Expression of Transgenes in Native Bladder Urothelium Using Adenovirus-Mediated Transduction

Summary

Methods are described for the generation of large amounts of recombinant adenoviruses, which can then be used to transduce the native rodent urothelium allowing for expression of transgenes or downregulation of endogenous gene products.

Abstract

In addition to forming a high-resistance barrier, the urothelium lining the renal pelvis, ureters, bladder, and proximal urethra is hypothesized to sense and transmit information about its environment to the underlying tissues, promoting voiding function and behavior. Disruption of the urothelial barrier, or its sensory/transducer function, can lead to disease. Studying these complex events is hampered by lack of simple strategies to alter gene and protein expression in the urothelium. Methods are described here that allow investigators to generate large amounts of high-titer adenovirus, which can then be used to transduce rodent urothelium with high efficiency, and in a relatively straightforward manner. Both cDNAs and small interfering RNAs can be expressed using adenoviral transduction, and the impact of transgene expression on urothelial function can be assessed 12 h to several days later. These methods have broad applicability to studies of normal and abnormal urothelial biology using mouse or rat animal models.

Introduction

The urothelium is the specialized epithelium that lines the renal pelvis, ureters, bladder, and proximal urethra¹. It comprises three strata: a layer of highly differentiated and polarized often bi-nucleate umbrella cells, whose apical surfaces are bathed in urine; an intermediate cell layer with a population of bi-nucleate transit-amplifying cells that can give rise to superficial umbrella cells in response to their acute loss; and a single layer of basal cells, a subset of which function as stem cells that can regenerate the entirety of the urothelium in response to chronic injury. Umbrella cells are chiefly responsible for forming the high-resistance urothelial barrier, components of which include an apical membrane (rich in cholesterol and cerebrosides) with low permeability to water and solutes, and a high-resistance apical junctional complex (comprised of tight junctions, adherens junctions, desmosomes, and an associated actomyosin ring)¹. Both the apical surface of the umbrella cell and its junctional ring expand during bladder filling and return to their pre-filled state rapidly after voiding^1,2,3,4,5. In addition to its role in barrier function, the urothelium is also hypothesized to have sensory and transducer functions that allow it to sense changes in the extracellular milieu (e.g., stretch) and transmit this information via release of mediators (including ATP, adenosine, and acetylcholine) to underlying tissues, including suburothelial afferent nerve processes^6,7,8. Recent evidence of this role is found in mice lacking urothelial expression of both Piezo1 and Piezo2, which results in altered voiding function⁹. Additionally, rats overexpressing the tight-junction pore-forming protein CLDN2 in the umbrella cell layer develop inflammation and pain analogous to that seen in patients with interstitial cystitis¹⁰. It is hypothesized that disruption of urothelial sensory/transducer or barrier function may contribute to several bladder disorders^6,11.

A better understanding of the biology of the urothelium in normal and disease states depends on the availability of tools that will allow investigators to readily downregulate endogenous gene expression or allow for the expression of transgenes in the native tissue. While one approach to downregulate gene expression is to generate conditional urothelial knockout mice, this approach depends on the availability of mice with floxed alleles, is labor intensive, and can take months to years to complete¹². Not surprisingly then, investigators have developed techniques to transfect or transduce the urothelium, which can lead to results on a shorter time scale. Published methods for transfection include the use of cationic lipids¹³, anti-sense phosphorothioated oligodeoxynucleotides¹⁴, or antisense nucleic acids tethered to the HIV TAT protein penetrating 11-mer peptide¹⁵. However, the focus of this protocol is on the use of adenoviral-mediated transduction, a well-studied methodology that is efficient at gene delivery to a broad range of cells, has been tested in numerous clinical trials, and most recently was used to deliver the cDNA encoding the COVID-19 capsid protein to recipients of one variant of the COVID-19 vaccine^16,17. For a more thorough description of the adenovirus life cycle, adenoviral vectors, and clinical applications of adenoviruses, the reader is directed to reference¹⁷.

An important milestone in the use of adenoviruses to transduce the urothelium, was a report by Ramesh et al. that showed short pretreatments with detergents, including N-dodecyl-β-D-maltoside (DDM) dramatically enhanced transduction of the urothelium by an adenovirus encoding β-galactosidase¹⁸. Using this proof-of-principle study as a guide, adenoviral-mediated transduction of the urothelium has now been used to express a variety of proteins, including Rab-family GTPases, guanine-nucleotide exchange factors, myosin motor fragments, pore-forming tight junction-associated claudins, and ADAM17^{10,19,20,21,22}. The same approach was adapted to express small interfering RNAs (siRNA), the effects of which were rescued by co-expressing siRNA-resistant variants of the transgene²². The protocol described here includes general methods to generate large amounts of highly concentrated adenovirus, a requirement for these techniques, as well adaptations of the methods of Ramesh et al.¹⁸ to express transgenes in the urothelium with high efficiency.

Protocol

Experiments involving the generation of adenoviruses, which requires BSL2 certification, were performed under approval from the University of Pittsburgh Environmental Health and Safety offices and the Institutional Biosafety Committee. All animal experiments performed, including adenoviral transduction (which requires ABSL2 certification), were done in accordance with relevant guidelines/regulations of the Public Health Service Policy on Humane Care and Use of Laboratory Animals and the Animal Welfare Act, and under the approval of the University of Pittsburgh Institutional Animal Care and Use Committee. Gloves, eye protection, and appropriate garb are worn for all procedures involving recombinant viruses. Any liquid or solid waste should be disposed of as described below. The bedding of the animals post transduction, and any resulting animal carcasses, should be treated as biohazardous materials and disposed of according to institutional policies.

