing AAV.
6. Remove the supernatant and resuspend the pellets with 10ml of Sarkosyl containing resuspension buffer (50 mM HEPES, 150 mM NaCl, 1% Sarkosyl, pH 8.0).
7. Transfer the pellet suspension to a 15 mL polypropylene tube and keep it at 4 °C overnight.
8. The following day, treat the pellet suspension with Serratia marcescens endonuclease (final concentration of 200 units/mL) and MgCl₂ (final concentration of 2 mM).
9. Incubate at 37 °C for 1h.
10. Centrifuge at 10,000 x g at 4 °C for 30 min.
11. Collect the supernatant and add cesium chloride (CsCl, final concentration of refractive index (RI) = 1.3710 +/- 0.0005) and EDTA (final concentration of 20 mM).
12. Load the sample into a 29.9 mL ultracentrifuge tube and add topping solution (50 mM HEPES, 150 mM NaCl, 20 mM EDTA, CsCl (RI = 1.3710 +/- 0.0005), pH 8.0) until it is full.
13. Centrifuge at 41,000 x g at 12 °C for 22 h.
14. Punch holes in the bottom and side of the ultracentrifuge tube to collect a total of 15 fractions of 2 mL each.
15. Use 10 µL of each fraction to check RI and perform the DNA dot blot assay⁴⁶.
16. Based on the RI and DNA dot blot assay results, determine the best 5 fractions that contain AAV vectors.
17. Combine the best 5 fractions (total about 10 mL) and adjust RI to 1.3710 +/- 0.0005.
18. Load the sample into an 11.2 mL ultracentrifuge tube and add the topping solution until it is full.
19. Centrifuge at 65,000 x g at 12 °C for 12 h.
20. Punch holes in the bottom and side of the 11.2 mL ultracentrifuge tube to collect a total of 18 fractions of 0.5 mL each.
21. Use 10 µL of each fraction to check RI and perform the DNA dot blot assay⁴⁶.
22. Based on the RI and DNA dot blot assay results, determine the best 6 fractions that contain AAV vectors.
23. Combine the best 6 fractions (total of about 3 mL) and load them into a dialysis cassette.
24. Float the dialysis cassette in 1.3 L of dialysis buffer (PBS buffer supplemented with 0.001% non-ionic surfactant) with a stir bar for 2 h to remove the CsCl.
25. Repeat the step 3.24 two times (total 6 h of dialysis).
26. Transfer the cassette to 500 mL of 5% sorbitol in PBS buffer supplemented with 0.001% surfactant for 2 h for buffer exchange.
27. Collect the sample from the dialysis cassette with a syringe and filter the collected sample using a 0.22 µm PVDF syringe filter.
28. Determined AAV titers by a quantitative dot blot assay⁴⁶.

4. Quality assessment of AAV vector preparations by SDS-PAGE followed by silver staining

1. Mix vector preparations, 1 x 10¹⁰ vg for CsCl AAV vectors, and 1 x 10⁹ and 1 x 10¹⁰ vg for CU AAV vectors, with 6x Laemmli buffer in a total 18 µL (vector preparation:6x Laemmli = 15:3) and heat them at 100 °C for 5 min.
2. Load the heat-treated AAV vector preparations onto a 7.5% SDS-PAGE gel and electrophorese the samples at 20 mA using Tris-glycine-SDS gel running buffer.
3. Remove the gel from the gel cassette, rinse it with ultrapure water, and stain it using a silver staining kit according to the manufacturer's instructions.
4. Incubate the gel loaded with CsCl AAV vectors in the developer solution for 7 min, whereas incubate the gel loaded with CU AAV vectors for 2 min to avoid signal saturation.
   NOTE: CU AAV vector preparations contain higher levels of protein impurities compared to CsCl AAV vectors. Therefore, gels loaded with CsCl AAV and CU AAV vectors were separated after gel running and silver-stained separately to prevent signal saturation due to the high sensitivity of the staining method.

5. Identification of AAV VP proteins in CU AAV vector preparations by western blot

1. Mix 1 x 10⁹ vg of CU AAV vector preparations with 6x Laemmli buffer in a total of 18 µL (vector preparation:6x Laemmli = 15:3) and heat them at 100 °C for 5 min.
2. Load the heat-treated AAV vector preparations onto a 7.5% SDS-PAGE gel and electrophorese the samples at 20 mA using Tris-glycine-SDS gel running buffer.
3. Electrotransfer proteins in the gel onto a polyvinylidene difluoride (PVDF) membrane, activated in methanol, using a semi-dry transfer system.
4. Block the membrane with 5% skimmed milk in 1x Tris-buffered saline with Tween 20 (TBST) at room temperature for 1 h.
5. Incubate the blocked membrane with primary antibody solution (rabbit polyclonal anti-VP antibody, diluted to 1:1000 with 5% skimmed milk in 1x TBST) at 4 °C overnight.
6. Wash the membrane with 1x TBST for 10 min, 3 times.
7. Incubate the membrane with secondary antibody solution (goat anti-rabbit IgG antibody conjugated with horseradish peroxidase, diluted to 1:5000 with 5% skimmed milk in 1x TBST) at room temperature for 1 h.
8. Wash the membrane with 1x TBST for 10 min, 3 times.
9. Visualize signals on the membrane using a western blot imaging system.

