CCR7 Deficiency Exacerbates Injury in Acute Nephritis Due to Aberrant Localization of Regulatory T Cells

Abstract

The homing of dendritic cells and T cells to secondary lymphoid organs requires chemokine receptor 7 (CCR7) expression on these cells. T cells mediate the pathogenesis of experimental accelerated nephrotoxic serum nephritis (NTS), including its suppression by regulatory T cells (Tregs), but the contribution of CCR7 to this disease is unknown. Here, we compared the development of NTS in CCR7-knockout (KO) and wild-type (WT) mice. Compared with WT mice, CCR7KO mice developed more severe disease with significantly more inflammatory cells infiltrating the kidney. These cells included FoxP3⁺ Tregs, which were virtually absent from WT kidneys. The adoptive transfer of WT Tregs into CCR7KO mice at the time of immunization protected the recipients from disease; these cells homed to secondary lymphoid organs but not to kidneys. Conversely, adoptive transfer of CCR7KO Tregs into WT mice did not inhibit development of NTS. These data suggest that NTS can develop without CCR7 expression, but Treg-mediated disease suppression, which seems to occur in secondary lymphoid organs, requires CCR7.

Chemokine receptor 7 (CCR7) is highly expressed on mature dendritic cells (DCs) and subpopulations of CD4⁺ T cells and naive, central memory and regulatory T cells (Tregs), and is necessary for their entry and positioning in the secondary lymphoid organs.^1,2 According to the prevalent paradigm, CCR7 ligands CCL19 and CCL21 guide DCs and T cells into the T cell zones of lymph nodes, where their interaction leads to antigen presentation required for the efficient initiation of adaptive immune responses as well as their regulation.¹ In CCR7-deficient mice (CCR7KO), the secondary lymphoid organs are hypocellular and their microarchitecture is scrambled with consequent functional defects and dysregulation of T cell immune responses.^1,3 The T cell–mediated immunity was convincingly shown to play a key role in the pathogenesis of different clinical and experimental autoimmune diseases, including murine accelerated nephrotoxic serum nephritis (NTS).⁴ This rapidly progressive disease is induced by the injection of rabbit anti-mouse glomerular basement membrane (GBM) antibody and accelerated by a preceding immunization against rabbit IgG. Animals with NTS present with proteinuria and proliferative and inflammatory glomerular changes, including crescent formation and leukocyte infiltrates.⁵ Not only the induction of NTS depends on T cells but also its regulation. Recently, we provided evidence that CD4⁺CD25⁺FoxP3⁺ Tregs control the onset and the course of NTS.⁶ Of note, Tregs exerted their inhibitory function in the draining lymph nodes and could not be detected in the renal tissue.⁶

To investigate whether CCR7-mediated cell homing to the secondary lymphoid organs is essential for either the induction or the regulation of NTS, we compared the development of this disease and its various parameters in CCR7KO and wild-type (WT) mice. Initially, CCR7KO mice were described to have a reduced ability to mount immune responses³; however, we recently showed that CCR7KO mice may display augmented immune responses as a result of a defective positioning of Tregs in the lymph nodes.⁷ Corresponding to the latter results, we show in this study that CCR7KO mice not only are susceptible to NTS but also develop a much more severe disease as compared with WT controls. The adoptive transfer of WT CD4⁺CD25⁺ Tregs into CCR7KO mice ameliorated the severity of the disease parameters to the level seen in WT animals. Thus, on the one hand, the pathogenic immune response in NTS develops in the absence of CCR7. On the other hand, CCR7 expression by Tregs is essential for their correct positioning in the secondary lymphoid organs and consequently in vivo suppressive function in this disease.

Results

CCR7KO Mice Are More Susceptible to NTS

To evaluate the role of CCR7 in the pathogenesis of NTS, the disease was induced in CCR7KO and WT controls. One week later, CCR7KO mice displayed significantly increased albuminuria as compared with WT controls (Figure 1A). As expected on the basis of previous studies, mild hypercellularity and focal deposition of periodic acid-Schiff (PAS)-positive material (Figure 1C) were detected in WT controls 7 d after induction of NTS. These abnormalities were found to be more severe in CCR7KO mice (Figure 1D). The quantification of PAS-positive deposits revealed their significant increase in CCR7KO mice as compared with WT controls (Figure 1B). Not any of the CCR7KO mice survived the follow-up period of 14 d, whereas five of six WT mice were still alive at this time point (Figure 1E). The increased susceptibility of CCR7KO mice to NTS was in accordance with a significant increase in the numbers of CD4⁺ and CD8⁺ T cells (Figure 2A) as well as F4/80⁺ cells (Figure 2B) in kidneys of CCR7KO mice as compared with WT controls. CD4⁺ T cells and F4/80⁺ cells were also found to be significantly increased intraglomerularly (Figure 2C), whereas CD8⁺ T cells were not detectable in the glomeruli. Further flow cytometric analysis of kidney-infiltrating CD4⁺ T cells revealed only very few CD4⁺ T cells to infiltrate the kidneys of WT mice, whereas a significantly increased proportion of activated CD4⁺CD69⁺ effector T cells were detected in kidneys of CCR7KO mice (Figure 2D).

Figure 1.

