RANK and its ligand RANKL play important roles in the development and regulation of the immune system. We show that mice transgenic for Rank in hair follicles display massive postnatal growth of skin-draining lymph nodes. The proportions of hematopoietic and nonhematopoietic stromal cells and their organization are maintained, with the exception of an increase in B cell follicles. The hematopoietic cells are not activated and respond to immunization by foreign Ag and adjuvant. We demonstrate that soluble RANKL is overproduced from the transgenic hair follicles and that its neutralization normalizes lymph node size, inclusive area, and numbers of B cell follicles. Reticular fibroblastic and vascular stromal cells, important for secondary lymphoid organ formation and organization, express RANK and undergo hyperproliferation, which is abrogated by RANKL neutralization. In addition, they express higher levels of CXCL13 and CCL19 chemokines, as well as MAdCAM-1 and VCAM-1 cell-adhesion molecules. These findings highlight the importance of tissue-derived cues for secondary lymphoid organ homeostasis and identify RANKL as a key molecule for controlling the plasticity of the immune system.

The immune system displays a great capacity to adapt to environmental changes to provide the best-tailored response to a given immune challenge. Secondary lymphoid organs (SLOs), such as lymph nodes (LNs), play an important part in this activity. They are the anatomical sites where information from tissue is integrated by the immune cells. Over the last years, it has emerged that nonhematopoietic stromal cells that support the parenchyma of SLOs play a key role in orchestrating the immune response (1, 2). Stroma comprises vascular blood endothelial cells (BECs), lymph endothelial cells (LECs), and fibroblastic reticular cells (FRCs), which are interconnected through a microconduit system (3). The FRC population consists of reticular cells of the T cell zone (TRCs) and the sinus marginal zone (MRCs), as well as B cell follicular dendritic cells (FDCs) (47). Vascular cells regulate entry and exit of lymphocytes (811), and FRCs function in the homeostasis of T and B cells (12, 13); together, they provide the necessary structural framework for immune cell encounter and activation (2, 3, 12). During an immune response, stromal cells possess the plasticity to adapt to rapid SLO expansion with massive increases in lymphocyte numbers (7, 14). However, it is not known how SLOs accomplish size increase and whether this process involves stroma.

During embryogenesis, stroma interacts with lymphoid tissue inducer (LTi) cells and orchestrates SLO development by producing chemokines and adhesion molecules (1519). Stroma expresses the lymphotoxin (LT) β receptor, which is an important molecule in this process (20, 21). LTβR and RANK are members of the TNFR superfamily (TNFRSF11a) and signal through the classical and the alternative NF-κB pathways (22, 23). RANK is activated by RANKL (TNFSF11 or TRANCE), which is blocked by the soluble decoy receptor osteoprotegerin (OPG) (22, 24). RANK- and RANKL-deficient mice lack most LNs (2527), which renders the study of RANK function in LN immune homeostasis difficult. LTi cells express RANK and RANKL, and their numbers and cluster formation are reduced in RANKL knockout (KO) mice (27). In addition, RANK stimulates LTα1β2 expression by LTi cells, the ligand for LTβR (28). This led to the proposal that RANK-RANKL signaling in LTi cells results in activation of LTβR on stromal cells and subsequent SLO formation. However, the rescue of LN organogenesis in Rankl−/− mice by transgenic RANKL expression in B or T cells does not restore LTi cell numbers to normal levels, and the engagement of stromal cell LTβR by an agonistic Ab appears unable to normalize LN formation (27). Moreover, Peyer’s patches and rescued mesenteric LNs of Rankl−/− mice are smaller (25, 27) and display disorganized B cell follicles (27, 29). Mice deficient in TRAF6, an adaptor molecule of RANK signaling, lack LNs; LN formation can be rescued by exogenous IL-7 but the SLOs then lack clearly defined B cell follicles (28). These findings underscore our incomplete knowledge of the prenatal and postnatal functions of RANK in SLO formation and homeostasis.

The hair follicle (HF), one of the defining features of mammals, is a miniorgan of the skin, which periodically regenerates by repetitive cycles of growth (anagen), regression (catagen), and relative quiescence (telogen) (30). It has a specialized association with the immune system, because it enjoys a relative immune privilege (31) and shares gene products with the thymus, most commonly exemplified by the Foxn-1−/−nude mice that lack hair and thymus. RANK and RANKL are expressed in the thymus and function in the formation of medullary thymic epithelial cells (3236). RANK also plays an important role in HF biology, because RANK stimulation of HF stem cells is required for entry into anagen. Anagen HFs produce soluble RANKL (37).

In this article, we present an analysis of a mouse transgenic (Tg) for Rank under the control of the S100A8 promoter, active in HF stem cells (37) and in macrophages (38). The mouse displays a massive postnatal expansion of skin-draining LNs. The proportions and organization of hematopoietic and stromal cell populations are well maintained, with the exception of an increase in small B cell follicles. The immune cells are not activated and respond rapidly to immunization. Using bone marrow (BM) transfers, skin transplantation, and genetic rescue experiments, we showed that Tg Rank expression in HFs is responsible for LN growth. RANKL is overproduced from the Tg HFs and is responsible for LN growth, because RANKL neutralization restores normal organ size. The vascular and reticular fibroblastic cells express RANK and undergo hyperproliferation, which is abrogated by RANKL neutralization. Moreover, stromal cells show increased transcription of chemokines and cell-adhesion molecules, which would contribute to LN growth. Our data showed that RANKL functions in SLO homeostasis by stimulating an equilibrated proliferation of all stromal cell subsets concomitant with increased lymphocyte recruitment.

Animals were kept in a pathogen-free barrier facility, and experimentation was performed in accordance with institutional guidelines and its animal bioethics accord (approval no. AL/01/20/11/08). All mutants were identified by PCR (Supplemental Table I), with the exception of CD45.1 mice, which were phenotyped by flow cytometry for the CD45 isoform. Mice aged 6–8 wk were used for all experiments, unless stated otherwise. Ultrasound biomicroscopy was performed with 24- and 47.8-MHz ultrasound systems on anesthetized mice (39). For adoptive BM transfer, 3-wk-old mice were lethally irradiated with 10-Gy x-rays (Particle accelerator, Saturn 41; General Electric); 24 h later, they received 2 × 106 BM cells i.v., harvested from congenic CD45.1+ wild-type (WT) or CD45.2+ Tg mice. Chimerism was verified every month by assessing the percentage of donor blood leukocytes. For skin transplantation, skin from age-matched mice was transplanted onto the back, below the neck, of 3 wk-old nude mice. A healing spray was applied, and mice were killed 7 wk later. The Rank transgene was backcrossed onto Msx-2lacZ/lacZ CD1 mice for more than eight generations, which maintained cutaneous LN (cLN) hyperplasia, although it was less pronounced. LN cell numbers were determined at 3 mo of age. To block in vivo RANK–RANKL interaction, the neutralizing anti-RANKL Ab IK22-5 (40) was injected s.c. into Rank-Tg mice every 3 d (10 μg at postnatal day [P] 10, 25 μg from P13 to P28, and 50 μg from P31 to P45). Mock-treated mice received azide-free rat IgG (Jackson Immunochemicals).

Stromal cells were isolated and identified using CD31 and podoplanin as markers, as described (12). Briefly, LNs were collected, and capsules were opened and incubated for 30 min at 37°C in RPMI 1640 (Lonza) containing 2% FCS (v/v; Lonza), Dispase (1 mg/ml; Roche), Collagenase D (1 mg/ml; Roche), and DNase (40 μg/ml; Sigma-Aldrich). Medium containing enzymes was then removed and renewed for another 20 min. The digest was gently pipetted every 10 min to break aggregates, until no visible fragment remained; the final pipetting was done in the presence of 5 mM EDTA. Cells were then passed through a 40-μm mesh and washed twice in RPMI 1640 10% FCS (v/v). In some experiments, CD45 stromal cells were isolated by negative selection using anti-CD45–coupled magnetic microbeads (Miltenyi Biotec). Dendritic cells (DCs) were enriched by an Optiprep gradient (Sigma-Aldrich) (41) after LN digestion with Collagenase D (1 mg/ml) for 45 min under agitation.

Single-cell suspensions obtained from different organs were stained with Abs (Supplemental Table II) and analyzed on a FACSCalibur (Becton Dickinson) using CellQuest and FlowJo software. Stromal cells were sorted on a FACSVantage SE (Becton Dickinson) by gating on CD45 (allophycocyanin) single cells and using podoplanin+ (FITC)- and CD31 (PE)-specific Abs (purity was 95 ± 2% for BECs, 95 ± 3% for LECs, and 93 ± 2% for FRCs). Regulatory T cells were labeled for Foxp3 following the manufacturer’s recommendations (eBioscience). Adult LTi-like cells, identified as lin (CD3, CD8, CD11c, TCRβ, TCRγδ, CD19, Ter119, NK1.1, Ly6C/G) RORγt+α4β7+CD127+, were first incubated with an anti-CD16/CD32 Ab before being labeled for LTα1β2 using mouse LTβR fused to human IgG-Fc, followed by fluorochrome-conjugated goat anti-human IgG. To ensure specificity, a negative control sample was preincubated with anti-LTα and LTβ mAbs before addition of mouse LTβR-human IgG-Fc.

LNs were frozen in Tissue-Tek O.C.T. Compound (Electron Microscopie Science), and sections were prepared. After fixation in cold acetone and blocking with goat serum (Sigma-Aldrich), they were incubated with primary Abs and, if necessary, secondary detection Abs (Supplemental Table II). Images were acquired with a Zeiss Axiovert 200M microscope and AxioVision Rel.4.8 software. Quantification of area and number of B cell follicles was done on images using ImageJ software. β-galactosidase activity was assessed, as described (42).

RNA from sorted cells was extracted with RNeasy kits (Qiagen), and cDNA was synthesized with oligo(dT)15 primers and ImProm-II (Promega) or Superscript III (Invitrogen). RT-PCR was performed using Promega GoTaq amplification mix and Eurogentec qPCR MasterMix Plus Low ROX. The gene-specific primers are listed in Supplemental Table I. Quantitative RT-PCR (qRT-PCR) was run on a Stratagene MX4000 thermal cycler, and threshold cycle values of target genes were normalized to GAPDH, hypoxanthineguanine phosphoribosyltransferase, and β-actin. The expression factor was calculated using the Relative Expression Software Tool (http://mmm.gene-quantification.de/rest.html).

Mice were injected i.p. with 0.5 mg BrdU (Sigma-Aldrich) diluted in PBS and given BrdU in drinking water (0.8 mg/ml). CD45 and CD45+ cells were isolated from LNs, and the incorporation of BrdU was measured by flow cytometry using the BrdU flow kit (BD Pharmingen). Stromal subsets were identified using CD31 and podoplanin as markers.

