Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2018 Jan 25;172(3):517-533.e20.
doi: 10.1016/j.cell.2017.11.036. Epub 2017 Dec 14.

Initiation of Antiviral B Cell Immunity Relies on Innate Signals from Spatially Positioned NKT Cells

Affiliations

Initiation of Antiviral B Cell Immunity Relies on Innate Signals from Spatially Positioned NKT Cells

Mauro Gaya et al. Cell. .

Abstract

B cells constitute an essential line of defense from pathogenic infections through the generation of class-switched antibody-secreting cells (ASCs) in germinal centers. Although this process is known to be regulated by follicular helper T (TfH) cells, the mechanism by which B cells initially seed germinal center reactions remains elusive. We found that NKT cells, a population of innate-like T lymphocytes, are critical for the induction of B cell immunity upon viral infection. The positioning of NKT cells at the interfollicular areas of lymph nodes facilitates both their direct priming by resident macrophages and the localized delivery of innate signals to antigen-experienced B cells. Indeed, NKT cells secrete an early wave of IL-4 and constitute up to 70% of the total IL-4-producing cells during the initial stages of infection. Importantly, the requirement of this innate immunity arm appears to be evolutionarily conserved because early NKT and IL-4 gene signatures also positively correlate with the levels of neutralizing antibodies in Zika-virus-infected macaques. In conclusion, our data support a model wherein a pre-TfH wave of IL-4 secreted by interfollicular NKT cells triggers the seeding of germinal center cells and serves as an innate link between viral infection and B cell immunity.

Keywords: B cells; CXCR3; IL-4; NKT cells; Zika virus; germinal center seeding; influenza; lymph node; macrophages; viral infection.

