Human monocyte-derived dendritic cells (MDDC) constitute a widely-used experimental model to study human dendritic cells. However, loss-of-function experiments in MDDC have been limited by the poor efficiency of traditional RNAi techniques. In our hands, lipid-based transfection of siRNA is largely ineffective. Fluorescently-labelled siRNA can be detected after lipofection of MDDC by microscopy or cytometry, but these siRNAs appear located in intracellular vesicles, and not in the cytoplasm, which is consistent with the highly endocytic nature of MDDC. Another technique, electroporation, can be effective with some siRNAs, but is extremely toxic by nature and leads to a large proportion of cell death. This can be a highly confounding factor in studying MDDC function, since MDDC can respond to the presence of apoptotic cells, among other factors.
We have developed a RNAi method in MDDC using shRNA lentiviral vectors. These vectors are typically derived from HIV-1. Due to a post-entry restriction to HIV-1 infection in MDDC, transduction of MDDC with HIV-1-derived vectors is highly inefficient. This restriction can be overcome by the Vpx accessory protein found in other lentiviruses such as SIVmac. SIVmac particles can be used as helper particles to facilitate MDDC transduction by alleviating the restriction to HIV-1 . We have used this property to transduce MDDC with shRNA vectors in a highly efficient manner.
Importantly, this method of MDDC transduction does not appear to interfere with the process of DC differentiation, as shown by the normal expression of DC markers (DC-SIGN, CD1a, HLA-DR) and the down-regulation of CD14 (see Figure). Furthermore, this method does not provoke MDDC maturation . As we show in the accompanying paper, HIV-1 induces MDDC maturation only after integration and de novo expression of viral proteins.
In brief, fresh monocytes are transduced with shRNA-encoding lentiviral particles and SIVmac virus-like particles. The SIVmac particles carry the Vpx protein but are devoid of genomic RNA. After 6 days of differentiation, MDDC expressing the shRNA are obtained. This process is highly efficient and >90% transduction is routinely obtained with non-concentrated, non-purified viral supernatants. In such cases, antibiotic selection or cell sorting are usually not required and the total cell population can be used as is for further experiments. We have used this protocol successfully to inhibit the expression of DC-SIGN, Cyclophilin A and IRF-3 at the RNA level, the protein level and at the functional level.
This protocol is focused on the use of shRNAs in the widely available LKO.1 backbone, with puroR or GFP as a marker. It is adaptable to other lentiviral shRNA vectors, to gain-of-function approaches (overexpression) and for differentiation into macrophages instead of dendritic cells.