Replication in cells of hematopoietic origin is necessary for Dengue virus dissemination

Abstract

Dengue virus (DENV) is a mosquito-borne pathogen for which no vaccine or specific therapeutic is available. Although it is well established that dendritic cells and macrophages are primary sites of DENV replication, it remains unclear whether non-hematopoietic cellular compartments serve as virus reservoirs. Here, we exploited hematopoietic-specific microRNA-142 (miR-142) to control virus tropism by inserting tandem target sites into the virus to restrict replication exclusively in this cell population. In vivo use of this virus restricted infection of CD11b+, CD11c+, and CD45+ cells, resulting in a loss of virus spread, regardless of the route of administration. Furthermore, sequencing of the targeted virus population that persisted at low levels, demonstrated total excision of the inserted miR-142 target sites. The complete conversion of the virus population under these selective conditions suggests that these immune cells are the predominant sources of virus amplification. Taken together, this work highlights the importance of hematopoietic cells for DENV replication and showcases an invaluable tool for the study of virus pathogenesis.

Figure 1. DENV-2 does not block miRNA function.

(A) Fluorescence microscopy of human fibroblasts cotransfected with a plasmid expressing GFP targeted by miR-142 (pEGFP-142t) and either a construct expressing miR-142 (p142) or an empty vector control (vector). Cells were mock treated or infected with DENV-2 at an MOI of 1 48 hrs prior to analysis. (B) Fluorescence-activated cell sorting of samples treated as in (A). (C) Quantitative RT-PCR on samples described in (A). Data depicted as NS5 over tubulin levels. (Statistical significance: ****, P<0.0001, ***, P<0.001).

Figure 2. Generation and characterization of miR-142-targeted DENV-2.

(A) Cloning strategy for the insertion of miR-142 target sites into the 3′untranslated region (UTR) of a T7-driven DENV-2 cDNA clone. A 157 nucleotide (nt) insert containing four tandem target sites were cloned into the variable region of DENV-2 (142t virus). A control (ctrl) virus containing four reverse sites is also depicted. (B) Northern blot for miR-142 expression in Aedes albopictus mosquito (C6/36) cells, mammalian baby hamster kidney cells (BHKs), murine bone-marrow-derived macrophages (BMMs), and a human B cell line (Raji). (C) Quantitative RT-PCR and western blot analysis on C6/36 cells infected with wt, ctrl, and 142t viruses at the indicated time points. Wild type (wt) virus refers to a clone encompassing no modifications. (D) Same as described in (C) for BHK cells.

Figure 3. In vitro knockdown of miR-142-targeted DENV-2 during exogenous and endogenous miR-142 expression.

(A) Northern blot confirming exogenous expression of miR-142 by plasmid transfection. BHKs and HEK293s were transfected with vector or p142 and analyzed 24 hrs post transfection (hpi). Bone marrow-derived macrophages (BMMs) were included as a positive control for miR-142 expression. (B) Western blot for NS5 and actin expression in BHKs transfected with vector (−) or p142 (+) and treated with wt, ctrl, or 142t in vitro transcription (IVT) products. (C) BHKs treated as in (B) were infected with wt, ctrl, or 142t viruses and analyzed as in (B). (D) Western blot for NS5 and actin expression in BHKs treated as in (B) with wt, ctrl, 142t, or 142tm IVT products. (E) Quantitative RT-PCR on DENV-2 transcript in BHKs treated as in (B) with ctrl, 142t, or 142tm IVT products. (F) Western blot for NS5 and actin expression in hematopoietic cells (Raji B cells and BMMs) and non-hematopoietic fibroblasts (HEK293s) infected with the indicated recombinant DENV-2 strains.

Figure 4. In vivo knockdown of miR-142-targeted DENV-2.

(A) Quantitative RT-PCR (qRT-PCR) on NS5 and tubulin from CD11b⁺ and CD11c⁺ cells from splenocytes of Ifnar1^−/−/Il28r^−/− mice inoculated with either ctrl or 142t strains via intravenous (IV) injection. (B) qRT-PCR for DENV capsid in CD45⁻ versus CD45⁺ splenocytes represented as a ratio for ctrl and 142t virus infections in Ifnar1^−/−/Il28r^−/− mice. (C) qRT-PCR on spleens from Ifnar1^−/−/Il28r^−/− mice infected intraveneously (IV), intraperitonealy (IP), and subcutaneously (SC) for 48 hrs. (D) Viral titers from spleens of Ifnar1^−/−/Il28r^−/− mice infected as in (A) for 24, 48, and 72 hpi. (Statistical significance: *, P<0.05, **, P<0.01) Each data point represents an individual animal.

Figure 5. In vivo escape mutants of miR-142-targeted DENV-2.

(A) RT-PCR on spleen samples from Ifnar1^−/−/Il28r^−/− mice infected with either ctrl or 142t viruses. (B) Mutations identified in Ifnar1^−/−/Il28r^−/− mice infected with ctrl and 142t viruses. Depicted mutations map to miR-142 complementary sites in the targeted and untargeted orientations. Notations in red designate nucleotide changes.