Twelve-week-old mice received s.c. injections of 200 μg chicken OVA with 5 μg LPS (Sigma-Aldrich), IFA, or 25 μg CpG (InvivoGen). A boost was administrated 2 wk later. Before and every 5 d thereafter, chicken OVA-reactive serum IgG was measured by direct ELISA. For Ab-affinity measures, mice received s.c. injections of 50 μg 4-hydroxy-3-nitrophenyl-acetyl (NP)30-keyhole limpet hemocyanin (KLH) (Biosearch Technologies) in 5 μg LPS (Sigma-Aldrich). A boost was administrated at 10 d. Serum anti-NP IgG was determined every 5 d by direct ELISA on plates coated with NP3-BSA or NP30-BSA (Biosearch Technologies). Titers were determined as highest serum dilution that gave a value of at least three SD above the average reading for secondary controls. The relative affinity was determined as the titer NP3/titer NP30 (43).

An unpaired two-tailed Student t test was used to determine statistically significant differences. The p values < 0.05 were considered statistically significant. GraphPad Prism version 5 for Windows (GraphPad software) was used for statistical data analysis.

Mice overexpressing murine Rank under control of the S100A8 promoter showed a massive hyperplasia of skin-draining LNs (inguinal, axial, brachial, and superficial parotid), whereas internal LNs (mesenteric or para-aortic), spleen, and thymus showed no such growth (Fig. 1A, 1B). Cell counting revealed an 8–10-fold increase in hematopoietic cell numbers of cLNs, whereas the counts for mesenteric LNs (mLNs), spleen, thymus, blood, and BM were unchanged or even slightly reduced (Fig. 1B, Table I). To determine whether the SLO hyperplasia was acquired postnatally, we assessed cLN cell numbers 1 wk after birth and at different time points thereafter. Growth started after week 2 and reached a plateau at week 13 (Fig. 1C). To better determine its onset, we used in vivo ultrasound imaging, which measures LN size in young mice more accurately than does cell counting or organ weighing (39). It fixed the start of inguinal LN (iLN) growth at P24 (Fig. 1D). Thus, Rank-Tg mice displayed a prominent postnatal growth of their cLNs.

FIGURE 1.

Postnatal cLN hyperplasia in Rank-Tg mice. A, Representative photographs of cLNs (left to right: inguinal, axial, brachial, and superficial parotid LN) and an mLN from 12-wk-old WT and Rank-Tg mice. Two-fold magnification. B, Fold difference in cell numbers of cLNs, mLNs, spleen, thymus, blood, and BM in 12-wk-old Rank-Tg versus WT mice. C, cLN cell numbers from week 1 to 16 in Rank-Tg and WT mice. D, iLN surface area, measured by ultrasound, in Rank-Tg and WT mice from P4–P58. Insets show iLN ultrasound images of 12-wk-old WT and Tg mice. Data in B are from 11 mice, collected in three experiments; in C and D, data are mean ± SD from six mice/genotype for each time point. **p < 0.01, ***p < 0.001.

FIGURE 1.

Postnatal cLN hyperplasia in Rank-Tg mice. A, Representative photographs of cLNs (left to right: inguinal, axial, brachial, and superficial parotid LN) and an mLN from 12-wk-old WT and Rank-Tg mice. Two-fold magnification. B, Fold difference in cell numbers of cLNs, mLNs, spleen, thymus, blood, and BM in 12-wk-old Rank-Tg versus WT mice. C, cLN cell numbers from week 1 to 16 in Rank-Tg and WT mice. D, iLN surface area, measured by ultrasound, in Rank-Tg and WT mice from P4–P58. Insets show iLN ultrasound images of 12-wk-old WT and Tg mice. Data in B are from 11 mice, collected in three experiments; in C and D, data are mean ± SD from six mice/genotype for each time point. **p < 0.01, ***p < 0.001.

Close modal
Table I.
Absolute cell numbers in organs (106)
AnimalcLNsmLNsSpleenThymusBloodBM
Rank-Tg (n = 11) 196 ± 39 6.3 ± 3.1 76 ± 14 96 ± 44 5.6 ± 4.8 5.6 ± 4.8 
WT (n = 8) 22 ± 3 7.2 ± 1.6 105 ± 20 125 ± 44 6.2 ± 4.0 6.2 ± 4.0 
 p < 0.001 NS p <0.01 NS NS NS 
AnimalcLNsmLNsSpleenThymusBloodBM
Rank-Tg (n = 11) 196 ± 39 6.3 ± 3.1 76 ± 14 96 ± 44 5.6 ± 4.8 5.6 ± 4.8 
WT (n = 8) 22 ± 3 7.2 ± 1.6 105 ± 20 125 ± 44 6.2 ± 4.0 6.2 ± 4.0 
 p < 0.001 NS p <0.01 NS NS NS 

Cell numbers (mean ± SD) of organ and blood cell suspensions from 12-wk-old Tg and WT animals were counted in a hemacytometer.

To investigate whether cLN hyperplasia was accompanied by abnormal overrepresentation of a hematopoietic cell lineage, we determined the proportion of the CD11b+ granulo-myeloid lineage, B220+ B cells, and CD3+ T cells by flow cytometry in cLNs, mLNs, and spleen of Tg and control animals. There was a significant increase in CD11b+ cells in both LN types and the spleen in mutant mice (Fig. 2A). An overrepresentation of these cells had also been noted in the BM (44); therefore, it is most likely the result of transgene expression by this lineage (44, 45) (data not shown). More restrictively, there was an increase in the proportion of B cells and a reduction in T cells in cLNs but not in mLNs or spleen. B cells do not express the transgene (45) (data not shown), suggesting that the change in B/T cell proportions was a reflection of the abnormal LN size. There was little alteration in B cell hematopoiesis, because the number of splenic transitional type 1 and type 2 B cells remained relatively stable in comparison with a strong reduction in mature B cells (Table II), suggesting a reinforced recruitment of mature B cells to the cLNs. Therefore, we visually inspected B cell organization in the cLNs by labeling sections for B220 and CD3. We noted normal B cell organization into follicles of the cortical area but the presence of small B cell follicles in the paracortex (Fig. 2B). All follicles comprised an FDC network expressing MAdCAM-1, FDC-M2, or CD35 (Fig. 2B). MAdCAM-1+ MRCs lined the subcapsular sinus of the outer margin of cortical follicles, with additional MRCs almost fully encircling the small paracortical follicles of the hyperplasic LNs (Fig. 2B). To assess whether cLN hyperplasia led to changes in the proportions of FRCs, BECs, and LECs, we compared their numbers and distribution in cLNs of Tg versus WT animals. The cells were identified on the basis of podoplanin and CD31 expression after gating on CD45 cells (12). FRCs and BECs were present in normal proportions, but the percentage of LECs showed a slight decrease (Fig. 2C). Absolute numbers demonstrated a robust increase in the abundance of all stromal subsets in the hyperplasic LNs. The subsets were normally distributed, with FRCs dispersed around cortical BECs, and LECs juxtaposed to B cell follicles (Fig. 2D). Peripheral LN addressin-expressing high endothelial venules were abundant and normally located in interfollicular regions (Fig. 2D). Thus, the cLN cell composition and its architecture showed remarkably little alteration for such a significant growth, with the exception of an increase in the number and location of B cell follicles and their associated stromal compartment.

FIGURE 2.

The enlarged cLNs display a well-conserved cell composition and organ architecture. A, Cell lineage proportions in cLNs, mLNs, and spleen (S) of CD11b+ granulo-myeloid cells, B220+ B cells, and CD3+ T cells in Rank-Tg versus WT mice were identified by flow cytometry (n = 11/genotype). B, Representative images of T cell zone and B cell follicle organization (CD3 and B220 staining), MRCs (MAdCAM-1), and FDCs (MAdCAM-1, FDC-M2, CD35) in brachial LNs of Rank-Tg and WT mice. Scale bars, 100 μm; insets, 50 μm. C, Identification of the CD45 FRC, LEC, and BEC stromal cell subsets by flow cytometry, using podoplanin and CD31 cell surface markers. The percentage ± SD is indicated next to the cell type. Graph (lower panel) depicts proportional changes. Table (lower panel) shows their absolute numbers in pooled inguinal, axial, and brachial LNs of Rank-Tg versus WT mice. Data are from 11–14 mice, accumulated in four experiments. D, Representative images showing BEC/FRC (CD31/podoplanin), LEC/B cell (Lyve-1/B220), and LEC/BEC (CD31/peripheral LN addressin) distribution on a brachial LN section. Scale bars, 100 μm; insets, 50 μm. *p < 0.05, **p < 0.01.

FIGURE 2.

The enlarged cLNs display a well-conserved cell composition and organ architecture. A, Cell lineage proportions in cLNs, mLNs, and spleen (S) of CD11b+ granulo-myeloid cells, B220+ B cells, and CD3+ T cells in Rank-Tg versus WT mice were identified by flow cytometry (n = 11/genotype). B, Representative images of T cell zone and B cell follicle organization (CD3 and B220 staining), MRCs (MAdCAM-1), and FDCs (MAdCAM-1, FDC-M2, CD35) in brachial LNs of Rank-Tg and WT mice. Scale bars, 100 μm; insets, 50 μm. C, Identification of the CD45 FRC, LEC, and BEC stromal cell subsets by flow cytometry, using podoplanin and CD31 cell surface markers. The percentage ± SD is indicated next to the cell type. Graph (lower panel) depicts proportional changes. Table (lower panel) shows their absolute numbers in pooled inguinal, axial, and brachial LNs of Rank-Tg versus WT mice. Data are from 11–14 mice, accumulated in four experiments. D, Representative images showing BEC/FRC (CD31/podoplanin), LEC/B cell (Lyve-1/B220), and LEC/BEC (CD31/peripheral LN addressin) distribution on a brachial LN section. Scale bars, 100 μm; insets, 50 μm. *p < 0.05, **p < 0.01.

