Figure 1.
IFN-γ-activated THP-1 cells inhibit intracellular replication of F. tularensis.
Quiescent (control) and IFN-γ-activated (100 U/ml) THP-1 cells were incubated for 2 h with GFP-expressing F. tularensis LVS, followed by 30 min gentamicin treatment to kill extracellular bacteria, and subsequent determination of the number of intracellular bacteria at 4 h and 24 h post-infection by (A) colony forming unit (CFU) assay. Multiplicity of infection (MOI) was 10∶1 (bacteria: macrophage). *P<0.05 compared to control. Data were pooled from three independent experiments and presented as mean±SD. (B) Representative images showing the number of intracellular GFP-LVS after 24 h infection by fluorescence microscopy (400×). MOI = 40∶1. (C) Quantification of the intracellular GFP-LVS bacteria after 24 h infection (MOI = 40∶1) by flow cytometry, with axes set to the same scales for all groups and10,000 cells counted per sample. Gray histograms represent uninfected cells. Green histograms represent cells infected with GFP-LVS, and overlap between the two in dark green.
Figure 2.
Protocol used for genome-wide RNAi screen: sorting for high GFP-positives.
Step 1: construct THP-1 library cells by transduction of human THP-1 macrophage cell line with GeneNet™ Human 50 k shRNA library. Step 2: differentiate THP-1 library cells with PMA, and activate the cells with IFN-γ. Step 3: infect the library cells with GFP-labeled F. tularensis live vaccine strain (GFP-LVS). Step 4: harvest cells at indicated time points, and fix the cells for biosafety prior to flow cytometry. Step 5: FACS sorting to isolate small population of cells with the highest green fluorescence (containing the most GFP-LVS). Step 6: identify siRNA targets and genes by PCR, cloning and DNA sequencing.
Figure 3.
(A) Top hits involved in ten significant networks identified by Ingenuity pathway analysis. The numbers surrounding the pie charts represent the number of the top hits involved in each network. (B) 16 hits (highlighted in green) are in a network linked to a category designated as ‘hematological disease, immunological disease, cell death’, which includes TNFRSF9/CD137. Connection lines ⁃⁃, represent direct interactions; ----, represent indirect interactions.
Figure 4.
Gene expression validation of top 168 hits.
Custom human qRT-PCR arrays (Array 1, top 84 hits;Array 2, top 85–164 genes) were used to measure gene expression in three human macrophages (THP-1 cell line, human monocyte-derived macrophages, and human primary AMs) with or without IFN-γ activation, and with or without F. tularensis LVS infection for 24 hours (MOI = 40∶1). All gene expression was normalized to three human housekeeping genes, ACTB, GAPDH and NONO. (A & B) Heat map showing IFN-γ–induced gene expression changes among top 84 (A) and 85–164 (B) hits in the three types of IFN-γ-activated macrophages ± LVS (left and right columns) compared to control cells without IFN-γ treatment. Green indicates lower, and red, higher expression. (C) Venn diagram of genes up-regulated in the three types of IFN-γ activated macrophages with and without F. tularensis challenge. Ten genes showed induced expression in all three macrophages.
Table 1.
Top 168 genes chosen for expression analysis by qRT-PCR arrays.
Table 2.
Top 10 genes with increased expression in all 3 human macrophage types.
Figure 5.
Validation of TNFRSF9 and SERPINI1 gene expression and function.
