Alloantigen-presenting Pdc Mediate Tolerance To Vascularized Grafts

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

ARTICLES

Alloantigen-presenting plasmacytoid dendritic cells mediate tolerance to vascularized grafts Jordi C Ochando1, Chiho Homma1, Yu Yang1, Andres Hidalgo2, Alexandre Garin3, Frank Tacke1, Veronique Angeli1, Yansui Li4, Peter Boros4, Yaozhong Ding1, Rolf Jessberger1,5, Giorgio Trinchieri6, Sergio A Lira4, Gwendalyn J Randolph1 & Jonathan S Bromberg1,4,7 The induction of alloantigen-specific unresponsiveness remains an elusive goal in organ transplantation. Here we identify plasmacytoid dendritic cells (pDCs) as phagocytic antigen-presenting cells essential for tolerance to vascularized cardiac allografts. Tolerizing pDCs acquired alloantigen in the allograft and then moved through the blood to home to peripheral lymph nodes. In the lymph node, alloantigen-presenting pDCs induced the generation of CCR4+CD4+CD25+Foxp3+ regulatory T cells (Treg cells). Depletion of pDCs or prevention of pDC lymph node homing inhibited peripheral Treg cell development and tolerance induction, whereas adoptive transfer of tolerized pDCs induced Treg cell development and prolonged graft survival. Thus, alloantigen-presenting pDCs home to the lymph nodes in tolerogenic conditions, where they mediate alloantigen-specific Treg cell development and allograft tolerance.

The main unsolved goal in clinical transplantation is the establishment of tolerance to transplant antigens. Although much has been learned over the past decade about the various cell populations involved in transplant rejection1–4, the precise cell subpopulations involved in the induction and maintenance of transplant tolerance remain unidentified. That is due in part to the difficulty in defining the precise interactions between diverse antigen-presenting cells (APCs) and T cell subsets with disparate trafficking patterns5. Antigen-specific immune responses are initiated after the acquisition of foreign molecules in the periphery by phagocytic APCs6. After alloantigen uptake, activated APCs undergo a shift in their expression of chemokine receptors and migrate to secondary lymphoid organs7. Entry into the spleen is considered nonspecific, whereas entry into the lymph node requires expression of the L-selectin CD62L and the chemokine receptor CCR7, which bind to peripheral node addressin and the chemokine CCL21, respectively, and allow extravasation across high endothelial venules (HEVs)8. Once in the secondary lymphoid organs, APCs present complexes of foreign peptide and major histocompatibility complex (MHC) molecules on their surfaces to alloantigen-specific CD4+ T cells, and signals arising from these interactions are key factors in the establishment of either immunity or tolerance. Central tolerance is established in the thymus by clonal deletion of potentially self-reactive T cells. In addition, CD4+CD25+Foxp3+ regulatory T cells (Treg cells) help to maintain peripheral tolerance by actively suppressing the activation and population expansion of

self-reactive T cells9. In several transplantation models, Treg cells have been shown to be important in the induction and maintenance of tolerance to foreign molecules10–13, although little is known about the identity of alloantigen-presenting cells that promote their development into antigen-specific suppressor cells or the location in which this process occurs. Plasmacytoid dendritic cells (pDCs) may be important in the in vivo generation of antigen-specific Treg cells, as their precursors have been reported to prolong graft survival14. Along with other published observations15,16, those results suggest that interactions between pDCs and CD4+ T cells in the lymph node may be required for Treg cell development and indefinite graft survival. Here we performed fully allogeneic vascularized cardiac transplants and subjected recipient mice to a well characterized tolerogenic protocol that is dependent on Treg cell development16,17. Using the YAe monoclonal antibody (mAb), which recognizes a donor-derived I-Ed peptide presented in the context of recipient MHC class II I-Ab molecules, we identified pDCs as alloantigen-presenting cells that mediate tolerance to vascularized allografts. Alloantigen-presenting pDCs acquired and processed cell-derived transplantation antigens and, after tolerogenic treatment, migrated ‘preferentially’ to the lymph node, where they induced antigen-specific CD4+CD25+Foxp3+ Treg cell development. These data are consistent with published results that have established that during tolerance, antigen-specific Treg cells must develop in the lymph node16 and provide direct evidence that pDCs can act as ‘professional’ tolerance-inducing alloantigenpresenting cells in vivo.

1Department of Gene and Cell Medicine, 2Hematology Center, 3Immunobiology Center and 4Recanati/Miller Transplantation Institute, Mount Sinai School of Medicine, New York, New York 10029-6574, USA. 5Department of Physiological Chemistry, Dresden University of Technology, Dresden, Germany. 6Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, Bethesda, Maryland 20892-8003, USA. 7Department of Surgery, Mount Sinai School of Medicine, New York, New York 10029-6574, USA. Correspondence should be addressed to J.C.O. ([email protected]) or J.S.B. ([email protected]).

Received 19 December 2005; accepted 10 March 2006; published online 23 April 2006; doi:10.1038/ni1333

652

VOLUME 7

NUMBER 6

JUNE 2006

NATURE IMMUNOLOGY

RESULTS pDCs acquire alloantigen in donor grafts To investigate whether a particular DC subtype is involved in alloantigen uptake and processing in allografts during tolerance, we characterized intracardiac alloantigen-presenting cells between 1 and 10 weeks after vascular cardiac transplantation. The YAe mAb recognizes donor-derived I-Ed peptide presented by recipient I-Ab (ref. 18). Two-color immunohistochemical analysis showed that intragraft YAe+ cells from tolerized recipient mice expressed CD11c, B220, the marker Gr-1 and pDC marker PDCA-1 but not CD19 (Fig. 1a). This surface phenotype is characteristic of mouse pDCs19. Using six-color flow

a

YAe

cytometry, we confirmed that YAe+ cells in both tolerized grafts (Fig. 1b) and rejecting grafts (data not shown) were pDCs. These data indicate that pDCs acquire and process alloantigen in vivo. This result was unexpected, as no reports indicate that pDCs, which are poorly endocytic20, can internalize cells, a process that would probably involve phagocytosis. To assess the phagocytic capacity of pDCs, we cultured pDCs together with fluorescein isothiocyanate (FITC)–labeled latex particles (beads) and monitored engulfment. We used naive splenic B cells and peritoneal macrophages as negative and positive controls, respectively. Immunofluorescence images (Fig. 1c) showed that both pDCs and

b

Merged

YAe 94

CD11c CD11c

5 97

B220 B220 2 96

Gr-1 Gr-1 3 98

PDCA-1

PDCA-1

1 6

CD19

CD19

102 101

0

pDC

pDC (PDCA-1)

101 102 103 104 105 5

10

7

86

104 103 102 101

3

MΦ (F4/80)

101 102 103 104 105

MGG

Phase

FITC-beads

Merged

105 15

83

104 103 102 101

0 101 102 103 104 105

FITC-beads

91

104 103 102 101

% of max.

103

105

8

100 80 60 40 20 0

101 102 103 104 105 5

10

11

104 103 102 101

100 80 60 40 20 0 100 80 60 40 20 0

88 101 102 103 104 105

105 103 102 101

53

Isotype

101 102 103 104 105

46

104

Isotype

101 102 103 104 105

% of max.

7

104

% of max.