Close modal
Table II.
Absolute cell numbers of hematopoietic cell types (106)
AnimalOrganCD3+ T CellsCD11b+ MacrophagesB220+ B CellsMature B IgDModIgMModT2 B IgDHiIgMHiT1/MZ B IgDLoIgMHi
Rank-Tg (n = 11) cLNsa 53.9 ± 17.7 2.7 ± 1.2 65.4 ± 30    
WT (n = 8)  8.5 ± 3.1 0.2 ± 0.06 5.6 ± 2.3    
  p < 0.001 p < 0.001 p < 0.001    
Rank-Tg (n = 11) mLNs 3.3 ± 1.4 0.08 ± 0.07 2.3 ± 1.3    
WT (n = 8)  4.3 ± 1.1 0.07 ± 0.06 2.4 ± 0.5    
  NS NS NS    
Rank-Tg (n = 11) Spleen 16.4 ± 4.7 4.4 ± 2 36.1 ± 9.5 28 ± 12 2.6 ± 1.6 8.8 ± 4.5 
WT (n = 8)  25.4 ± 6.1 3.3 ± 0.6 57 ± 13 39 ± 9 3.6 ± 0.6 7.4 ± 3.4 
  p < 0.001 NS p < 0.002 p < 0.03 p < 0.05 NS 
Rank-Tg (n = 11) Blood (ml) 2.2 ± 1.3 1.4 ± 1.0 2.9 ± 2.2    
WT (n = 8)  1.7 ± 0.7 1.0 ± 0.4 2.7 ± 1.5    
  NS NS NS    
AnimalOrganCD3+ T CellsCD11b+ MacrophagesB220+ B CellsMature B IgDModIgMModT2 B IgDHiIgMHiT1/MZ B IgDLoIgMHi
Rank-Tg (n = 11) cLNsa 53.9 ± 17.7 2.7 ± 1.2 65.4 ± 30    
WT (n = 8)  8.5 ± 3.1 0.2 ± 0.06 5.6 ± 2.3    
  p < 0.001 p < 0.001 p < 0.001    
Rank-Tg (n = 11) mLNs 3.3 ± 1.4 0.08 ± 0.07 2.3 ± 1.3    
WT (n = 8)  4.3 ± 1.1 0.07 ± 0.06 2.4 ± 0.5    
  NS NS NS    
Rank-Tg (n = 11) Spleen 16.4 ± 4.7 4.4 ± 2 36.1 ± 9.5 28 ± 12 2.6 ± 1.6 8.8 ± 4.5 
WT (n = 8)  25.4 ± 6.1 3.3 ± 0.6 57 ± 13 39 ± 9 3.6 ± 0.6 7.4 ± 3.4 
  p < 0.001 NS p < 0.002 p < 0.03 p < 0.05 NS 
Rank-Tg (n = 11) Blood (ml) 2.2 ± 1.3 1.4 ± 1.0 2.9 ± 2.2    
WT (n = 8)  1.7 ± 0.7 1.0 ± 0.4 2.7 ± 1.5    
  NS NS NS    

Cell numbers (mean ± SD) of organs and blood cell suspensions were counted in a hemacytometer and analyzed for T cells, macrophages, B cells, and B cell subsets by flow cytometry. The total number was determined using the percentage composition of the different cell types.

a

cLNs included inguinal, axial, brachial, and superficial parotid; for B cells and macrophages, cell numbers were determined excluding brachial LNs.

T1, transitional type 1, T2, transitional type 2, MZ, marginal zone.

To address the question of whether the hematopoietic cells of the cLNs were functional, we studied the cutaneous immune response. Skin- and blood-derived DCs, purified from brachial and axial LNs and identified on the basis of I-A/CD11c expression (41, 46), showed low CD86 levels but normally upregulated this activation marker in response to s.c. CpG injection (Fig. 3A). T cells of cLNs, mLNs, or spleen were resting but responded to CpG by CD62L downregulation and CD44 upregulation (Fig. 3B). Also, B cells showed no sign of activation in nonimmunized Tg mice (Fig. 3C) and were activated in response to s.c. immunization by OVA with LPS, CpG, or IFA (Fig. 3D). Of note, Ab-affinity maturation to NP-KLH in LPS proceeded faster in Tg versus control mice (Fig. 3E). To test for a possible rupture of self-tolerance, we assessed the presence of serum autoantibodies against actin, rheumatoid factor, and DNA in both nonimmunized and immunized Tg mice. However, no such Abs were detected (data not shown). In addition, there was no significant difference in B cell numbers in the autoimmune-sensitive salivary glands (WT: 30 ± 12 × 103, Tg: 35 ± 9 × 103; n = 8). In accord with these results, the proportion of Foxp3+ regulatory T cells was not markedly altered in cLNs, mLNs, or spleen (data not shown). Stromal cells express Aire, which is under RANK control in the thymus (3235), and regulate peripheral tolerance by expression of tissue-restricted peripheral tissue Ags (PTAs) (5, 4749). Therefore, we evaluated Aire and PTA expression in Tg mouse cLNs by qRT-PCR. No upregulation of Aire expression in CD45 stromal cells was seen (relative expression level, WT: 1, Tg: 0.642 ± 0.15; p = 0.092), and the stromal subset-restricted expression of Mlana and tyrosinase PTAs (48, 49) was also conserved (data not shown). Therefore, postnatal LN growth has no discernable negative influence on its immune cells, and peripheral tolerance is maintained.

FIGURE 3.

cLN immune cells are resting but respond rapidly to immunization. A, CD86 expression levels of skin- and blood-derived DCs, purified from brachial and axial LNs in nonimmunized WT and Rank-Tg mice, as well as from Rank-Tg mice 4 d after a s.c. immunization boost with CpG. Graph (left panel) shows the mean ± SD of three animals/genotype performed in duplicates. The flow cytometry dot blot (right panel) depicts the gate setting for I-AHi CD11c+ skin-derived DCs and for I-A+ CD11cHi blood-derived DCs. B, Percentage of CD44- and CD62L-expressing Th cells in cLNs, mLNs, or spleen in WT, Rank-Tg, and Rank-Tg mice 4 d after a s.c. immunization boost with CpG (left panel). Data are mean ± SD of six to eight animals/genotype. The flow cytometry dot plots (right panel) depict a representative analysis of CD62L and CD44 expression of nonimmunized and CpG-immunized WT and Tg Th cells from brachial LNs. C, CD86 and CD69 expression levels of B220+ B cells in cLNs, mLNs, and spleen from WT and Rank-Tg mice. Data are mean ± SEM of five animals/group. D, Serum anti-OVA IgG levels in WT and Rank-Tg mice, measured after a s.c. immunization boost of OVA with LPS, CpG, or IFA. Each symbol represents one animal, and the horizontal lines show the mean values. E, High-affinity anti-NP IgG levels after s.c. immunization of NP30 conjugated to KLH with LPS. Serum anti-NP IgG was measured by ELISA for NP3-BSA or NP30-BSA, and the relative affinity was determined as the titer NP3/titer NP30. Data are mean ± SEM of seven mice/group. **p < 0.01, ***p < 0.001.

FIGURE 3.

cLN immune cells are resting but respond rapidly to immunization. A, CD86 expression levels of skin- and blood-derived DCs, purified from brachial and axial LNs in nonimmunized WT and Rank-Tg mice, as well as from Rank-Tg mice 4 d after a s.c. immunization boost with CpG. Graph (left panel) shows the mean ± SD of three animals/genotype performed in duplicates. The flow cytometry dot blot (right panel) depicts the gate setting for I-AHi CD11c+ skin-derived DCs and for I-A+ CD11cHi blood-derived DCs. B, Percentage of CD44- and CD62L-expressing Th cells in cLNs, mLNs, or spleen in WT, Rank-Tg, and Rank-Tg mice 4 d after a s.c. immunization boost with CpG (left panel). Data are mean ± SD of six to eight animals/genotype. The flow cytometry dot plots (right panel) depict a representative analysis of CD62L and CD44 expression of nonimmunized and CpG-immunized WT and Tg Th cells from brachial LNs. C, CD86 and CD69 expression levels of B220+ B cells in cLNs, mLNs, and spleen from WT and Rank-Tg mice. Data are mean ± SEM of five animals/group. D, Serum anti-OVA IgG levels in WT and Rank-Tg mice, measured after a s.c. immunization boost of OVA with LPS, CpG, or IFA. Each symbol represents one animal, and the horizontal lines show the mean values. E, High-affinity anti-NP IgG levels after s.c. immunization of NP30 conjugated to KLH with LPS. Serum anti-NP IgG was measured by ELISA for NP3-BSA or NP30-BSA, and the relative affinity was determined as the titer NP3/titer NP30. Data are mean ± SEM of seven mice/group. **p < 0.01, ***p < 0.001.

Close modal

To elucidate the mechanisms for organized cLN expansion, we investigated the contribution of the Rank transgene in the CD11b+ granulo-myeloid cell lineage by performing adoptive BM transfers. Lethally irradiated 3-wk-old Tg mice were reconstituted with BM of congenic CD45.1+ C57BL/6 mice; the reciprocal transfer was also done. cLN cell numbers were determined 3 mo later. BM transfers were complete, because all hematopoietic cells, including the CD11b+ lineage, were of donor origin (Supplemental Fig. 1A, 1B). WT → Tg chimeras developed LN hyperplasia to the same extent as did control Tg → Tg animals (Fig. 4A). Reciprocally, Tg BM could not transfer LN growth to WT mice. These findings demonstrated that Rank transgene expression by the granulo-myeloid cell lineage is not responsible for cLN growth. We determined whether the radioresistant stromal cells expressed the Rank transgene by performing RT-PCR on cLNs recovered from WT → Tg chimeras. The transgene was no longer expressed, demonstrating that the cLN stroma is not transgenic for Rank (Supplemental Fig. 1C). We then considered the possibility that cLN growth is conveyed by Rank transgene expression in HFs (37). To address this issue, we first transplanted Tg or WT skin from 3-wk-old donors onto the backs of age-matched nude mice. Seven weeks later, cell numbers of brachial and axial LNs versus iLNs were determined. Because iLNs are not expected to drain from the graft, their cell counts served as internal control for natural LN size and experimental variations. We found a significant increase for Tg skin recipients compared with WT skin, indicating that LN growth is associated with Rank transgene expression in HFs (Fig. 4B). We verified this conclusion by reducing transgene expression in HFs. This was done by crossing the transgene onto Msx-2lacZ/lacZ mice, which have a shortened anagen phase (50) and, therefore, are expected to express lower levels of the Rank transgene (Supplemental Fig. 1D). This mutant gene combination resulted in strongly reduced cLN growth (Fig. 4C). To exclude the possibility that Msx-2 affected the organ size through local expression, we assessed β-galactosidase activity in brachial LNs of Rank Tg;Msx-2lacZ/+ animals. lacZ gene activity was not detected in the LNs, but it was found at the base of the HFs (Fig. 4D) (50). Taken together, these findings supported the conclusion that Rank transgene expression in HFs is necessary and sufficient for skin-draining LN hyperplasia.

FIGURE 4.