CCR7KO mice with NTS display significantly increased albuminuria and severe morphologic changes in the glomeruli. (A) Albumin and creatinine were evaluated on day 7 after induction of NTS. Urine albumin excretion (in μg) was determined and expressed per milligram of urinary creatinine to standardize for the urine concentration. CCR7KO mice (; n = 17) displayed significantly increased albuminuria as compared with WT mice (■; n = 17). *P < 0.05. Raw data (mean ± SEM) for urinary albumin-creatinine ratio in healthy WT opposed to nephritic WT on day 7: Healthy WT = 4.62 ± 0.86 μg/mg; nephritic WT = 4354 ± 1995 μg/mg. Raw data (mean ± SEM) for urinary albumin-creatinine ratio in healthy CCR7KO opposed to nephritic CCR7KO on day 7: Healthy CCR7KO = 4.67 ± 1.05 μg/mg; nephritic CCR7KO = 120233 ± 65412 μg/mg. (B) Kidney sections from day 7 of NTS were stained with PAS, and PAS-positive glomerular deposits were quantified as described in the Concise Methods section. Glomeruli in the kidneys of CCR7KO mice (; n = 21) contained significantly more PAS-positive deposits as compared with WT mice (■; n = 18). *P < 0.05. Raw data (mean ± SEM) for PAS score on day 7: WT = 0.86 ± 0.13 μg/mg; CCR7KO = 1.90 ± 0.15 μg/mg. (C) A representative glomerulus from a WT mouse with moderate enlargement, few PAS-positive deposits, and slight mesangial matrix formation. (D) A representative glomerulus from a CCR7KO mouse showing accentuated signs of active NTS. (E) NTS was induced in CCR7KO and WT mice (n = 6 per group), and their survival was monitored. None of the CCR7KO mice (open cycle) survived the 14-d follow-up period, whereas only one WT mouse (gray cycle) died during this time. Magnification, ×400.

Figure 2.

Infiltration of interstitial CD4⁺ and CD8⁺ T cells and macrophages and glomerular CD4⁺ and macrophages increased in CCR7KO mice with NTS. The infiltration of inflammatory cells was analyzed by immunohistochemistry. Kidneys of CCR7KO (; n = 21) and WT mice (■; n = 18) were stained for CD4⁺ and CD8⁺ T cells and macrophages 7 d after induction of NTS. (A through C) The interstitial accumulation of CD4⁺ and CD8⁺ T cells (A) and F4/80⁺ cells (B) as well as the intraglomerular infiltration of CD4⁺ T cells and F4/80⁺ cells (C) was significantly increased in CCR7KO mice. *P < 0.05. (D) Kidney-infiltrating CD4⁺ T cells were further characterized by flow cytometry analysis. Seven days after induction of NTS, significantly increased numbers of CD4⁺CD69⁺ effector T cells infiltrated kidneys of CCR7KO as compared with WT controls (n = 5 per group). Representative dot plots are shown.

The B Cell Response Is Increased in CCR7KO Mice

It was shown that in CCR7KO mice, a significant part of B cells retain the ability to migrate into the lymph nodes and the white pulp of the spleen.³ Accordingly, we detected a significant increase in the titers of mouse anti-rabbit IgG in the serum of CCR7KO mice as compared with respective control animals (Figure 3). This difference was also observed through all IgG isotypes (Figure 3); however, the deposition of the autologous mouse anti-rabbit IgG and the heterologous rabbit anti-mouse IgG on the GBM was comparable in both groups (Table 1).

Figure 3.

Anti-rabbit IgG production is significantly increased in CCR7KO mice with NTS. Total mouse anti-rabbit IgG as well as its subclasses (IgG1, IgG2b, and IgG3) were evaluated in CCR7KO (; n = 21) and WT mice (■; n = 18) on day 7 after NTS induction. Concentrations of mouse anti-rabbit IgG were significantly higher in CCR7KO mice. *P < 0.05.

Table 1.

Deposition of autologous and heterologous antibodies in glomeruli of mice with NTS

Deposition of	WT	CCR7KO
Mouse anti-rabbit IgG	1:6400	1:6400
Mouse anti-rabbit IgG1	1:1600	1:1600
Mouse anti-rabbit IgG2b	1:1600	1:1600
Mouse anti-rabbit IgG3	1:800	1:800
Rabbit anti-mouse IgG	1:3200	1:3200

NTS was induced in CCR7KO and WT mice. On day 7, the kidneys were removed and processed histologically. The glomerular deposition of mouse anti-rabbit total IgG and its isotypes as well as the rabbit anti-mouse IgG was measured because titers required their immunohistochemical detection by evaluating 20 glomeruli per kidney from six mice per group. The titers were identical in all samples, indicating no difference in antibody deposition on GBM in CCR7KO and WT mice.

Cytokine Transcripts in Kidneys and Lymph Nodes of CCR7KO versus WT Mice

We used real-time PCR to compare the Th1, Th2, and Th17 cytokine expression profiles in kidneys and the draining lymph nodes of CCR7KO and WT mice with NTS. We detected a significant increase of IL-10, IFN-γ, and IL-21 mRNA in kidneys of CCR7KO mice versus WT mice (Figure 4A) but no differences in the transcription of IL-6, TNF-α, and TGF-β (Figure 4A). IL-17 was found to be expressed in kidneys of CCR7KO, whereas it was absent in the majority of WT kidneys 7 d after induction of disease (data not shown).

Figure 4.