105 91

PE-Cy7 anti-FITC PE-Cy7 anti-FITC PE-Cy7 anti-FITC

d

B cell

c

B cell (B220)

93

MΦ

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

ARTICLES

Isotype

101 102 103 104 105

101 102 103 104 105

FITC-beads

PE-Cy7 anti-FITC

Figure 1 Phagocytic pDCs cells take up and process antigen in donor cardiac allografts. Fully mismatched vascularized BALB/c (H-2d) cardiac allografts were transplanted into C57BL/6 (H-2b) mice. Recipients were injected intravenously with 1
107 donor splenocytes on day –7 and 250 mg mAb to CD40L on days –7, –4, 0 and +4 (times relative to transplantation). Untreated rejecting control mice received hamster immunoglobulin in PBS and rejected their grafts within 1 week (n ¼ 6 mice per group). (a,b) Two-color microscopy (a; staining of first column indicated along left margin) or six-color flow cytometry gated on YAe cells (b) of tolerized (1-, 5- and 10-week) and rejecting (1-week) cardiac allografts. Data are representative results from ‘10-week tolerized’ mice. Original magnification,
250. (c,d) In vitro pDC phagocytic assay of the uptake of 1-mm FITC-labeled latex beads (FITC-beads) over 12 h of culture. (c) MayGru¨nwald Giemsa staining and live fluorescent images of splenic B cells, splenic pDCs and peritoneal macrophages (Mf). Original magnification,
1,000. Data are representative results of ten images. (d) Flow cytometry of phagocytic B cells, pDCs and macrophages. Gated FITC-labeled beads were stained with phycoerythrin-indotricarbocyanine-labeled (PE-Cy7) anti-FITC or the isotype control mAb. All experiments were done at least three times. Numbers in quadrants (b,d) indicate percentage of cells positive for markers along margins.

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 6

JUNE 2006

653

ARTICLES

3,000

7.87

2,000 0

3.11

0

0

0.78

Naive pDC

c

3,000

2,000

2,000

105

1,000

1,000

1,000

104 0.5%

101102103104105

0

101102103104105

0

101102103104105

YAe CD11b

B220

Gr-1

CD19

103

102 104

102 104

102 104

102 104

102 104

CD80

CD86

CD40

MHC-II

DEC-205 CD206

102 104

102 104

102 104

102 104

102 104

102 104

LFA-1

ICAM-1

VLA-2

VLA-4

α4β7

CD62L

103

Naive 0.1%

102

0.1%

0.1%

101

101 102 103 104 105

101 102 103 104 105

YAe

e 102 104

104 0.8%

4.4%

102

CD4

4

Expression (relative units × 103)

% of max. % of max.

CD11c

105

Tolerized

101

% of max.

d

YAe+ pDC

Blood

0.71

2,000 0

Naive mDC

101102103104105

4,000

3,000

0.63

0.77

1,000 101102103104105

4,000

3,000

3,000

b

2,000

1,000

4,000

SSC Lymph

3,000 2,000

101102103104105

Naive 4,000

Blood

Blood

4,000

1,000

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

Rejecting

Tolerized 4,000

PDCA-1

a

* *

3

2

*

* *

*

1

0 CCR2 Tolerized 102 104

102 104

102 104 102 104 Tolerized Rejecting

102 104

CCR7 Rejecting

ChemR23 Naive

102 104

Naive

Figure 2 Alloantigen-presenting pDCs circulate systemically through blood. (a) Percentage of total alloantigen-presenting YAe+ cells recovered from blood and lymph of tolerized mice (1-, 5- and 10-week), rejecting mice (1-week) and naive mice. Data are representative results of three independent experiments from ‘10-week tolerized’ mice (n ¼ 3 mice per group). (b) May-Gru¨nwald Giemsa staining (representative of two independent experiments) showing the morphology of freshly sorted blood naive pDCs (PDCA-1+CD11c+B220+), splenic naive mDCs (CD11c+CD11b+B220–) and ‘10-week tolerized’ blood circulating YAe+PDCA-1+ pDCs. Original magnification,
1,000; n ¼ 3 mice per group. (c) Flow cytometry of blood cells from tolerized mice (1- and 10-week). Representative results of three independent experiments from ‘10-week tolerized’ mice show that more than 90% of YAe+ cells express PDCA-1 (n ¼ 3). (d) Flow cytometry of the expression of cell surface molecules on freshly isolated blood YAe+PDCA-1+ cells from tolerized mice (1- and 10-week) and rejecting mice (1-week) compared with that of PDCA-1+ cells from naive mice. MHC-II, MHC class II. Data are representative results of three independent experiments from ‘10-week tolerized’ mice (n ¼ 3 mice per group). (e) Real-time PCR analysis of chemokine receptor expression on blood circulating YAe+PDCA-1+ cells from ‘10-week tolerized’ and ‘1-week rejecting’ mice compared with that of naive pDCs. *, P o 0.01 (one-way ANOVA). Data are from two independent experiments; n ¼ 3 mice per group.

macrophages acquired FITC-labeled beads, as demonstrated by fluorescent localization of FITC-coated beads in the cells. In contrast, FITC-labeled beads were bound only to the external cell membranes of B cells. We further confirmed the phagocytic capacity of pDCs by flow cytometry with a FITC-specific mAb (Fig. 1d), which showed that 88% of FITC+ cultured pDCs and 53% of FITC+ macrophages, but only 8% of FITC+ B cells, stained negative for cell surface FITC. Alloantigen-presenting pDCs circulate in blood To determine if alloantigen uptake and processing by pDCs results in subsequent maturation and egress from the allograft to lymphoid organs, we determined whether YAe+ cells were present in blood and/or lymph (Fig. 2a). YAe+ cells from tolerized recipient mice were present in blood but not in efferent lymph, suggesting that the main route for egress from the allograft is direct re-entry into blood. To further characterize blood-circulating YAe+ cells, we used WrightGiemsa staining of cytospin-sorted DC preparations. Freshly isolated blood pDCs (CD11c+PDCA-1+) were rounded cells with a smooth surface and large nuclei, whereas blood alloantigen-presenting plasmacytoid cells (YAe+PDCA-1+) had slightly indented nuclei and had pseudopods, and splenic myeloid DCs (mDCs) (CD11c+CD11b+B220–) contained more lobulated nuclei and demonstrated classical stellate morphology (Fig. 2b). More detailed

654

examination of the YAe+PDCA-1+ population showed that alloantigen-presenting cells had high expression of mRNA encoding interferon-a and Toll-like receptors 3, 7 and 9 (Supplementary Figure 1 online), consistent with a mature pDC lineage. We next defined the phenotype of the blood YAe+ cells by flow cytometry (Fig. 2c) and found that more than 90% of YAe+ blood cells were PDCA-1+, confirming their plasmacytoid phenotype. Detailed cell surface phenotypic analysis of YAe+ PDCA-1+ cells (Fig. 2d) showed that these pDCs had low expression of CD11c, high expression of Gr-1 and intermediate expression of B220 and lacked expression of CD19 and CD11b. More detailed examination of costimulatory molecule expression showed that YAe+PDCA-1+ cells had very low expression of CD80 and CD86 but high expression of MHC class II (Fig. 2d), indicative of a partially mature phenotype. Furthermore, YAe+PDCA-1+ cells had high expression of the phagocytic receptor CD206, consistent with the observation that pDCs acquired and processed allopeptides in the allograft. Further analysis of the phenotype of YAe+PDCA-1+ cells demonstrated large amounts of the adhesion molecules LFA-1, ICAM-1, VLA-4 and CD62L (Fig. 2d) and the lymph node–homing chemokine receptors CCR7, ChemR23 and CCR2 (Fig. 2e) in tolerized recipient mice. These findings collectively suggest that alloantigen-presenting pDCs migrate ‘preferentially’ to the lymph nodes in tolerized recipient mice but migrate to other lymphoid organs during rejection.