1. Preparation of high-titer adenovirus stocks

NOTE: Effective transduction of rodent bladders depends on the use of purified and concentrated viral stocks, typically 1 x 10⁷ to 1 x 10⁸ infectious viral particles (IVP) per µL. This portion of the protocol is focused on generating high-titer adenovirus stocks from existing virus preparations. All steps should be performed in a cell culture hood using sterile reagents and tools. While the available strains of adenovirus used today are replication defective, most institutions require approval to use adenoviruses and recombinant DNA. This often includes designation of a cell culture room as a BSL2 approved facility to produce and amplify adenoviruses. Some general considerations include use of masks, eye protection, gloves, and appropriate garb at all stages of virus production and purification. When performing centrifugation, safety caps are recommended if the centrifuge tubes lack tight-fitting caps. All non-disposable materials, including potentially contaminated centrifuge safety caps, bottles, and rotors are treated with an antiviral solution (see Table of Materials), and then rinsed with water or 70% ethanol. Liquid wastes are treated by adding bleach to a final concentration of 10% (v/v). Disposal of these liquid wastes will depend on institutional policies. Solid wastes are typically disposed of in biohazardous waste.

1. Culture HEK293T cells
   1. Thaw a frozen vial of HEK293T cells in a water bath at 37 °C and use a 5 mL pipette to transfer the cells to a 15 cm diameter cell culture dish. Using a 25 mL pipette, slowly add 20 mL of Dulbecco's Modified Eagle Medium (DMEM) containing 10% (v/v) fetal bovine serum and penicillin/streptomycin antibiotic (DMEM-FBS-PS) to the dish. Incubate the cells in a 37 °C cell-culture incubator gassed with 5% (v/v) CO₂ until they reach 80%-90% confluence (~2 x 10⁷ cells).
   2. Using a glass pipette attached to a vacuum source, aspirate the medium, and then rinse the cells by transferring 20 mL of sterile PBS (2.7 mM KCl, 1.5 mM KH₂PO₄, 136.9 mM NaCl, and 8.9 mM Na₂HPO₄) to the dish using a 25 mL pipette. Aspirate the spent PBS, and then use a 5 mL pipette to transfer 3 mL of warm (37 °C) proteinase solution (see Table of Materials) to the dish, incubating the dish in the cell-culture incubator until the cells detach (~3-4 min).
      NOTE: Effective proteolysis can be best assessed by slowly tilting the dish back and forth looking for release of cells from all portions of the dish into the moving fluid. HEK293T cells are sensitive to extended proteinase treatment and will die if left more than a few minutes in proteinase solution.
   3. Use a 10 mL pipette to transfer 7 mL of DMEM-FBS-PS to the dish of detached cells, and then use the same pipette to aspirate the cells and medium. Transfer the suspended cells to a 50 mL conical tube, and then pellet the cells by centrifuging the suspension in a low-speed clinical centrifuge at 200 x g for 5 min. Aspirate the supernatant using a glass pipette attached to a vacuum source. Use a 25 mL pipette to resuspend the cell pellet in 15 mL of DMEM-FBS-PS.
   4. Add 1 mL of the cell suspension to each of fifteen 15 cm dishes containing 19 mL of DMEM-FBS-PS.
   5. Grow the cells in a tissue culture incubator until they reach 85%-90% confluence (~3-4 days).
2. Prepare a diluted virus solution
   1. Add ~1.5 x 10⁹ to 3 x 10⁹ IVP of an existing virus stock (5-10 IVP/cell) to a 50 mL conical tube filled with 45 mL of DMEM lacking FBS or antibiotics.
      NOTE: It is possible to use a lower concentration of the virus (1-5 IVP/cell); however, it will take longer for virus production to ensue.
3. Infect the cultured cells with the virus.
   1. Using a glass pipette attached to a vacuum source, aspirate the medium from the almost confluent cells in step 1.1.5.
   2. Use a 25 mL pipette to transfer 3.0 mL of diluted virus solution (prepared in step 1.2.1) to each cell-culture dish. Then, use a 10 mL pipette to add 7.0 mL of DMEM medium (lacking FBS or antibiotics) to each dish. Incubate for 60 min in a cell-culture incubator, and then add 10 mL of DMEM containing 20% (v/v) FBS and 2x PS to each dish.
      NOTE: Using one of the 15 dishes as a control plate (in which virus is not added) makes it easier to identify virus-induced cell rounding and death in infected cells in the next step.
   3. Incubate the cells in the cell culture incubator for 2-4 days until the majority of them begin to round up and >60% of the cells have detached.
      NOTE: If using a lower titer of the virus, it may take up to a week for cell death to occur. If cell death does not occur within a week, it is likely the process will need to be repeated using a higher titer of the virus.
4. Recover virus from cell lysates
   1. Use a cell scraper to scrape the bottom of each dish, releasing attached cells into the medium.
   2. Using a 25 mL pipette, collect and pool the medium, cells, and cell debris from each cell culture dish into a 50 mL conical cell culture tube.
      NOTE: To save resources, the medium from two dishes can be combined into one 50 mL tube.
   3. Pellet the cellular material by centrifugation using a low-speed table-top centrifuge: 5 min at room temperature at 3,000 x g. Use a glass pipette attached to a vacuum device to aspirate the supernatant.
   4. Use a 10 mL pipette, combined with trituration, to consolidate all the resulting pelleted material in a total of 7 mL of sterile-filtered 100 mM Tris-HCl pH 7.4 containing 10 mM EDTA (Tris-EDTA solution). Transfer the pooled material to a sterile 15 mL cell culture conical tube and place on ice.
      NOTE: At this point, the virus prep can be frozen at -80 °C indefinitely.
5. Prepare cell lysates
   1. Perform three freeze-thaw cycles to disrupt the remaining cells, and thus further liberate the formed virus particles. Freeze the pooled material in step 1.4.4 by submerging the tube in liquid nitrogen (~30-60 s). Rapidly thaw the sample by placing the tube in a 37 °C incubator. Vortex the sample for 15 s, and then repeat the rapid freezing and thawing procedure an additional 2x.