6. Renal pelvis (RP) injection

1. Preoperative setup and anesthesia
   1. Measure the body weight of each mouse in a plastic beaker on a weight scale that is accurate to 0.1 g.
   2. Administer meloxicam (5 mg/kg) subcutaneously based on the body weight to provide presurgical analgesia.
   3. Dilute the AAV vector solution to a concentration of 6.0 x 10¹² vg/mL with 5% sorbitol in PBS and keep it on ice.
      NOTE: Prepare 40 μL more AAV solution than the total injection volume (50 μL x number of mice to be injected) to account for dead space in a gas-tight glass syringe and a silicone tube.
   4. Bend and break off only the needle tip from a 30 G needle using a needle holder.
   5. Under a stereomicroscope, connect the 30 G needle tip, blunt end of the 4 cm PE-10 tube, 15 cm silicone tube, a 30 G needle and a gas tight glass syringe in this order from the tip (Figure 1D). Sterilize both tubes prior to use and perform this assembly sequence on top of a sterile drape to maintain sterility. Sterilize the autoclave-compatible silicone tube by standard steam sterilization. Before sterilizing PE-10 tube, cut one end diagonally to create a beveled tip that facilitates insertion into the silicone tube. Sterilize the PE-10 tube by immersing it in 2% glutaraldehyde for more than 10 hours and then rinse it thoroughly with sterile PBS before use.
   6. Set the connected gas-tight glass syringe onto an infusion syringe pump.
   7. Aspirate 90 µL of AAV vector solution prepared in step 6.1.3 into the gas-tight glass syringe using an infusion syringe pump.
   8. Set the infusion mode of the syringe pump as follows: the target volume of 50 µL and a flow rate of 50 µL/min.
   9. Place a mouse in a plastic chamber filled with 4% isoflurane delivered in 100% oxygen at a flow rate of 1.5 to 3.0 L/min until it is fully anesthetized.
   10. Place the snout of the mouse into a nasal cone with 1.5% to 3.0% isoflurane flow. Monitor the depth of anesthesia by toe-pinch reflex throughout the surgical procedure.
   11. Cover the eyes with ophthalmic ointment using a cotton-tipped applicator to prevent corneal desiccation.
   12. Remove the fur from the left back side of the mouse with a depilatory cream using a cotton-tipped applicator.
       NOTE: Depilatory cream must be completely removed via rinsing to prevent skin irritation.
   13. Place the mouse in a lateral position on a closed-loop heat pad connected to a heat therapy pump to maintain body temperature at 37 °C (no need to secure it with adhesive tape).
   14. Disinfect the shaved area with three alternating applications of povidone-iodine and 70% ethanol using a cotton-tipped applicator in a circular motion beginning at the incision site.
   15. Cover the mouse with a sterile surgical drape.
2. RP injection to the left kidney
   1. Make an incision of approximately 1 cm to the skin layer with a sterile scalpel at the costovertebral angle level in the left kidney's location.
   2. After making a skin incision, cut the muscle layer at the same length.
      NOTE: If the incision length is too short, the kidney cannot be exposed, and instead, the spleen, which is smaller in size, may be accidentally exposed. On the other hand, if the incision length is too long, the kidney will not be kept in place by the incision itself and will continually slip back into the abdomen through the incision.
   3. Gently press on the abdomen with the fingers, not directly touching the kidney, so that the kidney is exposed through the incision site (Figure 1A).
   4. Minimally remove the surrounding fat so that the renal pelvis is visible as a small white area.
   5. First, clamp the ureter with a straight-type micro vessel clip, and then the renal artery and renal vein together with a curved-type micro vessel clip.
      NOTE: The kidney color turns dark red, confirming the successful blockade of blood flow into and out of the kidney (Figure 1B).
   6. Insert the 30 G needle connected to the syringe pump approximately 3 mm into the pelvic cavity with curved forceps and start an injection of 50 µL of AAV vector solution using an infusion syringe pump at a flow rate of 50 µL/min over 1 min (Figure 1C).
   7. Keep the needle and clamp for 5 min to increase the injected kidney's exposure to the AAV vector.
      NOTE: Appropriate dwelling time may vary and depend on the AAV capsids. A dwelling time of 5 min has been shown to be sufficient for transduction with AAV-KP1 or AAV-KP3 vector.
   8. Once the 5-min incubation is complete, remove the clamps.
      NOTE: The kidney color immediately changes from dark red to blood red, confirming the restoration of the blood flow in the kidney.
   9. Withdraw the injection needle carefully and slowly, simultaneously placing a cotton-tipped applicator over the injection site with pressure. Make sure that there is no bleeding from the injection site. Applying a hemostatic patch is not required.
   10. Gently place the kidney back into the peritoneal cavity.
   11. Suture the muscle layer with 6-0 absorbable suture and the skin with 5-0 monofilament suture.
   12. Administer 1 mL of saline subcutaneously into the back of mice for fluid support.
   13. Replace the used 30 G needle tip with a newly prepared one, discard the used 30 G needle according to the institution's guidelines for hazardous chemical waste management, and load 50 µL of the AAV vector solution using an infusion syringe pump for administration to the next mouse.
   14. Repeat steps 6.2.12-6.2.13 for all mice in the group, except for the last one, for which step 6.2.13 is not required.
   15. After completing all planned injections, discard the contaminated needles and tubes according to the institution's guidelines for hazardous chemical waste management.
3. Postoperative monitoring and care
   1. Administer meloxicam (5 mg/kg) subcutaneously into the back of mice for postsurgical analgesia 24 h and 48 h after RP injection.
   2. Administer 1 mL of saline subcutaneously into the back of mice for fluid support 24 h after RP injection.
   3. Monitor the body weight and general appearance of injected mice daily for 1 week postoperatively. If a 10% body weight loss occurs, administer 1mL of saline subcutaneously into the back of mice three times a week for a week. If no improvement is observed, affected mice are euthanized.