VOLUME 7

NUMBER 6

JUNE 2006

NATURE IMMUNOLOGY

mice bound four to five times more effectively than did YAe+ pDCs from rejecting mice to the same HEVs, consistent with the upregulation of CD62L on YAe+ pDCs from tolerized but not rejecting mice. We next monitored the in vivo localization of alloantigen-presenting YAe+ cells in the lymph node and spleen. In tolerized mice, YAe+ cells were distributed systemically throughout the lymphoid compartment (inguinal, axillary, cervical, paraortic and hepatic lymph nodes) yet were almost completely absent from the spleen (Fig. 3d). In contrast, YAe+ cells were localized mainly in the spleens of rejecting mice and were almost completely absent from the lymph node. Further phenotypic analysis of YAe+ cells in secondary lymphoid organs (Fig. 3e) indicated that more than 90% were pDCs during both rejection and tolerization.

pDCs migrate to lymph node during tolerance We next used laser-capture microdissection of lymph node sections from tolerized mice to characterize the function of HEVs in YAe+ pDC lymph node homing. CCR2 and CCR7 ligands CCL2, CCL7 and CCL21 were upregulated in HEVs during tolerization (Fig. 3a), which may favor lymph node extravasation of CCR2+CCR7+YAe+ pDCs. As CCL21 produced by HEVs mediates rapid binding of integrin to CD62L ligands21, which can be induced during inflammation22, we also investigated whether CD62L ligands were upregulated on HEVs during tolerization. Using a CD62L fusion protein (Fig. 3b), we found upregulation of CD62L ligands on HEVs in tolerized mice, which favors the adhesion of CD62L+ cells to the endothelial barrier. MAdCAM1 is one of the CD62L ligands specifically synthesized by HEVs23 that was upregulated during tolerization (Supplementary Fig. 2 online). To further determine the functional importance of CD62L and CD62L ligand upregulation on circulating YAe+ pDCs and HEVs from tolerized mice, we did a modified in vitro HEVbinding assay24 (Fig. 3c). Blood-derived YAe+ pDCs from tolerized

Localization of pDCs and Treg cells together We next characterized the specific localization of YAe+ cells in the lymph node. During tolerization, YAe+ cells localized specifically in the vicinity of HEVs and T cell zones outside the follicles and germinal

b

*

Tolerized HEV

Naive HEV

Rejecting HEV

EDTA control

*

100 50

1868±233

834±153

Tolerized HEV

+

870±126

0

CCL7

20

c

*

15 10 5

+

Tolerized YAe pDC

Rejecting YAe pDC

9±2

2±1

Merged YAe+ pDC bound/HEV

After

150

CCL2 Expression (relative units ×103)

Before

Expression (relative units)

a

0

CCL19

CCL21 Tolerized Rejecting Naive

Cap

e Axillary LN

Cervical LN

Inguinal LN

Axillary LN

** Tolerized

*

HEV

Naive

Paraortic LN

Hepatic LN

104

1.6%

YAe

0.8%

0.9%

5.0%

103 102

Spleen

0.4%

0.7%

105 104

1.1%

9.1%

0.1%

0.7%

103 102 101

YAe

R

Tolerized

PDCA-1

Spleen

Tolerized

Hepatic LN

T

YAe+ pDC

105

Cervical LN

101

Paraortic LN

Rejecting

Rejecting

PDCA-1

Inguinal LN

10 8 6 4 2 0 10 8 6 4 2 0 10 8 6 4 2 0

N

d

Rejecting

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

ARTICLES

0.3%

0.5%

101 102 103 104 105 101 102 103 104 105

YAe

YAe+

Figure 3 cells home to lymph nodes during tolerance and to the spleen during rejection. (a) Right, real-time PCR analysis of chemokine expression in HEVs isolated from ‘10-week tolerized’, ‘1-week rejecting’ and naive mice by laser-capture microdissection. Left, hematoxylin and eosin staining before (top) and after (middle) excision of tissue by laser-capture microdissection. Cap (bottom), excised tissue. Original magnification,
100. *, P o 0.01 (one-way ANOVA); n ¼ 3 mice per group. (b) Immunofluorescent analysis of CD62L ligands expressed on HEVs of tolerized mice (1-, 5- and 10-week), rejecting mice (1-week) and naive mice. Numbers in images indicate mean fluorescence intensity (determined over entire field). Data are representative results from ‘10-week tolerized’ and ‘1-week rejecting’ mice (n ¼ 3 mice per group). (c) Left, representative staining of Stamper-Woodruff assay of the binding of YAe+PDCA-1+ cells from ‘10-week tolerized’ mice (red) and ‘1-week rejecting’ mice (blue) to the same ‘10-week tolerized’ lymph node tissue section. Numbers in images indicate number of cells bound to HEV. Original magnification,
250. Right, quantitative analysis of the Stamper-Woodruff assay, determined from ten lymph node sections per slide. N, naive; T, tolerized; R, rejecting. **, P o 0.01, and *, P o 0.05 (one-way ANOVA). (d) Immunofluorescent analysis of YAe+ distribution in secondary lymphoid organs in tolerized mice (1-, 5- and 10-week) and rejecting mice (1-week). Data are representative results from ‘10-week tolerized’ mice. Original magnification,
100. (e) Flow cytometry of lymph node and splenic alloantigen-presenting cells from ‘10-week tolerized’ and ‘1-week rejecting’ mice. Numbers beside outlined areas indicate percentage of cells in each. Data are representative results from ‘10-week tolerized’ mice. All experiments were done three times.

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 6

JUNE 2006

655

centers, consistent with a pDC-like pattern of distribution25 (Fig. 4a). This ‘strategic’ positioning near the HEVs favors the encounter of tissue-derived DCs with antigen-specific T cells and results in retention of those T cells that recognize the alloantigen, which optimizes T cell activation26. We analyzed chemokine expression in YAe+ and YAe– lymph node and spleen pDCs and found considerable CCL17 mRNA in lymph node YAe+PDCA-1+ DCs (Fig. 4b, left). Conversely, CCL17+ pDCs were absent from the spleen (Fig. 4b, left), consistent with published results27. That result was further corroborated by

histological visualization of CCL17 in the YAe+ cells in the lymph node (Fig. 4b, right). The pDCs from rejecting mice did not express CCL17 (Supplementary Fig. 3 online). To determine whether CCL17+YAe+PDCA-1+ cells influence the migration or accumulation of T cells28, we isolated lymph node CD4+ T cells and analyzed them by PCR and multicolor flow cytometry. There was high expression of CCR4 mRNA and protein by CD4+CD25+Foxp3+ T cells in the lymph nodes but not the spleens of tolerized recipient mice (Fig. 4c). Further histological data demonstrated localization of Foxp3+ cells around the

a

d

Foxp3

HEV

B cell follicle

HEV HEV

Merged Merged

300

HEV

250 200 150 100

B cell follicle

50 YAe CCL17

YA e

+

Tolerized LN

Tolerized LN

e

100 Graft survival (%)

0 p YA DC e– pD C YA e+ pD C YA e– pD C

b

CCL17 (relative units)

YAe

Tolerized SPL

Tolerized spleen

80 60 WT

40

Ccr4 –/–

20 0 0

CD25

150 100 50 –

+

–

D

C

4+ C

D 2 4+ 5 C D 25 C D 4+ C D C 2 D 4+ 5 C D 25

+

0

Tolerized LN

Tolerized spleen

101 105

102

15

105

101 102 103 104 105

2

2

57

101

101 102 103 104 105

Tolerized spleen 12

104

13

16

103

7

102 101

103

105

102

104

101 102 103 104 105

CD4

Foxp3

21

101

9 101 102 103 104 105

3

102 101

103

105

102

102

105

Ccr4 –/– SPL

4

0

1

96

103

3

23

0 101 102 103 104 105

104 102

104

101 105

102

104

14

101

101 102 103 104 105

3

101

103 101

0

6

24 101 102 103 104 105

103

103

0 101 102 103 104 105

104

101

105

0

103

102

102

105

1

94

104

104

103

101 102 103 104 105

105

105

Ccr4 –/– LN

104

15

101

104

CD25

f

102

103

40

42

29

103

104

200

D

104

19

CD25

250

24

CD25

105

WT (DST + α-CD40L)