Tg RANK overexpression in HFs induces cLN hyperplasia. A, Analysis of cell numbers in cLNs (brachial, inguinal, and axial) after adoptive BM transfer. Data are from three mice/group compiled from seven different experiments. B, Relative cLN cell number increase in brachial and axial LNs, normalized to iLNs in nude mice after skin grafting. Data are from three mice/group compiled from three different experiments. C, cLN cell numbers in WT, Msx-2LacZ/LacZ, Rank-Tg, and Rank-Tg;Msx-2LacZ/LacZ mice. Data are from eight mice/group from four experiments. D, Representative β-galactosidase staining in a brachial LN (left panel) and skin (right panel) of Rank-Tg;Msx-2LacZ/+ mice. Arrowheads point to β-galactosidase enzymatic reactivity of HF matrix cells. Dashed line indicates the dermo–epidermal junction. Scale bar, 100 μm. Data are mean ± SD. *p < 0.05, ***p < 0.001.

FIGURE 4.

Tg RANK overexpression in HFs induces cLN hyperplasia. A, Analysis of cell numbers in cLNs (brachial, inguinal, and axial) after adoptive BM transfer. Data are from three mice/group compiled from seven different experiments. B, Relative cLN cell number increase in brachial and axial LNs, normalized to iLNs in nude mice after skin grafting. Data are from three mice/group compiled from three different experiments. C, cLN cell numbers in WT, Msx-2LacZ/LacZ, Rank-Tg, and Rank-Tg;Msx-2LacZ/LacZ mice. Data are from eight mice/group from four experiments. D, Representative β-galactosidase staining in a brachial LN (left panel) and skin (right panel) of Rank-Tg;Msx-2LacZ/+ mice. Arrowheads point to β-galactosidase enzymatic reactivity of HF matrix cells. Dashed line indicates the dermo–epidermal junction. Scale bar, 100 μm. Data are mean ± SD. *p < 0.05, ***p < 0.001.

Close modal

To identify the signal that is conveyed by the anagen HF and that is responsible for cLN growth, we considered RANKL as a probable candidate, because the anagen HF is an important source of soluble RANKL (37); soluble RANKL is a small (>10 kDa) globular protein, which can be expected to have direct access to LN stroma through its conduit system (3, 51, 52); and RANKL is known to play a critical role in SLO formation (2527). We measured soluble RANKL and OPG release from Tg and WT skin at different time points during HF morphogenesis and cycling. We confirmed that RANKL was highest at anagen, when OPG levels were low (37). Moreover, higher levels of RANKL were released from Tg skin at P14 and during anagen (Supplemental Fig. 2A, 2B). We then directly tested the role of RANKL by s.c. administration of a RANKL-blocking mAb. Tg mice received the mAb from P10 to wk 6, whereas littermates were administered control mAb. cLNs and spleen were weighed and cells counted, and B/T cell proportion was assessed. The RANKL-neutralizing mAb restored cLN weight and cell numbers (Fig. 5A). The proportion of B cells was reversed (Fig. 5B). The number of B cell follicles was recovered, and their size was restored (Fig. 5C). We then determined the presence of MRCs by labeling sections for MAdCAM-1 and discovered a distinct reduction (thin arrows, mouse #1) or even a complete absence of these B cell-associated stromal cells (mouse #2). In contrast, FDCs remained intact, as shown by labeling for MAdCAM-1 or FDC-M2 within the follicles (Fig. 5D, large arrows). This underscores the importance of MRCs in LN homeostasis and suggests a role for RANKL in MRC maintenance. RANKL neutralization did not lead to a reduction in splenocyte numbers (Supplemental Fig. 2C) nor were B or T cells activated, as assessed by plasmocyte and memory T cell formation (Supplemental Fig. 2D, 2E). The anti-RANKL treatment provoked a diminution in body weight (WT: 17.7 ± 0.5 g, Tg: 18.7 ± 1.1 g, treated Tg: 12.1 ± 1.6 g; n = 12), most probably attributable to perturbation of bone homeostasis (25, 26). These data showed that RANKL is responsible for LN growth and formation of small B cell follicles in the Rank-Tg mouse.

FIGURE 5.

RANKL induces cLN growth. A, Weight (left panel) and cell numbers (right panel) of cLNs, normalized to body weight, of WT, control Ig-administered Rank-Tg mice, and Rank-Tg mice, treated with anti-RANKL blocking mAb from P10 to 6 wk of age. B, Percentage of T and B cells in cLNs of WT, control Ig-administered Rank-Tg mice, and Rank-Tg mice, treated with anti-RANKL blocking mAb. C, Left panel, Representative visualization of B cell follicles in WT mice, control Ig-administered Rank-Tg mice, and Rank-Tg mice, treated with anti-RANKL blocking mAb by immunofluorescence on axial LN sections. Scale bars, 200 μm. B cell follicles visualized on B220-labeled axial LN cross-sections were counted (middle panel) and their area was quantified (right panel). Data are mean ± SEM of nine mice/group acquired in three separate experiments. D, MAdCAM-1+ MRCs in LNs of mock-treated and anti-RANKL-treated Rank-Tg mice (mouse #1 and #2). Arrows indicate MRCs at the outer margin of the B220+ B cell follicles. FDC-M2+ FDC networks on serial sections are indicated by large arrows. Dashed lines mark the limit of the LNs. Scale bars, 100 μm. **p < 0.01, ***p < 0.001.

FIGURE 5.

RANKL induces cLN growth. A, Weight (left panel) and cell numbers (right panel) of cLNs, normalized to body weight, of WT, control Ig-administered Rank-Tg mice, and Rank-Tg mice, treated with anti-RANKL blocking mAb from P10 to 6 wk of age. B, Percentage of T and B cells in cLNs of WT, control Ig-administered Rank-Tg mice, and Rank-Tg mice, treated with anti-RANKL blocking mAb. C, Left panel, Representative visualization of B cell follicles in WT mice, control Ig-administered Rank-Tg mice, and Rank-Tg mice, treated with anti-RANKL blocking mAb by immunofluorescence on axial LN sections. Scale bars, 200 μm. B cell follicles visualized on B220-labeled axial LN cross-sections were counted (middle panel) and their area was quantified (right panel). Data are mean ± SEM of nine mice/group acquired in three separate experiments. D, MAdCAM-1+ MRCs in LNs of mock-treated and anti-RANKL-treated Rank-Tg mice (mouse #1 and #2). Arrows indicate MRCs at the outer margin of the B220+ B cell follicles. FDC-M2+ FDC networks on serial sections are indicated by large arrows. Dashed lines mark the limit of the LNs. Scale bars, 100 μm. **p < 0.01, ***p < 0.001.

Close modal

To further address the mechanism of LN growth, we first considered adult LTi-like cells as a possible RANKL target. They have been shown to express RANK, and their role in stroma activation could depend on RANK–RANKL interactions (27, 53). Moreover, their implication in SLO expansion was observed in the case of enhanced survival by ectopic IL-7 expression (54). We compared adult LTi-like linRorγt+CD127+α4β7+ cells of WT and Rank-Tg mice in terms of phenotype, numbers, and frequency. Moreover, because RANK stimulates LTα1β2 expression by LTi cells (28), we measured expression of LTα1β2. The experiments were performed at P24 and P42, shortly after skin RANKL peaks and initiation of LN growth. No significant difference in cell numbers, phenotype, or LTα1β2 expression was noted (Fig. 6A, Supplemental Fig. 2F). We also measured LTα1β2 expression by B cells but did not find a significant difference (Fig. 6A). We then studied FRC and endothelial cells, in view of their role in coordinating lymphoid organ formation (2, 46). First, we determined Rank expression by RT-PCR on FACS-sorted FRCs, LECs, and BECs. Rank was transcriptionally active in all stromal subsets (Fig. 6B), suggesting that skin-derived RANKL could directly stimulate cLN stroma. Because chemokines and cell-adhesion molecules are expressed by stroma and play an important role in SLO growth (9, 55), we tested whether these type of molecules were upregulated. We performed qRT-PCR on sorted stromal subsets harvested from cLNs of 6-wk-old WT and Tg animals. Among the different genes tested (Supplemental Table I), we found in Tg mice that FRCs expressed more CXCL13 transcripts; LECs expressed more CCL19, MAdCAM-1, and VCAM-1 transcripts; and BECs expressed more CCL19 transcripts (Fig. 6C). This suggested that cLN growth is promoted by RANK-stimulated expression of chemokines and adhesion molecules. In light of the almost unchanged stromal cell proportion of the enlarged cLNs, together with the finding that RANKL induced medullary thymic epithelial cell proliferation (35), we next asked whether stromal cells proliferated in response to RANKL. To this end, we measured cell division by BrdU incorporation in 6-wk-old WT mice, as well as mock- and anti-RANKL–treated Tg mice. Mice were given BrdU for 6 d, and DNA incorporation of the nucleotide analog was measured by flow cytometry. The WT CD45 stroma had a reasonable proliferation rate, but it was much faster in Tg animals. Strikingly, when RANKL was neutralized, the proportion of BrdU+ stroma decreased to WT levels (Fig. 6D). In contrast, the BrdU incorporation rate for hematopoietic cells was similar between WT and Tg animals, and there was no change after administration of the blocking mAb. We next analyzed FRCs, LECs, and BECs separately. All subsets proliferated faster in Tg mice, and their greater BrdU incorporation was abrogated upon RANKL neutralization (Fig. 6E). Pulse-chase experiments were performed to address stromal cell viability. Loss of the BrdU label was not delayed in Tg mice, indicating that RANKL promoted stromal cell division without altering cell viability (Supplemental Fig. 2G). Thus, RANKL stimulates cLN growth by inducing the proliferation of its vascular and reticular fibroblastic stromal cells.

FIGURE 6.

RANKL activates stromal cell proliferation and chemokine/adhesion molecule expression. A, Left panel, Adult LTi-like cells (linRORγt+α4β7+CD127+) were identified in LNs of WT and Rank-Tg mice at P24 and P42, and absolute numbers were determined. Middle panel, LTα1β2 mean fluorescence intensity of adult LTi-like cells. Right panel, Relative expression levels of LTα1β2 mRNA in FACS-sorted adult LTi-like cells and CD19+ B cells at P42 (n = 6 animals). B, RT-PCR analysis of Rank and GAPDH mRNA expression by stromal subsets from cLNs of WT and Rank-Tg mice, sorted by flow cytometry on the basis of podoplanin/CD31 expression. Skin cDNA served as positive control, and water served as negative control. Image is representative of four separate experiments with 15 pooled WT mice and 5 pooled Tg mice. C, mRNA expression of chemokines and adhesion molecules in flow cytometry-sorted stromal subsets from WT and Rank-Tg mice. Measurements were done by qRT-PCR and are depicted as the fold increase with respect to WT. Data are the mean ± SEM of four independent experiments performed in duplicate with 15 pooled WT mice and 5 pooled Rank-Tg mice. D, BrdU incorporation measured by flow cytometry of CD45 cLN stromal and CD45+ hematopoietic cells from WT, control Ig-administered Rank-Tg mice, and Rank-Tg mice treated with anti-RANKL blocking mAb. E, BrdU incorporation by cLN stromal subsets from WT, control Ig-administered Rank-Tg mice, and Rank-Tg mice treated with anti-RANKL blocking mAb. Data in D and E are mean ± SEM of six to nine mice per group, obtained by flow cytometry using podoplanin/CD31 markers to identify the subsets. **p < 0.01, ***p < 0.001.