7. Tissue harvesting and cryosectioning

1. Preparation of fixation and cryoprotection solution.
   1. 0.2 M phosphate buffer (total 1 L)
      1. Prepare 800 mL of ultrapure water in a 1 L glass beaker.
      2. Add 6.4 g of sodium phosphate monobasic (NaH₂PO₄) and 21.8 g of sodium phosphate dibasic (Na₂HPO₄) to the solution.
      3. Adjust the pH to 7.4 with hydrochloric acid.
      4. Fill with ultrapure water to 1 L.
      5. Store at 4 °C until use.
   2. 4% paraformaldehyde (PFA) solution (total 40 mL)
      1. Mix 10 mL of 16% PFA, 20 mL of 0.2M phosphate buffer prepared in step 7.1.1, and 10 mL of ultrapure water.
      2. Aliquot 10 mL into a 15 mL polypropylene tube.
         NOTE: Make 4% PFA solution just before use for perfusion fixation.
   3. 30% sucrose solution (total 40 mL).
      1. Mix 30 mL of PBS with 12 g of sucrose.
      2. After dissolving completely, fill with PBS to 40 mL.
2. Perfusion fixation and tissue harvesting.
   CAUTION: Perfusion fixation should be performed in a fume hood as 4% PFA solution is a potent harmful chemical if inhaled.
   1. Anesthetize the mouse with isoflurane, place it in a supine position, and confirm the complete absence of responsiveness by gently pinching the extremities.
      NOTE: The mouse is deeply anesthetized with isoflurane in an induction chamber in accordance with institutional guidelines. For terminal perfusion, anesthesia is continuously maintained by transferring the animal to a nose cone delivering inhalant anesthetic while positioned supine. Adequate depth of anesthesia is confirmed prior to thoracotomy by the complete absence of response to gentle pinching of the extremities. Inhalant anesthesia is maintained without interruption until perfusion is completed and death is ensured.
   2. Once the mouse is anesthetized, make an incision using surgical scissors from the caudal abdomen to the xiphoid. Then, make two lateral incisions along either side of the sternum and retract the chest wall anteriorly to expose the heart and major vessels.
   3. Open the right atrium via sharp dissection to decompress the circulatory system. Insert a 25 G butterfly needle into the left ventricle and infuse 50 mL of PBS at a rate of approximately 10 mL/min, directing the flow toward the ascending aorta.
      NOTE: Exsanguinating is important to obtain high-quality immunofluorescence (IF) images because blood exhibits autofluorescence. To ensure complete removal of blood, 50 mL of PBS was used in this study, although 20-30 mL is commonly used.
   4. Infuse 20 mL of 4% PFA solution prepared in step 7.1.2 at the same rate, ensuring no air is introduced during the switch of solutions.
   5. Harvest tissues and keep them at 4 °C overnight in a 15 mL polypropylene tube filled with 4% PFA solution for post-fixation. When fluorescent protein is expressed in the collected tissue, cover the tube with aluminum foil to shield it from light.
3. Cryoembedding
   1. Transfer the tissue to a 15 mL polypropylene tube filled with 30% sucrose solution prepared in step 7.1.3 for cryoprotection.
   2. Keep at 4 °C until tissues completely sink to the bottom of the tubes.
   3. Transfer the tissue to a plastic cryomold with standard pattern forceps.
   4. Place the OCT compound over the tissue while ensuring that the tissue is correctly oriented and flush with the bottom of the cryomold.
   5. Fill the top with OCT compound. Remove any bubbles with a pipette tip.
   6. Place the plastic cryomold in a cryogenic bath containing 2-methylbutane, chilled with crushed dry ice in an ice bucket, until the OCT compound is completely frozen (the OCT compound will turn white).
   7. Store the tissue blocks at -80 °C until use.
4. Cryosectioning
   1. Transfer the frozen tissue blocks to a cryostat microtome at -20 °C prior to sectioning and allow the temperature of the frozen tissue blocks to equilibrate.
   2. Section each frozen tissue block into a desired thickness (typically 5 µm) using the cryostat microtome. Place the tissue sections onto a microscope glass slide for immunofluorescence staining.
   3. Sections can be stored at -80 °C for later use.

8. Immunofluorescence staining

1. Dry frozen slides at room temperature for 30 min.
2. Isolate the section with a liquid blocker pen and dry at room temperature for 5 min.
3. Rehydrate sections with 150 µL of PBS for 10 min.
4. Place the slides on a flat surface and add 150 µL of 1% bovine serum albumin (BSA) diluted with PBS per section onto the slides. Incubate at room temperature for 45 min for blocking.
5. Add 150 µL of primary antibody solution (anti-Wilms Tumor 1 (WT1) antibody, diluted to 1:100 with 1% BSA; anti-Na-K-2Cl cotransporter 2 (NKCC2) antibody, diluted to 1:100 with 1% BSA; or anti-aquaporin 2 (AQP2) antibody, diluted to 1:100 with 1% BSA) and incubate at 4 °C overnight under a humid chamber.
6. Wash the sections with PBS at room temperature for 5 min, 3 times.
7. Add 150 µL of secondary antibody (anti-rabbit IgG Alexa Fluor 647 antibody, diluted to 1:500 with 1% BSA) and Lotus tetragonolobus lectin (LTL)-fluorescein (diluted to 1:200 with 1% BSA), and incubate at room temperature for 60 min.
8. Wash the sections with PBS at room temperature for 5 min.
9. Place the slides on a flat surface and add 150 µL of Hoechst 33342 (diluted to 1:10,000 with PBS) for nuclear staining. Incubate at room temperature for 10 min, shielded from light.
10. Wash the sections with PBS at room temperature for 5 min.
11. Mount the slides with 20 µL of the mounting medium and a cover glass. Keep the slides in the dark at 4 °C until imaging is performed.

