Dendritic cells (DCs) populate virtually all organs, including the kidney. Together with macrophages, they constitute the renal mononuclear phagocyte system,1 , 2 3 4 5 6 7 , 8 9 10 7 + regulatory T cells (Tregs ) play another important counter–regulatory role, and Treg depletion aggravated cGN.11–13 regs is not exactly clear but involves the immunosuppressive cytokine IL-10.14 regs into the nephritic kidney requires the chemokine receptor CCR6 on Tregs ,15 16
More than 90% of kidney DCs belong to a DC subset characterized by expression of the integrin CD11b and the chemokine receptor CX3 CR1, which shares phenotypic and functional characteristics with tissue macrophages.17–19 18 , 20–22 + DCs constitute 20%–50%, whereas they represent a minority of <5% of DCs in the kidney.20 + DCs largely depend on the NF-κ B component RelB and only partially depend on Flt3L.20 , 23–26 27 28 + T cells.27 , 29–31 32 33 + DCs served classic DC functions in the induction of adaptive immunity after migration to draining lymph nodes.34 , 35 reg differentiation ex vivo , suggesting that they might regulate tissue homestasis.36–38 −/− mice that lack CD103+ DCs showed unaltered inflammatory bowel disease.39 + DCs is poorly understood. Here, we addressed this question in cGN and discovered an immunoregulatory role of these DCs in the kidney.
Results
CD103+ DCs Accumulate in the Kidney during cGN
We induced cGN in C57BL/6 wild–type (WT) mice and enumerated DC subsets in kidney single–cell suspensions from healthy and nephritic mice at different time points using flow cytometry. Although CD103+ DCs represented a minority among the renal immune cells under homeostatic conditions, they accumulated appreciably during cGN (Figure 1B ). The numbers of CD103+ DCs did not change much in the first 4 days of cGN but more than tripled between days 4 and 10, peaking at day 7 (Figure 1C ). CD11b+ DCs increased similarly (Figure 1D ) as described before.2 + DCs but also, CD103+ DCs accumulate in kidney inflammation, suggesting that the latter might play a functional role from day 7 onward.
Figure 1.: CD103+ DCs accumulate in the kidney in cGN. (A) Gating strategy of kidney DC subsets defined as CD11c+ MHC II+ cells and subset analysis on the basis of CD11b and CD103. (B) Representative flow cytometric analysis of mice before (control [Ctrl.]) and 4 (d4) and 7 days (d7) after NTS injection. Quantification of (C) renal CD103+ DCs and (D) CD11b+ DCs in nephritic WT mice. Results represent two to four experiments, with 4–12 mice per group. Statistical significance was tested with the Kruskal–Wallis test with Dunn's multiple comparison test. *P <0.05; **P <0.01; ***P <0.001.
CD103+ DCs Are Protective in cGN
To clarify the role of CD103+ DCs in cGN, we used Batf3−/− mice, which lack these DCs but not CD11b+ DCs.26 −/− mice do not show spontaneous renal pathology.26 + DC numbers were reduced by 99.9% in kidneys of nephritic Batf3−/− mice, whereas CD11b+ DC numbers were unchanged (Figure 2A ). On day 7 after disease induction, cGN was notably aggravated in Batf3−/− mice compared with WT controls, which was evident by diminished creatinine clearance and a tendency toward higher proteinuria (Figure 2, B and C ). Compared with WT mice, the histologic analysis showed significantly increased numbers of glomeruli with extracapillary proliferates, which were more severe (e.g. , often circumferential in Batf3−/− versus segmental in WT mice) (Figure 2, D and E ). On day 12 of cGN, kidney function was even more reduced (Figure 2, F and G ). Histologic analysis revealed more severe glomerular damage (Figure 2, H and I ). These results indicated that CD103+ DCs are protective in cGN.
Figure 2.: Batf3 deficiency aggravates cGN. (A) Percentages and absolute cell counts of CD103+ DCs and CD11b+ DCs in WT and Batf3−/− mice 7 days after cGN induction. (B and F) Creatinine clearance, (C and G) albuminuria, and (D and H) percentage of crescentic glomeruli in WT and Batf3−/− mice (A–E) 7 and (F–I) 12 days after NTS challenge. (E and I) Representative periodic acid–Schiff– (PAS) or Gömöri Methamine silver–stained histologic sections. Results are combined from two to three individual experiments, with 5–12 mice per group. Significance was tested by two–tailed Mann–Whitney U test. Scale bars, 50 μ m. *P <0.05; **P <0.01; ***P <0.001.
Reduced IL-10 Production in Nephritic Batf3−/− Mice
When we analyzed parameters for intrarenal inflammation, we noted higher levels of the proinflammatory cytokine TNFα and a trend toward higher levels of IL-6 and more IFNγ –producing Th1 cells (Figure 3A ). Although the anti–inflammatory cytokine TGFβ was unchanged, IL-10 was markedly reduced (Figure 3B ). IL-10 plays an important regulatory role in cGN, and it can be produced by Tregs in this model.14
Figure 3.: Reduced intrarenal anti-inflammatory cytokine production in nephritic Batf3−/− mice. (A) Concentrations of TNFα and IL-6 in supernatants of kidneys of nephritic Batf3−/− and WT mice measured by bead assay. Intracellular staining for IFNγ in CD45+ CD4+ TCRβ + cells from kidney single–cell suspensions. (B) Quantitative RT-PCR analysis of TGFβ and IL-10 expression in kidneys from nephritic WT and Batf3−/− mice 7 days after cGN induction. mRNA levels are expressed as x -fold of healthy controls. (C) Tregs from kidneys of Foxp3IRES-mRFP × IL-10 ires gfp–enhanced reporter (Fir/Tiger) mice were stained for CD4+ and RFP+ , indicating Foxp3+ cells; DCs were stained for MHC II+ CD11c+ and further subdivided into CD103+ and CD11b+ DCs. Representative contour plots depicting IL-10− and IL-10+ Tregs , IL-10− and IL-10+ CD103+ kidney DCs, and IL-10− and IL-10+ CD11b+ kidney DCs. (D) Absolute numbers of renal IL-10+ Foxp3+ Tregs , IL-10+ CD103+ kidney DCs, and IL-10+ CD11b+ kidney DCs. (E) IL-10 production measured by bead assay in supernatants of FACS–sorted CD103+ DCs, CD11b+ DCs, and CD45– nonimmune cells from WT mice 7 days after cGN induction. Data are representative of two independent experiments using (A–D) four to eleven or (E) 10 mice per group. Statistical significance was tested by (A and B) two–tailed Mann–Whitney U test or (D and E) Kruskal–Wallis test with Dunn's multiple postcomparison test. Ctrl., control; n.d., not detectable. *P <0.05; **P <0.01.