The cytokine expression pattern in kidneys and lymph nodes varies between CCR7KO and WT mice. (A and B) The expression of IL-10, TGF-β, IL-6, TNF-α, IFN-γ, IL-17, and IL-21 mRNA was evaluated by real-time PCR on total RNA from kidneys (A) or lymph nodes (B) of CCR7KO (; n = 21) or WT (■; n = 18) mice 7 d after anti-GBM injection. The data are shown as fold increase in comparison with healthy tissue of WT mice. The IL-17 expression was evaluated only in lymph nodes, because healthy controls do not express IL-17 in kidneys. CCR7KO mice express significantly more mRNA of IL-6, IL-10, and IL-21 in kidneys, whereas mRNA of IL-6, IFN-γ, and IL-17 was significantly increased in the lymph nodes of CCR7KO mice as compared with WT controls. *P < 0.05.

The proinflammatory cytokines IL-6, IFN-γ, and IL-17 were found to be significantly increased in the draining lymph nodes of CCR7KO as compared with WT mice 7 d after the induction of NTS (Figure 4B). No significant difference between the two groups was detected in the levels of IL-10, TGF-β, TNF-α, and IL-21 mRNA (Figure 4B).

Tregs Infiltrate Kidneys of CCR7KO Mice

We showed recently that in NTS, Tregs exert their suppressive effects in the draining lymph nodes and are undetectable in kidneys during the entire course of the disease.⁶ Surprisingly, we observed FoxP3⁺ cells in kidneys of CCR7KO mice but not in WT mice. On day 7 after induction of NTS, FoxP3 was detected on transcriptional level by real-time PCR (Figure 5A) and on protein level by immunohistochemistry (Figure 5, B and C), Western blot (Figure 5D), and flow cytometry analysis (Figure 5E). The infiltrating Tregs were mainly found close to the blood vessels (Figure 5C).

Figure 5.

Significantly increased numbers of FoxP3⁺ Tregs infiltrate kidneys of CCR7KO mice. (A) FoxP3 mRNA expression was evaluated by performing real-time PCR of whole-kidney extracts of CCR7KO (; n = 21) and WT (■; n = 18) mice 7 d after induction of NTS. Data are shown as fold increase as compared with healthy kidneys. *P < 0.05. (B) Immunohistochemical detection of FoxP3 was performed in kidney sections of CCR7KO (; n = 21) and WT mice (■; n = 18) 7 d after induction of NTS. The average number of kidney-infiltrating FoxP3⁺ cells in six high-power fields is given. *P < 0.05. (C) Localization of Tregs in kidneys was evaluated by immunohistochemistry for FoxP3. The representative micrographs of WT and CCR7KO mice show the FoxP3⁺ cells (black arrows) accumulating in CCR7KO kidneys around a blood vessel (V) and some found around the glomerulus (G). The control isotype stainings of CCR7KO kidneys showed no immunoreactivity. (D) For the evaluation of the FoxP3 protein expression, Western blotting of kidney extracts of WT and CCR7KO mice was performed. The results of one of three experiment performed on samples from four mice are shown. FoxP3 was found to be increased in kidneys of CCR7KO mice 7 d after induction of NTS. (E) Seven days after induction of NTS, single-cell suspensions of kidneys were performed, and cells were stained for CD4⁺CD25⁺FoxP3⁺ and analyzed by flow cytometry. Significantly increased CD4⁺CD25⁺FoxP3⁺ cells were detected in kidneys of CCR7KO as compared with WT controls (n = 5 per group). Representative dot plots are shown. Magnification, ×400.

Tregs Are Decreased in Secondary Lymphoid Organs of CCR7KO Mice

Healthy CCR7KO mice have significantly reduced numbers of Tregs in their peripheral lymph nodes.⁷ Tregs exert their suppressive activity in NTS in the regional draining lymph nodes⁶; therefore, we quantified by flow cytometry FoxP3⁺ cells in secondary lymphoid organs 7 d after the induction of the disease. Significantly decreased numbers of CD4⁺FoxP3⁺ cells were detected in spleens and lymph nodes of CCR7KO as compared with WT controls (Figure 6, A and B).

Figure 6.

Numbers of FoxP3⁺ cells were significantly decreased in lymph nodes and spleen of CCR7KO mice with NTS. (A) The number of CD4⁺FoxP3⁺ cells was evaluated by performing flow cytometry of lymph node and spleen single-cell suspensions from CCR7KO (; n = 4) and WT mice (■; n = 4) 7 d after induction of NTS. Significantly decreased numbers of CD4⁺FoxP3⁺ cells were detected in the secondary lymphoid organs of CCR7KO mice. *P < 0.05. (B) Representative dot plots of CD4⁺FoxP3⁺ stainings of spleen and lymph node suspensions of CCR7KO and WT mice 7 d after induction of NTS are shown.

