Fate mapping reveals origin and dynamics of lymph node follicular dendritic cells

Abstract

Follicular dendritic cells (FDCs) regulate B cell function and development of high affinity antibody responses but little is known about their biology. FDCs associate in intricate cellular networks within secondary lymphoid organs. In vitro and ex vivo methods, therefore, allow only limited understanding of the genuine immunobiology of FDCs in their native habitat. Herein, we used various multicolor fate mapping systems to investigate the ontogeny and dynamics of lymph node (LN) FDCs in situ. We show that LN FDC networks arise from the clonal expansion and differentiation of marginal reticular cells (MRCs), a population of lymphoid stromal cells lining the LN subcapsular sinus. We further demonstrate that during an immune response, FDCs accumulate in germinal centers and that neither the recruitment of circulating progenitors nor the division of local mature FDCs significantly contributes to this accumulation. Rather, we provide evidence that newly generated FDCs also arise from the proliferation and differentiation of MRCs, thus unraveling a critical function of this poorly defined stromal cell population.

Figure 1.

Tracking the ontogeny of FDCs. (A) Schematic showing that in Wnt-1Cre Ubow^+/− and ^+/+ mice, neural crest–derived head and neck mesenchymal cells stochastically express YFP or CFP at different ratios while the rest of the cells, including all hematopoietic cells, express dTomato (Ghigo et al., 2013). (B) Auricular and popliteal LN sections from nonirradiated Wnt-1Cre Ubow^+/− and ^+/+ mice were stained for CD21/35 expression and analyzed by confocal microscopy. F indicates monocolored foci of FDCs. Insets display high-magnification views of FDCs. (C) All confocal pictures from the auricular/cervical LNs of Wnt-1Cre Ubow^+/+ mice were processed to generate Voronoi tessellated pictures (see Materials and methods and Fig. S1) amenable to computational simulation. (D) The mean numbers of the rarest colored CD21/35⁺ FDCs (i.e., CFP⁺⁺ FDCs) per CFP⁺⁺ cluster were then compared with Monte Carlo–simulated pictures in which the same colored FDC populations were randomly distributed within the same region of interest (2 simulations displayed out of 10,000). Horizontal bar = median. Each dot represents one follicle. (E) CD21Cre Ubow^+/− and ^+/+ mice were irradiated and reconstituted with WT BM cells to generate chimeric mice in which B cell follicles would be composed of colored FDCs and nonfluorescent B cells. 8 wk later, their peripheral LNs were sectioned, stained for CD21/35 and IgD expression, and imaged by confocal microscopy. (F) Quantification of CFP⁺⁺ CD21/35⁺ FDC clustering index in the follicles of CD21Cre Ubow^+/+ chimera. Horizontal bar = median. Each dot represents one follicle. In D and F, a two-tailed Student’s t test was used to determine significance. *, P < 0.05. n.s. = nonsignificant. Bars, 25m. Data are representative of 4 different experiments (2 mice per experiment, at least 4 analyzed LNs per mouse).

Figure 2.

Reactive FDCs do not originate from circulating progenitors. (A) Wnt-1Cre Ubow^+/− mice were injected with an emulsion of CFA/OVA in their ears and rear footpads. 3 wk later, inflamed auricular and popliteal draining LNs were sectioned, stained for CD21/35 and IgD expression, and analyzed by confocal microscopy. Insets display high-magnification views of FDCs. Data are representative of 3 different experiments (2 mice per experiment, 4 analyzed LNs per mouse). (B) WT mice were joined surgically with naive CD21Cre Ubow^+/− µMt mice to create parabiotic mice. 3 mo later, when full chimerism was achieved, the WT partner was injected with CFA/OVA in its rear footpads. 3 wk later, popliteal LNs of both mice were sectioned, stained for CD21/35 and IgD expression, and the presence of colored FDCs in their LNs was analyzed by confocal microscopy. The development of mature FDC networks in the µMt mouse combined to the appearance of dTomato⁺ leukocytes in the WT partner indicates successful chimerism. Data are representative of 3 different experiments (2 mice per experiment). (C) CD21Cre Ubow^+/+ chimeras were injected with an emulsion of CFA/OVA in their ears and rear footpads. 3 wk later, inflamed draining LNs were sectioned, stained for CD21/35, and IgD expression and analyzed by confocal microscopy. (D) Quantification of CFP⁺⁺ CD21/35⁺ FDC clustering index in the follicles of CD21Cre Ubow^+/+ chimera. Horizontal bar = median. Data are representative of 4 different experiments (2 mice per experiment, 4 LNs analyzed per mouse). Each dot represents one follicle. In A and D, a two-tailed Student’s t test was used to determine significance. n.s = nonsignificant. Bars, 25 µm.