9. DNA extraction from tissue

1. DNA extraction from tissues using a commercially available kit (see Table of Materials).
   1. Transfer the tissue into a ceramic bead tube and add 400 µL of DNA extraction buffer (100 mM NaCl, 20 mM EDTA, 1% SDS, and 50 mM Tris-HCl, pH 8.0).
   2. Homogenize the tissue at a speed setting of 5 m/s for 20 s in 3 cycles using a homogenizer.
   3. Centrifuge at 2,000 x g at room temperature for 2 min.
   4. Collect 200 µL of the supernatant using a P200 pipette and add 10 µL of Proteinase K solution (final concentration of 1 mg/mL).
   5. Mix by vortexing for 15 s and incubate at 55 °C for 1 h.
   6. Add 200 µL of lysis buffer from the DNA extraction kit.
   7. Mix by vortexing for 15 s and incubate at 55 °C for 15 min.
   8. Add 250 µL of 100% ethanol.
   9. Mix by vortexing for 15 s and incubate at room temperature for 5 min.
   10. Apply the lysate onto the DNA extraction spin column.
   11. Centrifuge at 6,000 x g at room temperature for 1 min.
   12. Place the spin column in a new 2 mL collection tube. Add 500 µL of the first wash buffer from the DNA extraction kit.
   13. Centrifuge at 6,000 x g at room temperature for 1 min.
   14. Place the spin column in a new collection tube. Add 500 µL of the second wash buffer from the DNA extraction kit.
   15. Centrifuge at 6,000 x g at room temperature for 1 min.
   16. Place the spin column in a new 2 mL collection tube. Add 500 µL of 100% ethanol.
   17. Centrifuge at 6,000 x g at room temperature for 1 min.
   18. Place the spin column in a new 2 mL collection tube and centrifuge at 20,000 x g at room temperature for 3 min.
       NOTE: This step ensures complete drying of the spin column membrane, as ethanol carryover into the eluate can interfere with downstream applications.
   19. Place the spin column in a new 1.5 mL microcentrifuge tube. Add 35 µL of elution buffer from the DNA extraction kit to the spin column.
       NOTE: The volume of recovered eluate is typically about 5 µL less than the volume of elution buffer initially applied to the column.
   20. Incubate at room temperature for 5 min.
   21. Centrifuge at 20,000 x g at room temperature for 1 min.
   22. Measure the DNA concentration.
2. Vector genome copy number quantification by quantitative PCR (qPCR)
   1. Dilute the extracted DNA prepared in step 9.1 to a concentration of 25 ng/µL with molecular-grade water.
   2. Design a qPCR primer pair for the vector genome quantification (tdTomato was used in this study) and a qPCR primer pair for the mouse reference gene that will be used to quantify the diploid genomes in the extracted DNA samples (the mouse agouti gene was used in this study).
      NOTE: The primer sequences used in this study are as follows;
      tdTomato forward: 5'-TATGAACGGAGCGAGGGGAG-3'
      tdTomato reverse: 5'-TTCCACTAGACCCTGAGCCG-3'
      mouse agouti forward: 5'-GGCGTGGTCAGTGGTTGTG-3'
      ​mouse agouti reverse: 5'-TTTAGCTTCCACTAGGTTTCCTAGAAA-3'
   3. For a 25 µL qPCR reaction, mix 4 µL of diluted DNA sample in step 9.2.1 (a total of 100 ng DNA in a 25 µL reaction mixture), 12.5 µL of 2x qPCR reaction buffer, 0.5 µL of both 20 µM forward and reverse primes prepared in step 9.2.2 (final concentration of 400 nM of each primer), and 7.5 µL of molecular grade water.
   4. Measure the DNA concentration of plasmid standards for qPCR four times and determine the average value of the DNA concentration.
      NOTE: Supercoiled circular pAAV-CAG-tdTomato plasmid and a supercoiled circular plasmid containing the PCR target sequence derived from the mouse agouti gene were used for vector genome quantification and normalization, respectively, in this study.
   5. Prepare a series of 1:10 serial dilutions of plasmid DNA standards covering a dynamic range of 10⁶.
      NOTE: Each sample and standard should be analyzed at least in duplicate.
   6. Perform qPCR using the following thermal cycling conditions: an initial heat activation step at 95 °C for 10 min, followed by 35 cycles of denaturation at 95 °C for 15 s, and annealing/extension at 60 °C for 30 s.
   7. After completing the qPCR reaction and obtaining the double-stranded (ds) copy number values of vector genomes and the mouse reference gene in the samples, determine ds vector genome copy numbers per diploid genomic equivalent (ds-vg/dge) by normalizing the ds vector genome copy numbers to the reference gene copy number in each sample based on standard curves generated from known concentrations of both the vector genome and the reference gene.
      NOTE: Ds-vg/dge can be determined based on the following approximation that 100 ng of mouse genomic DNA represents 16,900 diploid (2N) mouse cells (i.e., 5.92 pg of DNA per diploid mouse cell).

Results

A recent study has identified a small subset of AAV capsids that exhibits remarkable enhancement of renal transduction compared to AAV9 following RP injection³⁸. The identified subset includes AAV-KP3, a chimeric AAV capsid generated by DNA shuffling⁴¹. Therefore, AAV-KP3 was chosen for RP injection in this study, along with AAV9 as the benchmark. AAV9-CAG-tdTomato vector (AAV9) and AAV-KP3-CAG-tdTomato vector (KP3) were produced in HEK293 cells by an adenovirus-free plasmid transfection method and purified by CsCl density-gradient ultracentrifugation (CsCl AAV) or centrifugal ultrafiltration (CU) with 100 kDa molecular weight cut-off filters (CU AAV). The procedure for preparing CsCl AAVs has been previously described⁴⁷.