Adoptive Transfer of WT Tregs Alleviates the NTS in CCR7KO

To evaluate whether WT Tregs can ameliorate the NTS in CCR7KO mice, CD4⁺CD25⁺ we isolated and adoptively transferred Tregs into CCR7KO mice immediately after immunization of mice. Seven days after the induction of NTS CCR7KO mice that received WT Tregs had a significantly decreased albuminuria (Figure 7A) and significantly improved histomorphologic disease parameters (Figure 7, B and C) as compared with CCR7KO mice. Both of these disease indices were comparable to those seen in WT mice (Figure 7, A through C). Total kidney and intraglomerular infiltrates of CD4⁺ T cells, CD8⁺ T cells, and F4/80⁺ cells were significantly diminished in CCR7KO mice after transfer of WT Tregs, but only F4/80⁺ cell infiltration was reduced to the numbers seen in WT mice (Table 2). In CCR7KO mice reconstituted with WT Tregs, the serum levels of total mouse anti-rabbit IgG (Figure 7D) and its subclasses (Table 2) remained higher than in WT mice and as high as in nonreconstituted CCR7KO mice. No difference in the deposition of autologous mouse-anti rabbit IgG and heterologous rabbit anti-mouse IgG on the GBM was detected (data not shown). Recently, we provided evidence that Tregs exert their suppressive activity in the draining lymph nodes rather than in kidneys.⁶ To evaluate the migratory behavior of the transferred WT Tregs in the CCR7KO mice, we performed reverse transcriptase–PCR (RT-PCR) for the detection in spleen, lymph nodes, and kidneys of the part of exon 3 of CCR7 known to be deleted in CCR7KO mice.³ Using this assay, the adoptively transferred Tregs were detected in lymph nodes and spleens but not in kidneys of the recipient CCR7KO mice (Figure 7E).

Figure 7.

The adoptive transfer of WT Tregs rescues the NTS in CCR7KO mice. (A) Albumin and creatinine were evaluated on day 7 after induction of NTS. Urine albumin excretion (in μg) was determined and expressed per milligram of urinary creatinine to standardize for the urine concentration. CCR7KO mice with NTS (; n = 6) displayed significantly increased albuminuria as compared with their WT controls (■; n = 6). **P < 0.025. When WT Tregs were adoptively transferred into CCR7KO mice (■; n = 5), the albuminuria was significantly less, as compared with nontransferred CCR7KO mice. #P < 0.025. (B) Kidney sections 7 d after induction of NTS were stained with PAS, and the presence of PAS-positive material in glomeruli was scored. Kidneys of CCR7KO mice with NTS (; n = 6) contained significantly more PAS-positive deposits as compared with their WT controls (■; n = 6). **P < 0.025. When WT Tregs were adoptively transferred into CCR7KO mice (■; n = 5), the PAS score was significantly less, as compared with nontransferred CCR7KO mice. #P < 0.025. (C) Representative PAS-stained sections from day 7 are shown. CCR7KO mice that received WT Tregs display glomerular changes comparable to the WT mice. (D) Mouse anti-rabbit IgG and their subclasses were evaluated in WT (■; n = 6), CCR7KO (; n = 6), and CCR7KO that received WT Tregs (■; n = 6) on day 7 after anti-GBM injection. Concentrations of mouse anti-rabbit IgG were significantly increased in CCR7KO mice (**P < 0.025) and CCR7KO mice transferred with WT Tregs (#P < 0.025) as compared with WT mice. (E) Five CCR7KO mice received WT Tregs on the day of immunization. Seven days after induction of NTS kidneys, lymph nodes and spleen were harvested from each mouse and CCR7 was detected by PCR (lanes 1 through 5). As a negative and positive control, cDNA of a lymph node of a CCR7KO mouse (−) and of a WT mouse (+), respectively, was used. Equal cDNA template was confirmed by RT-PCR for β-actin (data not shown). Magnification, ×400.

Table 2.

Mouse anti-rabbit IgG titers and kidney infiltrates of CD4⁺ T cells, CD8⁺ T cells, and F4/80⁺ cells in mice with NTS

Parameter	WT	CCR7KO	CCR7KO + WT Tregs
Mouse anti-rabbit IgG1	0.40 ± 0.08	0.83 ± 0.02a	0.77 ± 0.04
Mouse anti-rabbit IgG2b	0.20 ± 0.04	0.62 ± 0.06a	0.56 ± 0.08
Mouse anti-rabbit IgG3	0.09 ± 0.01	0.51 ± 0.09a	0.31 ± 0.07
Total kidney CD4+ T cells	5.00 ± 0.53	35.50 ± 3.08a	16.17 ± 1.88b
Intraglomerular CD4+ T cells	0.67 ± 0.30	11.17 ± 0.93a	4.83 ± 0.50b
Total kidney CD8+ T cells	2.67 ± 0.56	28.83 ± 4.17a	18.67 ± 2.40
Kidney F4/80+ cell score	1.08 ± 0.14	2.58 ± 0.14a	1.29 ± 0.17b
Intraglomerular F4/80+ cells	9.00 ± 1.22	24.33 ± 3.07a	6.00 ± 1.25b

Mouse anti-rabbit serum titers of total IgG and its isotypes (IgG1, IgG2b, and IgG3) and kidney infiltration by CD4⁺ T cells, CD8⁺ T cells, and F4/80⁺ cells were compared in CCR7KO mice, WT mice, and CCR7KO mice after transfer of WT Tregs (n = 6 per group) 7 d after induction of NTS. Titers are given as means ± SEM of absorption values of 1:400 dilution of serum read at OD₄₅₀. Kidney F4/80⁺ cell score was obtained as described in the Concise Methods Section. Total kidney leukocyte numbers are given per six high-power fields and those of intraglomerular leukocytes per 50 glomeruli (all mean ± SEM).

^aP < 0.025 CCR7KO versus WT.

^bP < 0.025 CCR7KO versus CCR7KO + WT Tregs.