(A) qPCR analysis of expression of TNFRSF9 and SERPINI1 genes in THP-1 macrophages after transduction with individual lentiviral shRNAs compared to controls transduced with non-targeting shRNA. PCR reactions were normalized against β-actin and plotted relative to expression levels in control cells. Error bars indicate± SD of triplicates. (B) Flow cytometry analysis of the effect of lentiviral shRNA knockdown of TNFRSF9 and SERPINI1 genes on F. tularensis GFP-LVS infection after 24 h in THP-1 cells ± IFN-γ, 40 U/ml. Controls were transduced with non-targeting shRNA. Quantification of GFP-LVS bacteria at 24 h post-infection (MOI = 82∶1) is shown by green fluorescence histograms, using identical scale axes for all groups; uninfected cell green fluorescence histograms are shown in gray, GFP-LVS infected cell histograms in light green, and overlap between the two in dark green. (C) CFU assay results of the effect of lentiviral shRNA knockdown of TNFRSF9 and SERPINI1 genes on GFP-LVS infection in THP-1 cells ± IFN-γ, 40 U/ml at 24 h (MOI = 82∶1). Control THP-1 cells were transduced with non-targeting shRNA. Mean ± SD, *P<0.05 compared to control+IFN-γ group.
Figure 6.
Validation of TNFRSF9 and SERPINI1 gene expression and function with gene knockdown by siRNA.
(A) Comparison of TNFRSF9 gene expression using qPCR assay after treatment of THP-1 cells with 100 nM specific or control siRNA for 48 hours and 72 hours. (B) Flow cytometry analysis of siRNA TNFRSF9 knockdown in THP-1 cell with F. tularensis GFP-LVS infection for 24 h. THP-1 macrophages were treated with 50 nM or100 nM of siRNA for 72 h, and pre-activated with IFN-γ (40 U/ml) overnight prior to GFP-LVS infection (MOI = 35∶1). Y-axis represents mean green fluorescence level (GFL). *P<0.05 compared to control Ctr_siRNA+IFN-γ group. (C) Flow cytometry analysis of GFP-LVS infection for 24 h in HMDMs following 100 nM siRNA knockdown of TNFRSF9 and SERPINI1, respectively. *P<0.05 compared to control Ctr_siRNA+IFN-γ group.
Table 3.
Summary of functional validation results for top candidates.
Table 4.
Top 5 hits as therapeutic leads.
Figure 7.
Validation of TNFRSF9 and SERPINI1 genes using virulent F. tularensis.
(A) IFN-γ-activated THP-1 cells inhibited SchuS4 growth at 4 h and 24 h infection. IFN-γ, 40 U/ml (SchuS4, MOI = 30∶1). (B) CFU assay of TNFRSF9 and SERPINI1 shRNA knockdown THP-1 macrophages infected with SchuS4 strain. Control (THP-1 cells transduced with negative control lentiviral shRNA), TNFRSF9 knockdown cells (clone D) and SERPINI1 knockdown cells (clone1) were infected with virulent SchuS4 strain for 24 h (MOI = 30∶1). *P<0.05 compared to +IFN-γ group.
Figure 8.
Effect of TNFRSF9 and SERPINI1 knockdown on Listeria growth in THP-1 macrophages.
Flow cytometry analysis of THP-1 macrophages (transduced with TNFRSF9 and SERPINI1 shRNA or non-target control shRNA) 20 hours following infection with GFP-Listeria (MOI = 150∶1). Y-axis indicates mean green fluorescence index (GFL). *P<0.05 compared to Non-target shRNA+IFN-γ group.
Figure 9.
CD137 modulates clearance of F. tularensis.
Adherent HMDMs were pre-treated with anti-CD137 antibody or isotype control prior to infection with F. tularensis GFP-LVS, and analyzed for bacterial load at 2 h and 24 h by flow cytometry, scanning cytometry and CFU assay. (A) Flow cytometry fluorescence histograms for a single experiment at 2 and 24 h. Gray histograms represent uninfected cells. Green histograms represent cells infected with GFP-LVS. (B) Representative scanning cytometry images of GFP-LVS-containing HMDMs at 24 h post infection. (C–E) Quantitative results at 2 h and 24 h for (C) flow cytometry (N = 4), (D) scanning cytometry (N = 5), and (E) CFU assay (N = 3). *P<0.05, **P<0.01, ** P<0.001 (symbols directly above bars indicate significance relative to untreated control; symbols between bars indicate significance between the two bars indicated by horizontal lines).