CCR4 (relative units)

Tolerized LN 300

10 20 30 Time after transplantation (d)

CD4

101 102 103 104 105

0 101 102 103 104 105 0

6

103

Foxp3

105

Ccr4 –/– (DST + α-CD40L)

c

C

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

ARTICLES

102

1

101

101 102 103 104 105

CCR4

CCR4

Figure 4 ‘Strategic’ localization of YAe+ cells in the lymph nodes during tolerance. (a) Immunohistochemical analysis of YAe+ staining around HEVs and B cell follicles (stained with MECA-79) in the lymph nodes of tolerized mice (1-, 5- and 10-week). Data are representative results from ‘10-week tolerized’ mice. Original magnification,
100; n ¼ 4 mice per group. (b) Left, CCL17 mRNA expression in ‘10-week tolerized’ lymph node and splenic YAe+PDCA-1+ and YAe–PDCA-1+ cells (n ¼ 4 mice per group). Right, immunofluorescent analysis of YAe+ cells expressing CCL17. Original magnification,
250; n ¼ 2. SPL, splenic. (c) Left, CCR4 expression in ‘10-week tolerized’ lymph node and splenic CD4+CD25+ and CD4+CD25– T cells. Right, CCR4 and Foxp3 expression in CD4+CD25– and CD4+CD25+ T cell populations from tolerized lymph nodes and spleen (n ¼ 4 mice per group). WT, wild-type; a-, anti-. (d) Immunohistochemical analysis of Foxp3+ cell distribution around HEVs and B cell follicles (stained with MECA-79) in the lymph nodes of ‘10-week tolerized’ mice. Original magnification,
100; n ¼ 2 mice per group. (e) Graft survival in CCR4-deficient (Ccr4–/–) and wild-type (WT) recipient mice after treatment with DST and anti-CD40L (n ¼ 4 mice per group). (f) CCR4 and Foxp3 expression on day of rejection in CD4+CD25– and CD4+CD25+ T cell populations from CCR4-deficient lymph nodes and spleens after treatment with DST plus mAb to CD40L. Data are representative results of three independent experiments. Numbers in quadrants (c,f) indicate percentage of cells in each.

656

VOLUME 7

NUMBER 6

JUNE 2006

NATURE IMMUNOLOGY

ARTICLES 105

CD25

a

Splenic naive CD4+ TEa

4,000

105

3,000

104

2,000

Blood

1.3%

103

104

Blood

102

103

101 101 102 103 104 105

102

1,000

105

101 0

97%

101

4,000

101 102 103 104 105

CD4

105

Graft

25% 75%

Graft

102

105

15% 85%

4,000

4,000

3,000

3,000

1.3%

98%

2,000

2,000

1,000

1,000

0

0

102

101 102 103 104 105 105

LN

LN

101 102 103 104 105

CFSE

CFSE

Foxp3

101 102 103 104 105

LN

105 LN CD4+CD25– 104

103 102

101 102 103 104 105

CFSE TEa

101

22%

104 2%

YAe

103

101 101 102 103 104 105

98%

SSC

SSC

0

101 102 103 104 105

101 102 103 104 105 105 LN CD4+CD25+ 104

102

1,000

20

101

103

2,000

40

102

SPL

104

3,000

60

0

105

SPL

CD25

% of max.

80

101 102 103 104 105

101 102 103 104 105

4,000

Foxp3 Isotype

SPL

SPL

103

103

101 101 102 103 104 105

CD4

Foxp3

0

Foxp3

104

101

100

CFSE TEa

101 102 103 104 105

102

1,000

YAe

101

103

2,000

Sorted CD4+CD25–

103

104

3,000

Graft

104

101 102 103 104 105

101 102 103 104 105

b

Blood

103

104 25% 75%

102

102 101 101 102 103 104 105

CCR4

Blood

5.6% 72 h

3,000

CD25

104

103

1,000

102

4,000

1.1%

103

2.9%

98%

101 101 102 103 104 105

CD4 Sorted CD4+CD25–

103

1,000

102

0

101 101 102 103 104 105

+ 4,000

60

105

SPL 0.5%

1,000

102

0

101

20 0

3,000

SSC

101 102 103 104 105 105 72 h

103

1,000

102

0

Graft

80 60 40

6

4

2

20 0 101 102 103 104 105 100

SPL

0 Stimulators C57BL/6 BALB/c BALB/c

SPL

80

5.5%

Responders (CD4+CD25–)

60 40 20

CBA

C57BL/6 C57BL/6 C57BL/6 + + Cultured Cultured CD4+CD25+ CD4+CD25+

0 101 102 103 104 105 100

LN 12.1%

104

2,000

101 102 103 104 105

100 2.9%

101 102 103 104 105

LN 7.2%

0

Graft

104 103

4,000 101 102 103 104 105

72 h

2,000

40

20

101 102 103 104 105

CD25

80

Foxp3 Isotype

40

104

2,000

3,000

100

72 h

8

60

101 102 103 104 105 105

Graft

10

Blood

80

101 101 102 103 104 105

3,000

102

6.7%

104

2,000

0

105

100

105 Blood

% of max.

Splenic naive CD4+ TEa

d

Cultured YAe+ + CD4+ TEa cells

Tolerized YAe 4,000

[3H]TdR uptake (SI)

c

% of max.

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

105

101

101 102 103 104 105

101 102 103 104 105

YAe

CD4

LN

100

80

80

60

60

30%

40

40

20

20 0

LN α-TGF-β

CD4+CD25+ CD4+CD25–

7%

0 101 102 103 104 105

101 102 103 104 105

Foxp3

Figure 5 Lymph node YAe+ pDCs induce the generation of Foxp3+CD4+CD25+ cells from of Foxp3–CD4+CD25– T cells. (a) Flow cytometry of CFSElabeled TCR-transgenic CD4+CD25–Foxp3– TEa T cells isolated from various organs 72 h after adoptive transfer into ‘10-week tolerized’ mice (n ¼ 3 mice per group). (b) Foxp3 (red) immunofluorescent analysis of adoptively transferred CFSE-labeled TCR-transgenic TEa CD4+ T cells (green) in proximity with YAe+ cells (blue). Original magnification,
250. (c) Flow cytometry of CFSE-labeled TCR-transgenic TEa CD4+ T cells cultured for 72 h with YAe+PDCA-1+ cells from various organs of ‘10-week tolerized’ mice with and without 10 mg/ml of anti-TGF-b1,2,3 (n ¼ 3 mice per group). (d) Suppressive function of CD4+CD25+ TCR-transgenic TEa cells cultured with lymph node YAe+PDCA-1+ cells. C57BL/6 responder CD4+CD25– T cells (4
104) were cultured together with irradiated T cell–depleted splenocyte samples (mouse strains, below graphs). Results are expressed as stimulation index (SI) and represent the mean of triplicate determinations ± s.e.m. Numbers in quadrants (a,c) indicate percentage of cells in each. All experiments were done three times.