      NOTE: The virus solution can be rapidly thawed in a 37 °C in a water bath, but caution is warranted as the temperature swings can cause the tube to crack, releasing virus solution into the water bath. To prevent this from occurring, place the tube containing the viral supernatant into a larger tube, which is then set in the water bath.
   2. Use a 10 mL pipette to transfer the thrice freeze-thawed cellular material to a superspeed centrifuge tube. Centrifuge the material in the tube for 30 min at 4 °C super-speed centrifuge at ~18,500 x g.
   3. Recover the ~7 mL of virus-rich supernatant with a 10 mL pipette and transfer it to a 15 mL conical tube. Retain the sample on ice until the next step.
      NOTE: At this point, consider keeping an aliquot of the unpurified viral supernatant as a preamble to start a new virus preparation (or as a backup). Store this aliquot at -80 °C.
6. Isolate and purify the virus using density gradient centrifugation.
   1. Prepare a discontinuous gradient of CsCl in a 12 mL PET thin-wall, clear ultracentrifuge tube (see Table of Materials) or equivalent. Use a 3 mL syringe outfitted with an 18 G needle to carefully introduce 2.5 mL of 1.4 g/mL CsCl solution into the bottom of the tube, and then use a new syringe/needle to layer it with 2.5 mL of 1.25 g/mL CsCl solution.
      NOTE: Dropping solutions directly on top of a preexisting layer will cause significant and unwanted mixing. Instead, place the beveled portion of the needle against the edge of the tube, and then very slowly press the syringe plunger, raising the position of the needle as the solution fills the tube.
   2. Load the ~7 mL of viral supernatant on top of the gradient in a similar fashion using a 10 mL syringe. If there is more than 2-3 mm of space between the viral supernatant and the top of the tube, add additional Tris-EDTA solution to fill the tube until only 2-3 mm of space remains.
   3. Make a similarly prepared balance tube, containing layers of CsCl, but substituting Tris-EDTA solution for the viral supernatant.
      NOTE: The two tubes must have identical weights (and similar densities) to prevent a potentially dangerous unbalanced load situation in the ultracentrifuge.
7. Isolate the virus using rate-zonal ultracentrifugation.
   1. Load the gradients formed in steps 1.6.2-1.6.3 into the buckets of an SW41 rotor or equivalent. Screw in bucket caps, place the rotor into an ultracentrifuge (precooled to 4 °C), and centrifuge for 1 h at ~150,000 x g.
   2. During centrifugation, equilibrate the column described in step 1.9.1 below.
8. Recover isolated virus particles
   1. At the end of the centrifugation step, carefully detach the buckets from the rotor, and in a cell-culture hood, remove the bucket caps, and then remove the tubes and place them in a rack.
   2. Collect the banded material, rich in virus particles, that floats at the interface between the 1.25 g/mL and 1.4 g/mL CsCl solution. Holding the tube containing the gradient over the bottom half of a cell culture dish (which will catch any spilled materials), use a 1 inch 18 G needle attached to a 3 mL syringe to carefully puncture the tube, just below the banded virus. Slowly aspirate the virus, which is typically recovered in ~1 mL.
   3. Remove the needle from the tube, which will result in the remaining material in the gradient to flow out of the tube into the bottom half of the cell-culture dish (any liquids should be treated as hazardous waste).
   4. Transfer the virus solution in the syringe into a sterile microcentrifuge tube on ice.
      NOTE: When recovering the virus in this step, position the needle so that the lumen of the needle is facing upward with the needle opening a few mm below the virus-rich band. Avoid contamination by the thin band of improperly assembled virus that is sometimes observed 2-3 mm above the banded virus (see thin black arrow in Figure 1A).
9. Remove CsCl from the sample by gel filtration.
   1. Equilibrate a PD-10 column (pre-packed with Sephadex G-25M), clamped to a support stand, with 50 mL of 0.2 µm sterile-filtered PBS containing 10% (v/v) glycerol.
      NOTE: The initial equilibration step takes 2-3 h to complete and is necessary to ensure complete washout of preservatives that are used by the manufacturer to stabilize the column.
   2. Allow the wash solution to recede below the frit (a protective disc of white material at the top of the column medium), and then carefully transfer the purified virus solution collected in step 1.8 to the top of the column. Allow the virus-rich solution to recede below the frit, and then begin filling the column with PBS-glycerol.
      NOTE: One function of the frit is to prevent the column from running dry. As a result, short delays are tolerated before more eluant is added to the column.
   3. Collect the eluate in 12 sterile microcentrifuge tubes, 0.5 mL per fraction.
10. Determine the viral yield.
    1. Determine the peak virus fractions using spectrophotometry. Prepare a 1:100 dilution of each fraction in PBS and measure the OD₂₆₀ in a spectrophotometer, using a 1:100 dilution of buffer alone as a blank. The virus particles should elute in the void volume, starting around fraction 6. Pool the fractions containing the highest OD₂₆₀ readings.
    2. Make a 1:100 dilution of the pooled viral fractions and remeasure the OD₂₆₀. Calculate the final concentration of virus particles and the number of IVP in the pooled fractions using the following formula: Virus particles per mL = OD₂₆₀ × 100 (this corrects for the dilution factor) × 10¹². A general estimate is that 1% of the virus particles are IVP, as such: IVP/mL = OD₂₆₀ × 100 × 10¹⁰ or IVP/µL = OD²⁶⁰ × 100 × 10⁷.
11. Aliquot the pooled virus fractions (containing 5 x 10⁷ to 1 x 10⁸ IVP) into sterile microcentrifuge tubes or cryovials. Store the samples at -80 °C.