Figure 3.

Lineage tracing of FDC progenitors. Ubow⁺⁺-CreERT2 mice were irradiated and reconstituted with WT BM cells to generate chimeric mice with fluorescent stromal cells and nonfluorescent hematopoietic cells. Reconstituted chimeras were treated with tamoxifen, maintained under tamoxifen-free regimen during 1 wk, and injected with an emulsion of CFA/OVA in their ears and rear footpads (left side only). 3 wk later, inflamed popliteal and auricular LNs (B and C) and contralateral LNs (A) were sectioned, stained for CD21/35 (dashed line) and IgD (A and B) or RANK-L (D) expression, and analyzed by confocal microscopy. (C) Quantification of CFP⁺⁺ YFP⁺⁺ CFP⁺YFP⁺ CD21/35⁺ FDC clustering index. Horizontal bar = median. Data are representative of 3 different experiments (2 mice per experiment, 4 LNs analyzed per mouse). Each dot represents one follicle. (E) Comparison of MRC/FDC cluster formation as measured by the clustering index (see Materials and methods section and Fig. S2 for details) in original and simulated data. Each dot represents the cell clustering index of a cell cluster composed of MRCs and FDCs sharing a similar color, in original and Monte Carlo–simulated data. See also Video 1. (F) Auricular/cervical LN sections from Wnt-1Cre Ubow^+/+ mice were stained for CD21/35 and RANK-L expression and analyzed by confocal microscopy. Insets in D and F represent magnified views of juxtaposed MRCs and FDCs sharing a similar color. Data are representative of 3 different experiments (2 mice per experiment, 2 LNs analyzed per mouse). In E and C, a two-tailed Student’s t test was used to determine significance. *, P < 0.08; **, P < 0.01. n.s. = nonsignificant. Bars, 25 µm.

Figure 4.

Cellular filiation in the Ubow mouse. (A) Schematic describing that when Ubow cells express Cre, they simultaneously lose dTomato and acquire CFP or YFP expression. Applied to a precursor/product relationship and in combination with the proper Cre reporter strain, this system unravels the transitional stage between progenitors and their differentiated cells. (B) Non-chimeric Ubow^+/−-CreERT2 mice were injected with a single dose of tamoxifen. Mice were bled every day and the relative proportion of CFP-, YFP-, and dTomato-expressing B220⁺ B cells were assessed by flow cytometry (same results were observed in CD3⁺ T cells). Data are representative of 3 different mice from 3 different experiments. (C) Schematic showing the expected phenotypes and colors of MRCs (RANK-L^hi CD21^neg) and immature (RANK-L^med CD21^med) and mature FDCs (RANK-L^neg CD21^hi) in CD21Cre Ubow mice according to the model in which MRCs give rise to FDCs.

Figure 5.

Tracking CD21⁻ FDC progenitors. (A) CD21Cre Ubow^+/+ chimeras were injected with an emulsion of CFA/OVA in their ears and rear footpads to trigger FDC remodeling in their draining LNs. 3 wk later, inflamed draining LNs were sectioned and stained for CD21/35 and RANK-L expression. (B) CD21Cre Ubow^+/− µMT mice were adoptively transferred with 6 × 10⁷ WT polyclonal B cells to trigger FDC development. 1 wk later, peripheral LNs were sectioned and stained for CD21/35 and RANK-L expression. In A and B, data are representative of 3 different experiments (3 mice per experiment, 4 LNs analyzed per mouse). Insets display high-magnification views of MRC and transitional and mature FDCs. (C) Confocal pictures acquired in all the experiments described in A and B were used to determine the colors and absolute numbers of MRCs, transitional and mature FDCs, and other cells present in B cell follicles. Bars, 25 µm.