CCR7KO Tregs Fail to Inhibit NTS

To investigate further the requirement of CCR7 expression by Tregs for their aptitude to regulate NTS, we isolated CD4⁺CD25⁺ Tregs from CCR7KO and WT mice and adoptively transferred them into WT mice immediately after immunization of mice. The control group received an injection of WT CD4⁺CD25⁻ cells. Whereas the transfer of WT Tregs significantly decreased albuminuria 7 and 14 d after induction of NTS as compared with controls, the transfer of CCR7KO Tregs had no effect (Figure 8A). In line, we found that PAS-positive deposits significantly decreased in glomeruli of mice treated with WT Tregs as compared with control mice (Figure 8B), whereas the amount of PAS-positive deposits in glomeruli of mice treated with CCR7KO Treg was not different from that in control mice 14 d after induction of NTS (Figure 8B). WT Treg transfer significantly decreased the infiltration of CD4⁺ T cells, CD8⁺ T cells, and F4/80⁺ cells into the kidney, as compared with control mice and mice receiving CCR7KO Tregs (Table 3). Of note, no difference between the groups was detected in the autologous antibody titers in the serum or autologous and heterologous IgG deposits on the GBM (data not shown).

Figure 8.

CCR7KO Tregs fail to alleviate NTS. (A) Urine albumin and creatinine were evaluated on days 7 and 14 after induction of NTS. Urine albumin excretion (in μg) was determined and expressed per milligram of urinary creatinine to standardize for the urine concentration. WT mice adoptively transferred with WT Tregs (; n = 5) displayed significantly decreased albuminuria as compared with their WT controls (■; n = 5). **P < 0.025. In contrast, albuminuria in WT mice that received CCR7KO Tregs (■; n = 5) did not decrease significantly. (B) Kidney sections 14 d after induction of NTS were stained with PAS, and PAS-positive deposits in glomeruli were scored. Mice that received WT Tregs (; n = 5) had significantly fewer PAS-positive deposits as compared with control mice (■; n = 5). **P < 0.025. The adoptive transfer of CCR7KO Tregs (■; n = 5) did not significantly change the PAS score.

Table 3.

Kidney infiltrates of CD4⁺ T cells, CD8⁺ T cells, and F4/80⁺ cells after transfer of CCR7KO Tregs

Parameter	WT Tregs	CCR7KO Tregs	Control Cells
Total kidney CD4+ T cells	7.50 ± 1.35a,b	21.75 ± 2.77	25.00 ± 1.84
Intraglomerular CD4+ T cells	4.40 ± 1.19a,b	11.20 ± 0.91	10.80 ± 1.45
Total kidney CD8+ T cells	11.80 ± 2.94b	19.60 ± 1.80	24.20 ± 2.34
Kidney F4/80+ cell score	0.90 ± 0.15a,b	1.80 ± 0.27	1.90 ± 0.30
Intraglomerular F4/80+ cells	10.00 ± 2.33a,b	26.40 ± 3.73	29.60 ± 3.01

Infiltration of CD4⁺ T cells, CD8⁺ T cells, and F4/80⁺ cells was evaluated in WT mice with NTS receiving WT Tregs, CCR7KO Tregs, or control cells (n = 5 per group) 2 wk after induction of disease. Kidney F4/80⁺ cell score was obtained as described in the Concise Methods section. Total kidney leukocyte numbers are given per six high-power fields and those of intraglomerular leukocytes per 50 glomeruli. Data are means ± SEM.

^aP < 0.025 WT Tregs versus CCR7KO Tregs.

^bP < 0.025 WT Tregs versus control cells.

Discussion

The interaction of CCL19 and CCL21 with CCR7 expressed by T cells and DC is essential to guide these cells into the T cell regions of secondary lymphoid organs, where, according to a classical paradigm, antigen presentation and initiation of the primary immune response takes place.^1,3 Also, the CCR7-mediated homing of Tregs into the lymph nodes was shown to be necessary for their antigen-driven expansion and suppressive function in a subset of immune reactions in vivo.⁷ Thus, CCR7 is directly involved in both the induction of immunity and its silencing.¹ In the experimental model of accelerated NTS, on one hand, T cell–mediated immune responses drive the development of disease,⁸ and, on the other hand, regulatory mechanisms that take place in the regional lymph nodes suppress the pathologic changes.^6,9 We used CCR7KO mice to investigate whether this chemokine receptor is involved primarily in the initiation or the regulation of NTS. We found that CCR7KO mice are susceptible to experimental NTS. Moreover, they develop a more severe disease than their WT counterparts. Thus, classical CCR7-mediated pathways of induction of primary immune responses can be bypassed in the absence of this receptor. The initial study of the CCR7KO phenotype revealed scrambled architecture of the secondary lymphoid organs and, as a result, an apparent failure to initiate efficient primary immune responses.³ In light of those findings, our data on an increased susceptibility of CCR7KO mice to NTS are unexpected; however, our results are in accordance with more recent publications that noted that, despite the lack of normal lymph node architecture, functional immune responses still can develop in the absence of CCR7 or its ligands.^7,10,11 Currently, we cannot answer how and where primary immunity that leads to an overt NTS can arise in CCR7KO mice. It is likely that productive encounters between sparse DCs and naive T cells nonetheless take place in disorganized lymph nodes or spleen of CCR7KO. Alternatively, it is possible that in CCR7KO mice, immune responses are initiated outside the secondary lymphoid organs, in the case of NTS, in the affected kidneys. It has been demonstrated that, contrary to the prevalent view, naive T cells regularly enter peripheral sites, including kidneys, where they can be primed by the antigens presented by the resident cells.¹² The homing of naive T cells into peripheral tissues does not depend on CCR7 or, in fact, any other chemokine receptor because it is pertussis toxin insensitive,¹² yet the departure of T cells from peripheral tissues via lymphatic vessels was shown to involve CCR7.^13,14 Thus, it is possible that in the absence of CCR7, T cells continue to home to kidneys but cannot efficiently leave, leading to their accumulation within the parenchyma. This mechanism may contribute to the exaggerated course of NTS in CCR7KO mice and account, in part, for the increased number of T cells detected in the diseased kidneys of CCR7KO mice as compared with WT controls.