2. Transduction of rodent bladder

NOTE: If new to this technique, it is recommended that the number of animals transduced at one time be limited to 2-4. This can be accomplished by staggering the start times for each animal, particularly during the detergent treatment in step 2.2, and then the virus incubation in step 2.3. Experienced investigators can transduce up to six animals at a time.

1. Catheterize the bladder
   1. Anesthetize female C57Bl/6J mice (typically 8-10 weeks old, ~20-25 g) or female Sprague Dawley rats (typically 2-3 months old, ~250 g) using a vaporizer with an attached nose cone. Calibrate the vaporizer to produce 3.0% (v/v) isoflurane, 97% (v/v) O₂ for mice or 4.0% (v/v) isoflurane, 96% (v/v) O₂ for rats. Confirm that the animals are anesthetized by ensuring they are unresponsive to toe pinch (typically after 1-2 min).
   2. Maintain the animal's body temperature by placing the animals on a heated pad. Monitor the animal throughout the transduction protocol to ensure the animals are anesthetized and do not experience any pain during this procedure.
   3. Reduce isoflurane to 1.5% (v/v) for mice or 2.0% (v/v) for rats and maintain the animals under anesthesia for the duration of the protocol.
   4. To prevent introducing air into the bladder, fill the plastic catheter portion of an IV catheter (see Table of Materials) and associated hub with sterile PBS using a transfer pipette.
   5. With the animal in the supine position, swab the external meatus with 70% alcohol, and insert the sterile catheter into the external meatus, then the urethra, and then into the bladder.
      1. To perform this task, use fine forceps to gently grab the tissue forming the external meatus and extend it vertically, away from the animal. Using the other hand, carefully insert the catheter vertically about 3-4 mm into the urethral meatus (a mound of flesh just above the opening of the vagina). Then, because of a bend in the urethra, lower the catheter, inserted into the external meatus, toward the tail of the animal, which eases its entry into the portion of the urethra that passes below the pubic bone, and ultimately into the bladder
         .NOTE: Especially in the case of the mouse, the catheter may be too long, and no more that 1.0-1.1 cm of it should be inserted into the animal. Otherwise, damage to the bladder mucosa will ensue. To prevent this, mark the catheter ~1 cm below its tip, and then do not insert the catheter beyond this marking.
   6. Allow the urine in the bladder to leak out. Remove any residual urine by performing Credé's maneuver: massage and gently press down on the bladder bump in the lower abdominal area.
2. Make the urothelium receptive to transduction by treating it with a detergent solution.
   1. Wash the mouse or rat bladder by attaching a sterile 1 mL syringe filled with PBS to the catheter hub. Inject 100 µL of sterile PBS into the mouse bladder (or 450 µL in the case of the rat bladder). Detach the syringe from the catheter fitting and allow the PBS drain. If necessary, perform Crede's maneuver to remove excess bladder fluid.
   2. Instill 100 μL of 0.1% (w/v) N-dodecyl-β-D-maltoside (DDM) (dissolved in PBS and 0.2 µm filter-sterilized) into the mouse bladder using a sterile 1 mL syringe. Retain the DDM in the bladder for 10 min by leaving the syringe in place. In rats, the volume of DDM is increased to 450 µL.
   3. Remove the DDM from the bladder by detaching the syringe and allowing it to drain out. Perform Crede's maneuver if necessary.
3. Introduce the virus into the bladder.
   1. Attach a sterile 1 mL syringe to the catheter hub and instill 0.5 x 10⁷ to 1 x 10⁸ IVP of adenovirus (prepared in step 1), diluted in 100 µL of sterile PBS for mice or 450 µL for rats, into the bladder. Leave the syringe attached to the catheter to prevent the virus solution from escaping.
   2. After 30 min, detach the syringe and allow the virus solution to evacuate the bladder onto a disposable pad. Blot any residual virus solution with an absorbent wipe, and discard the pad and wipe in biohazardous waste.
      NOTE: Glycerol is cytotoxic. As such, the maximal volume of virus solution instilled in mice is limited to 5 µL (which when diluted into 100 µL of PBS would result in a final glycerol concentration of ~0.5% [v/v]). Additionally, it is possible to enhance transduction efficiency by increasing the number of instilled IVP and by extending the virus incubation to 45 min.
   3. Optional step: The bladder can be rinsed with PBS as described in step 2.2.1 above; however, this is not required.
4. Allow the animal to recover.
   1. Cease the flow of isoflurane and allow the animal to recover and be fully mobile before returning it to its cage, particularly if the animals are group housed.
      NOTE: As virus transduction per se does not cause observable lower urinary tract symptoms or pain, there is usually no need for post-surgical treatment. However, if the encoded transgene is toxic, post-procedure analgesia or antibiotics may be necessary as required by the institution.
5. Analyze the effects of transgene expression 12-72 h post treatment using methods such as mRNA in situ hybridization, western blot, or immunofluorescence^9,10,23 (see the representative results).