Figure 6.

MRCs proliferate during inflammation. (A) Flow cytometry gating strategies used to identify MRCs (gp38⁺ MAdCam-1⁺ CD45⁻ CD31⁻ CD21/35⁻) and FDCs (gp38⁺ CD45⁻ CD31⁻ CD21/35⁺) in LN cellular suspensions. WT mice (B) and CD21Cre Ubow⁺⁺ chimeras (C) were subcutaneously injected or not with an emulsion of CFA/OVA. (B) 4, 9, or 20 d later, mice received a single i.p. injection of EdU to label proliferating cells. 1 d later, the percentages of EdU⁺ MRCs and EdU⁺ FDCs present in the peripheral inflamed LNs of WT mice were analyzed by flow cytometry. (C) The percentage of EdU⁺ MRCs (arrowheads) and EdU⁺ FDCs was determined at day 5 by confocal imaging in the inflamed LNs of CD21Cre Ubow⁺⁺ chimeras. Insets display high-magnification views of EdU⁺ MRCs. Bars, 25 µm. Data are representative of 3 experiments (at least 4 mice pooled per group in B and 2 LNs analyzed per mouse in C). (D) WT mice were injected s.c. with CFA/OVA in ears and footpads, followed by BrdU injection (i.p) on day 4. This pulse of BrdU was followed by a chase period of 1 and 3 d. At the end of the chase period, the percentage of BrdU-labeled MRCs and FDCs was determined by flow cytometry. Data are representative of 3 different experiments (5 mice pooled per group). (E) RAG-2°^/° mice were adoptively transferred or not with 6 × 10⁷ WT polyclonal B cells. 1 wk later, mice received a single injection of EdU. The proportion of EdU⁺ cells among the MRCs and FDCs of peripheral LNs were analyzed one day later by flow cytometry. At this stage, no mature FDCs could be recovered from the LNs of both types of mice. Data are representative of 3 different experiments (5 mice pooled per group).

Figure 7.

Postmitotic LN MRC network is remodeled upon initial B cell colonization. (A) RAG-2°^/° Ubow mice were irradiated and reconstituted with RAG-2°^/° nonfluorescent BM cells to generate chimeric mice with dTomato-expressing stromal cells. Reconstituted chimeras were adoptively transferred or not with 6 × 10⁷ CMFDA-labeled polyclonal WT B cells to trigger FDC development in recipient LNs. 1 d later, peripheral LNs were sectioned, stained for RANK-L expression, and analyzed by confocal microscopy. L displays the width of the MRC network based on RANK-L staining. (B) The number of RANK-L⁺ dTomato⁺ cellular bodies was manually counted on tissue sections and used to calculate the density of the dTomato⁺ RANK-L⁺ MRC network in the two types of LNs, either in its globality (left) or in the first 30 µm below the SCS (right). Data are representative of 3 different experiments (2 mice per experiment, 6 LNs analyzed per mouse). (C) Representative high-magnification views of the MRC network present in the 2 types of LNs. (D) Auricular/cervical LN sections from Wnt-1Cre Ubow^+/+ RAG-2°^/° mice were stained for RANK-L expression and analyzed by confocal microscopy. Note the presence of MRCs Foci (F) in the SCS of these LNs devoid of B cells. Insets on the right display high-magnification views of MRC clusters. (E) Quantification of CFP⁺⁺ RANK-L⁺ MRC clustering index in the auricular/cervical LNs of Wnt-1Cre Ubow^+/+ RAG-2°^/° mice. A two-tailed Student’s t test was used to determine significance. *, P < 0.05; ***, P < 0.001. Bar = median. Data are representative of 5 individual mice (4 LNs analyzed per mouse) obtained in 2 independent experiments. Each dot represents one follicle. Bars, 25 µm.