FIGURE 6.

RANKL activates stromal cell proliferation and chemokine/adhesion molecule expression. A, Left panel, Adult LTi-like cells (linRORγt+α4β7+CD127+) were identified in LNs of WT and Rank-Tg mice at P24 and P42, and absolute numbers were determined. Middle panel, LTα1β2 mean fluorescence intensity of adult LTi-like cells. Right panel, Relative expression levels of LTα1β2 mRNA in FACS-sorted adult LTi-like cells and CD19+ B cells at P42 (n = 6 animals). B, RT-PCR analysis of Rank and GAPDH mRNA expression by stromal subsets from cLNs of WT and Rank-Tg mice, sorted by flow cytometry on the basis of podoplanin/CD31 expression. Skin cDNA served as positive control, and water served as negative control. Image is representative of four separate experiments with 15 pooled WT mice and 5 pooled Tg mice. C, mRNA expression of chemokines and adhesion molecules in flow cytometry-sorted stromal subsets from WT and Rank-Tg mice. Measurements were done by qRT-PCR and are depicted as the fold increase with respect to WT. Data are the mean ± SEM of four independent experiments performed in duplicate with 15 pooled WT mice and 5 pooled Rank-Tg mice. D, BrdU incorporation measured by flow cytometry of CD45 cLN stromal and CD45+ hematopoietic cells from WT, control Ig-administered Rank-Tg mice, and Rank-Tg mice treated with anti-RANKL blocking mAb. E, BrdU incorporation by cLN stromal subsets from WT, control Ig-administered Rank-Tg mice, and Rank-Tg mice treated with anti-RANKL blocking mAb. Data in D and E are mean ± SEM of six to nine mice per group, obtained by flow cytometry using podoplanin/CD31 markers to identify the subsets. **p < 0.01, ***p < 0.001.

Close modal

In this study, we showed that RANK stimulation resulted in the proliferation of LN stromal subsets and their expression of chemokines and adhesion molecules, culminating in LN growth. LN organization was preserved, with a notable increase in B cell follicles. RANK stimulation did not provoke an immune activation; no autoreactivity could be detected, and immune cells responded rapidly and strongly to immunization by foreign Ags.

Although recognized as an important regulator of immunity, the lack of almost all SLOs in Rank or Rankl KO mice has hampered elucidation of RANK function in the development and homeostasis of the immune system. Therefore, we studied a Rank-Tg mouse, which displayed greatly oversized skin-draining LNs, whereas other SLOs, spleen, thymus, or BM were normal. LN growth occurred postnatally and, at its plateau, the lymphocyte number surpassed that of the spleen. We could not detect other major abnormalities of the immune system, aside from a general overrepresentation of Rank-Tg CD11b+ cells, presumably attributable to the cell survival action of RANK (5658). A condition that resembles this phenotype is found in a IL-7–Tg mouse (54). In this mutant, SLO growth and ectopic SLO formation were linked to increased LTi cell numbers. Dissimilar to the IL-7–Tg mouse model, the SLO expansion described in this article appeared much more prominent, and no ectopic SLOs could be found. Despite massive organ growth, hematopoietic and stromal cell compositions, as well as the overall architecture, were remarkably well conserved. The only notable alterations were increases in the number of small B cell follicles and the proportion of B cells. Support for a role for RANK signaling in regulating peripheral immune cell numbers stems from a study of Opg−/− mice. In these animals, LN B and T cell numbers were likewise augmented, and the B cell compartment was boosted, particularly in the spleen (59). This phenotype was attributed to an accelerated B cell development. We did not find a prominent increase in splenic transitional type 1 or type 2 B cells, discounting the possibility that the additional cLN B cells are the result of accelerated B cell hematopoiesis. Instead, the higher B cell proportion is likely due to reinforced recruitment or maintenance of mature B cells. Formation of the paracortical follicles was accompanied by colonization of MRCs in their outer margin, evoking the proposal that MRCs are implicated in de novo B cell follicle formation. In support of this is the loss of MRCs, as well as B cells, after RANKL neutralization.

The greatly expanded SLO size could have been the consequence of a dysfunctional hematopoietic cell compartment. However, T cells, B cells, and DCs showed no signs of activation, and no autoantigen reactivity could be detected. Moreover, T cells did not spontaneously proliferate in culture, as described in a different model of SLO hyperplasia (60), and transfer of cLN CD4+ T cells into Rag-2 KO mice did not result in abnormal expansion (data not shown). The cLN immune cells of the Rank-Tg animals were not anergic, because they responded to immunization. The immune response elicited from the Tg mice was even moderately superior to that of WT mice, and a more rapid Ab-affinity maturation was seen. The increased efficiency of the humoral response may reflect increased numbers of B and T cells, as well as more B cell follicles comprising MRCs and FDCs. Despite the relationship between RANKL and thymic epithelial cells (3235), and although LN Aire appears to contribute to peripheral T cell tolerance (49, 61), we could not detect such an association between RANKL and Aire in cLN stroma of the Tg animals.

By performing BM transfer, skin transplantation, and genetic rescue experiments, we discovered that the SLO growth signal originated in the HF. This finding provides an explanation for why only skin-draining LNs were affected and is in accord with initiation of LN growth at the start of the HF anagen phase after P24. The growth plateau is likely due to loss of transgene expression in the HF as the mouse develops alopecia (37). It is interesting to note that the postnatal period during which skin-draining LN expansion surpasses body growth occurs during HF morphogenesis and the first anagen phase (39). This is suggestive of a naturally operating link between HFs and cLNs, an association that appears to be amplified in the Tg mouse. We considered the possibility that DCs mediate the HF–LN link; however, in analogy to dermatopathic lymphadenopathy (62), more numerous, long-lived, or mature DCs would principally lead to an expanded T cell zone. This was not seen, and the proportion of DCs and their maturation status were unchanged. Because shortening of anagen by loss of Msx-2 abrogated LN growth, we deduced that the signal arises during the anagen phase of the hair cycle. High levels of soluble RANKL released from skin prompted us to propose the concept that HF RANKL conveys cLN growth. In fact, the level of HF RANKL is higher in the mutant animal, and the skin undergoes additional anagen phases, leading to a continuous RANKL release. The underlying reason for the increase in HF RANKL is unclear. One possibility is the existence of a self-amplifying RANK–RANKL loop, similar to what was described for thymic epithelial cells (36). Another argument in favor of RANKL as mediator of SLO growth is that RANKL is a small globular protein, which can be expected to have unobstructed access to all stromal cells through their conduits (3). We found that RANKL neutralization by a blocking mAb restored normal LN size, B/T cell ratio, and follicle area and numbers. We cannot formally exclude the possibility that blocking RANKL from other sources contributed to normalization; however, skin RANKL was definitely affected, because abnormal HF cycling and epidermal dysplasia were corrected (37). Loss of B cells and reduction of FRC proliferation beyond WT levels in the treated mice may be due to blocking RANKL produced by MRCs (7). Indeed, because RANKL neutralization led to the disappearance of MRCs from the preserved B cell follicles, a role for endogenous RANKL in the maintenance of MRCs can be evoked. Among other potential sources of SLO RANKL, we can exclude T cells, which synthesize RANKL only when activated (57). Although the mAb was administrated s.c., bone homeostasis was affected, because the treated mice displayed growth retardation. However, a disturbed hematopoietic cell output by reduction of BM niches (63) was not seen, because spleen cell numbers were not diminished. In a different model of skin RANKL overexpression by Rankl transgenesis under control of the keratin 14 promoter, cLN hyperplasia was not reported (64). Although this difference may suggest that factors other than RANKL are implicated in cLN growth, the cardinal difference between the two mouse models is that RANKL is not soluble in keratin 14 promoter Rankl Tg mice and, therefore, cannot freely reach the draining LNs. As proposed by the investigators, it is likely that the Tg RANKL principally affects Langerhans cells.

To identify the cellular RANKL target, we considered adult LTi-like cells and stromal cells as candidates. However, the proportion of adult LTi-like cells was not altered, nor was the phenotype and expression of LTα1β2 by adult LTi-like cells or B cells different. We also could not detect an increase in IL-7 synthesis by FRCs, an important SLO source of IL-7 (12) (data not shown). We could also rule out Rank transgene expression by the radioresistant stromal cells, because its mRNA was no longer detectable after WT → Tg BM transfer. However, all stromal subsets expressed endogenous RANK. It is known that endothelial cells (6567) and mesenchymal cells (56, 68) express RANK, but whether their SLO counterpart likewise carries RANK has not been studied. The cLN stroma upregulated chemokines and cell-adhesion molecules; however, more strikingly, it showed a dramatic hyperproliferation in the Tg mice, which was fully abolished by RANKL neutralization. Because of podoplanin coexpression it was not possible to differentiate between TRCs and MRCs within the FRC population. Other attempts to identify MRCs within the FRC population were not successful because of technical challenges, likely caused by the incompatibility of enzymatic liberation of cells with preservation of cell-adhesion molecules, such as MAdCAM-1. It is also possible that MRCs are concealed within the CD45+ cluster, given their close association with B cells (7). However, if MRCs were the only RANK-expressing cell type, one could expect a specific enrichment of MRCs versus TRCs. The unchanged qRT-PCR measures of Rankl mRNA, expressed by MRCs but not TRCs (7) (data not shown), do not support a scenario of MRC-restricted proliferation. Whether FDCs also express RANK is uncertain, owing to the known difficulties with obtaining a sufficiently homogeneous FDC population. RANK-mediated FDC proliferation would support the increase in B cell follicles. Indeed, the Rank gene was expressed in a transcriptome study of FDCs (69) (C. Berek, personal communication). However, in keeping with the idea that MRCs may be FDC precursors (7), it remains a possibility that RANK stimulation of MRC translates into de novo B cell follicle formation comprising FDCs. We observed that the proportion of LECs is diminished in Tg LNs, although proliferation and turnover rate are similar to those of BECs/FRCs. Occasionally, older mice (>20 wk) present with lymphedema in cLNs. This suggests an impaired function in the LEC compartment, so that reduction of egress rates may contribute to LN hyperplasia (11). The reasons for the loss of LECs are unclear and necessitate further investigations.