Representative Results

Virus preparation
An example of virus purification by density gradient centrifugation is shown in Figure 1A. The light pink band, found at the interface of the loaded cellular material and the 1.25 g/mL CsCl layer, is primarily composed of disrupted cells and their debris (see magenta arrow in Figure 1A). It derives its pinkish color from the small amount of culture medium that is carried over from step 1.5 in the protocol. The virus particles of interest, which appear as a milky white band, are found at the interface of the 1.25 g/mL CsCl and 1.40 g/mL CsCl solutions (see yellow arrow in Figure 1A). One may also observe a band of material that floats 2-3 mm above the enriched virus particles (see thin black arrow in Figure 1A). It comprises unassembled viruses and debris, contains few IVP, and should be avoided when collecting the virus sample.

Further purification of the virus and buffer exchange using a PD10 column is depicted in Figure 1B, and the OD₂₆₀ readings of the resulting fractions after elution are shown in Figure 1C. The void volume of these columns is approximately 3 mL, and thus the virus begins to appear in fraction 6 and peaks in fraction 9. In this experiment, fractions 6-9 were pooled. While fraction 6 and 7 are relatively low in virus particles, they contain sufficient virus to merit recovery, and they serve to dilute the very high titer fractions to a more reasonable concentration. Fractions 10-12 were not included as the increase in OD₂₆₀ in fraction 11 indicates the possible presence of a second contaminating peak, which is variably observed in these preparations. The pooled fraction will typically have 1 x 10⁷ to 1 x 10⁸ IVP/µL, and the expected yield would be in the order of 1 x 10¹⁰ to 2 x 10¹¹ total IVP. This is sufficient amount of virus to perform hundreds of transductions. While it is possible to titer the virus by plaque assays or by counting colonies of cells expressing fluorescent proteins²², the 1% rule is sufficient in most cases. This rule states that 1% of purified virus particles, estimated by measuring the OD₂₆₀ of the sample, are IVP. Aliquots of virus stored at -80 ˚C have a shelf-life of 2-5 years, although infectivity decreases over the long term. Thawed aliquots of virus can be refrozen at -80 °C one time without significant loss of infectivity. However, repeated thawing and freezing negatively impacts viral infectivity.

Bladder transduction
An important first step when assessing the impact of transduction is to confirm transgene expression. This can be assessed using several techniques, including tools that detect mRNA (e.g., RNAScope), western blot analysis, or use of immunofluorescence^9,10,23. Figure 2 is an example of mouse urothelium that was transduced with an adenovirus that encodes V5-epitope tagged human growth hormone (V5-hGH)²³. This protein is packaged into discoidal/fusiform vesicles and can be exocytosed during bladder filling^19,23. Western blot analysis of urothelial lysates revealed V5-hGH expression in the urothelium of transduced bladders, but not in untransduced ones (Figure 2A). Expression was also confirmed by immunofluorescence, in this case using antibodies that recognized hGH or the V5 epitope tag (untransduced bladders lacked signal, not shown) (Figure 2B).

An additional example is transduction of the rat bladder with a virus encoding CLDN2, a pore-forming tight junction-associated protein^24,25. CLDN2 increases paracellular flux of cations (including K⁺) and its overexpression results in inflammation and development of visceral pain¹⁰. Western blot analysis confirmed expression of CLND2 in rat bladders transduced with adenovirus encoding Cldn2, but not those transduced with a control GFP-encoding virus (Figure 2C). In general, GFP is not considered to be toxic when expressed in cells and it thus serves as a useful control. The use of GFP also allows the investigator to confirm that transduction is working. Immunofluorescence analysis further confirmed exogenous CLDN2 expression in urothelial umbrella cells transduced with virus encoding Cldn2 cDNA (red signal in Figure 2D). Furthermore, similar to endogenous CLDN2 (not shown), the expressed CLDN2 is localized to TJP1-labeled tight junctions, as well as the basolateral surfaces of the umbrella cells (CLDN2 is labeled green in Figure 2E)¹⁰. In the case of CLDN2 expression, it was highest one day after transduction, but then trailed off, and was barely detectable after 15 days.

A second consideration is, which cell types will be targeted by adenoviral transduction. While in rats it is possible to mostly transduce the umbrella cell layer¹⁹, in mouse, all the layers of the urothelium can be transduced, although transduction of the intermediate and basal cell layers can be variable. Importantly, in the entirety of the bladder wall, only the urothelium is transduced and no other tissue is targeted by the instilled adenovirus (Figure 3).

While analyses of transduced cells can be performed at the single cell level, when exploring an overall bladder phenotype, one must transduce the majority of the urothelial cells. Thus, it is important to define the efficiency of transduction (i.e., what fraction of urothelial cells are transduced). For example, in the case of the Cldn2-expressing adenovirus, >95% of umbrella cells were transduced (see Figure 2D). An additional example is the image field shown in Figure 3, where counting the number of transduced umbrella cells (in this case, expressing the Ca²⁺ sensor GCAMP5G), reveals an efficiency that approaches 95%. However, one must examine cells in random fields taken throughout the bladder wall to achieve an accurate and unbiased estimate of the transduction efficiency.