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 6

JUNE 2006

657

ARTICLES

α-CD62L LN α-CD62L DST + α-CD40L DST + α-CD40L + α-CD62L

WT LN

WT Spleen

20 0

0

DST + α-CD40L DST + α-CD40L + 120G8

Untreated DST + α-CD40L

60

DST + α-CD40L + 120G8

0.02%

SSC

1.08%

80

40

86%

54%

20 0

15 5 10 20 25 Time after transplantation (d)

13%

20

30

40

129 WT Spleen

YAe

45%

120G8

e

Untreated + naive pDC

f

100

% 0.5 11

80

86

60 Untreated + tolerized YAe+ pDC Untreated + naive pDC

40 20

% 0.1 14 85

CD4

0

0

15 5 10 20 25 Time after transplantation (d)

30

Foxp3

Untreated + tolerized YAe+ pDC % 11

LN

0

10

Time after transplantation (d)

Graft survival (%)

d

100

129 WT LN

20

YAe

YAe

c

129 WT (DST + α-CD40L) Swap70 –/– (DST + α-CD40L)

40

0

20 40 60 80 Time after transplantation (d)

120G8

Graft survival (%)

45%

10

% of max.

Spleen

Figure 6 Alloantigen-presenting pDC lymph node homing is necessary for Treg cell development and tolerance induction. 77 (a,b) Graft survival (left) and immunofluorescent analysis of YAe+ cells on day of rejection (right) in mice treated with mAb to % 18% 6 CD62L (+ a-CD62L; a) or Swap70–/– mice (b). Original magnification,
250; n ¼ 4 mice per group. 129, 129/Sv. (c) Graft 11 survival in mice receiving 500 mg of pDC-specific mAb 120G8 on days 0, 1, 3 and 4 after transplantation (downward arrows; 82 n ¼ 4 mice per group). (d) Flow cytometry of YAe+ pDCs in lymph nodes of recipient mice treated with mAb 120G8 on day 9 CD4 Foxp3 after transplantation. Numbers in quadrants and beside outlined areas indicate percent cells in each. SSC, side scatter. Data are 6 representative of two independent experiments. (e) Graft survival of untreated mice after adoptive transfer of 2
10 naive pDCs or YAe+PDCA-1+ cells from the lymph nodes of ‘10-week tolerized’ mice (n ¼ 3 mice per group). (f) Flow cytometry (two independent experiments) of the CD4+CD25+Foxp3+ cell population 72 h after adoptive transfer of 2
106 naive pDCs or YAe+PDCA-1+ cells from ‘10-week tolerized’ mice. Numbers beside outlined areas and bracketed lines indicate percent cells in each area. CD25

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

0

60

% of max.

40

Swap70–/– Spleen

80

LN

60

Swap70–/– LN

b 100

Spleen

80

α-CD62L Spleen

CD25

Graft survival (%)

100

Graft survival (%)

a

HEVs, adjacent to the B cell follicles (Fig. 4d), similar to that of YAe+ cells (Fig. 4a), which suggested that interactions between CCL17+ YAe+ and CCR4+CD4+ T cells in the lymph nodes of tolerized mice might influence Treg cell development. To further investigate the importance of CCR4-dependent interactions during transplantation, we transplanted cardiac allografts into CCR4-deficient mice and found that Treg cells did not develop and rejection occurred despite administration of the tolerogenic regimen (Fig. 4e,f).

only the YAe+PDCA-1+ cells isolated from tolerized lymph nodes were able to induce Foxp3 expression in CD4+CD25+ T cells, and this Foxp3 induction occurred in a transforming growth factor-b (TGF-b)–dependent way (Fig. 5c). CD4+CD25+Foxp3+ Treg cells generated from antigen-specific CD4+CD25–Foxp3–TEa cells suppressed the in vitro population expansion of CD4+CD25– responder T cells in an alloantigen-specific way (Fig. 5d), demonstrating their specific regulatory nature.

pDCs induce Treg cell development in lymph nodes To determine whether alloantigen-presenting pDCs mediate tolerance by inducing Treg cell development in the lymph node, we monitored expression of the transcription factor Foxp3 in T cell receptor (TCR)– transgenic CD4+CD25–Foxp3– TEa T cells. We isolated TEa T cells, which recognize the same I-Ed–I-Ab complex of peptide and MHC as the YAe mAb, labeled the cells with carboxyfluoroscein succinimidyl ester (CFSE) and transferred them into tolerized mice. At 72 h after adoptive transfer, 98% of TEa cells in the lymph nodes had proliferated, in contrast to 15% in the spleen, 25% in the allograft and 25% in blood (Fig. 5a). Monitoring of CD25 expression showed that 20–30% of the lymph node TEa cells became CD25+ and FoxP3+, whereas spleen, graft and blood TEa T cells remained CD25–. Because Foxp3+ cells also expressed CCR4 (Fig. 4c), we monitored these markers and found that only the CD25+ TEa cells had high expression of both Foxp3 and CCR4 (Fig. 5a). Those results were further confirmed by fluorescent immunohistochemistry, which showed that CFSE-labeled TEa cells expressing Foxp3 in the lymph nodes of tolerized mice were localized together in proximity to YAe+ cells (Fig. 5b). To further determine the unique function of lymph node alloantigen-presenting pDCs during Treg cell development, we cultured CD4+CD25–Foxp3– naive TEa cells with YAe+PDCA-1+ cells from lymph node, spleen, blood or allograft. YAe+ cells from each anatomic compartment were able to induce CD25 expression in CD4+ T cells (Fig. 5c). However,

pDCs in the lymph node are necessary for tolerance Published reports have demonstrated that lymph node occupancy by T cells is critical for the establishment of tolerance15,16. Interfering with normal T cell homing by CD62L blockade or deficiency of CD62L, CCL19, CCL21 or CCR2 (refs. 15,16) or deficiency of CCR4 (Fig. 4) prevented tolerization. To further demonstrate the essential function of lymph node homing and occupancy of blood alloantigenpresenting pDCs in Treg cell development and tolerance induction, we did cardiac transplantation and monitored the localization of YAe+ cells in the lymphoid tissues after administration mAb to CD62L. Treatment with mAb to CD62L abrogated tolerance within 1 week (Fig. 6a, left) and caused an almost complete absence of YAe+ cells in the lymph node (Fig. 6a, right). Because treatment with mAb to CD62L blocked lymph node homing of both alloantigen-presenting pDCs and CD4+ T cells, we used the 129/Sv EMS (H-2b) Swap70–/– mouse to further determine whether alloantigen-presenting pDC lymph node occupancy was necessary for Treg cell development. Swap-70 is involved in actin rearrangement and cell migration29,30 and is expressed in pDCs but not in CD4+ T cells (Supplementary Fig. 4 online). Steady-state lymph node homing was unimpaired in Swap70–/– mice (Supplementary Fig. 5 online), but after transplantation there was a profound reduction in the number of YAe+ cells (Fig. 6b, right) but not CD4+ T cells (Supplementary Fig. 6 online) in the lymph node. Treg cells did not develop in the lymph