We determined the cellular source of IL-10 in cGN with the use of double–knockin reporter mice (Foxp3IRES-mRFP × IL-10 ires gfp–enhanced reporter), which allow sensitive simultaneous detection of IL-10 (GFP+ ) and FoxP3 (mRFP+ ). Phenotypic characterization of mRFP+ Tregs revealed expression of CCR6 (Supplemental Figure 1 regs into the kidney.15 Supplemental Figure 1 + Tregs , as expected, and also, CD103+ DCs were positive, but there were no IL-10+ CD11b+ DCs (Figure 3C ). IL-10+ Tregs were more numerous than IL-10+ CD103+ DCs (Figure 3D ). These results suggests that CD103+ DCs maintain intrarenal IL-10 production indirectly via Tregs and to a smaller extent (also directly), by producing IL-10 themselves.
Reduction of Renal FoxP3+ Tregs but Not iNKT Cells in Nephritic Batf3−/− Mice
Tregs are potent anti–inflammatory regulators in cGN.11–13 + Tregs on day 7 of cGN, they were 65% less numerous in kidneys of Batf3−/− mice compared with WT controls (Figure 4, A and B ). By contrast, in CX3 CR1GFP/GFP mice, which lack renal CD11b+ DCs40 + DCs, Treg numbers were unaltered (Figure 4C ), indicating that CD103+ DCs but not CD11b+ DCs regulate intrarenal Treg numbers.
Figure 4.: Protective FoxP3+ Tregs but not iNKT cells are diminished in nephritic Batf3−/− mice. (A and B) Intrarenal Tregs detected by flow cytometry on day 7 after cGN induction. A shows representative contour plots, and B shows absolute numbers of CD4+ FoxP3+ Tregs per kidney. (C) Quantification of renal CD11b+ DCs, CD103+ DCs, and FoxP3+ Tregs in WT and CX3 CR1GFP/GFP mice 7 days after cGN induction. (D) Representative flow cytometry contour plots and absolute numbers of α -GalCer+ TCRβ + NKT cells in WT and Batf3−/− mice on day 7 after cGN induction. Results are combined from two to three individual experiments, with five to eight mice per group. Statistical significance was tested with the two–tailed Mann–Whitney U test. *P <0.05; **P <0.01.
We next asked whether CD103+ DCs also regulate the other previously described renoprotective T cell subsets, iNKT cells.10 + TCRβ + was not significantly reduced in Batf3−/− compared with WT mice (Figure 4D ), arguing against an impaired anti–inflammatory iNKT cell response in the absence of CD103+ DCs. It has to be kept in mind that this experiment does not permit conclusions on the baseline activation status of iNKT cells or their immunosuppressive capacity and therefore, does not absolutely exclude a role of iNKT cells in Batf3−/− mice. Nevertheless, these results support a role of CD103+ kidney DCs in the maintenance of FoxP3+ Tregs but not iNKT cells.
No Influence of Batf3 Deficiency on Systemic Nephritogenic Immune Responses
The reduction of intrarenal Tregs might theoretically result from a systemic reduction of these cells in Batf3−/− mice. We, therefore, enumerated Tregs in the spleen and excluded this alternative explanation by showing unchanged numbers of FoxP3+ Tregs in nephritic WT and Batf3−/− mice (Figure 5A ). Likewise, numbers of splenic IFNγ + CD4+ Th1 cells, CD11b+ DCs, and F4/80+ macrophages were unaltered, whereas CD8+ DCs were almost absent in spleens of Batf3-deficient mice (Figure 5A ).
Figure 5.: Batf3-deficiency does not affect the systemic nephritogenic immune response. (A) Quantification of splenic FoxP3+ Tregs , IFNγ + CD4+ Th1 cells, CD11b+ DCs, F4/80+ macrophages, and CD8+ DCs 7 days after NTS challenge. (B) Concentrations of TNFα and IFNγ in supernatants of kidneys of nephritic WT and Batf3−/− mice measured by bead assay. (C) Titers of mouse anti–sheep IgG and isotypes of IgG1 and IgG2a were measured in sera by ELISA. Analyses were performed 7 days after cGN induction. Results were combined from two experiments, with 4–11 mice per group. Statistical significance was tested with the two–tailed Mann–Whitney U test. ***P <0.001.
To corroborate these findings, we asked whether Batf3−/− mice mounted a stronger nephritogenic immune response in cGN. To address this possibility, we isolated splenocytes from nephritic WT and Batf3−/− mice and restimulated them in vitro with nephrotoxic sheep serum. Cytokine analysis of supernatants showed no significant difference in the production of TNFα and IFNγ between the two groups (Figure 5B ). As another parameter for the nephritogenic Th cell response, we measured serum antibody titers against sheep Ig, the nephritogenic antigen planted into the kidney. Titers of sheep Ig–specific IgG1, characterizing Th2 responses, and specific IgG2a, characterizing Th1 responses, were not altered between WT and Batf3−/− mice (Figure 5C ). Taken together, these data argued against an intensified systemic humoral or cellular immune response in nephritic Batf3−/− mice. Thus, a local effect of renal CD103+ DCs likely affected cGN in the kidneys of these mice.
Depletion of CD103+ DCs Reduces Intrarenal Tregs
To corroborate this interpretation, we decided to deplete CD103+ DCs with the use of Langerin–diphtheria toxin receptor (DTR) mice. Langerin is expressed by renal CD103+ DCs but not by CD8+ DCs,20 27 + kidney DCs and CD8+ splenic DCs. We induced cGN in these mice and depleted CD103+ DCs every second day by injecting diphtheria toxin (DT). As controls, WT mice were treated with the same amount of DT. On day 7 of nephritis, numbers of CD103+ kidney DCs in Langerin-DTR mice were 50% reduced (Figure 6A ), whereas the numbers of CD11b+ kidney DCs were not significantly changed (WT, 272,571±38,473; Langerin-DTR, 198,501±42,613; n =4; P >0.05). Intrarenal Treg numbers were reduced by about 50% as well (Figure 6B ). Creatinine clearance seemed to be lower in mice depleted of CD103+ DCs, albeit not to a statistically significant level (Figure 6C ). Crescent formation was not different (Figure 6D ). The less pronounced disease aggravation in Langerin-DTR mice compared with Batf3-deficient mice is likely because of the reduction of CD103+ DCs by only 50% in the former compared with >99% in the latter. Nevertheless, these findings supported the linkage between CD103+ DCs and Tregs and argued against a role of CD8+ lymphatic tissue DCs in the regulation of intrarenal Tregs .
