Mechanistic insights into the enhancement of adeno-associated virus transduction by proteasome inhibitors

Abstract

Proteasome inhibitors (e.g., bortezomib, MG132) are known to enhance adeno-associated virus (AAV) transduction; however, whether this results from pleotropic proteasome inhibition or off-target serine and/or cysteine protease inhibition remains unresolved. Here, we examined recombinant AAV (rAAV) effects of a new proteasome inhibitor, carfilzomib, which specifically inhibits chymotrypsin-like proteasome activity and no other proteases. We determined that proteasome inhibitors act on rAAV through proteasome inhibition and not serine or cysteine protease inhibition, likely through positive changes late in transduction.

Fig 1

Carfilzomib enhances rAAV2 transduction. (A) HeLa cells were cotreated with the indicated dose of bortezomib, carfilzomib, MG132, or a dimethyl sulfoxide (DMSO) vehicle control and transduced with 1,000 vg/cell rAAV2-luciferase. Transduction at 24 h is indicated as normalized luciferase activity and fold values to the vehicle-treated group. (B and C) HeLa cells were cotreated with 1 μM PI and 500 vg/cell rAAV2-EGFP, and transduction was analyzed by flow cytometry at 24 h. The percentage of cells transduced (B) and median fluorescence intensity of the transduced cells (C) are indicated. (D) Bright-field and EGPF fluorescence images at 24 h postransduction of cells treated as described for panels B and C, visually indicating transduction. Data shown are representative of three independent experiments. Error bars represent one standard deviation (SD). *, P < 0.05 versus the vehicle control based on the Kruskal-Wallis test.

Fig 2

Serine and cysteine protease inhibition does not enhance rAAV2 transduction. (A) HeLa cells were treated 3 h prior to and at the time of transduction with the indicated dose of PMSF, a serine protease inhibitor, or an ethanol vehicle control and transduced with 1,000 vg/cell rAAV2-luciferase. Transduction is indicated as normalized luciferase activity. The “Vehicle L” group corresponds to treatments with 10 to 1,000 μM, while the “Vehicle H” group corresponds to treatment with 10,000 μM. (B) Indicated concentrations of PMSF or vehicle were combined with 0.002% trypsin and incubated at 30 min at room temperature. Solutions were diluted 1:10 in assay buffer and combined with trypsin substrate in quadruplicate. Average absorbance at 415 nm is shown for 60 readings at 60-s intervals. (C) HeLa cells were treated as described for panel A, with E-64, a cysteine protease inhibitor, or a DMSO vehicle control. Transduction is indicated as normalized luciferase activity. (D) Indicated concentrations of E-64 or vehicle control were combined with 20 nM human calpain 1 (BioVision), incubated at room temperature for 10 min, and combined with Calpain-Glo luciferase reagent (Promega) with 2 mM CaCl₂. E-64 activity is indicated as relative light units. Data shown are representative of three independent experiments. Error bars represent one SD.

Fig 3

Bortezomib and carfilzomib act on rAAV2 transduction through the same mechanism. (A) HeLa cells were treated with 1 μM bortezomib, carfilzomib, or vehicle control and 20,000 vg/cell rAAV6-EGFP, 100,000 vg/cell rAAV8-EGFP, or 100,000 vg/cell rAAV9-EGFP, and transduction was assayed by flow cytometry at 24 h postransduction. Transduction is indicated as fold values of percentages of cells transduced and median fluorescence intensities of vehicle control groups. (B) HeLa cells were treated with the indicated doses (μM) of bortezomib and carfilzomib or a vehicle control and 1,000 vg/cell rAAV2-luciferase. Transduction is indicated as normalized luciferase activity. (C) HeLa cells were treated as described for Fig. 1B, and intracellular vector genome copy number was analyzed at 24 h postransduction by qPCR. Data are indicated as fold values of vg/cell of the vehicle control. Data shown in panels A and B are representative of three independent experiments. Data shown in panel C are the means from three independent experiments. Error bars represent one SD. *, P < 0.05 versus the vehicle control group based on the Kruskal-Wallis test.

Fig 4

Bortezomib is more efficient at enhancing rAAV2 transduction in vivo than carfilzomib. Female BALB/c mice were treated with 1E11 vg/mouse rAAV2-luciferase and 0.5 mg bortezomib/kg, 1 mg carfilzomib/kg, or DMSO vehicle control. (A) Serum was collected from mice at 24 h posttreatment, and aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels were measured. Individual animals are indicated by diamonds, and the mean is indicated by bars. Transduction was assayed by live imaging at 7 days postransduction (B), and light output from the area of the liver was quantified (C). (D) Transduction at 13 days was assayed by live imaging. (E) At 14 days, livers were harvested; transduction is indicated by normalized luciferase activity. (F) Vector genome copy number was assayed by qPCR. Error bars represent one SD. *, P < 0.05 versus the vehicle control based on the Kruskal-Wallis test.