658

VOLUME 7

NUMBER 6

JUNE 2006

NATURE IMMUNOLOGY

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

ARTICLES node (Supplementary Fig. 7 online) and tolerance was not induced, despite administration of the tolerizing regimen (Fig. 6a, left). As CD62L and Swap-70 deficiency also alter B cell migration, we investigated the specific function of pDCs during the induction of tolerance to vascularized allografts with a pDC-specific mAb (120G8)31. Treatment with 120G8 prevented tolerance induction in recipient mice subject to tolerogenic treatment (Fig. 6c) and caused peripheral depletion of pDCs (Fig. 6d). This provides definitive evidence that pDCs are required for the induction of tolerance to vascularized allografts. We next investigated the potential therapeutic use of toleranceinducing alloantigen-presenting pDCs by adoptive transfer of YAe+PDCA-1+ cells isolated from the lymph nodes of tolerized mice into untreated recipient mice. When untreated recipient mice received YAe+PDCA-1+ cells from the lymph nodes of tolerized donor mice, graft survival was prolonged for up to 24 d (Fig. 6e). In contrast, graft survival was not prolonged when untreated recipient mice received YAe–PDCA-1+ cells from naive donor mice, and rejection occurred 8–10 d after transplantation. We also studied the population expansion of CD4+CD25+Foxp3+ cells to further investigate the involvement of YAe+PDCA-1+ cells in Treg cell development and prolonged graft survival. In mice receiving naive control pDCs, CD4+ T cells in the lymph node and the spleen had low expression of CD25 and no expression of Foxp3 (Fig. 6f). However, when we adoptively transferred tolerized YAe+PDCA-1+ cells, we found a distinct population of CD25hiCD4+ T cells (Fig. 6f). This population expressed Foxp3 in the lymph node (45%) and spleen (18%), suggesting that adoptively transferred YAe+PDCA-1+ cells induce Treg cell population expansion and prolong graft survival. DISCUSSION Here we have reported that tolerance to vascular alloantigens is mediated at least in part by a distinct pDC subpopulation that is capable of acquiring and processing MHC class II–derived donor allopeptide in allografts; migrating to peripheral lymph node across HEVs due to their expression of a repertoire of integrins, selectins and chemokine receptors; and inducing the peripheral development of CD4+CD25+Foxp3+ Treg cells. Our findings indicate several potential therapeutic targets for pharmacological manipulation, including pDC antigen uptake and presentation, pDC migration and Treg cell development. This and other tolerizing protocols rely not only on the induction of Treg cells but also on the engagement other tolerance mechanisms, such as clonal anergy or deletion17. The tolerogenic regimen of donor-specific transfusion (DST) plus mAb to CD40L used here may contribute to pDC specialization for tolerization by ensuring proper trafficking of pDCs into the lymph node or unique interactions between pDCs and naive T cells. Indeed, pDCs activated through CD40L have reduced CD62L expression32, which may explain the localization of blood pDCs to the lymph nodes but not the spleen after treatment with mAb to CD40L. The identification of pDCs as phagocytic cells that transport foreign molecules from a donor allograft to recipient secondary lymphoid organs bears relevance to transplantation and demonstrates previously unknown functional attributes of this DC subpopulation. Although pDCs have been identified in peripheral tissue33, published work has suggested that human DCs derived from CD11c–CD4+ pDCs have poor phagocytic abilities20, indicating that pDCs only present cellassociated endogenous antigen and prime CD8+ T cells34. However, our in vitro experiments have demonstrated that freshly isolated mouse pDCs are capable of phagocytosis. That finding raises the issue of whether mouse pDCs act like human pDCs and demonstrates

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 6

JUNE 2006

that pDCs also present phagocytosed membrane-associated exogenous antigen and can prime CD4+ T cells. Supporting the idea of phagocytic functions for pDCs, and consistent with our observation that the C-type lectin mannose receptor CD206 was upregulated after capture of donor allograft MHC II molecules, a C-type lectin receptor on human pDCs that functions as an antigen-capturing molecule has been identified35. Defining the mechanisms of alloantigen-presenting pDC trafficking is critical for understanding tolerance to vascular transplants, as secondary lymphoid organs have different functions in the establishment of immunological recognition of or ‘ignorance to’ vascularized allografts36. In vascularized transplants, alloantigens are not presented in a localized way but are distributed systemically in many secondary lymphoid organs37,38, indicating that the designation of ‘draining’ is dependent on the type of allograft and its anatomic location and the surgical manipulation required for its placement. During tolerization, CD62L expression was upregulated on circulating alloantigen-presenting cells and CD62L ligand expression was upregulated on HEVs, and interactions between these two molecules facilitated binding of pDCs to HEVs and extravasation of pDCs into lymph nodes. We also found that LFA-1, VLA-4, ICAM-1 and chemokine receptors were upregulated on pDCs, whereas chemokines were upregulated in HEVs, during tolerization. Expression of a(1,3)-fucosyltransferase VII, a fucosyltransferase essential for the biosynthesis of selectin ligands, can be induced by TGF-b39. Therefore, it is possible that in tolerogenic conditions, TGF-b secreted by pDCs or other cell types upregulates L-selectin ligands on HEVs, which favor extravasation of alloantigenpresenting pDCs into the lymph node and subsequent Treg cell development40. Indeed, failure to visualize alloantigen-presenting cells in the lymph nodes after administration of mAb to CD40L and in Swap-70-deficient mice correlates with a failure of tolerance induction in these mice. Thus, tolerance may rely on an integrated signaling pathway orchestrated toward extravasation of alloantigenpresenting pDCs from the blood to the lymph node. These findings indicate that treatment with a new generation of immunomodulatory drugs directed against LFA-1 and ICAM-1, which prolong allograft survival41 by inhibiting the signals necessary for antigen-dependent T cell activation42 and trafficking into inflamed allografts, may also interfere with processes such as pDC homing, which are essential for tolerance43. It is unclear why the lymph node provides the anatomic location for the tolerogenic immune response to vascular alloantigen. It is noteworthy that the spleen is the main site where antigen-specific immune responses are initiated by systemic priming in mice and humans44,45 and, in contrast to lymph nodes, provides the anatomical site where B cells differentiate into immunoglobulin-secreting plasma cells46. It has been shown that splenic DCs ‘preferentially’ present alloantigen during the immune response against vascular antigens47. Tolerogenic DST treatment inhibits alloreactive proliferative responses in the spleen but not in the lymph nodes after vascular cardiac transplantation48 and induces a decrease in splenic alloantibody-secreting cells49. In addition, treatment with mAb to CD40L has been reported to disrupt germinal center structure50. We noted the most alloantigenpresenting pDCs in the spleens of rejecting mice and the lymph nodes of tolerized mice. Those findings suggest that the spleen may be a dominant location for immune priming and responsiveness, whereas the lymph node may be poised to provide regulatory signals depending on many factors, such as the presence of exogenous immunosuppression or whether splenic function has been disrupted. Important variables that determine priming versus tolerance include the DC subset that presents alloantigen. It has been proposed

659

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

ARTICLES that the particular DC subset that presents antigen determines whether immunity (mDCs) or tolerance (pDCs) is induced. Specifically, T cell priming has been noted when mDCs present FITC-labeled ovalbumin (OVA) protein in an inflammatory airway disease model51. In contrast, suppression of the asthmatic response occurs when the antigen is transported by pDCs. Skin-derived DCs present alloantigen and migrate through lymphatics to local draining lymph nodes, where they prime antigen-specific T cells52. Here we have reported that for a vascularized transplant, alloantigen is presented systemically by pDCs that circulate through blood. Thus, anatomical differences may determine the type of APC that enters the grafted tissue. Although pDCs had the ability to generate Treg cells in our model and expand CD4+CD25+Foxp3+ T cell populations in vitro53, pDCs may not be an exclusive tolerogenic DC subset, so other DCs may have similar functions in other settings54. We did not further investigate other DC subpopulations in our study here of the APC subpopulations involved in alloantigen presentation and mediating tolerance to vascularized allografts. The reason for that is that when studied at different time points, the only DC subpopulation that expressed the alloantigen recognized by the YAe mAb was the pDC subpopulation. We noted a very small population of YAe+ cells (o5%) that were not pDCs or mDCs; future experiments should be directed toward elucidating the phenotype of these cells. The observation that pDCs represent more than 95% of the alloantigen-presenting cells in the allograft challenges whether conventional mDC are able to migrate into the inflamed vascularized tissue and process alloantigen. Although homing of pDCs from blood into peripheral organs has been reported25, not much is known about the migration of blood circulating mDCs into inflamed vascularized tissue. It is possible that differences in graft size and type and site of engraftment may influence the type of APC that present alloantigen in different transplant settings. As for Treg cell induction, it has been shown that Foxp3 is not induced anew in non–Treg cells during pathogen-dependent immune responses, although the induction of Foxp3 in non–Treg CD4+ T cells using systemic tolerogenic protocols was not addressed55. In transplantation models, Treg cell development has been reported in the periphery of tolerant athymic mice12, suggesting that new CD25+CD4+ T cells induced in the periphery of transplanted mice mediate tolerance to transplant antigens after tolerogenic treatment13. Supporting that hypothesis, it has been reported that peripheral development of alloantigen-specific CD4+CD25+Foxp3+ Treg cells occurs in the lymph nodes, as blocking lymph node homing by treatment with mAb to CD40L abrogates Treg cell population expansion and tolerance16. The findings presented here suggest that pDCs mediate new Treg cell induction in the periphery in a transplantation setting. We have shown that an essential pathway for achieving tolerance to vascular allografts is provided by antigen-specific CD4+CD25+Foxp3+ regulatory cells that can be induced anew in the lymph nodes of transplanted mice by pDCs with systemic tolerogenic regimens. We have provided evidence that pDCs mediate Treg cell development in vivo. We suggest that the identification of pDCs as alloantigenpresenting cells that either activate antigen-specific effector T cells in the spleen or induce the population expansion of Treg cells in the lymph nodes may help in the design of specific protocols to accomplish indefinite graft survival. METHODS Mice. BALB/c, C57BL/6 and CBA mice 8–10 weeks of age were purchased from The Jackson Laboratory. C57BL/6 TCR-transgenic TEa mice were provided by