Figure 6.: Conditional depletion of CD103+ DCs reduces Tregs . Numbers of (A) CD103+ DCs and (B) Tregs , (C) creatinine clearance, (D) percentage of crescentic glomeruli, and representative immunohistology in Langerin-DTR or WT mice 7 days after cGN induction that were injected with DT on days −1, 1, 3, and 5 after cGN induction. Results are representative for two experiments, with four mice per group. Statistical significance was tested with the unpaired t test. Scale bars, 50 μ m. *P <0.05.
Aggravated cGN Severity and Fewer Intrarenal FoxP3+ Tregs in Flt3L-Deficient Mice
Because the aggravation of cGN in Langerin-DTR mice was not pronounced enough, we wished to corroborate the protective role of CD103+ DCs by using another mouse line harboring fewer of these cells. To this end, we induced cGN in Fl3L−/− mice, in which CD8+ and CD103+ DCs are reduced in other tissues.20 , 41 + DCs were reduced in kidneys of nephritic Fl3L−/− mice by >80% (Figure 7A ). We also noted a significant reduction of CD11b+ DCs, albeit to a lesser extent (Figure 7B ). Intrarenal FoxP3+ Tregs were decreased by >50% in nephritic Flt3L−/− mice compared with WT controls (Figure 7C ), whereas renal iNKT cells were not significantly altered (WT, 23,260±10,322; Flt3L−/− , 22,983±5608; n =3; P >0.05), matching the observations in Batf3−/− mice (Figure 4D ). Creatinine clearance was lower in Flt3L−/− mice than in WT controls (Figure 7D ). Albuminuria and glomerular crescents, however, did not statistically differ between nephritic WT and Flt3L−/− mice 7 days after disease induction (Figure 7, E and F ). The less severe phenotype in Flt3L−/− mice might be explained by the previous finding that this cytokine also fosters CD11b+ DCs, albeit not as effectively as CD103+ DCs.25
Figure 7.: Flt3L−/− mice harbor less intrarenal Treg and show aggravated cGN (A) CD103+ DCs, (B) CD11b+ DCs, (C) CD4+ FoxP3+ Tregs , (D) creatinine clearance, (E) albuminuria, (F) proportion of crescentic glomeruli, and (G) representative periodic acid–Schiff–stained histologic images of glomeruli 7 days after cGN induction in WT and Flt3L−/− mice. Results are combined from three individual experiments, with 6–13 mice per group. Statistical significance was tested with the two–tailed Mann–Whitney U test. Scale bars, 50 μ m. *P <0.05; **P <0.01; ***P <0.001.
Flt3L Supplementation Increases Intrarenal Tregs and Attenuates cGN When Combined with Depletion of CCR2+ CD11b+ DCs
Previous studies have applied Flt3L to increase protective CD103+ DC numbers in the intestine and succeeded to attenuate experimental inflammatory bowel disease.42 + DCs and Tregs (Figure 8, A and B ). However, CD11b+ DCs were increased (Figure 8C ), which is consistent with previous studies showing that Flt3L also exerts a minor effect on CD11b+ DCs.20 , 23–25 + DCs, which may explain why disease severity was unaltered (Figure 8D ).
Figure 8.: Flt3L increases renal CD103+ DCs and Tregs and attenuates cGN when inflammatory DCs are removed. (A–D) Melanoma cells (5×105 ) expressing Flt3L or OVA as a control were implanted subcutaneously into mice, and on day 9, cGN was induced. After another 7 days, (A) renal CD103+ , (B) Tregs , (C) CD11b+ DCs, and (D) creatinine clearance were determined. (E–K) Same experiment as in A–D, except that inflammatory CD11b+ DCs were depleted using CCR2 antibodies. (E) Renal CD103+ , (F) Tregs , (G) CD11b+ DCs, (H) creatinine clearance, (I) albuminuria, and (J) the proportion of crescentic glomeruli were determined. K shows representative histologic images stained by Gömöri Methamine silver. Results are representative of two individual experiments, with four to six mice per group in each experiment. Statistical significance was tested with the two–tailed Mann–Whitney U test. Scale bars, 50 μ m. *P <0.05; **P <0.01.
To overcome this problem, we decided to remove the recruited CD11b+ DCs with the use of CCR2 antibodies, which specifically targets inflammatory myeloid cells but not T cells.43 + DCs and Tregs (Figure 8, E and F ), but the number of CD11b+ DCs did not increase (Figure 8G ). Indeed, creatinine clearance was higher and albuminuria was lower in mice supplemented with Flt3L (Figure 8, H and I ). The proportion of crescentic glomeruli was not altered (Figure 8J ), despite the histologic impression of improved morphology (Figure 8K ). These findings showed that increasing only CD103+ DCs but not CD11b+ DCs increases renal Tregs and attenuates renal function in cGN.
CD103+ Kidney DCs Can Induce FoxP3+ Tregs and Produce the Treg –Recruiting Chemokine CCL20 Ex Vivo
To understand how CD103+ kidney DCs sustain intrarenal Tregs , we flow sorted CD103+ and as a control, CD11b+ kidney DCs and cocultured them with naive specific Th cells and antigen in vitro . As a model antigen, we used OVA, and as OVA–specific T cells, we used the well established DO11.10 transgenic T cells, which in contrast to OT-II cells, can be readily transformed into Tregs .44 reg conversion by measuring FoxP3 induction. Furthermore, DEREG mice allow for depletion of preexisting Tregs by injecting DT. This allows for the conclusion that FoxP3+ cells have arisen from naive Th cells and have not by expansion of preexisting Tregs from the donor mice. When we analyzed these cocultures, CD103+ kidney DCs but not CD11b+ kidney DCs readily induced differentiation of sorted naive CD4+ T cells into FoxP3+ CD25+ Tregs (Figure 9, A and B ). In contrast, CD11b+ kidney DCs induced more Th effector cells identified as CD4+ CD25+ FoxP3− DO11.10 cells (Figure 9C ).
