Proteasome Inhibition Increases the Efficiency of Lentiviral Vector-Mediated Transduction of Trabecular Meshwork

Abstract

Purpose: To determine if proteasome inhibition using MG132 increased the efficiency of FIV vector-mediated transduction in human trabecular meshwork (TM)-1 cells and monkey organ-cultured anterior segments (MOCAS).

Methods: TM-1 cells were pretreated for 1 hour with 0.5% dimethyl sulfoxide (DMSO; vehicle control) or 5 to 50 μM MG132 and transduced with FIV.GFP (green fluorescent protein)- or FIV.mCherry-expressing vector at a multiplicity of transduction (MOT) of 20. At 24 hours, cells were fixed and stained with antibodies for GFP, and positive cells were counted, manually or by fluorescence-activated cell sorting (FACS). Cells transduced with FIV.GFP particles alone were used as controls. The effect of 20 μM MG132 treatment on high- and low-dose (2 × 107 and 0.8 × 107 transducing units [TU], respectively) FIV.GFP transduction with or without MG132 was also evaluated in MOCAS using fluorescence microscopy. Vector genome equivalents in cells and tissues were quantified by quantitative (q)PCR on DNA.

Results: In the MG132 treatment groups, there was a significant dose-dependent increase in the percentage of transduced cells at all concentrations tested. Vector genome equivalents were also increased in TM-1 cells treated with MG132. Increased FIV.GFP expression in the TM was also observed in MOCAS treated with 20 μM MG132 and the high dose of vector. Vector genome equivalents were also significantly increased in the MOCAS tissues. Increased transduction was not seen with the low dose of virus.

Conclusions: Proteasome inhibition increased the transduction efficiency of FIV particles in TM-1 cells and MOCAS and may be a useful adjunct for delivery of therapeutic genes to the TM by lentiviral vectors.

Figure 1

Toxicity of MG132 in TM-1 cells. (A) TM-1 cells were incubated with various concentrations of MG132 in the presence of 0.5% DMSO for 1 hour followed directly by assay; 1 hour with PBS wash, replacement by media, and assay 24 hours later; or 24 hours; and toxicity was evaluated using the MTS assay. Toxicity was not evident in the cells exposed to MG132 for 1 hour and then assayed or treated for 1 hour followed by assay 24 hours later, up to concentrations of 50 μM, so a CC₅₀ value could not be calculated. MG132 was toxic after 24 hours of exposure with a CC₅₀ value of 28 μM. DMSO at 0.5% was not toxic (97% viability at 1 hour, 103% viability at 1 hour and wash, and 90% viability at 24 hours). The experiments were done two or three times and the results were averaged. (B) Representative images of cell cultures used to determine the effect of MG132 treatment on transduction efficiency. Note that TM-1 cells have a low level of autofluorescence in the GFP channel in the absence of the FIV.GFP vector, so only bright cells (denoted by arrows) were counted.

Figure 2

Quantification of TM-1 cell transduction by microscopy and qPCR. TM-1 cells were left untreated, pretreated with MG132 solvent only, or pretreated with various concentrations of MG132 for 1 hour followed by transduction with FIV.GFP. The final DMSO concentration was adjusted to 0.5% in all samples. (A) The percentage of GFP+ cells as determined by counting the cells in five random microscope fields at a magnification of ×40. A minimum of 250 total cells were counted for each condition in a masked fashion and the experiment was done twice. (B) qPCR of total cellular DNA isolated from parallel cultures with primers for copGFP normalized to β-actin. Note that the assay measures both integrated and episomal vector DNA. Asterisks indicate P < 0.05 compared to the FIV.GFP sample only. The experiment was done twice and the results were averaged.

Figure 3

Quantitative analysis of FIV.mCherry transduction efficiency in TM-1 cells using flow cytometry. (A) The percentage of mCherry-positive cells in control and MG132-treated TM-1 cells. (B) Quantitative PCR determination of FIV.mCherry vector DNA normalized to β-actin in control and MG132-treated TM-1 cells. Note that this does not represent vector genomes per cell. (C) The percentage of cells that were alive in control and MG132-treated TM-1 cells as determined by DAPI staining. Asterisks indicate significantly different from the FIV.mCherry-only sample (*P < 0.05). All studies were done twice and the results were averaged.

Figure 4

FIV.GFP transduction in the trabecular meshwork of MOCAS. (A, B) Representative examples of GFP expression in FIV.GFP-transduced trabecular meshwork of MOCAS that were used for quantification of GFP signal. (A) Control eye, 0.5% DMSO + FIV.GFP; (B) 20 μM MG132 + 0.5% DMSO + FIV.GFP. Note the wider expression band in the eye treated with 20 μM MG132 (B) compared to control eye (A) that was treated only with 0.5% DMSO and FIV.GFP. Images (×20) were taken with a Zeiss AxioVert 200M inverted fluorescence microscope. (C) Quantification of GFP density in the trabecular meshwork of MOCAS determined using ImageJ and AxioVision Rel. 4.8. (D) Quantitative PCR determination of GFP vector DNA normalized to β-actin in transduced TM segments. Asterisks indicate significantly different (P < 0.05) from control (FIV.GFP + DMSO) or baseline (FIV.GFP only).

Figure 5

Localization of lentiviral-mediated transduction. MOCAS were treated with 20 μM MG132 for 1 hour, then transduced with 2 × 10⁷ units of FIV.GFP. Sections were stained with antibody to copGFP (Alexa 488 fluorophore) to localize transduced cells and anti-CD31 antibody (Cy3 fluorophore) to identify Schlemm's canal cells. Nuclei (blue) were stained with DAPI. Note that transduction occurred only in the TM.