Fig 2: Dual-Specific B Cells Are Expanded in the Blood and Lack CXCR5(A) Gating strategy for virus-specific B cells showing their enrichment within the IgD-negative subset.(B) Comparison of the percentages of CXCR5+ and CXCR5- cells within virus-specific and total memory B cells.(C) Total B cells stained for RV-A16 and RV-A39 and colored for CXCR5 expression. Data in (A)-(C) are representative of six subjects.(D) Percentages of CXCR5+ and CXCR5- mono-specific and dual-specific B cells within total memory B cells (CD19+CD20+IgD-) (n = 6).(E) The percentage of each RV specificity within total B cells according to naive (IgD+) and memory (CXCR5+ or CXCR5-) phenotype (n = 6).(F) The percentage of naive, CXCR5+ memory, and CXCR5- memory B cells within RV-specific B cells (n = 6).(G) The percentage of mono-specific and dual-specific B cells within naive, CXCR5+ memory, and CXCR5- memory B cell subsets (n = 6).(H) Total B cells stained for RV-A16 and ICAM-1 without ICAM-1 blocking.(I) Comparison of cell marker MFI of CXCR5+ and CXCR5- memory subsets (n = 8).(J) Representative scatterplot of memory B cells analyzed for T-bet and CXCR5 expression.(K) Comparison of isotype and T-bet expression in CXCR5+ and CXCR5- memory subsets (n = 8).Significance was determined by Wilcoxon matched pairs signed-rank test (D, E, F, G, I, and K). Geometric mean ± geometric SD (D, E, G, and I). Mean ± SD (F and K). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001.

Fig 3: Comparison of RNAi efficiency mediated by miR-E versus miR-AB in vitro and in vivo Two top-ranked SplashRNA-designed shRNAmirs targeting human Nurd complex components and a control shRNAmir targeting human CD4 were cloned into both miR-AB and miR-E lentiviral vectors with Venus as reporter. The lentiviruses packaged by these constructs were used for transduction of multiple human cell lines. RNAi efficiency was determined by western blot (WB) analysis of target protein levels in the lysates of the transduced cells. ß-actin served as loading control.A shERWOOD-designed shRNAmir targeting mouse transcriptional factor Tbx21 and a control shRNAmir targeting mouse CD4 were cloned into both miR-AB and miR-E retroviral vectors with Ametrine as reporter. The packaged viruses were used for transduction of primary mouse P14 TCR transgenic CD8+ T cells, which were used for evaluation of CD8+ T cells differentiation in vivo as described previously(Wang et al, 2018) (see details in the Materials and Methods section).Representative FACS plots of differentiation marker CD127/KLRG1 staining of the gated P14 CD8+ T cells and the statistical summary of SLEC subsets were shown to indicate CD8+ T cell differentiation. Data are shown as mean ± SD and n = 5 mice/group. * indicates significant difference (P < 0.01, two-tailed unpaired Student's t-test). Data are from one representative experiment of two biological replicates.The Tbx21 RNAi efficiency mediated by miR-AB vs. miR-E in the transduced in vitro-expanded P14 CD8+ T cells was determined by western blot analysis of T-bet protein level. ß-actin served as loading control.The RNAi efficiency of Pten.1524 and Pten.932 shRNAmir mediated by miR-AB versus miR-E was tested in NIH-3T3 cells in the context of < 30% transduction efficiency (GFP as reporter). Representative western blots of Pten protein expression of the FACS-sorted GFP+ cells from two biological replicates were shown. ß-actin served as loading control.Three shRNAmirs targeting the GFP reporter of a lentiviral vector in miR-AB or miR-E backbone were transduced into MC38 or 293T cells. The RNAi efficiency was determined by FACS analysis of GFP in the context of < 30% transduction efficiency. Data are shown as mean ± SD and n = 3 technical replicates. * indicates significant difference (P < 0.05). NS indicates no significant difference (P > 0.05, two-tailed unpaired Student's t-test). Source data are available online for this figure.