PubMed Disclaimer

Figures

None
Graphical abstract
Figure 1
Figure 1
NKT Cell-Deficient Mice Exhibit Impaired Antiviral B Cell Responses (A and B) Flow cytometry analysis of mediastinal lymph nodes (LN) from control (black), wild-type (blue), or CD1d−/− (red) mice on day 9 of influenza infection. Contour plots display the percentage of (A) germinal center and (B) TfH cells. (C) Confocal microscopy images of mediastinal lymph node sections from mice treated as in (A) and (B), stained with antibodies to IgD (green) and GL7 (magenta). Scale bar, 450 μm. (D) Quantification of the number and follicular area occupied by germinal centers in confocal images shown in (C) by ImageJ. (E) ELISPOT analysis of IgG1 influenza-specific ASCs in mediastinal lymph nodes from wild-type and CD1d−/− mice treated as described in (A) and (B). (F and G) Flow cytometry analysis of mediastinal lymph nodes of from control (black), wild-type (blue), or CD1d−/− (red) mice on day 9 of vaccinia virus infection. Contour plots display the percentage of (F) germinal center and (G) TfH cells. Data show one representative result from three experiments, where each dot represents one mouse. Data represent mean ± SEM; two-tailed paired Student’s t test, p < 0.05, ∗∗p < 0.01, and ∗∗∗∗p < 0.0001.
Figure S1
Figure S1
NKT Cell-Deficient Mice Display Early Impairment in B Cell Immunity, Related to Figure 1 (A–D) Flow cytometry analysis of mediastinal lymph node cells from wild type or CD1d−/− mice that were intranasally infected with 200 PFU of influenza virus for (A-B) 7 or (C-D) 21 days. Representative contour plots display the percentage of (A-C) B220+ cells bearing biomarkers of germinal cell activity, Fas+GL7+, and (B–D) CD4+ T cells bearing biomarkers of TfH cells, CXCR5+PD-1+. Dot plots show quantification of (A-C) germinal center and (B-D) TfH cells. (E and F) Flow cytometry analysis of mediastinal lymph nodes harvested at day 10 from wild type mice that were intranasally challenged with (E) 10 μg of HA trimmer or (F) PBS in the presence or absence of poly I:C or α-GalCer. Representative contour plots display the percentage of HA-specific B220+ cells differentiated into Fas+GL7+ germinal center cells. (G and I) ELISPOT analysis of IgG1 HA-specific ASCs in mediastinal lymph nodes from wild type and CD1d−/− mice treated as in figure S1E. (H and J) Flow cytometry analysis of HA-specific germinal center cells in mediastinal lymph nodes from wild type and CD1d−/− mice treated as in figure S1E. In all quantification charts, each dot represents one mouse. Data are representative from two experiments. Statistical analysis two-tailed Student’s t test; p < 0.05.
Figure 2
Figure 2
NKT Cells Deliver Non-cognate Help to B Cells during Viral Infection (A and B) Flow cytometry analysis of wild-type mice infected with influenza virus. (A) Gating strategy for NKT and CD4+ T cells. B220+ cells were excluded from the analysis. (B) Analysis of the percentage of NKT (blue) and CD4+ T (red) cells that acquire CXCR5 and PD-1 markers after infection. (C) Confocal microscopy analysis of lymph nodes on day 9 of influenza infection incubated with CD1d tetramer (magenta) and GL7 (green). Charts show the number and density of NKT cells inside and outside of germinal centers. Each dot represents a single germinal center. Scale bar, 60 μm. (D) Schematic showing the CD1d locus targeted for deletion with two loxP sites flanking exon 3 of CD1d. (E) Representative contour plot shows the gating strategy for B cells. (F and G) Flow cytometry analysis of CD1d expression in B cells from (F) CD1dflox/floxCD19-Cre or (G) CD1dflox/floxMb1-Cre mice: Cre (blue) and Cre+ (red). (H–K) Flow cytometry analysis of germinal center and TfH cell formation in (H and I) CD1dflox/floxCD19-Cre or (J and K) CD1dflox/floxMb1-Cre mice on day 9 of influenza infection: Cre (blue dots) and Cre+ (red dots). All data are representative of 3 independent experiments. Unless specified, each dot represents one mouse. Horizontal bars, mean; error bars, SEM; statistical analysis, two tailed Student’s t test, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001.
Figure S2
Figure S2
Expression Analysis of TfH Markers by NKT Cells and Strategy for the Generation of CD1d-Floxed ESCs, Related to Figure 2 (A) Flow cytometry gating strategy for the analysis of NKT cells (TCRβ+ CD1d tetramer+), CD4+ T cells (TCRβ+ CD1d tetramer- CD4+) and TfH cells (TCRβ+ CD1d tetramer- CD4+ CXCR5+ PD-1+) in mediastinal lymph nodes at day 9 of influenza infection. (B) RT-qPCR analysis of IL-21 expression at day 3 and 9 of influenza infection by sorted NKT and TfH cells compared to CD4+ cells. (C and D) Representative contour plots show the expression of (C) Bcl-6 and (D) SLAM by NKT, TfH and CD4+ T cells at day 3 and 9 of infection. Quantification on the right charts shows the mean fluorescence intensity for (C) Bcl-6 and (D) SLAM. (E) Scheme showing the wild type and targeted CD1d locus containing two loxP sites flanking the exon 3 of CD1d as well as two FRT sites for the removal of the Neomycin resistance cassette. (F) Specific fragments amplified by long-range PCR for the 5′ end (7.1kb) and 3′ end (6.6kb) of the clones that correctly performed homologous recombination. (G) DNA from three positive clones was digested with SacI restriction enzyme and Southern blot performed using a labeled Southern probe. The upper band corresponds to the wild type locus (10.6kb) and the lower band to the targeted locus (7.3kb). (H) Southern blot of the three positive clones performed with a labeled Southern probe against the Neomycin cassette. Orange arrow indicates the single 5.2kb band recognized by this probe, indicating single insertion of the construct. The clone 2, in red, was chosen for injections. In all quantification charts, each dot represents one mouse. Data are representative from three experiments. Statistical analysis two-tailed Student’s t test; ∗∗p < 0.01, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001.