Figure 9.: CD103+ kidney DCs can induce FoxP3+ Tregs and produce the Treg –attracting chemokine CCL20. (A–C) Kidney single–cell suspensions from C57BL/6xBALB/c mice 1 hour after ip injection of 700 μ g OVA per mouse were stained for MHC II+ CD11c+ DCs and sorted according to CD103 and CD11b expression 7 days after cGN induction; 2×104 CD103+ or CD11b+ kidney DCs were cultured with 4×104 DO11.10 × DEREG CD4+ T cells purified from DT-treated mice. Shown are (A) absolute numbers of CD4+ FoxP3+ CD25+ KJ126+ Tregs after 6 days and (B) their intracellular FoxP3 expression and (C) absolute numbers of CD4+ FoxP3– CD25+ KJ126+ effector Th cells. Cytokine-induced Tregs served as positive controls. Negative controls were unstimulated T cells. (D–F) CCL20 protein expression shown as (D) representative histograms and (E) MFI of flow cytometric analysis or (F) ELISA analysis of culture supernatants of sorted cells from mice 7 days after cGN induction. Splenic DCs stimulated with LPS for 2 hours served as positive controls for the staining, and negative controls were stained using isotype antibodies. (G) CCL20 mRNA levels in WT and Batf3−/− mice 7 days after cGN induction. Statistical significance was tested using (A–F) Kruskal–Wallis test with Dunn's multiple comparison test and (G) unpaired t test. contr., Control; neg., negative; pos., positive. *P <0.05.
Finally, we wondered whether CD103+ DCs might be able to recruit Tregs into the kidney. Because the chemokine receptor CCR6 on Tregs is essential for such recruitment,15 + MHC II+ kidney DCs from nephritic mice for the only known CCR6 ligand, CCL20.16 + DCs, we noted a subset that produced considerable levels of this chemokine, whereas no CD11b+ DCs did so (Figure 9, D and E ). When we isolated CD103+ DCs from nephritic kidneys, CCL20 was secreted into the supernatant by CD103+ DCs but not by CD11b+ DCs (Figure 9F ). CCL20 was also produced by nonhematopoietic cells (Figure 9F ). CCL20 levels in nephritic kidneys were reduced in Batf3−/− mice compared with WT controls (Figure 9G ), showing that CD103+ DCs contributed a significant part of chemokine production in cGN. These findings suggest that CD103+ kidney DCs might regulate Tregs by inducing their differentiation and/or recruiting them into the nephritic kidney.
Discussion
Tissue DCs of both the CD11b+ and the CD103+ subsets are being studied extensively in many disease settings. The CD103+ DC subset is remarkably rare in the kidney, compared with other organs.20 + kidney DCs.19 + DCs and their role in cGN by using a variety of genetic tools. We found that cGN was severely aggravated in mice that lack CD103+ DCs. Mechanistic analysis revealed a substantial reduction of two factors previously shown to be protective in cGN: FoxP3+ Tregs and IL-10. The latter was mainly produced by intrarenal Tregs but also, to some extent, by CD103+ DCs themselves. Systemic antirenal immune responses were unchanged in Batf3-deficient mice, validating a local effect by CD103+ DCs in the kidney rather than through Batf3–dependent CD8+ DCs in lymphatic tissues. This was corroborated by our results in Langerin-DTR mice, which allow for distinguishing between these two Batf3–dependent DC subsets.
The severity of cGN and Treg numbers in the three mouse lines examined here was tightly correlated with the efficiency and specificity of the CD103+ DC reduction. This was best in Batf3-deficient mice, and consistently, these mice showed the lowest Treg numbers and the most severe nephritis symptoms. In Langerin-DTR mice, specificity was equally high, but efficiency was moderate along with the effect on Tregs and cGN symptoms. DC reduction was better in Flt3L-deficient mice, but specificity for CD103+ DCs was not, so that some proinflammatory CD11b+ DCs were lost as well. Consistently, cGN was aggravated but not as severely as in Batf3-deficient mice. Thus, CD103+ renal DCs regulate the abundance of FoxP3+ Tregs in the kidney and thereby, regulate the progression of cGN.
In the intestine, lung, and skin, the ability of CD103+ DCs to induce Tregs has been documented, suggesting usefulness for the modulation of autoimmune diseases by harnessing their protective immunoregulatory functions.36 , 37 , 45 , 46 −/− mice lack CD103+ DCs, and some of them documented fewer Tregs and more inflammation and tissue damage (for example, in the aorta), leading to exacerbated atherosclerosis.47 + DCs and increased Tregs , thereby attenuating murine inflammatory bowel disease.42 + DCs and Tregs in the kidney. Flt3L is known to increase not only CD103+ but also, CD11b+ tissue DC numbers to a lesser extent,20 , 23–25 + DCs are 10 times more numerous in the kidney, so that a small relative increase raises absolute cell numbers as much as the great increase of the scarcer CD103+ DCs. This is different in intestine and lung, where CD11b+ DCs and CD103+ DCs are similarly frequent, so that Flt3L caused a comparatively stronger increase of CD103+ DCs.
The concomitant increase of both kidney DCs types might have balanced their anti- and proinflammatory effects, and cGN was unchanged. This unexpected hurdle was overcome by preventing the accumulation of proinflammatory CD11b+ DCs using CCR2-depleting antibodies, so that cGN was attenuated. It is unclear whether this approach is feasible in the human situation, but nevertheless, these findings show that increasing CD103+ DCs but not CD11b+ DCs increases intrarenal Treg numbers, supporting causality between these cell types.
We found two hypothetic mechanisms by which CD103+ renal DCs might sustain Tregs . First, in coculture experiments, only these DCs but not CD11b+ kidney DCs induced FoxP3+ Tregs ex vivo . This is consistent with previous studies showing that intestinal CD103+ DCs can drive Treg differentiation after migrating to mesenteric lymph nodes when retinoid acid and TGFβ are present.45 , 46 + kidney DCs, but it is likely that Treg induction takes place in the renal lymph nodes given that naive T cells hardly enter nonlymphoid tissues. Second, another hypothetic mechanism is the production of CCL20, the only ligand of CCR6, which is required for recruiting Tregs into the inflamed brain48 , 49 50 48 , 49 51 , 52 regs prevailed, so that kidneys of CCR6-deficient mice contained fewer Tregs and more severe cGN was noted.15 −/− mice. Moreover, CCL20 production was abrogated when CD103+ DCs were lacking. These findings suggest that CD103+ renal DCs regulate Treg numbers by promoting their differentiation and/or recruiting them into nephritic kidneys, but future studies are required to validate these hypotheses.