It was found that RANK ligation on vascular cells stimulates angiogenesis and ICAM-1 and VCAM-1 expression (66, 67, 70). In an artificial SLO-like cell system, high fluid pressure activates stromal cell chemokine synthesis (71). This invites speculation on the existence of an interlinked direct RANK action on endothelial cells and reticular fibroblasts. Importantly, LTβR and TNFR signaling are mandatory for FDC formation and maintenance (72, 73), and LTβR signaling contributes to SLO growth in response to viral infection (14). Also, CXCL13 and MAdCAM-1 expression on BECs are increased upon LTβR engagement (9). The possibility that LTβR signaling plays a part in cLN growth cannot be excluded, but we did not detect an upregulation of LTβR mRNA by FRCs (data not shown), nor could we find evidence for an increase in LTα1β2 by adult LTi-like cells or B cells. Elucidation of the precise roles of RANK and LTβR in SLO homeostasis is complicated by the finding that LTβR stimulates RANK and RANKL expression (36, 74). It would be of interest to test the effect of RANKL neutralization in a model of viral-induced SLO growth (14). Our results, together with the current understanding of RANK and LTβR actions, raise the question of whether RANK and LTβR functions are redundant. However, specificity in biological systems may be introduced by variations in spatial-temporal ligand availabilities.

The production of RANKL by activated T cells (75, 76) invites the idea that, during an immune response, T cells sustain SLO expansion and contribute to tertiary lymphoid tissue by stimulating RANK+ stroma. Conversely, CD40-stimulated B cells release OPG, which would keep RANKL in check (77). The findings presented in this article should prompt novel investigations into the role of T cell RANKL in regulating the quality and quantity of the immune response through stromal cells. The question of whether RANK-directed therapy can find novel applications in boosting the immune system or, on the contrary, in the treatment of inflammatory diseases, now also merits close examination. Our results support the idea that tissue-derived RANKL contributes to lymphoid tissue formation and growth by stimulating stromal cells, as well as highlight the importance of environmental cues for immune regulation (78).

We thank Monique Duval and Anne Soley for technical support and Claudine Ebel of the Institut de Génétique et de Biologie Moléculaire et Cellulaire flow cytometry facility for sorting. We appreciate critical reading of the manuscript by Antonino Nicoletti and Marc Coles.

This work was supported by Centre National de la Recherche Scientifique, University of Strasbourg, and La Ligue Contre le Cancer (to E.H.). V.D. and E.H. were recipients of fellowships from Centre National de la Recherche Scientifique Bourse Doctorat Ingénieur and Allocation Ministérielle de Recherche, respectively.

The online version of this article contains supplemental material.

Abbreviations used in this article:

BEC

blood endothelial cell

BM

bone marrow

cLN

cutaneous lymph node

DC

dendritic cell

FDC

B cell follicular dendritic cell

FRC

fibroblastic reticular cell

HF

hair follicle

iLN

inguinal lymph node

KLH

keyhole limpet hemocyanin

KO

knockout

LEC

lymph endothelial cell

LN

lymph node

LT

lymphotoxin

LTi

lymphoid tissue inducer

mLN

mesenteric lymph node

MRC

reticular cell of sinus marginal zone

NP

4-hydroxy-3-nitrophenyl-acetyl

OPG

osteoprotegerin

P

postnatal day

PTA

peripheral tissue Ag

qRT-PCR

quantitative RT-PCR

SLO

secondary lymphoid organ

Tg

transgenic

TRC

reticular cell of T zone

WT

wild-type.