Our data indicate, however, that an additional mechanism is responsible for more severe NTS in CCR7KO mice. We found that the adoptive transfer of WT Tregs into CCR7KO mice at the time of immunization dramatically alleviated the disease and rendered its indices comparable to those in WT controls. Additional direct comparison of the therapeutic effects in NTS of transferred WT and CCR7KO Tregs revealed that the latter lack in vivo regulatory activity. These data suggest that in CCR7KO mice, the inadequate function of Tregs is responsible for the exacerbated course of NTS. It was shown that despite intact in vitro suppressive activity, CCR7KO Tregs suppress inefficiently in vivo,^7,15,16 because they cannot home to the lymph nodes, a key site of their activity.^7,15–17 Accordingly, we found dramatically reduced numbers of Tregs in the secondary lymphoid organs of CCR7KO mice as compared with WT both before and after the induction of NTS. The development and regulation of immunity depends on equilibrium of specialized cellular responses with characteristic cytokine profiles. According to a current paradigm, the immunoinflammatory axis entails Th1 and Th17 cells and is counterbalanced by regulatory and allergic axes driven by Tregs and Th2 cells, respectively.¹⁸ In CCR7KO mice, the lack of Tregs in the lymph nodes may allow for increased expansion and activation of pathogenic Th1 and Th17 cells. Next, these cells migrate into kidneys, where they induce foci of autoimmunity and together with other subsequently recruited inflammatory cells ultimately lead to the destruction of kidney parenchyma.⁴ This scenario is supported by significant increases in the Th1- and Th17-specific cytokines in the lymph nodes and kidneys of diseased CCR7KO mice as well as increased numbers of pathogenic CD4⁺ T cells and macrophages in their kidneys. Peculiar, Tregs constituted a part of T cell infiltrates in the diseased kidneys of CCR7KO mice. Conversely, neither in the current nor in a previous study⁶ could we detect Tregs in kidneys of WT mice with NTS. The adoptive transfer of WT Tregs into CCR7KO mice rendered NTS disease indices comparable to those in WT mice but failed to reduce the size of T cell infiltrates in kidneys of CCR7KO mice. This suggests that WT Tregs transferred into CCR7KO mice cannot influence the homing properties of the endogenous effector T cells but effectively interfere with their pathogenicity. It is unclear which molecular pathogenic pathways in NTS are influenced by adoptively transferred Tregs.

On the basis of their expression of chemokine receptors and adhesion molecules, Tregs comprise separate functional subpopulations.¹⁹ The naive-like CCR7⁺ Tregs home to the lymph nodes, where, upon antigen stimulation, they expand and suppress the proliferation and activation of the effector T cells.^7,15 The CD103⁺ effector-like Tregs express receptors for inflammatory chemokines and home to the peripheral inflammatory sites, where they can suppress the ongoing immune reactions.^15,19,20 A part of Tregs express both CCR7 and inflammatory chemokine receptors. The effector-like lineage of Tregs was shown to develop in the absence of CCR7 but, without additional in vitro stimulation, cannot efficiently regulate.^1,7,15 Accordingly, in our study, Tregs infiltrating kidneys of CCR7KO mice seemed to be ineffective in limiting the ongoing autoimmunity. The WT Tregs adoptively transferred into CCR7KO mice were detected in the secondary lymphoid organs but not kidneys and were highly efficient in limiting the NTS. Conversely, the adoptively transferred CCR7KO Tregs do not home to lymph nodes and fail to inhibit NTS. It is possible that the CCR7-mediated homing of Tregs to the lymph nodes not only is required to suppress at this site the initiation of the immune response^6,7 but also enables them to acquire putative armament for their subsequent suppressive activity at the peripheral sites. This is contradicted by the lack of Tregs in kidneys either in “regulated” NTS in WT mice or after the curative adoptive transfer of WT Tregs into CCR7KO hosts. These findings support the notion that unlike, for example, inflammatory bowel disease,²⁰ the NTS is regulated not at the periphery but primarily in the lymph nodes. Because CCR7 guides thymocyte migration within the thymus,²¹ it is also possible that in the absence of this receptor, Tregs fail to mature and lack yet-unknown characteristics required for their suppressive activity in vivo but not in vitro. Also, we cannot rule out that the lack of CCR7 on DCs may contribute to the more severe phenotype of NTS in CCR7KO mice. The role of DCs in glomerulonephritis is controversial. On the one hand, DCs have been shown to attenuate NTS by promoting the infiltration of IL-10–producing cells into the kidneys.^22,23 On the other hand, lymph node DCs as well as resident DCs in the tubulointerstitium were shown to activate pathogenic T cells in an experimental glomerulonephritis in transgenic mice that selectively expressed ovalbumin and hen egg lysozyme in glomerular podocytes.²⁴

A limitation of our study might be that we used, because of availability, mice on BALB/c background. BALB/c mice are known to develop decreased disease indexes and infiltration of T cells and macrophages as compared with the usually used C57BL/6 mice.²⁵ This is in accordance with our findings, but when the two-fold dose of antiserum is used, BALB/c mice develop NTS with kidney infiltrates consisting of T cells and macrophages comparable to C57BL/6 mice.

