PMC

(a) Hy10 B cells were adoptively transferred together with OT-II T-cells and immunized with DEL-OVA. Seven days later mice were re-challenged with either cognate HEL-PE (red, left panel) or non-cognate NP-PE (red, right panel). 8 hours after challenge LNs were stained with CD35 (green) and IgD (blue). (b) QM B cells were transferred and recipients challenged with NP-CGG. Seven days later mice were rechallenged and analysed as in (a). (a) and (b) Data are representative of 3 experiments. (c) Three-dimensional reconstruction of follicle in Cr2^−/− BM chimera 7 days following DEL-OVA immunization, imaged in online. The capsule is visible in blue, GC B cells are green and adoptively transferred WT follicular B cells are cyan. HEL-PE was injected 3 hours before the lymph node was harvested and imaged. White track represents movement of a HEL-PE IC⁺ CFP⁺ B cell. (d) Immunohistochemical assessment of antigen distribution in WT and Cr2^−/− BM chimeras. Mice were challenged 7 days after immunization with HEL-OVA and 8 hours later draining LNs were stained for HEL (blue) and IgD (brown). Data are representative of 3 experiments. Scale bars (a), (b) and (d) indicate 100 µm. Co-ordinates in (c) indicate distance in µm.

(a) Mosaic tiled fluorescent microscopy of inguinal LN stained with mAbs to CD169 (green), F4/80 (red) and B220 (blue) to identify SCS and medullary macrophages in situ. Scale bar indicates 200 µm. Inset scale bar indicates 50 µm. Data are representative of 3 experiments. (b) and (c) Flow cytometric analysis of the same single cell suspension of protease-digested peripheral LNs stained with mAbs to CD11b, CD11c, CD169 and F4/80. (b) CD11c^hi classical DCs can be distinguished from CD11b⁺ CD11c^lo macrophages which can be further resolved into four subpopulations based on CD169 and F4/80. (c) Alternative gating strategy showing CD169^hiCD11c^lo cells comprise CD11b⁺ cells which are either F4/80⁻ SCS macrophages or F4/80⁺ medullary macrophages. (d) Forward and side scatter profiles of SCS and medullary macrophages gated as in (c). (e) and (f) PE-IC capture by SCS and medullary macrophages 2 hours after PE injection. (b)–(d) Data are representative of more than 5 experiments. (e) Representative pseudocolor plots from 3 experiments of PE-IC capture by CD169^hiCD11c^intF4/80⁻ SCS and CD169^hiCD11c^intF4/80⁺ medullary macrophages. Ly5.2⁺ cells serve as ex vivo mixing controls. Numbers indicate the percentage of cells in each quadrant. (f) Graph of pooled data from 3 experiments, n = 5 mice, showing PE mean fluorescence intensity (MFI) of medullary macrophages and SCS macrophages 2 hours after PE injection.

Wild-type mice that had been reconstituted with Cr2^−/− or WT BM received Hy10 B cells and OT-II T-cells on day-1. Mice were immunized on day 0 with the low affinity antigen DEL-OVA in adjuvant and analysed on day 7 and 14. (a) Flow cytometric analysis of draining LNs for Ly5.1, Ly5.2, Fas, IgD, HEL-binding BCR and IgG_2b. GC B cells are identified as Fas^hi IgD^lo and transferred cells as Ly5.2⁺. Middle- and far-right plots show HEL-binding and DEL-binding BCR and IgG_2b gated on donor-derived Hy10 B cells as indicated. Numbers indicate frequencies of cells in the gates or quadrants. Data are representative of 4 experiments. (b) Enumeration of GC response, class switching to IgG_2b and affinity maturation (DEL-binding) on day 14 using the gates shown in (a). Circles indicate individual LNs (4 experiments, n = 9 mice). (c) Serum anti-HEL and anti-DEL Igκ ELISA for low and high affinity antibodies, respectively, on day 7 (2 experiments, n = 5) and 14 of the response (4 experiments, n = 9 mice). HyHEL9 mAb which recognises a distinct epitope from Hy10 B cells was used to construct a standard curve for quantitation of antibody levels. (d) Somatic hypermutation (SHM) data on day 14. Mice were immunized as above and donor-derived GC B cells sorted on day 14 for single-cell PCR and sequence analysis. Arrow indicates the position of Y53. Replacement mutations are in red and silent mutations in blue bars. Frequency of Y53F mutations is significantly higher for WT (34/46) than Cr2^−/− BM chimera recipients (23/53, P = 0.002, Fischer’s exact test). Sequence data are from a single sorting experiment.

