Optimization and functional effects of stable short hairpin RNA expression in primary human lymphocytes via lentiviral vectors

Abstract

Specific, potent, and sustained short hairpin RNA (shRNA)-mediated gene silencing is crucial for the successful application of RNA interference technology to therapeutic interventions. We examined the effects of shRNA expression in primary human lymphocytes (PBLs) using lentiviral vectors bearing different RNA polymerase III promoters. We found that the U6 promoter is more efficient than the H1 promoter for shRNA expression and for reducing expression of CCR5 in PBLs. However, shRNA expression from the U6 promoter resulted in a gradual decline of the transduced cell populations. With one CCR5 shRNA this decline could be attributed to elevated apoptosis but another CCR5 shRNA that caused cytotoxicity did not show evidence of apoptosis, suggesting sequence-specific mechanisms for cytotoxicity. In contrast to the U6 promoter, PBLs transduced by vectors expressing shRNAs from the H1 promoter could be maintained without major cytotoxic effects. Since a lower level of shRNA expression appears to be advantageous to maintaining the shRNA-transduced population, lentiviral vectors bearing the H1 promoter are more suitable for stable transduction and expression of shRNA in primary human T lymphocytes. Our results suggest that functional shRNA screens should include tests for both potency and adverse metabolic effects upon primary cells.

FIG. 1

The U6 promoter is more potent than the H1 promoter in driving the expression of shRNAs in primary human PBLs. PHA/IL-2-stimulated PBLs were transduced at a m.o.i. of 5 with lentiviral vectors expressing various shRNAs under the control of a U6 or an H1 promoter as indicated. The transduced cells were further cultured in IL-2-containing medium for 7 days before flow-cytometric analysis for CCR5 and EGFP expression. The x axis indicates GFP expression; the y axis indicates CCR5 expression of the cell populations in the live cell gate. Each graph shows the CCR5/EGFP plot from the U6- and the H1-shRNA vector-transduced PBLs for each shRNA against CCR5 (186) (a, b), CCR5 (13) (c, d), or firefly luciferase (e, f). (g) Cells transduced with vector without shRNA expression unit (no shRNA) and (h) mock-transduced PBLs are also shown. The quadrant lines were defined by mock transduction (h) and isotype-control stain. Percentage CCR5 positivity and mean fluorescence intensity (MFI) of CCR5 in the EGFP-positive population are shown at the top. The percentage of cells in each quadrant is also indicated. The data are representative of four independent experiments. The U6-shRNA vector-transduced PBLs had a greater CCR5 reduction than H1-shRNA vector-transduced PBLs in the four experiments (P = 0.004) by the Wilcoxon rank sum test.

FIG. 2

The U6 promoter expresses higher levels of shRNA than the H1 promoter. CEM.NKR-CCR5 cells were transduced at an m.o.i. of 1 with lentiviral vectors bearing either the U6 or the H1 promoter. As CEM.NKR-CCR5 cells are fivefold more susceptible to lentiviral vector transduction than primary PBLs, we used an m.o.i. of 1 to achieve a similar percentage EGFP expression between CEM.NKR-CCR5 and primary PBLs. The transduced cells were cultured for 12 days and total RNA was isolated from the cells. The total RNA was blotted with probes against the CCR5-shRNA (186) or (13). Known amounts of synthetic CCR5-siRNAs (186) or (13) were blotted in parallel as standards for the quantitation. The data are representative of two independent experiments.

FIG. 3

(A) Slower cell growth of PBLs expressing U6-promoter-driven shRNA. Vector-transduced PBLs were plated at 50 × 10⁴/ml and counted at 4, 7, and 12 days posttransduction. After each cell count, cells were divided 1:3 and fresh medium was added. Each graph shows the cell count (×10⁴/ml) from the H1- and the U6-shRNA vector-transduced PBLs for shRNA against CCR5 (186) (a), CCR5 (13) (b), firefly luciferase (c), or LacZ (d). No shRNA indicates cells transduced with vector that did not express shRNA. (B) Decline in EGFP-positive population in primary PBLs expressing U6-promoter-driven shRNA. EGFP populations were monitored by FACS analysis in the shRNA-transduced PBLs at days 4, 7, and 12 post-vector transduction. Each graph shows percentage EGFP positivity from the H1- and the U6-shRNA vector-transduced PBLs for shRNA against CCR5 (186) (a), CCR5 (13) (b), firefly luciferase (c), or LacZ (d). No shRNA indicates cells transduced with vector that did not express shRNA. The data are representative of four independent experiments. The U6-shRNA vector-transduced PBL had an overall greater decrease (greater negative slope) than H1-shRNA vector-transduced PBL in the four experiments (P = 0.0001) by the Wilcoxon rank sum test.

FIG. 4

IFN-induced genes OAS1 and ISG15 are not up-regulated in shRNA-transduced PBLs. RNA was isolated from transduced PBLs 4 days posttransduction and subjected to Northern blot analysis using specific probes for human OAS1 and ISG15. As a control, poly(I:C) was transfected into PBLs by electroporation. The data are representative of two independent experiments.

FIG. 5

Reduction of CCR5 in NOD SCID-Hu PBL model. PBLs (1 × 10⁷) transduced with the lentiviral vector expressing shRNA as indicated were injected into NOD SCID mice. Peripheral blood was obtained from the mice at 30 days postinjection and treated with red blood cell lysis buffer and the remaining cells were analyzed by FACS for human CD45 and CCR5 expression. The human CD45-positive population was gated and analyzed for EGFP and CCR5 expression. The percentage CCR5 positivity in the EGFP-positive population was calculated and is shown at the top. The percentage of cells in each quadrant is also shown. The data shown are a representative flow-cytometric analysis data set. N/A, not applicable.