Briefly, the 5-day culture media containing cell pellets that were frozen -80 °C (step 1.3.11) were thawed and underwent PEG precipitation, Sarkosyl resuspension, Serratia marcescens endonuclease treatment, and two rounds of CsCl density gradient ultracentrifugation. CsCl density gradient fractions containing AAV were identified by refractive index and DNA dot blot assay, pooled, dialyzed against 5% sorbitol in PBS, and filter sterilized. In the CU procedure, intermediate process samples (Fractions 1 to 7) were collected as well as the final CU AAV preparations (Fraction 8). The titration of nuclease-resistant AAV vector genome (vg) DNA in each fraction was performed by a quantitative DNA dot blot assay⁴⁶ (Figure 2A). AAV9 and KP3 vector titers in the crude lysates obtained from four T225 flasks (Fraction 1) were 1.6 x 10¹³ vg (168 mL of 9.5 x 10¹⁰ vg/mL) and 6.0 x 10¹² vg (168 mL of 3.6 x 10¹⁰ vg/mL), respectively (Figure 2B, Fraction 1). After concentration of crude lysates and buffer exchange to 5% sorbitol in PBS, CU-purified AAV9 and KP3 vectors were concentrated to final volumes of approximately 0.3 mL with 3.2 x 10¹³ vg/mL and 8.8 x 10¹² vg/mL, respectively (Figure 2B, Fraction 8).

To assess the purity of the CsCl AAV and CU AAV vectors, we performed silver staining accompanied by a western blot to detect VP protein bands (Figure 3). The CsCl AAV vectors loaded at 1 x 10¹⁰ vg also showed VP1, VP2 and VP3 bands clearly with minimal background, indicating high purity (Figure 3A). The CU AAV vectors loaded at 1 x 10¹⁰ vg showed a smearing pattern showing a high level of protein contamination (Figure 3B), while the three bands comigrating with VP1, VP2 and VP3 protein bands identified by western blot were clearly visible with minimal background when a lower amount of the CU AAV vectors (1 x 10⁹ vg) was loaded (Figure 3C). This observation indicates that the predominant proteins contained in the CU AAV vector preparations were indeed AAV capsid proteins although there were more empty capsids in CU AAV vectors than in CsCl AAV vectors.

Each of the two CU AAV vectors (CU AAV9 and CU KP3) was then injected into eight-week-old C57BL/6J male mice by RP injection at a dose of 3.0 x 10¹¹ vg/mouse delivered in 50 µL of vector solution. During the RP injection procedure, mice were placed in a lateral position, and the left kidney was exposed (Figure 1A). To increase the AAV vector exposure for enhancing AAV transduction, the renal artery, the renal vein, and the ureter were clamped during AAV vector injection (Figure 1B). The injection into the renal pelvis was performed at a low speed (50 µL/min) using a syringe pump to prevent injection-related damage in the renal parenchyma (Figure 1C, D). Two weeks post-injection, injected kidneys were harvested, and the AAV transduction efficiency in the kidneys was assessed by immunofluorescence microscopy. The microscopic analysis revealed that RP injection of the CU KP3 vector achieved transduction in the proximal tubules in the kidney, whereas the CU AAV9 vector transduced only mesangial cells in the glomeruli, with renal tubular transduction barely observed (Figure 4A). The percentages of tdTomato-positive areas in the proximal tubules of KP3 and AAV9 were 12.9% and 0.3%, respectively. Importantly, there was no remarkable difference in renal tubular transduction pattern between CU AAV vectors and CsCl AAV vectors (Figure 4B). The kidney section from mice injected with 50 µL of 5% sorbitol in PBS was used as a negative vehicle control (Figure 4C).

Quantification of vector genomes in the kidney using DNA qPCR showed that there were 73 times more vector genomes in the CU KP3-injected kidneys than the CU AAV9-injected kidneys (Figure 4D). The average vector genome copy number in the CU KP3-injected kidneys was 73% of that of CsCl KP3-injected kidneys, demonstrating that, while CU AAV vectors are slightly less active with statistical significance (adjusted p < 0.0001, a two-way ANOVA followed by Tukey's post hoc test), they remain effective at transducing cells. Additional fluorescence microscopic analysis revealed that the KP3 vector also transduced collecting duct cells (Figure 5A) and AAV9 transduced thick ascending limb cells (Figure 5B), as previously reported³⁸. While off-target liver transduction was not completely prevented by RP injections, the KP3 vector showed much less liver transduction than the AAV9 vector (Figure 5C). Finally, hematoxylin and eosin (H&E) staining demonstrated that there was no apparent tissue damage in the RP-injected kidneys (Figure 6).

Figure 1: Illustration of surgical procedures for slow retrograde renal pelvis (RP) injection of AAV vectors. (A) The left kidney was exposed in the lateral position. In this orientation, the right side corresponds to the head of the mouse, while the bottom side is the back of the mouse. (B) Two microvessel clips were used to clamp the urinary tract (1) and both the renal artery and vein (2). Please note that the darker color of the kidney after clamping is a result of the temporary blockade of the renal blood flow. (C) The 30 G needle tip (3) was inserted into the renal pelvis. (D) Injection was performed using a device connecting the 30 G needle tip (4), PE-10 polyethylene tube (5), silicone tube (6), 30 G needle (7), 100 μL gas-tight glass syringe (8), and a syringe pump (9) in this order. Please click here to view a larger version of this figure.