Figure 1: Purification of adenovirus. (A) Adenovirus particles produced by infected HEK293T cells were purified by centrifugation on a discontinuous CsCl gradient made up of a layer of 1.4 g/mL CsCl, a layer of 1.25 g/mL CsCl, and a layer of sample (S) diluted in Tris-EDTA solution. The cellular material (pink arrow) accumulates at the S/1.25 interface, while the purified adenovirus floats at the 1.25/1.4 interface (yellow arrow). A small band of improperly assembled virus floats above the latter (thin black arrow). (B) The adenovirus-rich band in the gradient is recovered using a needle, and the CsCl is removed from the adenovirus by gel filtration using a G25 Sephadex-filled PD10 column equilibrated with PBS containing 10.0% (v/v) glycerol. The surface of the Sephadex is protected by a porous, plastic frit. (C) Fractions, 0.5 mL, were collected from the PD10 column. The OD₂₆₀ of the fractions were measured in a spectrophotometer and the values plotted. Virus-rich fractions elute in the void volume, which begins at fraction 6 and extends to fraction 9. Pooled fractions for this representative experiment are shaded in blue. Please click here to view a larger version of this figure.

Figure 2: Transduction of rodent bladder urothelium with adenoviruses encoding V5-hGH or CLDN2. (A) Mouse urothelium was left untransduced (UT) or transduced with adenoviruses encoding V5 epitope tagged human growth hormone (hGH). After 24 h, the bladders were recovered, urothelial lysates were prepared and subjected to sodium dodecyl sulfate polyacrylamide gel electrophoresis, and V5-hGH expression confirmed using western blots probed with an antibody against hGH. (B) Detection of V5-hGH in mouse urothelium sectioned and stained using antibodies to hGH (green) or the V5 epitope (red) and fluorophore-tagged secondary antibodies. Immunofluorescence was captured using confocal microscopy. Samples were counterstained with TO-PRO3 to label nuclei. (C-F) Rat urothelium was transduced with a virus encoding rat CLDN2 (common name claudin-2) or GFP (common name green fluorescent protein) as a control. (C) Detection of exogenous CLDN2 by western blotting using an antibody again CLDN2. Note that the endogenous CLDN2 is expressed at low levels and is not detected in this experiment. (D) Transduction of the umbrella cell layer is revealed by immunofluorescence and confocal microscopy. The borders of the cell are revealed by co-staining the tissue with FITC-phalloidin, which labels the cortical actin cytoskeleton. (E) Detection of exogenous CLDN2 and TJP1 (common name ZO1) by immunofluorescence and confocal microscopy of whole-mounted or cross-sectioned urothelium. Small white arrows indicate the location of the tight junction and small magenta arrowheads mark the location of intracellular accumulations of CLDN2 in the cell cytoplasm. These were previously revealed to be Golgi-associated CLDN2¹⁰. (F) Following transduction, the animals were euthanized on the indicated days post infection. Exogenous CLDN2 expression was detected using western blotting. Animals expressing GFP were euthanized after day 1. The data in Figure 2C-E have been modified from Montalbetti et al.¹⁰, and are reproduced with permission of the American Physiological Society. Please click here to view a larger version of this figure.

Figure 3: Efficiency of adenoviral transduction. Mouse bladder urothelium was transduced with an adenovirus encoding the calcium sensor GCaMP5G. The upper panels show staining for GCaMP5G (detected using an antibody to green fluorescent protein; GFP, green), the middle panels show the distribution of DAPI-stained nuclei (blue) and rhodamine-phalloidin-labeled actin (red), and the bottom panels are a merger of the three signals. The white arrowheads in the lower panel are the rare umbrella cells that are not transduced. The overall efficiency of transduction of umbrella cells in this image is ~95%. Note that only the urothelium is transduced. The boxed regions are magnified in the panels to the right. The region in box 1 primarily comprises transduced umbrella cells. The region in box 2 is urothelium showing efficient transduction of umbrella cells, but less efficient transduction of the underlying cell layers. LP = lamina propria; ME = muscularis externa; Se = serosa; Ut = urothelium. Images were acquired using a confocal microscope (see Table of Materials). Please click here to view a larger version of this figure.

Discussion

While Ramesh et al. were focused on developing strategies to use adenoviral transduction in the treatment of bladder cancer¹⁸, more recent reports have demonstrated the utility of these techniques in studying normal urothelial biology/physiology and pathophysiology^{10,18,19,20,21}. The important features of this approach include the following: (i) In the entirety of the rodent bladder wall, only the urothelium is transduced^{10,19,20,21,22}. In the case of rats, transduction is mostly limited to the umbrella cell layer, while in mice, the entirety of the urothelium can be targeted; (ii) This approach can be used to express proteins or siRNAs^{10,19,20,21,22}; (iii) The efficiency of transduction can approach 70%-95% (e.g., see Figure 3)^18,20; (iv) One day after transduction, the ultrastructure of the urothelium, the integrity of the tissue, presence of a high-resistance barrier (assessed by measuring trans-epithelial resistance), and the expression of urothelial differentiation markers is "normal"^10,19. The only morphological effect noted thus far is a decrease in umbrella cell diameter (when viewed from above); however, they revert to their normal size after 3-4 days^10,19; (v) The urothelium can be transduced with up to three different adenoviruses simultaneously¹⁸; (vi) Transgene expression can be altered by titrating the number of IVP, which affects the amount of protein expression, lengthening or shortening the time of incubation before the animal is sacrificed, or use of different promoters such a tet-regulated system (tet-off)¹⁹. In the latter case, one can co-express a virus encoding the transactivator/tet-repressor, and then modulate the expression of the transgene by altering the concentration of doxycycline in the animal's diet.