Figure 3
Figure 3
NKT Cells Generate a Pre-TfH Cell Wave of IL-4 (A) Flow cytometry analysis of CD69 levels in the TCRβ+ CD1d tetramer+ population after 0–6 days of influenza infection. MFI, mean fluorescence intensity. (B) Representative plots showing the percentage of NKT cells producing IFN-γ (blue) and IL-4 (red) after 3 days of influenza infection. (C) Flow cytometry analysis of IFN-γ (blue) and IL-4 (red) production on day 3 of influenza infection. Gates were drawn on IFN-γ+ and IL-4+ lymph node cells to analyze the percentage of NKT cells inside each cytokine+ population. (D–F) Flow cytometry analysis of NKT (blue) and TfH (red) cells from IL-4 GFP mice infected with influenza virus. (D) Numbers of NKT and TfH cells at different stages of influenza infection. (E) GFP expression in NKT and TfH cells at different times of infection. (F) Percentage of CD1d-tetramer+ NKT cells and CXCR5+ TfH cells inside the IL-4 GFP+ lymph node cell population at different stages of influenza infection. In all charts, each dot represents one mouse. In graphs showing statistical analysis, data are representative of three independent experiments. Horizontal bars, mean; error bars, SEM. Student’s t test, ∗∗∗∗p < 0.0001.
Figure S3
Figure S3
Analysis of the NKT Cell Source of IL-4, Related to Figure 3 (A) Quantification of IFN-γ+ (blue) and IL-4+ (red) producing lymph node (left chart) or NKT cells (right chart) in mediastinal lymph nodes after 3 days of influenza infection. (B) RT-qPCR analysis of IL-4, IL-5 and IL-13 expression at day 3 of influenza infection by sorted NKT cells compared to CD4+ T cells. Each dot represents one mouse. (C) Adoptive transfer strategy for the analysis of the source of IL-4+ NKT cells after influenza infection. (D and E) Flow cytometry analysis at day 3 of influenza infection of (D) mediastinal lymph node or (E) lung NKT cells from mice that were treated as in (C) to detect IL4-GFP expression. Data are representative from two experiments.
Figure 4
Figure 4
The Early IL-4 Wave Is Localized at the Periphery of B Cell Follicles (A–C) Confocal microscopy analysis of (A and B) wild-type and CD1d−/− and (C) IL4-GFP mice on day 3 of influenza infection. Lymph nodes were labeled with CD1d tetramer (magenta) and anti-B220 antibody (green in A and B and white in C). The arrows in (C) indicate CD1d tetramer+ cells expressing IL-4 (IL-4 GFP, green). Scale bars, 300 μm (lymph node) and 60 μm (section). (D) Flow cytometry analysis of IL-4 GFP+ cells in mediastinal lymph nodes on day 3 of influenza infection, showing CD1d-tet and CD1d-tet+ cells. (E) t-SNE plots of CD1d-tet and CD1d-tet+ subsets. The CD1d-tet+ population separates into two clusters (1, dark purple; 2, light purple), as does the CD1d-tet population (3, light green; 4, dark green). (F and G) Expression distribution (violin plots) in each population (horizontal axes) for (F) CD3e, Zbtb16, CXCR6, and CXCR3 and (G) Gata3, Tbx21, and Rorc. Expression levels are log2(tpm+1). (H) Flow cytometry analysis of CXCR3 levels in IL-4-GFP (red) and GFP+ (blue) NKT cells on day 3 of influenza infection. Lines connect cells from the same mouse. (I) Flow cytometry analysis of IL-4 production by CXCR3 (red) and CXCR3+ (blue) NKT cells on day 3 of influenza infection. Lines connect cells from the same mouse. (J) ELISPOT analysis of IL-4-producing CXCR3 (red) and CXCR3+ (blue) NKT cells sorted on day 3 of influenza infection. Each dot represents pooled NKT cells from 6 mice. In graphs showing statistical analysis, data are representative of at least two independent experiments. Horizontal bars, mean; error bars, SEM. Student’s t test, p < 0.05, ∗∗p < 0.01.
Figure S4
Figure S4
Single-Cell RNA-Seq Analysis of IL-4+ Lymph Node Cells, Related to Figure 4 (A) Confocal microscopy of mediastinal lymph nodes sections at day 3 of influenza infection. Lymph nodes were labeled with PBS-57 loaded CD1d tetramer (magenta) and antibodies to B220 (white) and CD169 (green). IF, interfollicular. Scale bars, 300 μm (left); 60 μm (right). Chart on the right shows the quantification of NKT cells-macrophages contacts observed in the different areas of the lymph node at day 0, 1, 2 and 3 of infection in mediastinal lymph nodes. Data are representative from three experiments. (B) Flow cytometry analysis of IL4-GFP+ CD1d-tet- cells in mediastinal lymph nodes at day 3 of influenza infection showing TCRβ- and TCRβ + cells. (C) Single-cell RNA-seq heatmap showing the expression of 25 significant marker genes (Table S1) for the CD1d-tet+ clusters (columns 1 and 2) and CD1d-tet- clusters (columns 3 and 4). Single cells and genes were determined on the basis of unsupervised hierarchical clustering (Table S1). (D) Expression distribution (violin plots) in each population (horizontal axes) for Ccr7, Sell, S1pr1, CD4, CD8a and Bcl6. Expression levels are Log2(tpm+1). (E) Flow cytometry analysis of IL4-GFP+ TCRβ+ CD1d-tet- cells in mediastinal lymph nodes at day 3 of influenza infection showing CD4+ and CD8a + cells. (F) Representative contour plot show the expression of GATA-3 by NKT cells, ILC2s (Live/Dead- B220- CD3- CD5- CD11b- CD11c- CD127+ CD45+ GATA-3+) and CD4+ T cells at day 3 of infection. Quantification on the right chart shows the mean fluorescence intensity for GATA-3. (G) Flow cytometry analysis of RORγt+ and T-bet+ NKT cells in mediastinal lymph nodes at day 3 of influenza infection. Histogram shows the levels of CXCR3 in RORγt+ and T-bet+ populations. Dot graphs show the quantification of CXCR3+ NKT cells in the RORγt+ (red dots) and T-bet+ populations (blue dots), with lines connecting NKT cells from the same mouse. In graphs showing statistical analysis, data are representative of at least two independent experiments. Statistical analysis two-tailed Student’s t test; ∗∗p < 0.01, ∗∗∗p < 0.001 and ∗∗∗∗p < 0.0001.
Figure 5
Figure 5
Resident Macrophages Promote Early IL-4 Production by NKT Cells and Antiviral B Cell Immunity (A–C) Flow cytometry analysis of IL-4 production by NKT cells in (A) CD1dflox/floxMb1-Cre, (B) CD1dflox/floxLyz2-Cre, and (C) CD1dflox/exCD11c-Cre mice on day 3 of influenza infection: Cre (blue dots) and Cre+ (red dots). (D and E) Confocal microscopy analysis of CD169DTR/+ mice after 4 days of (D) PBS or (E) diphtheria toxin (DT) administration. Sections were stained with antibodies to B220 (white) and CD169 (green). Scale bars, 300 μm (left) and 60 μm (center and right). (F) Flow cytometry analysis of IL-4 production by NKT cells from CD169DTR/+/PBS-treated or CD169DTR/+/DT-treated mice on day 3 of influenza infection: PBS (blue dots) and DT (red dots). (G and H) Flow cytometry analysis of germinal center (G) and TfH (H) cells in CD169DTR/+/PBS-treated and CD169DTR/+/DT-treated mice on day 9 of influenza infection: PBS (blue dots) and DT (red dots). In all panels, each dot represents one mouse. Data show one representative result from three experiments. Data represent mean ± SEM; two-tailed paired Student’s t test, ∗∗p < 0.01, ∗∗∗∗p < 0.0001.
Figure S5
Figure S5
Conditional Targeting of CD1d in the Myeloid Lineage, Related to Figure 5 (A–C) Flow cytometry plots showing the gating strategy for (A) Gr-1+CD11b+ neutrophils (NFs) and myeloid-derived suppressor cells (MDSCs) and (C) MHCII+CD11c+ dendritic cells (DCs). Red and blue histograms depict the levels of CD1d in the gated populations in (A) CD1dflox/floxLyz2-Cre- versus Cre+ and (B) CD1d−/− mice or (C) CD1dflox/exCD11c-Cre- versus Cre+ mice. These quantitative differences are depicted via dot plots; blue and red dots show CD1d+ cells in the Cre- and Cre+ animals, respectively. (D) Confocal microscopy of mediastinal lymph nodes sections from wild type, CD1d−/−, CD1dflox/floxLyz2-Cre+ and CD1dflox/exCD11c-Cre mice. Sections were stained with antibodies to CD169 (green) and CD1d (magenta). Scale bars, 60 μm. (E–H) Flow cytometry analysis of mediastinal lymph node cells at day 9 of influenza infection. Representative contour plots show the percentage of B cells and T cells from (E-F) CD1dflox/floxLyz2-Cre- and Cre+ or (G-H) CD1dflox/exCD11c-Cre- and Cre+ mice that acquire (E and G) germinal center markers, Fas and GL7, or (F and H) TfH cell markers, CXCR5 and PD-1. Quantifications are shown in dot plots, Cre- mice (blue dots) and Cre+ mice (red dots). In all panels, each dot represents a single mouse. Horizontal bars – mean; error bars – SEM. Data are representative of three independent experiments. Statistical analysis, two tailed Student’s t test, p < 0.05, ∗∗∗∗p < 0.0001.
Figure 6
Figure 6
Resident Macrophages Promote Early NKT Cell IL-4 Production through CD1d and IL-18 (A) Experimental scheme showing the strategy for the generation of chimeras lacking CD1d specifically on CD169+ macrophages. (B) Flow cytometry analysis of IL-4 production by NKT cells from mice subjected to the experimental scheme show in (A): wild-type chimeras (blue dots) and CD1d−/− chimeras (red dots). (C and D) Confocal microscopy analysis on day 2 of influenza infection showing (C) SCS and (D) medullar areas. Sections were stained with antibodies to CD169 (green) and IL-1β/IL-18/IL-33 (magenta). Scale bars, 60 μm. (E–I) Flow cytometry analysis of wild-type and (E) MyD88−/−, (F) TLR7−/−, (G) IL-1R−/−, and (H) IL-18R−/− mice or (I) IL-33R−/− chimeric mice on day 3 of influenza infection. Charts show IL-4+ NKT cells in wild-type (blue dots) or mutant (red dots) populations. (J) ELISPOT analysis of IL-4 production by sorted NKT cells incubated ex vivo with IL-18. Each dot represents one mouse. Data show mean ± SEM and are representative of 3 independent experiments. Statistical analysis: two-tailed Student’s t test, ∗∗p < 0.01, ∗∗∗p < 0.001.
Figure S6
Figure S6