During glomerular inflammation, FoxP3+ Tregs are essential for the regulation of Th1 immune responses,11–13 14 reg numbers, we observed a profound reduction of IL-10 in kidneys of nephritic Batf3−/− compared with WT mice, and FoxP3+ Tregs , indeed, were a major source of this cytokine. However, also, CD103+ DCs produced some IL-10 and may, thus, potentially attenuate nephritis directly. This would explain previous studies showing a large proportion of IL-10+ immune cells other than Tregs in the kidney.14 , 53 7 + DCs are especially numerous. However, the superior IL-10 production by Tregs will hamper the formal experimental proof that CD103+ DC–derived IL-10 is protective.
CD103+ DCs can play a proinflammatory role under infections conditions (for example, by activating antiviral CD8+ T cell responses through crosspresentation).30 , 54 + DCs aggravated some models of kidney inflammation9 , 55 3 , 5 , 7 + DCs or harmful CD11b+ DCs were operative.
In summary, we showed that CD103+ kidney DCs are protective in cGN and antagonize the proinflammatory functions of CX3 CR1–dependent CD11b+ kidney DCs. These findings identify a major role of a minor DC subset in immune-mediated GN. Strategies to improve the functions of CD103+ DCs might be of therapeutic use in this disease and potentially, in other inflammatory diseases.
Concise Methods
Mice, Reagents, and cGN Model
We bred 8- to 12-week-old C57BL/6, Batf3−/− , Flt3L−/− , Langerin-DTR, Foxp3IRES-mRFP × IL-10 ires gfp–enhanced reporter, and DO11.10 × DEREG mice and kept them under specific pathogen–free conditions at the animal facility of the University Clinic Bonn. cGN was induced by intraperitoneal (ip) injection of 0.35–0.45 ml nephritis serum per mouse as previously described.9 , 56 + DCs were depleted in Langerin-DTR mice by ip injection of 8 ng DT per 1 g body wt.57 + Tregs .58 5 B16 melanoma cells expressing either OVA or Flt3L59 , 60 43
Histology
Light microscopy was performed on 3- to 5-μ m paraffin sections of paraformaldehyde-fixed tissue stained by periodic acid–Schiff or Gömöri Methamine silver according to standard protocols. Kidney damage was histologically determined and documented by a nephropathologist blinded to the experimental groups.
Isolation of DCs from Murine Kidneys
Kidneys were digested with collagenase (Roche Diagnostics, Indianapolis, IN) and DNAse-I as described previously.2 + DCs or CD103+ DCs from kidneys were sorted with a BDAria Sorter (Becton Dickinson, San Diego, CA). Purity was usually 90%–95%.
Flow Cytometry
After treatment with Fc-blocking antibodies (Privigen; CLS Behring GmbH), cells were stained for 20 minutes at 4°C with fluorochrome-labeled mAbs against CD45 (clone 30-F11), CD11c (HL3), I-Ab (AF6–120.1), CD11b (M1/70), CD103 (2E7), F4/80 (BM8), Ly6G (1A8), CD4 (GK1.5), TCRβ (H57–597), CD25 (pC61), and FoxP3 (FJK-16s). All antibodies were from BioLegend (San Diego, CA) if not indicated otherwise. Absolute cell numbers were determined with Calibrite APC Beads (BD Biosciences, San Jose, CA), and cells were analyzed with a BDCanto II (Becton Dickinson).
Cytokine Measurement
Cytokines in kidney supernatants were determined by Flowcytometry Bead Assay (eBioscience, San Diego, CA) according to the manufacturer’s instructions. For cell type–specific measurements, single-cell suspensions were incubated in RPMI (10% FCS; Gibco, Carlsbad, CA) with 1 μ l/ml GolgiPlug (BD Biosciences) for 4 hours at 37°C. After cell surface staining, cells were fixed in 2% paraformaldehyde for 10 minutes on ice and permeabilized with PBS containing 2% BSA and 0.5% Saponin for 20 minutes at room temperature. Intracellular staining was performed in that buffer for 20 minutes at room temperature using fluorochrome-labeled mAbs against TNFα (clone MP6-XT22), IL-6 (MP5–20F3), and IFNγ (XMG1.2; BD Biosciences). CCL20 was measured by ELISA (Bio-Techne).
In Vitro Treg InductionNTS-treated mice were injected with 700 μ g OVA 1 hour before DC isolation; 2×104 sorted kidney DCs were cocultured with 4×104 naïve Th cells in 200 μ l RPMI medium (10% FCS; Gibco) in 96-well plates for 6 days. Naïve Th cells were obtained by sorting CD4+ CD25− GFP– T cells from spleens of DT–treated DO11.10 × DEREG mice with an Aria Sorter (Becton Dickinson). Purity was usually about 95%. Because DO11.10 cells are H2-Kd restricted, we used C57BL/6 × Balb/c F1 mice as DC donors, ensuring that DO11.10 cells can recognize antigen. As a positive control for Treg generation, we coated wells with α CD3 (1 μ g/ml) and added sorted T cells, α CD28 (2 μ g/ml; BD Biosciences), TGFβ (5 ng/ml; Peprotech), and IL-2 (100 U/ml; Proleukin; Novartis, Basel, Switzerland).
RNA Preparation and RT-PCR
Whole-tissue RNA was extracted using the RNA NucleoSpin Kit (Macherey-Nagel) according to the manufacturer’s instructions. RNA was reverse transcribed into cDNA using the High-Capacity Reverse Transcription Kit and amplified by 40 PCR cycles using SYBR Green (Applied Biosystems, Foster City, CA). The following primer sequences were used:
IL-10 (forward): 5′-CCAGTTTTACCTGGTAGAAGTGATG-3′;
IL-10 (reverse): 5′-TGTCTAGGTCCTGGAGTCCAGCAGACTC-3′;
TGFβ (forward): 5′-GCTCGCTTTGTACAACAGCACC-3′; and
TGFβ (reverse): 5′-GCGGTCCACCATTAGCACG-3′.
All samples were run in triplicate and normalized to GAPDH.