Figure 2: The titration of centrifugally ultrafiltered (CU) AAV vectors using a quantitative DNA dot blot assay. (A) A DNA dot blot membrane was hybridized with a ³²P-labeled CAG promoter probe. pAAV-CAG-tdTomato plasmid DNA was used as the DNA standard (STD). Centrifugally ultrafiltered (CU) AAV9-CAG-tdTomato vector (AAV9), AAV-KP3-CAG-tdTomato vector (KP3), and their intermediate process samples collected during the CU procedure were loaded as follows: crude lysate sample fractions as Fraction 1, 0.45 µm-filtrated sample fractions as Fraction 2, centrifugal ultrafiltrate flow-through fractions during the concentration process as Fraction 3 to Fraction 5, centrifugal ultrafiltrate flow-through fractions during the buffer exchange process as Fraction 6 and Fraction 7, and CU AAV vector as Fraction 8. The volumes of each Fraction used for each dot were as follows: 10 µL for Fractions 1 and 2, 100 µL for Fractions 3 to 7, and 0.06 µL for Fraction 8. Original samples with no dilution are marked as 1x, and 10 times diluted samples are marked as 0.1x. (B) AAV vector concentrations in Fractions 1 to 8 collected during the CU AAV purification process. The concentrations of AAV9 and KP3 vectors (vg/mL) in each Fraction were determined based on the signal intensity of the data shown in (A). Please click here to view a larger version of this figure.

Figure 3: Quality assessment of AAV vector preparations by SDS-PAGE. CsCl AAV and CU AAV preparations (1 x 10⁹ or 1 x 10¹⁰ vg/lane as indicated in the figure) were electrophoresed on 7.5% SDS-polyacrylamide gels under reducing conditions, followed by silver staining (A-C) or by western blot probed with a rabbit polyclonal anti-VP antibody (D). Arrowheads indicate positions of VP1, VP2, and VP3. Please click here to view a larger version of this figure.

Figure 4: AAV vector transduction in the RP-injected kidneys. Representative fluorescence microscopic images of the kidneys of mice treated with RP injection of (A) centrifugally ultrafiltered AAV-KP3-CAG-tdTomato or AAV9-CAG-tdTomato vector (CU KP3 or CU AAV9, respectively), (B) CsCl-purified AAV-KP3-CAG-tdTomato vector or AAV9-CAG-tdTomato vector (CsCl KP3 or CsCl AAV9, respectively), or (C) 5% sorbitol in PBS as a negative vehicle control (Vehicle). Green, magenta, yellow, and blue signals represent Lotus tetragonolobus lectin (LTL, a marker for proximal tubules), Wilms Tumor 1 (WT1, a marker for podocytes), native tdTomato (tdT) fluorescence, and nuclei stained with Hoechst 33342, respectively. Scale bars: 50 µm. (D) Quantification of AAV vector genome copy numbers in the kidneys using DNA qPCR. Data are presented as double-stranded vector genome copies per diploid genomic equivalent (ds-vg/dge). Error bars indicate the standard error of the mean (SEM). Statistical analysis was performed using two-way ANOVA followed by Tukey's post hoc test. ****, adjusted p < 0.0001. Please click here to view a larger version of this figure.

Figure 5: Assessment of vector transduction in the kidney and liver tissues from the mice treated with RP injection of AAV vectors. Representative fluorescence microscopic images of the kidneys (A,B) and livers (C) from the mice treated with RP injection of centrifugally ultrafiltered AAV-KP3-CAG-tdTomato vector (KP3), centrifugally ultrafiltered AAV9-CAG-tdTomato vector (AAV9), or 5% sorbitol in PBS as a vehicle control (Vehicle). In (A), magenta, green, yellow, and blue signals represent Lotus tetragonolobus lectin (LTL, a marker for proximal tubules), aquaporin 2 (AQP2, a marker for collecting duct cells), native tdTomato (tdT) fluorescence, and nuclei stained with Hoechst 33342, respectively. White arrows indicate AAV-KP3 vector-transduced collecting duct cells. In (B), magenta, green, yellow, and blue signals represent LTL, Na-K-2Cl cotransporter 2 (NKCC2, a marker for thick ascending limb cells), native tdT fluorescence, and nuclei, respectively. The white arrow indicates AAV9 vector-transduced thick ascending limb cells. In (C), yellow and blue signals represent native tdT fluorescence and nuclei, respectively. Scale bars: 50 µm. Please click here to view a larger version of this figure.

Figure 6: Hematoxylin and eosin (H&E) staining of the RP-injected kidney tissues. H&E-stained images of the pelvis, papilla, and cortex of the kidneys from the mice treated with RP injection of centrifugally ultrafiltered AAV-KP3-CAG-tdTomato vector (KP3) or 5% sorbitol in PBS as a vehicle control (Vehicle injection) or left untreated (No injection). Asterisks indicate the renal pelvic cavity. Scale bars: 50 µm. Please click here to view a larger version of this figure.

Supplementary File 1: Plasmid sequences used in this study. Plasmids used for the production of the AAV9-CAG-tdTomato vector: pAAV-CAG-tdTomato, pRepCap-AAV9, and pHelper. Plasmids used for the production of the AAV-KP3-CAG-tdTomato vector: pAAV-CAG-tdTomato, pRepCap-KP3, and pHelper. Please click here to download this File.

Discussion

This report has detailed slow retrograde RP injection of AAV vectors as a local administration method for gene delivery to the mouse kidney. One critical step for safely performing RP injection is the adjustment of the injection rate of AAV vectors to avoid parenchymal damage to the injected kidneys. In a previous report on hydrodynamic renal pelvis injection, 100 µL of plasmid DNA solution was injected into the mouse kidney within only 3 s²⁴. This fast injection is essential for increasing the transient cell membrane permeability for effective DNA transfection but can result in tissue injury²⁴. To mitigate this risk, vectors were injected slowly via RP at a controlled rate of 50 µL/min using a syringe pump. This procedure could successfully prevent tissue damage in the RP-injected kidney.