While the overall protocol is relatively simple, there are some critical steps that must be considered when performing these experiments. One of these is the availability of high-titer virus stocks that are of reasonable purity. Adenoviruses can be obtained from commercial vendors, from institutional virus production cores, from other investigators, or they can be produced in house. In the case of the latter, the Bert Vogelstein laboratory AdEasy system is recommended because its components can be purchased from Addgene, and adenoviruses generated using this approach work well in urothelial transduction⁴. This technique allows one to generate adenoviruses encoding cDNAs as large as 8 Kb using the pAdEasy-1 packaging plasmid (the insert replaces viral genes E1 and E3), the pAdEasier-1 bacterial cells (which encode the adenoviral genes necessary to produce virus), and production of replication-defective adenoviruses in HEK293T cells (which express the required viral E1 protein). However, substituting the pAdEasy-1 packaging plasmid with the pAdEasy-2 packaging plasmid increases the packaging size by an additional 2.7 kB, but necessitates using a different cell line (E1-transformed human embryonic retinal 911 cells) as the viruses lack early gene E4 in addition to E1 and E3⁴. Expression of siRNAs can be accomplished using several systems, including pAdloss, which is described in detail in Kasahara et al.²⁶. While it may be possible to use virus-rich cell culture medium to transduce the urothelium, this method can be unreliable. Instead, large-scale virus amplification, CsCl gradient purification, followed by gel filtration provides the most reliable way to generate large amounts of high-titer virus. The relative stability of virus at -80 °C, coupled with relatively large yields of virus, makes this an ideal way to have a consistent stock of virus that can be used over a large number of experiments.

Additional critical steps in this protocol include those associated with the transduction protocol. These include use of PBS (without divalent cations) as a washing agent and diluent, and the use of DDM detergent. As the junctional complex is dependent on Ca²⁺ for proper function, PBS likely promotes viral entry by disrupting the junction-associated barrier. DDM is also critical; as Ramesh et al. reported, the efficiency of adenovirus transduction is exceedingly low in the absence of detergent treatment¹⁸. However, the mode of action of the detergent is not clear. A 10 min incubation with DDM is ideal, but this can be shortened to 5 min; however, it is not recommended that the incubation extend beyond 10 min as the detergent may cause unexpected damage to underlying tissues. The amount of virus used is a critical parameter, with higher concentrations generally leading to greater amounts of transgene expression and higher transduction efficiencies. However, the optimal virus concentration that gives the best combination of expression, efficiency, and phenotype must be empirically determined. As a starting concentration, a range of 5.0 x 10⁶ to 2.0 x 10⁷ IVP per animal is recommended. Depending on the protein being expressed, this results in efficiencies in the 70%-95% range^{10,19,20,21,22}; however, some dominant-negative GTPase constructs are less efficient (30%-50%) and require a single-cell analysis approach^4,20. These differences in efficiency may reflect the turnover of the transgene, its toxicity, or the purity and yield of virus used for transduction. The latter can be ruled out by performing plaque assays. Although not hyper-critical, the time the virus remains in the bladder must be long enough for the virus to attach to its CXADR receptor. Times less than 30 min can lower the efficiency of transduction, while extending the incubation period to 45 min can increase efficiency; however, this can be transgene-dependent. The final critical step in this protocol is determining how long the animal will be held before a phenotype is assessed. Transgenes exhibit the highest expression after 1-2 days¹⁹, but the expression can decrease after that. In the case of siRNAs, it depends on the turnover rate of the protein itself. For example, when expressing siRNAs that target expression of Rab11a, a Rab-family GTPase, efficient downregulation is only observed 72 h after transduction²³. Thus, long-lived proteins (with half-lives measured in days) may not be suitable targets using this approach.

The investigator should also be aware of the caveats associated with this approach. First, its utility may be limited to female rodents because of difficulties in catheterizing male rodents through their penis. However, it may be technically possible to perform this protocol in males by introducing a catheter into the dome of the bladder, similar to the preparation employed when performing cystometry⁹. This would allow one to introduce a detergent or a virus, and perform washes as needed. A second caveat is that overexpression of proteins can exhaust cell pathways and resources, which can lead to events such as protein aggregation, activation of cell stress pathways, and death²⁷. Thus, it is generally wise to limit transgene expression to non-toxic levels (unless of course that is the intention of transgene expression), as assessed using markers of cell stress and cell death. A third caveat is that in some settings, long-term adenovirus expression can trigger immune responses²⁸. While there is no evidence that adenoviral transduction of the urothelium per se stimulates immune responses¹⁰, this is transgene-dependent and as noted above CLDN2 overexpression leads to cystitis.