660

A.Y. Rudensky (University of Washington, Seattle, Washington). C57BL/6 CCR4-deficient mice were provided by C. Hogaboam (University of Michigan, Ann Arbor, Michigan), and Swap-70-deficient 129/Sv EMS and congenic 129/ Sv EMS mice have been described29,31. All mice were housed in a specific pathogen–free facility in microisolator cages. All experiments used age- and sex-matched mice in accordance with protocols approved by the Mount Sinai Institutional Animal Care and Utilization Committee. Reagents. The YAe hybridoma was a gift from A.Y. Rudensky. The MR1 hamster anti–mouse CD40L hybridoma was a gift from R.J. Noelle (Dartmouth University, Hanover, New Hampshire). The MEL-14 rat immunoglobulin G2a (IgG2a) anti–mouse CD62L hybridoma was purchased from the American Type Culture Collection. The 120G8 rat anti-mouse hybridoma has been described31. All hybridomas were grown in culture and supernatants were purified over protein G or protein A columns (Amersham Pharmacia Biotech). Vascularized cardiac transplantation. BALB/c hearts were transplanted as fully vascularized heterotopic grafts into C57BL/6 mice as described56. BALB/c cardiac grafts were transplanted by suturing of donor aorta and donor pulmonary artery end-to-side to the C57BL/6 recipient lower abdominal aorta and inferior vena cava, respectively. Recipient mice received intravenous injections in 0.5 ml PBS at various times. For tolerance, mice were treated with DST (1
107 donor splenocytes intravenously) on day –7 and 250 mg mAb to CD40L on days –7, –4, 0 and +4 (times relative to transplantation). One group received 100 mg mAb to CD40L 30 d after tolerization and mice rejected at 37–40 d. Graft function was monitored every other day by abdominal palpation. Tolerizing mice were studied at 1, 5 and 10 weeks after transplantation. Mice that had graft survival 10 weeks or more were considered ‘tolerized’ (called ‘10-week tolerized’ here). Untreated control mice received hamster IgG in PBS and rejection, defined as complete cessation of a palpable beat and confirmed by direct visualization at laparotomy, occurred 1 week after transplantation. Cell preparations. Single-cell suspensions were prepared as described16. Lymph (20 ml/mouse) was collected from the cisterna chyli and was resuspended in icecold Hank’s balanced-salt solution medium until further use. Mixed-lymphocyte reaction and in vitro suppression assay. A total of 2
104 sorted YAe+ cells from tolerized and rejecting lymph nodes or spleens were cultured in triplicate with 5
104 with TCR-transgenic TEa CD4+ T cells. For testing of the antigen-specific suppressive properties of CD4+CD25+ T cells derived from those cultures, 4
104 freshly isolated responder CD4+CD25– T cells from naive C57BL/6 mice were cultured together in triplicate with 5
104 irradiated (1,500 rads) BALB/c splenocyte samples that had been depleted of T cells by negative selection with Mouse pan T Dynabeads according to the manufacturer’s protocol (Dynal), along with 2
104 CD4+CD25+ TEa T cells that had been previously cultured for 72 h with tolerized YAe+ cells. Cells were then cultured for 3 d in 96-well plates. At 18 h before termination of culture, wells were pulsed with 1 mCi [3H]thymidine and incorporation was quantified with a scintillation counter. Results are expressed as stimulation index, determined from mean of triplicate determinations ± s.e.m. Adoptive transfer and culture conditions of TCR-transgenic CD4+ TEa cells. CD4+ T cell subsets were isolated from the spleens of TCR-transgenic TEa mice using the Mouse T-cell CD4 Subset Column Kit (R&D Systems) according to the manufacturer’s protocol and were purified to 85–90% purity. Enriched CD4+ cells were stained with CyChrome-labeled antibody to CD4 (anti-CD4) and phycoerythring-indotricarbocyanine–anti-CD25, and CD4+CD25– cells were isolated by sorting on a MoFlo cell sorter (DakoCytomation) to 97–99% purity. Cells were then stained with 5 mM CFSE (Molecular Probes), and 1
107 cells were adoptively transferred into transplanted C57BL/6 mice. In vitro CFSElabeled TCR-transgenic TEa CD4+ T cells were cultured together for 72 h at a ratio of 10:1 with YAe+PDCA-1+ cells from the blood, lymph node, spleen or graft from ‘10-week tolerized’ mice in 96 well-plates with 200 ml of complete RPMI medium with or without anti-TGF-b1,2,3 (10 mg/ml; R&D Systems). Laser-capture microdissection. Lymph node frozen sections 10 mm in thickness were fixed in acetone for 1 min and were washed with PBS containing