Resident glomerular mesangial cells have also been shown to express CCR7, which seems to be is involved in their migration and proliferation as well as glomerular homeostasis.²⁶ Thus, the lack of CCR7 on mesangial cells may also have an impact on glomerular pathology in our model of NTS. Additional studies including electron microscopy evaluations of WT and CCR7KO kidneys after induction of NTS as well as evaluating NTS in cell-specific KO animals will be needed to dissect the importance of CCR7 on various cell types.

Of note, CCR7KO mice with NTS developed higher titers of mouse anti-rabbit IgG antibodies than their WT counterparts. Nevertheless, the amount of antibody deposited on the GBM did not differ between the groups. Despite the known ability of Tregs to inhibit B cell responses,²⁷ the elevated antibody titers in CCR7 mice were unlikely to be the consequence of a Treg defect. The adoptive transfer of WT Tregs that “cured” NTS did not reduce the elevated antibody titers. These findings are in line with previous observations showing that B cell responses are not essential for the induction of NTS. For example, the B cell–deficient μ-chain KO mice are susceptible to NTS.²⁸

CCR7KO mice were shown to develop spontaneous manifestations of autoimmunity including lymphocyte infiltrates in several organs, circulating autoantibodies against tissue antigens, and IgG deposition in the glomeruli.²⁹ In contrast, we found no signs of kidney disease in nonimmunized CCR7KO mice. This discrepancy may be explained by the use in our study of only young mice before they could develop the disease spontaneously.

Taken together, we provide evidence that experimental NTS takes an exaggerated course in CCR7KO mice. This is primarily because Tregs fail in the absence of CCR7 to migrate and exert their suppressive activity in the lymph nodes. Instead, Tregs home to diseased kidneys, where they cannot control the autoimmune processes. Together with our previous findings, this clearly demonstrates that Tregs regulate the NTS in the draining regional lymph nodes.

Concise Methods

Induction of Accelerated NTS

CCR7KO mice were backcrossed with BALB/c mice for 12 generations.³ The resulting strain is referred to as CCR7KO. Control WT were obtained from Charles River Laboratories (Sulzfeld, Germany). Eight- to 12-wk-old female mice were used in all studies. Accelerated NTS was induced as described previously.⁵ In brief, mice were preimmunized subcutaneously with 100 μl of 2 mg/ml rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA) dissolved in incomplete Freund's adjuvant (Sigma, St. Louis, MO) and nonviable desiccated Mycobacterium tuberculosis H37a (Difco Laboratories, Detroit, MI). After 3 d, heat-inactivated rabbit anti-mouse GBM antiserum was injected via the tail vein. All animal experiments were approved by the institutional and municipal animal welfare committees and were performed according to the laws of Austria (GZ 66.011/87-BrGT/2005). Experiments evaluating NTS in CCR7KO, WT, and CCR7KO receiving WT Tregs were performed at Novartis Institutes for BioMedical Research (Vienna, Austria) and, in part, Medical University Innsbruck (Innsbruck, Austria). The experiments evaluating the in vivo functionality of CCR7KO Treg versus WT Tregs in WT recipients were performed exclusively at the Medical University Innsbruck.

Detection of Urinary Albumin and Creatinine

Urinary albumin was determined by a double-sandwich ELISA (Abcam, Cambridge, MA) as reported previously.⁵ Urinary creatinine was quantified spectrophotometrically using a picric acid–based method (Sigma).

Histo- and Immunomorphologic Evaluation of Renal Pathology

Formalin-fixed renal tissue was embedded in paraffin, cut in 4-μm sections, and stained with PAS. The three-layer immunoperoxidase staining of frozen tissue sections (4 μm) was used for the detection of macrophages and T cell subpopulations in the kidney sections.⁵ Macrophages were stained with a rat anti-mouse antibody (clone F4/80; Serotec, Oxford, UK). A semiquantitative scoring system was performed as follows: 0 = 0 to 4 cells stained positive, 1+ = 5 to 10 cells, 2+ = 10 to 50 cells, 3+ = 50 to 200 cells, and 4+ = >200 cells stained positive per low-power field. For the detection of CD4⁺ T cells, CD8⁺ T cells, and FoxP3⁺ Tregs, we used rat anti-mouse mAb clone YTS191.1 (Serotec), clone KT15 (Serotec), and clone FJK-16s (eBiosciences, San Diego, CA), respectively. Quantification of T cells was done by counting the number of cells in six adjacent high-power fields of renal cortex and medulla. Furthermore, CD4⁺ T cells and F4/80⁺ cells were counted intraglomerularly in at least 50 glomeruli per cross-section. For the detection of heterologous and autologous IgG deposition, sections were stained for direct immunofluorescence with a FITC-conjugated goat anti-rabbit IgG or anti-mouse IgG (Jackson ImmunoResearch Laboratories). For detection of deposited mouse IgG isotypes, horseradish peroxidase (HRP)-conjugated goat anti-mouse IgG1, IgG2b, and IgG3 antibodies (all from Jackson ImmunoResearch Laboratories) were used.