Miscellaneous Assays
For the analysis of albumin excretion and creatinine clearance, urine was collected over 18 hours using metabolic cages. Albumin concentrations in the urine were measured using a commercial mouse albumin ELISA kit according to the manufacturer’s instructions (Bethyl Laboratories, Inc., Montgomery, TX). Serum and urine creatinine concentrations were analyzed in the central laboratory of the University Hospital of Bonn. Serum levels of murine IgG1 and IgG2a specific for sheep Ig were determined by ELISA (Bethyl Laboratories, Inc.).
Statistical Analyses
Results are expressed as means±SEMs. Comparisons were drawn using a two–tailed unpaired t , Kruskal–Wallis, or two–sided nonparametric Mann–Whitney U test as indicated. P values <0.05 were considered significant.
Disclosures
None.
We thank Magdalena Esser for excellent technical assistance, Ken Murphy for Batf3−/− mice, Tim Sparwasser for DEREG mice, and Björn Clausen for Langerin–diphtheria toxin receptor mice.
This work was supported by Bonfor (to B.D.G.E.); projects 4, 1, and 5 of Deutsche Forschungsgemeinschaft (DFG) KFO228 (to G.T., U.P., and C.K., respectively); the SFBTR57, the SFB1192; DFG grant BO 3755/2-1 (to P.B.); and German Federal Ministry of Education and Research grant BMBF 01GM1518A (to P.B.). C.K. is a member of the ExcellenceCluster ImmunoSensation. We acknowledge support from the Flow Cytometry Core Facility and the House for Experimental Therapy.
References
1. Soos TJ, Sims TN, Barisoni L, Lin K, Littman DR, Dustin ML, Nelson PJ: CX3CR1+ interstitial
dendritic cells form a contiguous network throughout the entire kidney. Kidney Int 70: 591–596, 200616760907
2. Krüger T, Benke D, Eitner F, Lang A, Wirtz M, Hamilton-Williams EE, Engel D, Giese B, Müller-Newen G, Floege J, Kurts C: Identification and functional characterization of
dendritic cells in the healthy murine kidney and in experimental glomerulonephritis. J Am Soc Nephrol 15: 613–621, 200414978163
3. Dong X, Swaminathan S, Bachman LA, Croatt AJ, Nath KA, Griffin MD: Resident
dendritic cells are the predominant TNF-secreting cell in early renal ischemia-reperfusion injury. Kidney Int 71: 619–628, 200717311071
4. Tittel AP, Heuser C, Ohliger C, Knolle PA, Engel DR, Kurts C: Kidney
dendritic cells induce innate immunity against bacterial pyelonephritis. J Am Soc Nephrol 22: 1435–1441, 201121757770
5. Mulay SR, Kulkarni OP, Rupanagudi KV, Migliorini A, Darisipudi MN, Vilaysane A, Muruve D, Shi Y, Munro F, Liapis H, Anders HJ: Calcium oxalate crystals induce renal inflammation by NLRP3-mediated IL-1β secretion. J Clin Invest 123: 236–246, 201323221343
6. Dong X, Bachman LA, Miller MN, Nath KA, Griffin MD:
Dendritic cells facilitate accumulation of IL-17 T cells in the kidney following acute renal obstruction. Kidney Int 74: 1294–1309, 200818974760
7. Hochheiser K, Engel DR, Hammerich L, Heymann F, Knolle PA, Panzer U, Kurts C: Kidney
dendritic cells become pathogenic during crescentic glomerulonephritis with proteinuria. J Am Soc Nephrol 22: 306–316, 201121164025
8. Pindjakova J, Hanley SA, Duffy MM, Sutton CE, Weidhofer GA, Miller MN, Nath KA, Mills KH, Ceredig R, Griffin MD: Interleukin-1 accounts for intrarenal Th17 cell activation during ureteral obstruction. Kidney Int 81: 379–390, 201221975862
9. Scholz J, Lukacs-Kornek V, Engel DR, Specht S, Kiss E, Eitner F, Floege J, Groene HJ, Kurts C: Renal
dendritic cells stimulate IL-10 production and attenuate nephrotoxic nephritis. J Am Soc Nephrol 19: 527–537, 200818235094
10. Riedel JH, Paust HJ, Turner JE, Tittel AP, Krebs C, Disteldorf E, Wegscheid C, Tiegs G, Velden J, Mittrücker HW, Garbi N, Stahl RA, Steinmetz OM, Kurts C, Panzer U: Immature renal
dendritic cells recruit regulatory CXCR6(+) invariant natural killer T cells to attenuate crescentic GN. J Am Soc Nephrol 23: 1987–2000, 201223138484
11. Wolf D, Hochegger K, Wolf AM, Rumpold HF, Gastl G, Tilg H, Mayer G, Gunsilius E, Rosenkranz AR: CD4+CD25+ regulatory T cells inhibit experimental anti-glomerular basement membrane glomerulonephritis in mice. J Am Soc Nephrol 16: 1360–1370, 200515788479
12. Paust HJ, Ostmann A, Erhardt A, Turner JE, Velden J, Mittrücker HW, Sparwasser T, Panzer U, Tiegs G: Regulatory T cells control the Th1 immune response in murine crescentic glomerulonephritis. Kidney Int 80: 154–164, 201121525855
13. Ooi JD, Snelgrove SL, Engel DR, Hochheiser K, Ludwig-Portugall I, Nozaki Y, O’Sullivan KM, Hickey MJ, Holdsworth SR, Kurts C, Kitching AR: Endogenous foxp3(+) T-regulatory cells suppress anti-glomerular basement membrane nephritis. Kidney Int 79: 977–986, 201121248715
14. Ostmann A, Paust HJ, Panzer U, Wegscheid C, Kapffer S, Huber S, Flavell RA, Erhardt A, Tiegs G: Regulatory T cell-derived IL-10 ameliorates crescentic GN. J Am Soc Nephrol 24: 930–942, 201323641052
15. Turner JE, Paust HJ, Steinmetz OM, Peters A, Riedel JH, Erhardt A, Wegscheid C, Velden J, Fehr S, Mittrücker HW, Tiegs G, Stahl RA, Panzer U: CCR6 recruits regulatory T cells and Th17 cells to the kidney in glomerulonephritis. J Am Soc Nephrol 21: 974–985, 201020299360
16. Power CA, Church DJ, Meyer A, Alouani S, Proudfoot AE, Clark-Lewis I, Sozzani S, Mantovani A, Wells TN: Cloning and characterization of a specific receptor for the novel CC chemokine MIP-3alpha from lung
dendritic cells . J Exp Med 186: 825–835, 19979294137
17. Steinman RM, Idoyaga J: Features of the dendritic cell lineage. Immunol Rev 234: 5–17, 201020193008
18. Geissmann F, Manz MG, Jung S, Sieweke MH, Merad M, Ley K: Development of monocytes, macrophages, and
dendritic cells . Science 327: 656–661, 201020133564