Intrinsic Role of MyD88 and IL-18R in NKT Cell Production of IL-4, Related to Figure 6 (A) Quantification of IL-1β+, IL-18+ and IL-33+ cells in the SCS and medullar areas from the images shown in Figures 6C and 6D. (B) Confocal microscopy of mediastinal lymph nodes from wild type mice that were infected with 200 PFU of PR8 NS1-GFP and sacrificed after 3 days. Sections were stained with antibodies to CD169 (red), B220 (gray) and Langerin (magenta). Scale bars, 60 μm. Quantification of GFP+ cells is shown in the right chart. (C–E) Flow cytometry analysis of chimeric mice that were generated by transferring a 1:1 bone marrow mixture of wild type (CD45.1+) and (C) wild type (CD45.2+), (D) MyD88−/− (CD45.2+) or (E) IL-18R−/− (CD45.2+) to sub-lethally γ-irradiated CD45.1 recipient mice. Eight weeks after bone marrow transplant, mice were infected with 200 PFU of influenza virus and mediastinal lymph nodes were harvested after further 3 days. Flow cytometry panels show the production of IL-4 by CD45.1+ and CD45.2+ NKT cells in the different sets of chimeras. Dot graphs show the quantification of IL-4+ NKT cells in the CD45.1+ population (blue dots) and the CD45.2+ population (red dots), with lines connecting the two NKT cell subpopulation coming from the same chimera. In all panels, each dot represents a single mouse. Horizontal bars – mean; error bars – SEM. Data are representative of two independent experiments. Statistical analysis, two tailed Student’s t test, p < 0.05, ∗∗p < 0.01.
Figure 7
Figure 7
A Conserved Early IL-4 Wave Is Critical for Induction of B Cell Immunity (A) Flow cytometry analysis on day 3 of influenza infection. Graphs show the number of IL-4+ lymph node cells in wild-type (blue dots) and CD1d−/− (red dots) animals. (B–D) Flow cytometry analysis on day 8 of influenza infection. Graphs show the percentage of (B) germinal center, (C) TfH, and (D) IgG1+ cells in wild-type (blue dots) and IL-4−/− mice (red dots). (E and F) Flow cytometry analysis on day 8 of influenza infection of wild-type mice treated with PBS or blocking α-IL-4 antibody during the initial 3 days of infection. Graphs show (E) germinal center and (F) IgG1+ cells in wild-type (blue dots) and α-IL-4 treated mice (red dots). (G and H) Flow cytometry analysis on day 8 of influenza infection of CD1d−/− mice treated with PBS or IL-4 complex (IL-4c) during the initial 3 days of infection. Graphs show (G) germinal center cells and (H) IgG1+ cells in wild-type (blue dots) and IL-4c-treated mice (red dots). (I) Experimental scheme showing the Zika virus infection strategy. (J) Linear regression model of gene signatures for NKT cells, TfH cells, IL-4, STAT6, and IL-18 from harvested lymph nodes on day 3 with neutralizing antibodies measured in plasma on day 7. Red squares, positive correlation; blue squares, negative correlation; white squares, no significant correlation. (K) Gene-interacting network inference representing interconnectivity of NKT cells, IL-4, STAT6, and IL-18 pathways. In (A)–(H), each dot represents a single mouse. Data are representative of three independent experiments. Statistical analysis: two-tailed Student’s t test, p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001.
Figure S7
Figure S7
Importance of Early IL-4 Production on B Cell Immunity, Related to Figure 7 (A) Representative confocal microscopy images of mediastinal lymph node sections from mice that were treated as in Figure 7E. Sections were stained with antibodies to the B lymphocyte marker, IgD (green) and the germinal center marker, GL7 (magenta). Scale bar, 450 μm. Quantification of the germinal center area in confocal images by ImageJ is shown in the right chart. (B) Flow cytometry analysis at day 8 of mediastinal lymph nodes from wild type mice that were treated with PBS or anti-IL4 blocking antibody on day −3, and infected with 200 PFU of influenza virus at day 0. Contour plots show Fas+GL7+ germinal center cells. The responses are quantified in right charts. (C) Experimental scheme showing the strategy for the generation of antigen-specific TfH cell precursors. (D–F) Flow cytometry analysis of mediastinal lymph node cells from mice treated as in (C). Representative contour plots display (D) the gating strategy used to differentiate endogenous CD4+ T cells from adoptively transferred OT-II cells, (E) the percentage of endogenous and exogenous cells bearing the TfH cell markers PD-1 and CXCR5 and (F) the percentage of B cells that acquire GL7 marker. Quantifications are shown on the right charts. (G and H) OCR at baseline and after sequential treatment with oligomycin, FCCP and Rotenone of B cells that were (G) stimulated ex vivo with anti-IgM and/or IL-4 or (H) isolated from lymph node cells of influenza-infected mice (day 3). Bar charts show the respiratory capacity obtained as (OCR after FCCP)- (basal OCR). Each dot represents an individual measurement. (I) Experimental scheme showing Zika virus infection strategy and time points at which lymph nodes and serum were collected from macaques. (J) Linear regression model of gene signatures for NKT cells, TfH cells, B cells, IL-4, STAT6 and IL-18 from harvested lymph nodes at day 14 with neutralizing antibodies measured in plasma at day 17. Red squares represent positive correlation, blue squares represent negative correlation and white squares represent no significant correlation. (K) Gene interacting network inference using GeneMANIA software representing interconnectivity of NKT, TfH cells, B cells, IL-4, STAT6 and IL-18 pathways and the co-expression of their leading genes. Unless stated, each dot represents a single mouse. Horizontal bars – mean; error bars – SEM. Data are representative of two independent experiments. Statistical analysis, two tailed Student’s t test, p < 0.05, ∗∗p < 0.01 and ∗∗∗∗p < 0.0001.