Currently, investigators have a variety of methods that they can use to modulate gene and protein expression in the urothelium, including use of transgenic or conditional urothelial knockout mice, use of transfection reagents, or use of adenoviral transduction. The latter method, the subject of this protocol, is relatively easy, efficient, and reproducible. Other than the initial expense of cell culture reagents needed to generate the virus stocks, the very large amount of virus produced (enough to perform hundreds of transductions), results in a relatively low cost on a per animal basis. Adenoviral transduction can be exploited to study urothelial biology. For example, adenoviral transduction was used to define the importance of Rho-family and Rab-family GTPases, guanine-nucleotide exchange factors, myosin motor fragments, and ADAM17 in exocytosis and endocytosis^{10,19,20,21,22}, and adenoviral transduction was recently used to study urothelial mechanotransduction⁹. Likewise, adenoviral transduction may be relevant when exploring and treating human disease. For example, Ramesh et al. initially proposed adenoviral transduction may be a useful way to target tumor cells¹⁸. While their studies were more a proof-of-principle approach, one can imagine that expression of toxins in cancer cells or expression of epitopes that could be recognized by the immune system would be useful strategies. Adenoviral transduction may also be useful in understanding diseases where overexpression or underexpression of gene products leads to disease. As one example, biopsies from the bladders of patients with interstitial cystitis exhibit a 90-fold increase in CLDN2 expression²⁹. Interestingly, over-expressing CLDN2 using adenoviral transduction replicates many of the symptoms of this disease in rats¹⁰. Thus, CLDN2 could be one target in the treatment of patients with this disorder.

Materials

Name	Company	Catalog Number	Comments
10 mL pipette	Corning Costar (Millipore Sigma)	CLS4488	sterile, serological pipette, individually wrapped
12 mL ultracentrifuge tube	ThermoFisher	06-752	PET thinwall ultracentrifuge tube
15 mL conical centrifuge tube	Falcon (Corning)	352097	sterile
18 G needle	BD	305196	18 G x 1.5 in needle
20 mL pipette	Corning Costar (Millipore Sigma)	CLS4489	sterile, serological pipette, individually wrapped
50 mL conical centrifuge tube	Falcon (Corning)	352098	sterile
5 mL pipette	Corning Costar (Millipore Sigma)	CLS4487	sterile, serological pipette, individually wrapped
Cavicide	Henry Schein	6400012	Anti-viral solution
Cell culture dish - 15 cm	Falcon (Corning)	353025	sterile, tissue-culture treated  (150 mm x 25 mm dish)
Cell scraper	Sarstedt	893.1832	handle length 24 cm, blade length 1.7 cm
CsCl	Millipore Sigma	C-4306	Molecular Biology grade ≥ 98%
DMEM culture medium (high glucose)	Gibco (ThermoFisher)	11965092	with 4.5 g/L glucose + L-glutamine + phenol red
EDTA	Millipore Sigma	EDS	Bioiultra grade ≥ 99%
Fetal bovine serum	Hyclone (Cytiva)	SH30070.03	defined serum
Glass pipette	Fisher Scientific	13-678-20A	5.75 in glass pipette, autoclaved
Glycerol	Millipore Sigma	G-5516	Molecular Biology grade ≥ 99%
HEK293 cells	ATCC	CRL-3216	HEK293T cells are a variant of HEK293 cells that express the SV40 large T-antigen
Isoflurane	Covetrus	29405
IV catheter - mouse	Smith Medical Jelco	3063	24 G x 3/4 in Safety IV catheter  radiopaque
IV catheter - rat	Smith Medical Jelco	3060	22 G x 1 in Safety IV catheter radiopaque
KCl	Millipore Sigma	P-9541	Molecular Biology grade ≥ 99%
KH2PO4	Millipore Sigma	P5655	Cell culture grade  ≥ 99%
Na2HPO4•7 H2O	Millipore Sigma	431478	≥ 99.99%
NaCl	Millipore Sigma	S3014	Molecular Biology grade ≥ 99%
N-dodecyl-β-D-maltoside	Millipore Sigma	D4641	≥ 98%
Nose cone for multiple animals	custom designed		commercial options include one from Parkland Scientific (RES3200)
PD-10 column	GE Healthcare	17-085-01	Prepacked columns filled ith Sephadex G-25M
Penicillin/streptomycin antibiotic (100x)	Gibco (ThermoFisher)	15070063	100x concentrated solution
Spectrophotometer	Eppendorf	BioPhotometer
Stand and clamp	Fisher Scientific	14-679Q and 05-769-8FQ	available from numerous suppliers
Sterile filter unit	Fisher Scientific (Nalgene)	09-740-65B	0.2 µm rapid-flow filter unit (150 mL)
Sterile filter unit 0.2 µm (syringe)	Fisher Scientific	SLGV004SL	Millipore Sigma Milex 0.22 µm filter unit that attaches to syringe
Super speed centrifuge	Eppendorf	5810R	with Eppendorf F34-6-38 fixed angle rotor (12,000 rpm)
Syringe (1 mL)	BD	309628	1-mL syringe Luer-lok tip - sterile
Syringe (3 mL)	BD	309656	3-mL syringe slip tip - sterile
Table-top centrifuge (low speed)	Eppendorf	5702	with swinging bucket rotor
Transfer pipettes	Fisher Scientific	13-711-9AM	polyethylene 3.4 mL transfer pipette
Tris-base	Millipore Sigma	648310-M	Molecular Biology grade
TrypLE select protease solution	Gibco (ThermoFisher)	12604013	TrypLE express enzyme (1x), no phenol red
Ultracentrifuge	Beckman Coulter	Optima L-80 XP	with Beckman SW41 rotor (41,000 rpm)
Vaporizer	General Anesthetic Services, Inc.	Tec 3	Isoflurane vaporizer
Vortex Mixer	VWR	10153-838	analog vortex mixer