Detection of Circulating Mouse Anti-rabbit Total IgG and Its Isotypes

For detection of circulating mouse anti-rabbit IgG, 96-well plates (Greiner, Kremsmuenster, Austria) were coated with 100 μg/ml rabbit IgG (Jackson ImmunoResearch Laboratories) in carbonate/bicarbonate buffer (pH 9.5). After blockade with 1% BSA, plates were incubated with serial-doubling dilutions of mouse serum. Bound mouse IgG was detected with HRP-conjugated goat anti-mouse IgG (Dako, Glostrup, Denmark). In addition, Ig isotypes were evaluated by using HRP-conjugated goat anti-mouse IgG1, IgG2b, and IgG3 antibodies (all from Jackson ImmunoResearch Laboratories).

Real-Time RT-PCR

Total RNA was isolated using TRIzol (Sigma) according to a standard protocol. Thereafter, 2 μg of total RNA was reverse-transcribed using Superscript III Transcription Kit (Invitrogen, Carlsbad, CA) and random primers (Roche, Basel, Switzerland). Real-time PCR was performed on an ABI Prism 7700 (Applied Biosystems, Foster City, CA). For linear amplification of TNF-α, FoxP3, and β-actin (reference gene), SYBR Green Master Mix (Invitrogen) and the primers shown in Table 4 were used. For quantification of IFN-γ, IL-10, IL-6, IL-17, and IL-21, TaqMan Mastermix (Applied Biosystems) and the gene expression assays Mm00801778_m1, Mm00439616_m1, Mm00446190_m1, Mm00439619_m1, and Mm00517640_m1 (Applied Biosystems) were used.

Table 4.

Primer pairs

Gene	Forward Primer	Reverse Primer
TNF-α	5′-GAA CTG GCA GAA GAG GCA CT-3′	5′-AGG GTC TGG GCC ATA GAA CT-3′
TGF-β	5′-ATT CAG CGC TCA CTG CTC TT-3′	5′-CGC ACA CAG CAG TTC TTC TC-3′
FoxP3	5′-TCT TGC CAA GCT GGA AGA CT-3′	5′-AGC TGA TGC ATG AAG TGT GG-3′
β-actin	5′-GAA GTG TGA CGT TGA CAT CCG-3′	5′-TGC TGA TCC ACA TCT GCT GGA-3′
CCR7	5′-TAA GGG CAT CTT TGG CAT CT-3′	5′-CAG TGA GCA TCT CAG CGT GT-3′

Measurement of FoxP3 by Western Blotting

Kidney samples were homogenized and lysed on ice by using Triton lysis buffer (37.6 mM KCl, 24.8 mM Tris base, and 1% Triton X-100) supplemented with 1% protease inhibitor cocktail (Sigma). Twenty-five micrograms of each sample was analyzed on 10% SDS-PAGE and blotted on hydrophobic polyvinylidene difluoride membrane (Amersham Biosciences Corp., Piscataway, NJ). Rabbit anti-FoxP3 antibody (clone eBio7979; eBiosciences) was used at a final concentration of 1 μg/ml. As a secondary antibody, HRP-conjugated goat anti-rabbit IgG (Dako) at a dilution of 1:2000 was used. After chemiluminescent reaction using an electrochemiluminescence kit (ECL) Western blot reagent (Amersham Biosciences Corp.), the blots were exposed to Hyperfilm ECL (Amersham Biosciences Corp.).

Flow Cytometry of FoxP3⁺ Cells

Cell suspensions from lymph nodes, spleens, and kidneys were stained for FoxP3⁺ Tregs using co-staining with CD4 antibody (BD Biosciences, San Diego, CA) and a FoxP3⁺ antibody (eBiosciences) according to the protocol provided by eBiosciences. Samples were analyzed on a FACSCalibur (BD Biosciences).

Isolation and Transfer of CD4⁺CD25⁺ Tregs

CD4⁺CD25⁺ Tregs were isolated from minced spleens obtained from BALB/c and CCR7KO mice using magnetic bead separation (CD4⁺CD25⁺ regulatory T cell kit; Miltenyi Biotec GmbH, Bergisch Gladbach, Germany). The purity of both populations was controlled by flow cytometric analysis and routinely reached >90%. CCR7KO or WT mice received intravenous 4 × 10⁶ CD4⁺CD25⁺ Tregs on the day of immunization.

Detection of Transferred Tregs in CCR7KO Mice

Tregs from WT controls transferred into CCR7KO were detected in spleen, lymph nodes, and kidney by RT-PCR for CCR7 mRNA. RNA isolation and reverse transcription were performed as described above. PCR was performed with the HotStarTaq polymerase (Qiagen, Hilden, Germany) and primers, which were designed to amplify a part of exon 3 that is known to be deleted in CCR7KO.³ We used the primer pair shown in Table 4. β-Actin PCR was performed as a loading control.

Statistical Analysis

When comparing two groups, the nonparametric Mann-Whitney U test was performed, and P < 0.05 was considered significant. When comparing three groups, we performed the Kruskal-Wallis test. When significances were detected, groups were compared by Mann-Whitney U test. The level of significance was corrected to the number of groups, and P < 0.025 was considered significant. Significances in survival were evaluated by using the log-rank test. All statistical analyses were done using SPSS 13.0.1 for Windows (SPSS, Inc., Chicago, IL).

Disclosures

At the time of the study, M.P. and A.R. were employed by Novartis Institutes for BioMedical Research (Vienna, Austria).