19. Nelson PJ, Rees AJ, Griffin MD, Hughes J, Kurts C, Duffield J: The renal mononuclear phagocytic system. J Am Soc Nephrol 23: 194–203, 201222135312
20. Ginhoux F, Liu K, Helft J, Bogunovic M, Greter M, Hashimoto D, Price J, Yin N, Bromberg J, Lira SA, Stanley ER, Nussenzweig M, Merad M: The origin and development of nonlymphoid tissue CD103+ DCs. J Exp Med 206: 3115–3130, 200920008528
21. Guilliams M, Henri S, Tamoutounour S, Ardouin L, Schwartz-Cornil I, Dalod M, Malissen B: From skin
dendritic cells to a simplified classification of human and mouse dendritic cell subsets. Eur J Immunol 40: 2089–2094, 201020853491
22. Merad M, Sathe P, Helft J, Miller J, Mortha A: The dendritic cell lineage: Ontogeny and function of
dendritic cells and their subsets in the steady state and the inflamed setting. Annu Rev Immunol 31: 563–604, 201323516985
23. Burkly L, Hession C, Ogata L, Reilly C, Marconi LA, Olson D, Tizard R, Cate R, Lo D: Expression of relB is required for the development of thymic medulla and
dendritic cells . Nature 373: 531–536, 19957845467
24. Maraskovsky E, Brasel K, Teepe M, Roux ER, Lyman SD, Shortman K, McKenna HJ: Dramatic increase in the numbers of functionally mature
dendritic cells in Flt3 ligand-treated mice: Multiple dendritic cell subpopulations identified. J Exp Med 184: 1953–1962, 19968920882
25. McKenna HJ, Stocking KL, Miller RE, Brasel K, De Smedt T, Maraskovsky E, Maliszewski CR, Lynch DH, Smith J, Pulendran B, Roux ER, Teepe M, Lyman SD, Peschon JJ: Mice lacking flt3 ligand have deficient hematopoiesis affecting hematopoietic progenitor cells,
dendritic cells , and natural killer cells. Blood 95: 3489–3497, 200010828034
26. Hildner K, Edelson BT, Purtha WE, Diamond M, Matsushita H, Kohyama M, Calderon B, Schraml BU, Unanue ER, Diamond MS, Schreiber RD, Murphy TL, Murphy KM: Batf3 deficiency reveals a critical role for CD8alpha+
dendritic cells in cytotoxic T cell immunity. Science 322: 1097–1100, 200819008445
27. Shortman K, Heath WR: The CD8+ dendritic cell subset. Immunol Rev 234: 18–31, 201020193009
28. Mashayekhi M, Sandau MM, Dunay IR, Frickel EM, Khan A, Goldszmid RS, Sher A, Ploegh HL, Murphy TL, Sibley LD, Murphy KM: CD8α(+)
dendritic cells are the critical source of interleukin-12 that controls acute infection by Toxoplasma gondii tachyzoites. Immunity 35: 249–259, 201121867928
29. den Haan JM, Lehar SM, Bevan MJ: CD8(+) but not CD8(-)
dendritic cells cross-prime cytotoxic T cells in vivo. J Exp Med 192: 1685–1696, 200011120766
30. del Rio ML, Rodriguez-Barbosa JI, Kremmer E, Förster R: CD103- and CD103+ bronchial lymph node
dendritic cells are specialized in presenting and cross-presenting innocuous antigen to CD4+ and CD8+ T cells. J Immunol 178: 6861–6866, 200717513734
31. Gottschalk C, Damuzzo V, Gotot J, Kroczek RA, Yagita H, Murphy KM, Knolle PA, Ludwig-Portugall I, Kurts C: Batf3-dependent
dendritic cells in the renal lymph node induce tolerance against circulating antigens. J Am Soc Nephrol 24: 543–549, 201323411785
32. Kurts C, Robinson BW, Knolle PA: Cross-priming in health and disease. Nat Rev Immunol 10: 403–414, 201020498667
33. Wang Y, Wang YP, Tay YC, Harris DC: Role of CD8(+) cells in the progression of murine adriamycin nephropathy. Kidney Int 59: 941–949, 200111231349
34. Schulz O, Jaensson E, Persson EK, Liu X, Worbs T, Agace WW, Pabst O: Intestinal CD103+, but not CX3CR1+, antigen sampling cells migrate in lymph and serve classical dendritic cell functions. J Exp Med 206: 3101–3114, 200920008524
35. Denning TL, Wang YC, Patel SR, Williams IR, Pulendran B: Lamina propria macrophages and
dendritic cells differentially induce regulatory and interleukin 17-producing T cell responses. Nat Immunol 8: 1086–1094, 200717873879
36. Coombes JL, Siddiqui KR, Arancibia-Cárcamo CV, Hall J, Sun CM, Belkaid Y, Powrie F: A functionally specialized population of mucosal CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-beta and retinoic acid-dependent mechanism. J Exp Med 204: 1757–1764, 200717620361
37. Sun CM, Hall JA, Blank RB, Bouladoux N, Oukka M, Mora JR, Belkaid Y: Small intestine lamina propria
dendritic cells promote de novo generation of Foxp3 T reg cells via retinoic acid. J Exp Med 204: 1775–1785, 200717620362
38. Scott CL, Aumeunier AM, Mowat AM: Intestinal CD103+
dendritic cells : Master regulators of tolerance? Trends Immunol 32: 412–419, 201121816673
39. Edelson BT, Kc W, Juang R, Kohyama M, Benoit LA, Klekotka PA, Moon C, Albring JC, Ise W, Michael DG, Bhattacharya D, Stappenbeck TS, Holtzman MJ, Sung SS, Murphy TL, Hildner K, Murphy KM: Peripheral CD103+
dendritic cells form a unified subset developmentally related to CD8alpha+ conventional
dendritic cells . J Exp Med 207: 823–836, 201020351058
40. Hochheiser K, Heuser C, Krause TA, Teteris S, Ilias A, Weisheit C, Hoss F, Tittel AP, Knolle PA, Panzer U, Engel DR, Tharaux PL, Kurts C: Exclusive CX3CR1 dependence of kidney DCs impacts glomerulonephritis progression. J Clin Invest 123: 4242–4254, 201323999431
41. Waskow C, Liu K, Darrasse-Jèze G, Guermonprez P, Ginhoux F, Merad M, Shengelia T, Yao K, Nussenzweig M: The receptor tyrosine kinase Flt3 is required for dendritic cell development in peripheral lymphoid tissues. Nat Immunol 9: 676–683, 200818469816
42. Collins CB, Aherne CM, McNamee EN, Lebsack MDP, Eltzschig H, Jedlicka P, Rivera-Nieves J: Flt3 ligand expands CD103⁺
dendritic cells and FoxP3⁺ T regulatory cells, and attenuates Crohn’s-like murine ileitis. Gut 61: 1154–1162, 201222068168
43. Mack M, Cihak J, Simonis C, Luckow B, Proudfoot AE, Plachý J, Brühl H, Frink M, Anders HJ, Vielhauer V, Pfirstinger J, Stangassinger M, Schlöndorff D: Expression and characterization of the chemokine receptors CCR2 and CCR5 in mice.