1
Junt
T.
,
Scandella
E.
,
Ludewig
B.
.
2008
.
Form follows function: lymphoid tissue microarchitecture in antimicrobial immune defence.
Nat. Rev. Immunol.
8
:
764
775
.
2
Mueller
S. N.
,
Germain
R. N.
.
2009
.
Stromal cell contributions to the homeostasis and functionality of the immune system.
Nat. Rev. Immunol.
9
:
618
629
.
3
Lämmermann
T.
,
Sixt
M.
.
2008
.
The microanatomy of T-cell responses.
Immunol. Rev.
221
:
26
43
.
4
Roozendaal
R.
,
Mebius
R. E.
.
2011
.
Stromal cell-immune cell interactions.
Annu. Rev. Immunol.
29
:
23
43
.
5
Fletcher
A. L.
,
Malhotra
D.
,
Turley
S. J.
.
2011
.
Lymph node stroma broaden the peripheral tolerance paradigm.
Trends Immunol.
32
:
12
18
.
6
Luther
S. A.
,
Vogt
T. K.
,
Siegert
S.
.
2011
.
Guiding blind T cells and dendritic cells: A closer look at fibroblastic reticular cells found within lymph node T zones.
Immunol. Lett.
138
:
9
11
.
7
Katakai
T.
,
Suto
H.
,
Sugai
M.
,
Gonda
H.
,
Togawa
A.
,
Suematsu
S.
,
Ebisuno
Y.
,
Katagiri
K.
,
Kinashi
T.
,
Shimizu
A.
.
2008
.
Organizer-like reticular stromal cell layer common to adult secondary lymphoid organs.
J. Immunol.
181
:
6189
6200
.
8
Palframan
R. T.
,
Jung
S.
,
Cheng
G.
,
Weninger
W.
,
Luo
Y.
,
Dorf
M.
,
Littman
D. R.
,
Rollins
B. J.
,
Zweerink
H.
,
Rot
A.
,
von Andrian
U. H.
.
2001
.
Inflammatory chemokine transport and presentation in HEV: a remote control mechanism for monocyte recruitment to lymph nodes in inflamed tissues.
J. Exp. Med.
194
:
1361
1373
.
9
Browning
J. L.
,
Allaire
N.
,
Ngam-Ek
A.
,
Notidis
E.
,
Hunt
J.
,
Perrin
S.
,
Fava
R. A.
.
2005
.
Lymphotoxin-beta receptor signaling is required for the homeostatic control of HEV differentiation and function.
Immunity
23
:
539
550
.
10
Pham
T. H.
,
Baluk
P.
,
Xu
Y.
,
Grigorova
I.
,
Bankovich
A. J.
,
Pappu
R.
,
Coughlin
S. R.
,
McDonald
D. M.
,
Schwab
S. R.
,
Cyster
J. G.
.
2010
.
Lymphatic endothelial cell sphingosine kinase activity is required for lymphocyte egress and lymphatic patterning.
J. Exp. Med.
207
:
17
27
.
11
Schwab
S. R.
,
Cyster
J. G.
.
2007
.
Finding a way out: lymphocyte egress from lymphoid organs.
Nat. Immunol.
8
:
1295
1301
.
12
Link
A.
,
Vogt
T. K.
,
Favre
S.
,
Britschgi
M. R.
,
Acha-Orbea
H.
,
Hinz
B.
,
Cyster
J. G.
,
Luther
S. A.
.
2007
.
Fibroblastic reticular cells in lymph nodes regulate the homeostasis of naive T cells.
Nat. Immunol.
8
:
1255
1265
.
13
Kosco
M. H.
,
Pflugfelder
E.
,
Gray
D.
.
1992
.
Follicular dendritic cell-dependent adhesion and proliferation of B cells in vitro.
J. Immunol.
148
:
2331
2339
.
14
Kumar
V.
,
Scandella
E.
,
Danuser
R.
,
Onder
L.
,
Nitschké
M.
,
Fukui
Y.
,
Halin
C.
,
Ludewig
B.
,
Stein
J. V.
.
2010
.
Global lymphoid tissue remodeling during a viral infection is orchestrated by a B cell-lymphotoxin-dependent pathway.
Blood
115
:
4725
4733
.
15
Cupedo
T.
,
Mebius
R. E.
.
2005
.
Cellular interactions in lymph node development.
J. Immunol.
174
:
21
25
.
16
Honda
K.
,
Nakano
H.
,
Yoshida
H.
,
Nishikawa
S.
,
Rennert
P.
,
Ikuta
K.
,
Tamechika
M.
,
Yamaguchi
K.
,
Fukumoto
T.
,
Chiba
T.
,
Nishikawa
S. I.
.
2001
.
Molecular basis for hematopoietic/mesenchymal interaction during initiation of Peyer’s patch organogenesis.
J. Exp. Med.
193
:
621
630
.
17
Cupedo
T.
,
Vondenhoff
M. F.
,
Heeregrave
E. J.
,
De Weerd
A. E.
,
Jansen
W.
,
Jackson
D. G.
,
Kraal
G.
,
Mebius
R. E.
.
2004
.
Presumptive lymph node organizers are differentially represented in developing mesenteric and peripheral nodes.
J. Immunol.
173
:
2968
2975
.
18
White
A.
,
Carragher
D.
,
Parnell
S.
,
Msaki
A.
,
Perkins
N.
,
Lane
P.
,
Jenkinson
E.
,
Anderson
G.
,
Caamaño
J. H.
.
2007
.
Lymphotoxin a-dependent and -independent signals regulate stromal organizer cell homeostasis during lymph node organogenesis.
Blood
110
:
1950
1959
.
19
Coles
M. C.
,
Veiga-Fernandes
H.
,
Foster
K. E.
,
Norton
T.
,
Pagakis
S. N.
,
Seddon
B.
,
Kioussis
D.
.
2006
.
Role of T and NK cells and IL7/IL7r interactions during neonatal maturation of lymph nodes.
Proc. Natl. Acad. Sci. USA
103
:
13457
13462
.
20
Tumanov
A. V.
,
Kuprash
D. V.
,
Nedospasov
S. A.
.
2003
.
The role of lymphotoxin in development and maintenance of secondary lymphoid tissues.
Cytokine Growth Factor Rev.
14
:
275
288
.
21
Browning
J. L.
2008
.
Inhibition of the lymphotoxin pathway as a therapy for autoimmune disease.
Immunol. Rev.
223
:
202
220
.
22
Wada
T.
,
Nakashima
T.
,
Hiroshi
N.
,
Penninger
J. M.
.
2006
.
RANKL-RANK signaling in osteoclastogenesis and bone disease.
Trends Mol. Med.
12
:
17
25
.
23
Dejardin
E.
,
Droin
N. M.
,
Delhase
M.
,
Haas
E.
,
Cao
Y.
,
Makris
C.
,
Li
Z. W.
,
Karin
M.
,
Ware
C. F.
,
Green
D. R.
.
2002
.
The lymphotoxin-beta receptor induces different patterns of gene expression via two NF-kappaB pathways.
Immunity
17
:
525
535
.
24
Walsh
M. C.
,
Choi
Y.
.
2003
.
Biology of the TRANCE axis.
Cytokine Growth Factor Rev.
14
:
251
263
.
25
Kong
Y. Y.
,
Yoshida
H.
,
Sarosi
I.
,
Tan
H. L.
,
Timms
E.
,
Capparelli
C.
,
Morony
S.
,
Oliveira-dos-Santos
A. J.
,
Van
G.
,
Itie
A.
, et al
.
1999
.
OPGL is a key regulator of osteoclastogenesis, lymphocyte development and lymph-node organogenesis.
Nature
397
:
315
323
.
26
Dougall
W. C.
,
Glaccum
M.
,
Charrier
K.
,
Rohrbach
K.
,
Brasel
K.
,
De Smedt
T.
,
Daro
E.
,
Smith
J.
,
Tometsko
M. E.
,
Maliszewski
C. R.
, et al
.
1999
.
RANK is essential for osteoclast and lymph node development.
Genes Dev.
13
:
2412
2424
.
27
Kim
D.
,
Mebius
R. E.
,
MacMicking
J. D.
,
Jung
S.
,
Cupedo
T.
,
Castellanos
Y.
,
Rho
J.
,
Wong
B. R.
,
Josien
R.
,
Kim
N.
, et al
.
2000
.
Regulation of peripheral lymph node genesis by the tumor necrosis factor family member TRANCE.
J. Exp. Med.
192
:
1467
1478
.
28
Yoshida
H.
,
Naito
A.
,
Inoue
J.
,
Satoh
M.
,
Santee-Cooper
S. M.
,
Ware
C. F.
,
Togawa
A.
,
Nishikawa
S.
,
Nishikawa
S.
.
2002
.
Different cytokines induce surface lymphotoxin-alphabeta on IL-7 receptor-alpha cells that differentially engender lymph nodes and Peyer’s patches.
Immunity
17
:
823
833
.
29
Knoop
K. A.
,
Kumar
N.
,
Butler
B. R.
,
Sakthivel
S. K.
,
Taylor
R. T.
,
Nochi
T.
,
Akiba
H.
,
Yagita
H.
,
Kiyono
H.
,
Williams
I. R.
.
2009
.
RANKL is necessary and sufficient to initiate development of antigen-sampling M cells in the intestinal epithelium.
J. Immunol.
183
:
5738
5747
.
30
Schneider
M. R.
,
Schmidt-Ullrich
R.
,
Paus
R.
.
2009
.
The hair follicle as a dynamic miniorgan.
Curr. Biol.
19
:
R132
R142
.
31
Paus
R.
,
Nickoloff
B. J.
,
Ito
T.
.
2005
.
A ‘hairy’ privilege.
Trends Immunol.
26
:
32
40
.
32
Rossi
S. W.
,
Kim
M. Y.
,
Leibbrandt
A.
,
Parnell
S. M.
,
Jenkinson
W. E.
,
Glanville
S. H.
,
McConnell
F. M.
,
Scott
H. S.
,
Penninger
J. M.
,
Jenkinson
E. J.
, et al
.
2007
.
RANK signals from CD4(+)3(−) inducer cells regulate development of Aire-expressing epithelial cells in the thymic medulla.
J. Exp. Med.
204
:
1267
1272
.
33
Akiyama
T.
,
Shimo
Y.
,
Yanai
H.
,
Qin
J.
,
Ohshima
D.
,
Maruyama
Y.
,
Asaumi
Y.
,
Kitazawa
J.
,
Takayanagi
H.
,
Penninger
J. M.
, et al
.
2008
.
The tumor necrosis factor family receptors RANK and CD40 cooperatively establish the thymic medullary microenvironment and self-tolerance.
Immunity
29
:
423
437
.
34
White
A. J.
,
Withers
D. R.
,
Parnell
S. M.
,
Scott
H. S.
,
Finke
D.
,
Lane
P. J.
,
Jenkinson
E. J.
,
Anderson
G.
.
2008
.
Sequential phases in the development of Aire-expressing medullary thymic epithelial cells involve distinct cellular input.
Eur. J. Immunol.
38
:
942
947
.
35
Hikosaka
Y.
,
Nitta
T.
,
Ohigashi
I.
,
Yano
K.
,
Ishimaru
N.
,
Hayashi
Y.
,
Matsumoto
M.
,
Matsuo
K.
,
Penninger
J. M.
,
Takayanagi
H.
, et al
.
2008
.
The cytokine RANKL produced by positively selected thymocytes fosters medullary thymic epithelial cells that express autoimmune regulator.
Immunity
29
:
438
450
.
36
Mouri
Y.
,
Yano
M.
,
Shinzawa
M.
,
Shimo
Y.
,
Hirota
F.
,
Nishikawa
Y.
,
Nii
T.
,
Kiyonari
H.
,
Abe
T.
,
Uehara
H.
, et al
.
2011
.
Lymphotoxin signal promotes thymic organogenesis by eliciting RANK expression in the embryonic thymic stroma.
J. Immunol.
186
:
5047
5057
.
37
Duheron
V.
,
Hess
E.
,
Duval
M.
,
Decossas
M.
,
Castaneda
B.
,
Klöpper
J. E.
,
Amoasii
L.
,
Barbaroux
J. B.
,
Williams
I. R.
,
Yagita
H.
, et al
.
2011
.
Receptor activator of NF-kappaB (RANK) stimulates the proliferation of epithelial cells of the epidermo-pilosebaceous unit.
Proc. Natl. Acad. Sci. USA
108
:
5342
5347
.
38
Lagasse
E.
,
Weissman
I. L.
.
1997
.
Enforced expression of Bcl-2 in monocytes rescues macrophages and partially reverses osteopetrosis in op/op mice.
Cell
89
:
1021
1031
.
39
Bosisio
M. R.
,
Maisonneuve
C.
,
Gregoire
S.
,
Kettaneh
A.
,
Mueller
C. G.
,
Bridal
S. L.
.
2009
.
Ultrasound biomicroscopy: a powerful tool probing murine lymph node size in vivo.
Ultrasound Med. Biol.
35
:
1209
1216
.
40
Kamijo
S.
,
Nakajima
A.
,
Ikeda
K.
,
Aoki
K.
,
Ohya
K.
,
Akiba
H.
,
Yagita
H.
,
Okumura
K.
.
2006
.
Amelioration of bone loss in collagen-induced arthritis by neutralizing anti-RANKL monoclonal antibody.
Biochem. Biophys. Res. Commun.
347
:
124
132
.
41
Ruedl
C.
,
Koebel
P.
,
Bachmann
M.
,
Hess
M.
,
Karjalainen
K.
.
2000
.
Anatomical origin of dendritic cells determines their life span in peripheral lymph nodes.
J. Immunol.
165
:
4910
4916
.
42
Aïoub
M.
,
Lézot
F.
,
Molla
M.
,
Castaneda
B.
,
Robert
B.
,
Goubin
G.
,
Néfussi
J. R.
,
Berdal
A.
.
2007
.
Msx2−/− transgenic mice develop compound amelogenesis imperfecta, dentinogenesis imperfecta and periodental osteopetrosis.
Bone
41
:
851
859
.
43
