Human Correlation Experimental Data

Isolation and transduction of primary human peripheral blood mononuclear cells Blood is collected from normal human donors and used to prepare mononuclear cells and subsequently to isolate T cells (CD4 and CD8), B lymphocytes, NK cells and monocytes which can be further differentiated to dendritic cells and macrophages. Gene knock down approaches used by the core utilize commercially available reagents and include Lentiviral transduction and expression of shRNA (Openbiosystems), Accell siRNA (Dharmacon) and Nucleofection (Lonza / Amaxa). Before addition of lentivirus the cells are stimulated using appropriate agonists, e.g. immobilized anti-CD3 / anti-CD28 and IL-2 for CD4 and CD8 cells or CpG, anti-Immunoglobulin, anti-CD40 and IL-2 for B lymphocytes. After exposure to Lentivirus, transduced cells are selected using puromycin, and GFP expression is observed using fluorescence microscopy and flow cytometry. Specific knock down of targeted genes is evaluated using qRT-PCR and Western blot analysis, and changes in cell phenotype are observed using polychromatic flow cytometry, qRT-PCR and western blot. Protocols for viral transduction of T cells and B cells are shown in Figures 1 and 10 respectively.

Figure 1 shows the standard protocol we have used for Lentiviral transduction of human T cells. Two major determinants are critical for high transduction efficiency: having suitably activated T cells and using high titer purified packaged Lentivirus. The Dynabead anti-CD3/CD28 T cell expander reagent is particularly effective, with cells binding beads, developing prominent cytoplasmic extensions and forming aggregates within 2-4 hours. The best Lentiviral preparations in our hands are obtained by limiting exposure of the HEK293 packaging cells to transfection mixture (reagent + plasmids) to 3.5 hours, harvesting virus-containing supernatant at 48 hours post-transduction and using ultracentrifugation through 20% sucrose to concentrate and purify the virus. With concentrated virus stocks on hand, it is easy to overdose the target cells, and it is critical to determine virus titer and optimal MOI using suitable targets, e.g. HT-1080 cells as well as the cells to be used in gene knock-down studies. Use of Retronectin-coated plates appears to improve transduction efficiency but is not a critical step. Low serum conditions (i.e. 0.5% FBS) consistently yields higher transduction efficiency, but satisfactory results can be obtained in 10% serum where experimental conditions preclude using LsDMEM. Puromycin is needed to prevent overgrowth of transduced cells by uninfected lymphocytes. Alternatively flow cytometry can be used to gate (and sort) the GFP positive population. Depending of the time of addition, Puromycin does leave behind substantial numbers of dead cell which persist as the culture continues to grow. If necessary, dead cells can be removed by centrifugation over ficoll.

Human CD4 cells are relatively easy to culture and tranduce using good quality packaged Lentivirus. Shown in Figure 2, the T cells are present in large aggregates which begin to form within minutes to hours after addition of co-stimulatory beads (anti-CD3 / anti-CD28) and IL-2. An MOI of 1-3 is optimal to avoid introduction of multiple copies of the vector into the target cells, but the culture might be overgrown by untransduced cells unless Puromycin is added within a few days of addition of virus (lower panel of Figure 2). Adding Puromycin at 1.5-3 days post-transduction will help minimize having large numbers of dead cells in the cultures.

Figure 3 shows a typical PBMC preparation containing CD4-positive T lymphocytes and monocytes (top left plot) and also CD8 positive lymphocytes (bottom left plot). The unstimulated CD4 and CD8 cells have minimal expression of CD25 and CD45RO. However when exposed to co-stimulating Dynal beads (anti-CD3 / anti-CD28) over hours to days, both cell types proliferate and upregulate expression of CD25 and also expression of CD45RO.