J Immunol 166: 4697–4704, 2001
44. Croce AC, Ferrigno A, Santin G, Vairetti M, Bottiroli G: Bilirubin: An autofluorescence bile biomarker for liver functionality monitoring. J Biophotonics 7: 810–817, 201423616471
45. Idoyaga J, Fiorese C, Zbytnuik L, Lubkin A, Miller J, Malissen B, Mucida D, Merad M, Steinman RM: Specialized role of migratory
dendritic cells in peripheral tolerance induction. J Clin Invest 123: 844–854, 201323298832
46. Khare A, Krishnamoorthy N, Oriss TB, Fei M, Ray P, Ray A: Cutting edge: Inhaled antigen upregulates retinaldehyde dehydrogenase in lung CD103+ but not plasmacytoid
dendritic cells to induce Foxp3 de novo in CD4+ T cells and promote airway tolerance. J Immunol 191: 25–29, 201323733880
47. Choi J-H, Cheong C, Dandamudi DB, Park CG, Rodriguez A, Mehandru S, Velinzon K, Jung I-H, Yoo J-Y, Oh GT, Steinman RM: Flt3 signaling-dependent
dendritic cells protect against atherosclerosis. Immunity 35: 819–831, 201122078798
48. Villares R, Cadenas V, Lozano M, Almonacid L, Zaballos A, Martínez-A C, Varona R: CCR6 regulates EAE pathogenesis by controlling regulatory CD4+ T-cell recruitment to target tissues. Eur J Immunol 39: 1671–1681, 200919499521
49. Elhofy A, Depaolo RW, Lira SA, Lukacs NW, Karpus WJ: Mice deficient for CCR6 fail to control chronic experimental autoimmune encephalomyelitis. J Neuroimmunol 213: 91–99, 200919535153
50. Kurts C, Panzer U, Anders HJ, Rees AJ: The immune system and kidney disease: Basic concepts and clinical implications. Nat Rev Immunol 13: 738–753, 201324037418
51. Paust HJ, Turner JE, Steinmetz OM, Peters A, Heymann F, Hölscher C, Wolf G, Kurts C, Mittrücker HW, Stahl RA, Panzer U: The IL-23/Th17 axis contributes to renal injury in experimental glomerulonephritis. J Am Soc Nephrol 20: 969–979, 200919339380
52. Summers SA, Steinmetz OM, Li M, Kausman JY, Semple T, Edgtton KL, Borza DB, Braley H, Holdsworth SR, Kitching AR: Th1 and Th17 cells induce proliferative glomerulonephritis. J Am Soc Nephrol 20: 2518–2524, 200919820122
53. Reantragoon R, Corbett AJ, Sakala IG, Gherardin NA, Furness JB, Chen Z, Eckle SB, Uldrich AP, Birkinshaw RW, Patel O, Kostenko L, Meehan B, Kedzierska K, Liu L, Fairlie DP, Hansen TH, Godfrey DI, Rossjohn J, McCluskey J, Kjer-Nielsen L: Antigen-loaded MR1 tetramers define T cell receptor heterogeneity in mucosal-associated invariant T cells. J Exp Med 210: 2305–2320, 201324101382
54. Bedoui S, Whitney PG, Waithman J, Eidsmo L, Wakim L, Caminschi I, Allan RS, Wojtasiak M, Shortman K, Carbone FR, Brooks AG, Heath WR: Cross-presentation of viral and self antigens by skin-derived CD103+
dendritic cells . Nat Immunol 10: 488–495, 200919349986
55. Tadagavadi RK, Reeves WB: Renal
dendritic cells ameliorate nephrotoxic acute kidney injury. J Am Soc Nephrol 21: 53–63, 201019875815
56. Gotot J, Piotrowski E, Otte MS, Tittel AP, Linlin G, Yao C, Ziegelbauer K, Panzer U, Garbi N, Kurts C, Thaiss F: Inhibitor of NFκB kinase subunit 2 blockade hinders the initiation but aggravates the progression of crescentic GN [published online ahead of print November 16, 2015]. J Am Soc Nephrol doi:ASN.201506069926574045
57. Bennett CL, van Rijn E, Jung S, Inaba K, Steinman RM, Kapsenberg ML, Clausen BE: Inducible ablation of mouse Langerhans cells diminishes but fails to abrogate contact hypersensitivity. J Cell Biol 169: 569–576, 200515897263
58. Lahl K, Loddenkemper C, Drouin C, Freyer J, Arnason J, Eberl G, Hamann A, Wagner H, Huehn J, Sparwasser T: Selective depletion of Foxp3+ regulatory T cells induces a scurfy-like disease. J Exp Med 204: 57–63, 200717200412
59. Mach N, Gillessen S, Wilson SB, Sheehan C, Mihm M, Dranoff G: Differences in
dendritic cells stimulated in vivo by tumors engineered to secrete granulocyte-macrophage colony-stimulating factor or Flt3-ligand. Cancer Res 60: 3239–3246, 200010866317
60. Zou T, Caton AJ, Koretzky GA, Kambayashi T:
Dendritic cells induce regulatory T cell proliferation through antigen-dependent and -independent interactions. J Immunol 185: 2790–2799, 201020686126