Another critical step is to completely occlude the renal artery and vein at the time of injection, which can increase the exposure to injected AAV vectors and enhance their transduction efficiency. It has been previously shown that RP injection without blockade of the renal blood flow resulted in reduced kidney transduction in mice³⁸. For confirming the proper clamping of the renal artery and vein, rapid darkening of the kidney can be a useful visual clue. It has also been previously demonstrated that there was no difference in transduction efficiency of RP injection of AAV-KP1 vectors between 5-min, 10-min, and 15-min blockade of the renal arterial and venous flow³⁸. Therefore, the RP injection shown here was performed by a total of 8-min blockade (2-min blockade before the start of RP injection, 1-min blockade during the RP injection, and another 5-min blockade after RP injection for dwelling). This resulted in effective AAV-KP3 vector transduction in the injected kidney. Since the local retention efficiency of AAV vectors may vary depending on the AAV capsid used, it might be necessary to determine the optimal dwelling time for RP injection for each distinct AAV capsid.

According to the previous studies assessing renal transduction of common AAV serotypes following IV^28,30 and RP^{15,16,18,20,21} injections, it has been generally considered that AAV9 is one of the most efficient capsids regardless of administration routes. However, a recent study has demonstrated that effective AAV capsids are not necessarily the same between systemic IV injection and local injections³⁸. The following six AAV capsids, AAV-KP1, AAV-KP2, AAV-KP3⁴¹, AAV-DJ⁴², AAV2G9⁴³, and AAV2.7m8⁴⁴, have been shown to outperform AAV9 only when administered by RV or RP injection³⁸. The identification of a small subset of AAV capsids capable of significantly enhancing renal transduction through local vector administration, such as retrograde renal pelvis (RP) injection, marks a major advancement in renal gene transfer. This finding paves the way for the development of more efficient strategies for kidney-targeted gene delivery and, ultimately, gene therapy for kidney diseases.

The RP injection-based gene delivery technology can be employed not only for AAV-mediated kidney gene therapy research but also for the generation of genetically engineered mice without time-consuming transgenesis and breeding. In particular, the establishment of conditional knock-in and knock-out mice typically involves the cumbersome process of crossing transgenic mice carrying the gene of interest flanked by loxP sequences (floxed mice) and mice expressing Cre recombinase under a cell type-specific promoter of interest⁴⁸. By contrast, the injection of AAV-Cre vectors into floxed mice enabled rapid cell type-specific gene knock-in and knock-out in adult mice^49,50. Besides the Cre-based technologies, the cluster regularly interspaced short palindromic repeats (CRISPR)-Cas9 system has emerged as a powerful tool for genome editing^51,52. A recent study demonstrated that AAV-mediated gRNA delivery into Cas9-expressing mice could establish knock-out mice without breeding⁵³. Therefore, gene delivery via RP injection will greatly facilitate the study of genetic modifications targeting the kidneys in mice and help further the understanding of renal molecular biology. Furthermore, it has been recently demonstrated that slow retrograde RP injection of AAV vectors can be safely performed in non-human primates, resulting in robust proximal tubule transduction even in the presence of pre-existing anti-AAV neutralizing antibodies and without the need for renal blood flow blockade³⁸. This underscores the promising translatability of slow retrograde RP injection as an effective and safe method for renal gene transfer in clinical settings.

Conventional AAV purification methods involve time-consuming processes and require expertise and specialized equipment such as an ultracentrifuge or chromatography, which may prevent small laboratories from preparing AAV vectors on their own^54,55,56. Here, a simplified AAV purification method using centrifugal ultrafiltration was employed to prepare AAV vector preparations on a small scale (i.e., AAV mini preps)⁴⁵, sufficient for generating preliminary data in mouse experiments. Although four T225 flasks were used in this study, two T225 flasks would have been sufficient to produce the AAV9 and AAV-KP1 vectors required for RP injections at a dose of 3.0 x 10¹¹ vg/mouse in small groups of mice. This method requires only common equipment and reagents that are readily available in most laboratories. However, the CU process is not effective in eliminating empty capsids or cellular contaminants, as demonstrated by the quality assessment of the CU AAVs, resulting in a modest reduction in transduction efficiency in the kidney. In addition, the higher levels of contamination of empty capsids and host cell-derived human proteins in CU AAVs compared to CsCl AAVs may elicit stronger immune responses when administered at equivalent vector genome doses. Despite these limitations, no significant differences were observed in the transduction profiles between CU AAVs and CsCl AAVs in the kidney and other organs. Therefore, RP injection of CU AAV miniprep vectors is a valuable method for quickly assessing the kidney transduction efficacy of individual AAV vectors and generating mouse models. This approach has the potential to accelerate the identification of renal cell type-tropic AAV capsids and the generation of preliminary yet crucial data for kidney research. In summary, this report highlights the utility and safety of slow retrograde RP injection of AAV vectors as an efficient local gene delivery platform for the mouse kidney.