Matsumoto
M.
,
Lo
S. F.
,
Carruthers
C. J.
,
Min
J.
,
Mariathasan
S.
,
Huang
G.
,
Plas
D. R.
,
Martin
S. M.
,
Geha
R. S.
,
Nahm
M. H.
,
Chaplin
D. D.
.
1996
.
Affinity maturation without germinal centres in lymphotoxin-alpha-deficient mice.
Nature
382
:
462
466
.
44
Castaneda
B.
,
Simon
Y.
,
Jacques
J.
,
Hess
E.
,
Choi
Y. W.
,
Blin-Wakkach
C.
,
Mueller
C.
,
Berdal
A.
,
Lézot
F.
.
2011
.
Bone resorption control of tooth eruption and root morphogenesis: Involvement of the receptor activator of NF-κB (RANK).
J. Cell. Physiol.
226
:
74
85
.
45
Lagasse
E.
,
Weissman
I. L.
.
1992
.
Mouse MRP8 and MRP14, two intracellular calcium-binding proteins associated with the development of the myeloid lineage.
Blood
79
:
1907
1915
.
46
Jakubzick
C.
,
Bogunovic
M.
,
Bonito
A. J.
,
Kuan
E. L.
,
Merad
M.
,
Randolph
G. J.
.
2008
.
Lymph-migrating, tissue-derived dendritic cells are minor constituents within steady-state lymph nodes.
J. Exp. Med.
205
:
2839
2850
.
47
Gardner
J. M.
,
Devoss
J. J.
,
Friedman
R. S.
,
Wong
D. J.
,
Tan
Y. X.
,
Zhou
X.
,
Johannes
K. P.
,
Su
M. A.
,
Chang
H. Y.
,
Krummel
M. F.
,
Anderson
M. S.
.
2008
.
Deletional tolerance mediated by extrathymic Aire-expressing cells.
Science
321
:
843
847
.
48
Cohen
J. N.
,
Guidi
C. J.
,
Tewalt
E. F.
,
Qiao
H.
,
Rouhani
S. J.
,
Ruddell
A.
,
Farr
A. G.
,
Tung
K. S.
,
Engelhard
V. H.
.
2010
.
Lymph node-resident lymphatic endothelial cells mediate peripheral tolerance via Aire-independent direct antigen presentation.
J. Exp. Med.
207
:
681
688
.
49
Fletcher
A. L.
,
Lukacs-Kornek
V.
,
Reynoso
E. D.
,
Pinner
S. E.
,
Bellemare-Pelletier
A.
,
Curry
M. S.
,
Collier
A. R.
,
Boyd
R. L.
,
Turley
S. J.
.
2010
.
Lymph node fibroblastic reticular cells directly present peripheral tissue antigen under steady-state and inflammatory conditions.
J. Exp. Med.
207
:
689
697
.
50
Ma
L.
,
Liu
J.
,
Wu
T.
,
Plikus
M.
,
Jiang
T. X.
,
Bi
Q.
,
Liu
Y. H.
,
Müller-Röver
S.
,
Peters
H.
,
Sundberg
J. P.
, et al
.
2003
.
‘Cyclic alopecia’ in Msx2 mutants: defects in hair cycling and hair shaft differentiation.
Development
130
:
379
389
.
51
Gretz
J. E.
,
Norbury
C. C.
,
Anderson
A. O.
,
Proudfoot
A. E.
,
Shaw
S.
.
2000
.
Lymph-borne chemokines and other low molecular weight molecules reach high endothelial venules via specialized conduits while a functional barrier limits access to the lymphocyte microenvironments in lymph node cortex.
J. Exp. Med.
192
:
1425
1440
.
52
Roozendaal
R.
,
Mempel
T. R.
,
Pitcher
L. A.
,
Gonzalez
S. F.
,
Verschoor
A.
,
Mebius
R. E.
,
von Andrian
U. H.
,
Carroll
M. C.
.
2009
.
Conduits mediate transport of low-molecular-weight antigen to lymph node follicles.
Immunity
30
:
264
276
.
53
Schmutz
S.
,
Bosco
N.
,
Chappaz
S.
,
Boyman
O.
,
Acha-Orbea
H.
,
Ceredig
R.
,
Rolink
A. G.
,
Finke
D.
.
2009
.
Cutting edge: IL-7 regulates the peripheral pool of adult ROR gamma+ lymphoid tissue inducer cells.
J. Immunol.
183
:
2217
2221
.
54
Meier
D.
,
Bornmann
C.
,
Chappaz
S.
,
Schmutz
S.
,
Otten
L. A.
,
Ceredig
R.
,
Acha-Orbea
H.
,
Finke
D.
.
2007
.
Ectopic lymphoid-organ development occurs through interleukin 7-mediated enhanced survival of lymphoid-tissue-inducer cells.
Immunity
26
:
643
654
.
55
Mebius
R. E.
2003
.
Organogenesis of lymphoid tissues.
Nat. Rev. Immunol.
3
:
292
303
.
56
Anderson
D. M.
,
Maraskovsky
E.
,
Billingsley
W. L.
,
Dougall
W. C.
,
Tometsko
M. E.
,
Roux
E. R.
,
Teepe
M. C.
,
DuBose
R. F.
,
Cosman
D.
,
Galibert
L.
.
1997
.
A homologue of the TNF receptor and its ligand enhance T-cell growth and dendritic-cell function.
Nature
390
:
175
179
.
57
Wong
B. R.
,
Josien
R.
,
Lee
S. Y.
,
Sauter
B.
,
Li
H. L.
,
Steinman
R. M.
,
Choi
Y.
.
1997
.
TRANCE (tumor necrosis factor [TNF]-related activation-induced cytokine), a new TNF family member predominantly expressed in T cells, is a dendritic cell-specific survival factor.
J. Exp. Med.
186
:
2075
2080
.
58
Cremer
I.
,
Dieu-Nosjean
M. C.
,
Maréchal
S.
,
Dezutter-Dambuyant
C.
,
Goddard
S.
,
Adams
D.
,
Winter
N.
,
Menetrier-Caux
C.
,
Sautès-Fridman
C.
,
Fridman
W. H.
,
Mueller
C. G.
.
2002
.
Long-lived immature dendritic cells mediated by TRANCE-RANK interaction.
Blood
100
:
3646
3655
.
59
Yun
T. J.
,
Tallquist
M. D.
,
Aicher
A.
,
Rafferty
K. L.
,
Marshall
A. J.
,
Moon
J. J.
,
Ewings
M. E.
,
Mohaupt
M.
,
Herring
S. W.
,
Clark
E. A.
.
2001
.
Osteoprotegerin, a crucial regulator of bone metabolism, also regulates B cell development and function.
J. Immunol.
166
:
1482
1491
.
60
Willerford
D. M.
,
Chen
J.
,
Ferry
J. A.
,
Davidson
L.
,
Ma
A.
,
Alt
F. W.
.
1995
.
Interleukin-2 receptor alpha chain regulates the size and content of the peripheral lymphoid compartment.
Immunity
3
:
521
530
.
61
Lee
J. W.
,
Epardaud
M.
,
Sun
J.
,
Becker
J. E.
,
Cheng
A. C.
,
Yonekura
A. R.
,
Heath
J. K.
,
Turley
S. J.
.
2007
.
Peripheral antigen display by lymph node stroma promotes T cell tolerance to intestinal self.
Nat. Immunol.
8
:
181
190
.
62
Shamoto
M.
,
Shinzato
M.
,
Qian
B.
,
Hosokawa
S.
,
Ishibashi
M.
.
1999
.
Paracortical hyperplasia of superficial lymph nodes in a new mutant strain of hairless rats (ISh): a lesion similar to human dermatopathic lymphadenopathy.
Pathol. Int.
49
:
305
309
.
63
Kollet
O.
,
Dar
A.
,
Shivtiel
S.
,
Kalinkovich
A.
,
Lapid
K.
,
Sztainberg
Y.
,
Tesio
M.
,
Samstein
R. M.
,
Goichberg
P.
,
Spiegel
A.
, et al
.
2006
.
Osteoclasts degrade endosteal components and promote mobilization of hematopoietic progenitor cells.
Nat. Med.
12
:
657
664
.
64
Loser
K.
,
Mehling
A.
,
Loeser
S.
,
Apelt
J.
,
Kuhn
A.
,
Grabbe
S.
,
Schwarz
T.
,
Penninger
J. M.
,
Beissert
S.
.
2006
.
Epidermal RANKL controls regulatory T-cell numbers via activation of dendritic cells.
Nat. Med.
12
:
1372
1379
.
65
Collin-Osdoby
P.
,
Rothe
L.
,
Anderson
F.
,
Nelson
M.
,
Maloney
W.
,
Osdoby
P.
.
2001
.
Receptor activator of NF-kappa B and osteoprotegerin expression by human microvascular endothelial cells, regulation by inflammatory cytokines, and role in human osteoclastogenesis.
J. Biol. Chem.
276
:
20659
20672
.
66
Kim
H. H.
,
Shin
H. S.
,
Kwak
H. J.
,
Ahn
K. Y.
,
Kim
J. H.
,
Lee
H. J.
,
Lee
M. S.
,
Lee
Z. H.
,
Koh
G. Y.
.
2003
.
RANKL regulates endothelial cell survival through the phosphatidylinositol 3′-kinase/Akt signal transduction pathway.
FASEB J.
17
:
2163
2165
.
67
Min
J. K.
,
Kim
Y. M.
,
Kim
Y. M.
,
Kim
E. C.
,
Gho
Y. S.
,
Kang
I. J.
,
Lee
S. Y.
,
Kong
Y. Y.
,
Kwon
Y. G.
.
2003
.
Vascular endothelial growth factor up-regulates expression of receptor activator of NF-kappa B (RANK) in endothelial cells. Concomitant increase of angiogenic responses to RANK ligand.
J. Biol. Chem.
278
:
39548
39557
.
68
Rifas
L.
,
Arackal
S.
,
Weitzmann
M. N.
.
2003
.
Inflammatory T cells rapidly induce differentiation of human bone marrow stromal cells into mature osteoblasts.
J. Cell. Biochem.
88
:
650
659
.
69
Wilke
G.
,
Steinhauser
G.
,
Grün
J.
,
Berek
C.
.
2010
.
In silico subtraction approach reveals a close lineage relationship between follicular dendritic cells and BP3(hi) stromal cells isolated from SCID mice.
Eur. J. Immunol.
40
:
2165
2173
.
70
Min
J. K.
,
Kim
Y. M.
,
Kim
S. W.
,
Kwon
M. C.
,
Kong
Y. Y.
,
Hwang
I. K.
,
Won
M. H.
,
Rho
J.
,
Kwon
Y. G.
.
2005
.
TNF-related activation-induced cytokine enhances leukocyte adhesiveness: induction of ICAM-1 and VCAM-1 via TNF receptor-associated factor and protein kinase C-dependent NF-kappaB activation in endothelial cells.
J. Immunol.
175
:
531
540
.
71
Tomei
A. A.
,
Siegert
S.
,
Britschgi
M. R.
,
Luther
S. A.
,
Swartz
M. A.
.
2009
.
Fluid flow regulates stromal cell organization and CCL21 expression in a tissue-engineered lymph node microenvironment.
J. Immunol.
183
:
4273
4283
.
72
Gommerman
J. L.
,
Browning
J. L.
.
2003
.
Lymphotoxin/light, lymphoid microenvironments and autoimmune disease.
Nat. Rev. Immunol.
3
:
642
655
.
73
Endres
R.
,
Alimzhanov
M. B.
,
Plitz
T.
,
Fütterer
A.
,
Kosco-Vilbois
M. H.
,
Nedospasov
S. A.
,
Rajewsky
K.
,
Pfeffer
K.
.
1999
.
Mature follicular dendritic cell networks depend on expression of lymphotoxin beta receptor by radioresistant stromal cells and of lymphotoxin beta and tumor necrosis factor by B cells.
J. Exp. Med.
189
:
159
168
.
74
Vondenhoff
M. F.
,
Greuter
M.
,
Goverse
G.
,
Elewaut
D.
,
Dewint
P.
,
Ware
C. F.
,
Hoorweg
K.
,
Kraal
G.
,
Mebius
R. E.
.
2009
.
LTbetaR signaling induces cytokine expression and up-regulates lymphangiogenic factors in lymph node anlagen.
J. Immunol.
182
:
5439
5445
.
75
Wong
B. R.
,
Rho
J.
,
Arron
J.
,
Robinson
E.
,
Orlinick
J.
,
Chao
M.
,
Kalachikov
S.
,
Cayani
E.
,
Bartlett
F. S.
 III
,
Frankel
W. N.
, et al
.
1997
.
TRANCE is a novel ligand of the tumor necrosis factor receptor family that activates c-Jun N-terminal kinase in T cells.
J. Biol. Chem.
272
:
25190
25194
.
76
Takayanagi
H.
,
Ogasawara
K.
,
Hida
S.
,
Chiba
T.
,
Murata
S.
,
Sato
K.
,
Takaoka
A.
,
Yokochi
T.
,
Oda
H.
,
Tanaka
K.
, et al
.
2000
.
T-cell-mediated regulation of osteoclastogenesis by signalling cross-talk between RANKL and IFN-gamma.
Nature
408
:
600
605
.
77
Yun
T. J.
,
Chaudhary
P. M.
,
Shu
G. L.
,
Frazer
J. K.
,
Ewings
M. K.
,
Schwartz
S. M.
,
Pascual
V.
,
Hood
L. E.
,
Clark
E. A.
.
1998
.
OPG/FDCR-1, a TNF receptor family member, is expressed in lymphoid cells and is up-regulated by ligating CD40.
J. Immunol.
161
:
6113
6121
.
78
Matzinger
P.
,
Kamala
T.
.
2011
.
Tissue-based class control: the other side of tolerance.
Nat. Rev. Immunol.
11
:
221
230
.

The authors have no conflicts of interest.