Comment in

References

    1. Adachi O., Kawai T., Takeda K., Matsumoto M., Tsutsui H., Sakagami M., Nakanishi K., Akira S. Targeted disruption of the MyD88 gene results in loss of IL-1- and IL-18-mediated function. Immunity. 1998;9:143–150. - PubMed
    1. Aid M., Abbink P., Larocca R.A., Boyd M., Nityanandam R., Nanayakkara O., Martinot A.J., Moseley E.T., Blass E., Borducchi E.N. Zika Virus Persistence in the Central Nervous System and Lymph Nodes of Rhesus Monkeys. Cell. 2017;169:610–620.e14. - PMC - PubMed
    1. Amigorena S., Drake J.R., Webster P., Mellman I. Transient accumulation of new class II MHC molecules in a novel endocytic compartment in B lymphocytes. Nature. 1994;369:113–120. - PubMed
    1. Artiaga B.L., Yang G., Hackmann T.J., Liu Q., Richt J.A., Salek-Ardakani S., Castleman W.L., Lednicky J.A., Driver J.P. α-Galactosylceramide protects swine against influenza infection when administered as a vaccine adjuvant. Sci. Rep. 2016;6:23593. - PMC - PubMed
    1. Bai L., Deng S., Reboulet R., Mathew R., Teyton L., Savage P.B., Bendelac A. Natural killer T (NKT)-B-cell interactions promote prolonged antibody responses and long-term memory to pneumococcal capsular polysaccharides. Proc. Natl. Acad. Sci. USA. 2013;110:16097–16102. - PMC - PubMed

Publication types