Knock-down of specific gene targets using lentiviral expression of siRNA One of the genes shown in forward genetics studies to be important for adaptive immune responses is the heterogeneous nuclear ribonucleoprotein, hnRNPLL, which is required for alternative splicing of CD45. HnRNPLL is a regulated gene product in lymphocytes, with the lowest expression in B cells which express B220 / full length CD45ABC compared to high expression in memory T cells which express CD45RO. The ENU mutant mouse, Thunder (Th-under) with a V136D substitution in hnrnpll was found to be unable to silence exons 4, 5 and 6 of CD45 (Wu, et al. Immunity 29:863, 2008). CD45 exon silencing was restored in thu/thu T cells that were transduced with wild-type hnrpll cDNA. In other studies, an shRNA screen identified hnRNPLL as necessary and sufficient for signal-induced exon skipping of CD45 (Oberdoerffer, et al, Science 321:686, 2008). Knock-down of hnRNPLL in primary CD4 and CD8 lymphocytes was performed using specific lentiviral transfer vector from OpenByosystems (Figures 4-6). In separate wells, CD4 and CD8 cells were infected with lentivius expressing shRNA silencing A20 (TNFAIP-3). These experiments were performed using blood from a normal human donor collected in ACD anticoagulant followed by purification of PBMC using Ficoll-Paque and direct immunomagnetic separation of CD4 and CD8 cells using Miltenyi reagents. The purified cells were cultured under conditions of costimulation with immobilized anti-CD3/anti-CD28 and stimulation with IL-2 (50 U/ml). After four hours, packaged lentivirus (purified and concentrated by ultracentrifugation) was added to the proliferating cells (MOI = 4) followed by overnight incubation at 37°C with 10% CO₂. Puromycin, 2 µg /ml final, was added 48 hours after transduction, and the cells were maintained in culture for several weeks, removing spent medium and replenishing with DMEM + 10% FBS containing Puromycin at 3-5 day intervals. After three weeks in culture the cells were essentially 100% GFP positive, and qRT-PCR demonstrated substantial knock-down of hnrpll message. Figure 4 shows GFP expression observed by fluorescence microscopy. All of the packaged lentivirus preparations successfully transduced the target cells as observed by expression of GFP and Puromycin resistance. Transduction was optimal in LsDMEM (0.5% FBS), but acceptable transduction efficiency was obtained using 10% FBS.

After several days in culture, CD4 and CD8 cells were processed for flow cytometry to evaluate expression of CD45 isoform, CD45RA versus CD45RO. As shown in Figure 5 three of the four vectors targeting hnRPLL caused marked diminution of CD45RO expression compared to the two controls, NSC (non-silencing) and EG5. In contrast, targeting of A20 (top row of Figure 5) caused increased expression of CD45RO. Similar results were observed in CD8 cell cultures (Figure 6).

Alternative methods for gene knock-down Packaging of lentivirus is labor-intensive and time-consuming, and target cells have to be activated and dividing in order for infection / transduction of occur. Alternatives that circumvent these issues include transfection of siRNA using Accell reagents (Dharmacon) or nucleofection (Amaxa). Accell siRNA is used at substantially higher concentration than would typically be used for transfection, but transfection reagent is not required, avoiding potential toxicity to the cells. In addition both of the methods can be used in proliferating cells as well as resting cells that are not actively dividing. Figure 7 shows data from CD4 cells that were exposed to co-stimulation followed by addition of Accell siRNA. Cells were harvested for FACS analysis after three days in culture. As shown in the graphs on the right side of the panel, all of the targeting siRNA induced diminished expression of CD45RO, requiring as little as 1 µm or as much as 4 µM siRNA for ≥ 80% reduction in CD45RO expression.

Accell siRNA is also effective in targeting gene expression in non-dividing cells. As shown in Figure 8, PBMC were exposed to fluorescent non-targeting siRNA or siRNA specific for MUNC13-4 at a final concentration of 1 µM in serum-free medium. After three days in culture the cells were examined by fluorescent microscopy (upper-right panel) or were processed for real-time PCR. All of the cells were diffusely stained by the FAM-siRNA, and MUNC13-4 expression was found to be reduced by exposure to the targeting siRNA but not by exposure to siRNA specific for cyclophilin or to non-targeting siRNA.

Nucleofection (Amaza / Lonza) provides another method for introduction of siRNA into non-dividing primary cells. Figure 9 shows fluorescent microscope images from PBMC that were expressing GFP after nucleofection to introduce pMAX (a relatively small vector) or pGIPZ (three times larger). This approach can be used for a number of approaches to modify gene expression, including introduction of siRNA as well as plasimids. Limitations include the relatively small amounts of cells that can be processed and cell injury that occurs during nucleofection.

Culture and lentiviral transduction of B lymphocytes Compared to CD4 cells, B lymphocytes are more challenging to isolate in useful quantities, maintain in culture and to transduce using lentivirus. For example using Miltenyi microbead purification, 1E8 input PBMC would typically yield 2.5E7 CD4 cells while as few as 5E6 B lymphocytes might be obtained. Using conditions for culture and transduction of B cells shown in Figure 10, B cells can be activated to proliferate and will become infected with virus, yielding GFP positive cells (Figures 11-12). However the rate of proliferation and the viability in culture is significantly less than that observed in CD4 cell cultures. An alternative approach currently under evaluation is to culture B cells with agonists (shown in Figure 10) in the presence of Accell siRNA, enabling generation of genetically altered primary cells within 4-6 days of isolation in adequate quantity for down-stream applications.