(a) Purification of SCS and medullary macrophages. Single cell suspensions of protease-digested peripheral LNs were enriched for large low density cells by Percoll density gradient. Single DAPI⁻ live cells were gated and B220⁺, CD4⁺, CD8⁺ cells dumped. CD169^hiCD11c^loCD11b⁺F4/80⁻ cells were sorted as SCS and CD169^hiCD11c^loCD11b⁺F4/80⁺ cells were sorted as medullary macrophages. Numbers indicate the percentage of cells in the gates shown. Pseudocolor plots are representative of 3 experiments, n = 10–15 mice for each sort. (b), (c) and (e) Comparison of the gene expression profiles of purified SCS and medullary macrophages showing (b) housekeeping genes, (c) macrophage transcription factors (e) lysosomal proteins, proteases and vacuolar ATPases. Each data point represents an independent experiment (three Affymetrix microarray experiments in total). In some cases there were multiple probe-sets for individual genes and hence more than 3 data points are shown. (d) Q-PCR analysis showing expression of Sfpi1 (PU.1) relative to housekeeping gene Hprt1 by purified SCS and medullary macrophages. T-cells serve as negative controls. Each data point represents one experiment. (f) Microscopy of purified SCS and medullary macrophages showing cell size and granularity by differential interference contrast microscopy (DIC, left panels), and immunofluorescence microscopy (IFM) to detect CD169 (green), LAMP-1 (red) and the nucleus (DAPI, blue). (g) Intracellular flow cytometric analysis for LAMP-1 and LAMP-2 protein in SCS macrophages (blue histogram) and medullary macrophages (red histogram) compared to isotype control staining. Data in (f) and (g) are representative of 3 experiments.

(a) and (b). Effect of acid stripping on macrophage retention of ICs. (a) PE and PE-specific rabbit IgG was injected to generate PE-ICs in vivo and macrophages isolated from draining lymph nodes 4 hours later. Representative pseudocolor plots from 3 experiments show flow cytometric analysis of CD169^hi CD11c^lo cells before and after acid stripping. (b) Graph shows a pool of MFI data for F4/80⁻ (SCS) and F4/80⁺ (medullary) cells where circles indicate individual LNs and bars indicate mean (3 experiments, n= 3 mice). (c) and (d) Measurement of IC internalization. (c) Cells prepared as in (a) were surface stained with anti-rabbit IgG to detect PE⁺ IgG⁻ cells. Representative pseudocolor plot from 6 experiments shows detection of PE-IC and rabbit IgG on CD169^hiCD11c^loF4/80⁻ SCS and CD169^hiCD11c^loF4/80⁺ medullary macrophages. (d) Graph of % PE⁺ IgG⁻ cells represents pooled data from 6 experiments, n = 6 mice. Numbers in (a) and (c) indicate percentage of cells in each gate. (e) In vivo endocytosis assay. Representative pseudocolor plots from 3 experiments shows gating for SCS and medullary macrophages with low (lo), intermediate (int) and high (hi) amounts of in vivo generated BSA-anti-BSA ICs 4 hours after DQ Green BSA injection. Ly5.2⁺ cells serve as a negative control. Overlay histogram shows amount of BSA degradation by SCS and medullary macrophages with the same level of BSA-anti-BSA ICs. (f) Time lapse-images from intravital two-photon microscopy shown in the first movie in online showing capture, disaggregation and unidirectional transport of discrete bright red PE-ICs along the cellular process of a SCS macrophage labeled green with CD169-specific mAb. Time stamp is in hh:mm:ss. Scale bars represent 10 µm.

(a), (b) and (c) analysis of SCS macrophages in B cell-deficient Igh-6^−/− mice. (a) Confocal microscopy of paracortical region from the inguinal LN of Igh-6^−/− mice stained with CD169 (green), F4/80 (red) and B220 (blue). Macrophages lining the SCS are all double positive for CD169 and F4/80. (b) Overlay histogram of flow cytometric analysis showing F4/80 expression by CD169^hiCD11c^lo of inguinal LN from Igh-6^−/− and Igh-6^+/− littermates. (a) and (b) Data are representative of 3 experiments. (c) Enumeration of SCS macrophages per inguinal LN of Igh-6^+/−, Igh-6^−/− and Igh-6^−/− reconstituted with WT or Lta^−/− B cells. (d) Enumeration of SCS macrophages per inguinal LN of WT and LTα transgenic mice. (c) and (d) Circles indicate individual inguinal LNs (3 experiments, n = 5 mice). (e) Confocal microscopy performed as in (a) of inguinal LN from LTα transgenic mouse. Data are representative of 3 experiments. (f) Q-PCR analysis of LTβR gene expression by purified SCS and medullary macrophages. Each data point represents a single experiment, n = 10–15 mice. B cells serve as negative control. (g) Overlay histogram of flow cytometric analysis for surface expression of LTβR by WT (blue) and Ltbr^−/− (red) SCS and medullary macrophages (left panel). Graph of LTβR MFI (right panel). Data are representative of and pooled from 3 experiments, n = 5 mice. (h) Enumeration of SCS macrophages in WT (blue) and Ltbr^−/− (red) mixed BM chimeras. The Ly5.2 congenic control is shown in white. Circles indicate individual inguinal LNs (3 experiments, n = 5 mice). (i) and (j) Mice were treated with LTβR-Fc or control human IgG for 4 weeks and lymph nodes stained to detect SCS and medullary macrophages in situ (i) or by flow cytometry (j). Left graph shows the enumeration of SCS macrophages and right panel of medullary macrophages. Circles indicate individual inguinal LNs (3 experiments, n = 6 mice). (k) Flow cytometric analysis of PE-IC capture by follicular B cells 2 hours after PE injection in mice treated for 2 weeks with control IgG or LTβR-Fc. Circles indicate individual inguinal LNs (3 experiments, n = 8 mice). (l) Reduced SCS macrophage and B cell IC capture following 3 day LTβR-Fc treatment. Circles indicate individual LNs (3 experiments, n = 8 mice). Scale bars in (a), (e) and (i) indicate 100 µm.