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
0.45 μm PVDF syringe filter	MilliporeSigma	SLHVR33RS
1.5 mL microcentrifuge tube	Fisher Scientific	05-408-129
15 mL polypropylene tube	Corning	352097
16% paraformaldehyde (PFA)	Electron Microscopy Sciences	15710
250 mL centrifuge bottle	Fisher Scientific	055641S24
2F Silicone tubing	Sai Infusion Technologies	SIL-2-50
25G butterfly needle	Medex Supply	26708
2-methylbutane	MilliporeSigma	MX0760
2x qPCR reaction buffer	Fisher Scientific	43-676-59	Power SYBR Green PCR Master Mix
30G needle	Becton-Dickinson & Co	305106
50 mL polypropylene tube	Corning	430291
5-0 monofilament suture	Ethicon	Y463G
5 M NaCl	Lonza	51202
6-0 absorbable suture	Ethicon	J489G
7.5% Mini-PROTEAN TGX Precast Protein Gels, 10-well, 30 µL	Bio-Rad	4561023
Amersham ImageQuant 800 Western Blot Imaging System (Cytiva)	Cytiva	29459405
Amicon 100 kDa MWCO, 15 mL tube	Millipore Sigma	UFC910024
Anti-aquaporin 2 (AQP2) antibody	Abcam	ab199975	A marker for collecting duct cells
Anti-Na-K-2Cl cotransporter 2 (NKCC2) antibody	StressMarq Biosciences	SPC-401	A marker for thick ascending limb cells
Anti-rabbit IgG Alexa Fluor 647 antibody	Jackson ImmunoResearch	111-605-144
Anti-Wilms Tumor 1 (WT1) antibody	Abcam	ab89901	A marker for podocytes
Beckman Coulter OptiSeal Tube, Polypropylene, 29.9 mL	Fisher Scientific	NC9691210
Beckman Coulter OPTISEAL TUBES	Fisher Scientific	NC9611575
Benzonase	Sigma-Aldrich	1016970001
Bovine serum albumin (BSA)	Sigma-Aldrich	5470
Centrifugal ultrafiltration tube	MilliporeSigma	UFC9100	Amicon 100 kDa MWCO, 15 mL
Ceramic bead tube	Fisher Scientific	15-340-154
Cesium chloride	Sigma-Aldrich	C3032-1KG
Clamp applying forceps	Fine Science Tools	00071-14
Closed-loop heat pad	Stryker Medical	8002-062-012
Cotton-tipped applicator	Puritan	806-WC
Cover glass	Fisher Scientific	12-541B
Cryostat microtome	Tanner Scientific	TN50
Curved forceps	Fine Science Tools	11052-10
Curved-type micro vessel clip	Kleinert-Kutz	65145	Clamp for the renal artery and renal vein
Dialysis cassette	Fisher Scientific	PIA52976
DNA extraction kit	Qiagen	57704	QIAamp MinElute Virus Spin Kit
D-Sorbitol	Sigma-Aldrich	S6021-1KG
Dulbecco's Modified Eagle Medium- high glucose	Corning	15-013-CV	DMEM-high glucose (4.5 g/L)
EDTA	Fisher Scientific	15-575-020
Ethanol	Decon Laboratories	2716GEA
Fetal bovine serum (FBS)	VWR	89510-186
Gas-tight glass syringe	Hamilton	1710 TLL
Glutaraldehyde solution	Sigma-Aldrich	G6257
Heat therapy pump	Kent Scientific	HTP-1500
HEPES	Sigma Aldrich	H4034-100G
Hoechst 33342	Invitrogen	H3570
Homogenizer	Fisher Scientific	15-340-163
Human embryonic kidney (HEK) 293 cells	Agilent	240073
Ice bucket	Fisher Scientific	07-210-104
Infusion syringe pump	Harvard Apparatus	70-4507
Isoflurane	Piramal Critical Care	NDC 66794-017-10
L-glutamine	Fisher Scientific	25030-081
Liquid blocker pen	Ted Pella	22309
Lotus tetragonolobus lectin (LTL)- Fluorescein	Vector Laboratories	FL-1321-2	A marker for proximal tubules
Magnesium chloride (MgCl2)	MilliporeSigma	M1028
Meloxicam	VetOne	NDC 86136-012-10	Analgesic
Microscope glass slide	Fisher Scientific	12-550-15
Mounting medium	SouthernBiotech	0100-01
Needle holder	Fine Science Tools	12501-13
OCT compound	Sakura Finetek	4583
Ophthalmic ointment	Bausch & Lomb	NDC 24208-313-34
PE-10 polyethylene tube	BD	427401	Inner diameter: 0.28 mm
Penicillin & Streptomycin Mix	Fisher Scientific	15140-122
Phosphate-buffered saline (PBS)	Fisher Scientific	10010-049
Plastic cryomold	Sakura Finetek	4566
Poloxamer 188 Non-ionic Surfactant (10%)	Fisher Scientific	24-040-032
Polyethylene Glycol 8000	Fisher Scientific	BP233-1
Polyethylenimine (PEI)	Polysciences	23966
Povidone iodine	Ricca Chemical Company	395516
Power Blotter–Semi-dry Transfer System	Fisher Scientific	PB0010
Proteinase K solution	Fisher Scientific	25530049
Rabbit anti-Anti-Adeno-Associated Virus (AAV), VP1, VP2, VP3	American Research Products	03-61084
Sarkosyl	Fisher Scientific	BP234-500
Screw cap 2 mL tube	Corning Axygen	SCT-200-SS-O-S
Silver Stain Kit	Sigma-Aldrich	PROTSIL1-1KT
Sodium chloride (NaCl)	Fisher Scientific	S271
Sodium phosphate dibasic (Na2HPO4)	MilliporeSigma	S0876
Sodium phosphate monobasic (NaH2PO4)	MilliporeSigma	S0751
Sorbitol	MilliporeSigma	S6021
Standard pattern forceps	Fine Science Tools	11000-12
Sterile surgical drape	SAI Infusion Technologies	PSS5-1519
Straight-type micro vessel clip	Fine Science Tools	00396-01	Clamp for the ureter
Sucrose	Fisher Scientific	S5
Surgical scissors	Fine Science Tools	14058-09
T225 flask	Corning	431082
Technocut Scalpel #11	Myco Medical Supplies Inc.	6008T-11
Tissue forceps	Roboz Surgical Instrument	RS-5155