VOLUME 7

NUMBER 6

JUNE 2006

NATURE IMMUNOLOGY

ARTICLES

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

RNA inhibitors (200 U/ml; GenHunter). Stained sections were microdissected with a PixCell II Laser Capture Microscope (Arcturus Engineering). Microdissected cells from at least ten tissue sections were placed together on the same cap until its entire surface was covered. The transfer film cap with the cells derived by laser-capture microdissection was then inserted into a microcentrifuge tube containing 50 ml of extraction buffer for RNA isolation (Epicentre) according to the manufacturer’s instructions. Flow cytometry. For YAe staining, naive untreated cells from C57BL/6 mice that lack I-Ad molecules were used as negative controls, and cells from C57BL/6
BALB/c F1 mice were used as positive controls. Anti–mouse CD16-32 and biotinylated mouse IgG2b isotype control (PharMingen) were used to block Fcg III/II receptors and as a YAe isotype control antibody, respectively. Fluorochrome-conjugated antibodies specific for CD4, CD25, CD11c, B220, Gr-1, CD19, CD62L, CD11b, GR1, B220 and CD11b and purified rat antibody specific for mouse a4b7 were purchased from PharMingen. Purified goat anti-mouse CCR4 was from Capralogics and was visualized by secondary staining with a phycoerythrin-conjugated monoclonal anti-goat (Jackson ImmunoResearch). Purified anti-CD62L was from PharMingen and was visualized by secondary staining with Alexa Fluor 594– conjugated goat anti-rat (Molecular Probes). FITC-conjugated rat anti–mouse DEC-205 was from Serotec. FITC-conjugated rat anti–mouse MAdCAM1 was from Southern Biotech. FITC-conjugated rat antibodies specific for mouse CD80, CD86, CD40, MHC-II, CD206, LFA-1, ICAM, VLA-2, VLA-4 and CD2 were from eBioscience. Internalization of FITC-labeled beads was confirmed with biotinylated anti-FITC (Jackson Laboratory) that recognizes the FITClabeled beads on the cell surface of cultured cells or with biotinylated mouse IgG isotype control (PharMingen), plus streptavidin-allophycocyanin (Caltag) to recognize the membrane-bound FITC-labeled beads. For intracellular analysis of Foxp3 expression, cells were stained with rat anti-Foxp3 or rat IgG2a as an isotype control, according to the manufacturer’s protocol (eBioscience). Anti-TGF-b1,2,3 was from R&D Systems. An LSR II (BD Biosciences) was used for flow cytometry, and data were analyzed with FlowJo software (Tree Star). Results are expressed as a percentage of cells with staining above background, and mAbs were ‘titered’ at regular intervals during the course of these studies to ensure that saturating concentrations were used. In vitro phagocytic assay. Sorted splenic B cells, splenic pDCs and peritoneal macrophages (495% purity) were cultured with 1-mm FITC-labeled beads (Polysciences) in 10% FBS RPMI medium for time periods ranging from 2 h to 12 h. After that time, cells were placed on glass slides and were examined with a fluorescence microscope. Images of cell suspensions were acquired with a Leica DMRA2 fluorescence microscope (Wetzlar) and a digital Hamamatsu chargecoupled device camera for monitoring internalization of FITC-labeled beads. In addition, two-color flow cytometry was done on the resuspended cell populations to demonstrate that the FITC-labeled beads were not bound to the external cell membranes of pDCs. Real-time PCR. Total RNA was extracted with the ArrayPure Nano-scale RNA purification Kit (Epicentre) according to the manufacturer’s instructions. Reverse transcription used 1 mg of RNA and the Sensiscript RT Kit (Qiagen) and random primers (Invitrogen) according to the manufacturer’s protocol. Quantitative real-time PCR was done in duplicate or triplicate using 25 mg cDNA with 0.4 mM of each primer in a final reaction volume of 30 ml of 1
SYBR Green PCR Master Mix (Applied Biosystems). PCR cycling conditions were as follows: 50 1C for 2 min and 95 1C for 10 min, and then 40 cycles of 95 1C for 15 s followed by 60 1C for 1 min. Relative expression was calculated as 2(C T ubiquitin – C T gene) (where CT is cycling threshold; ABI PRISM 7700 User Bulletin 2) with ubiquitin RNA as the endogenous control. All samples were run in triplicate. Immunofluorescence microscopy. Hamster anti–mouse CD11c (HL3), purified rat anti–mouse B220, purified rat anti–mouse Ly6G and purified rat anti– mouse CD19 were purchased from PharMingen. Phycoerythrin-conjugated rat anti–mouse PDCA and purified rat anti–mouse PDCA were purchased from Miltenyi. Purified rat anti–mouse TARC-CCL17 was purchased from R&D Systems. Indocarbocyanine-conjugated goat anti-hamster IgG and indocarbocyanine-conjugated goat anti-rat IgG were purchased from Jackson

NATURE IMMUNOLOGY

VOLUME 7

NUMBER 6

JUNE 2006

ImmunoResearch. Phycoerythrin-conjugated rat ant–mouse Foxp3 was purchased from eBioscience. Biotin-labeled YAe antibody was developed with streptavidin–horseradish peroxidase, biotinyl tyramide (NEN; PerkinElmer) and streptavidin-carbocyanine (Jackson ImmunoResearch). All slides were mounted with Vectashield (Vector Laboratories) to preserve fluorescence. Images were acquired with a Leica DMRA2 fluorescence microscope (Wetzlar) and a digital Hamamatsu charge-coupled device camera. Separate green, red and blue images were collected and analyzed with Openlab software (Improvision). Final image processing was done with Volocity software (Improvision). For imaging of CD62L ligands on HEVs, chimeric CD62L fused to human m-chain (a gift from J.B. Lowe, Case Western Reserve University, Cleveland, Ohio) followed by FITC–anti-human IgM (Jackson ImmunoResearch) was used. The fluorescence intensity of HEVs was measured with Slidebook 4.1 Digital Software (Intelligent Imaging Innovations) with masking of at least four representative areas for each section to obtain the mean fluorescence intensity over 10 mm2. The z-projection images of 8-mm sections were analyzed at maximum fluorescence intensities every 0.5 mm. In vitro HEV-binding assay. A modified Stamper-Woodruff assay was done24. Frozen sections 8 mm in thickness were overlaid with tolerized and rejecting blood YAe+ pDCs (1
104 in 100 ml) at a ratio of 1:1, labeled with PKH red (2.5 mM) and PKH green (2.5 mM), respectively (Sigma). Sections were incubated at 4 1C for 45 min with horizontal rotation at 60 r.p.m. Slides were then washed in cold PBS and then were fixed for 10 min in 3% paraformaldehyde on ice. Samples were stained with rat anti–mouse peripheral node addressin followed by FITC–goat anti-rat for HEV localization and were examined by fluorescent microscopy. The adhesion assay used ten lymph node sections per slide with triplicate slides. Results are expressed as number of cells bound to HEVs ± s.e.m. Statistics. For graft survival, one-way analysis of variance (ANOVA) was done. For chemokine receptor expression, one-way ANOVA and the Dunnett test were done to examine individual differences versus various controls. For binding assays, one-way ANOVA and the Dunnett test were done. For total cell percentages, one-way ANOVA was done. Note: Supplementary information is available on the Nature Immunology website.

ACKNOWLEDGMENTS We acknowledge the technical contributions of J. Llodra, H. Nikolayevskiy and S. Freeman, and discussions with M. Merad. Supported by National Institutes of Health (R01 AI41428, AI44929 and AI62765 to J.S.B.). COMPETING INTERESTS STATEMENT The authors declare that they have no competing financial interests. Published online at http://www.nature.com/natureimmunology/ Reprints and permissions information is available online at http://npg.nature.com/ reprintsandpermissions/ 1. Krieger, N.R., Yin, D.P. & Fathman, C.G. CD4+ but not CD8+ cells are essential for allorejection. J. Exp. Med. 184, 2013–2018 (1996). 2. Sayegh, M.H. et al. Allograft rejection in a new allospecific CD4+ TCR transgenic mouse. Am. J. Transplant. 3, 381–389 (2003). 3. Yamada, A. et al. Further analysis of the T-cell subsets and pathways of murine cardiac allograft rejection. Am. J. Transplant. 3, 23–27 (2003). 4. He, C. & Heeger, P.S. CD8 T cells can reject major histocompatibility complex class I-deficient skin allografts. Am. J. Transplant. 4, 698–704 (2004). 5. Hancock, W.W. Chemokine receptor-dependent alloresponses. Immunol. Rev. 196, 37–50 (2003). 6. Banchereau, J. & Steinman, R.M. Dendritic cells and the control of immunity. Nature 392, 245–252 (1998). 7. Sallusto, F. et al. Rapid and coordinated switch in chemokine receptor expression during dendritic cell maturation. Eur. J. Immunol. 28, 2760–2769 (1998). 8. Miyasaka, M. & Tanaka, T. Lymphocyte trafficking across high endothelial venules: dogmas and enigmas. Nat. Rev. Immunol. 4, 360–370 (2004). 9. Sakaguchi, S. Naturally arising Foxp3-expressing CD25+CD4+ regulatory T cells in immunological tolerance to self and non-self. Nat. Immunol. 6, 345–352 (2005). 10. Schenk, S. et al. Alloreactive T cell responses and acute rejection of single class II MHC-disparate heart allografts are under strict regulation by CD4+CD25+ T cells. J. Immunol. 174, 3741–3748 (2005). 11. Hara, M. et al. IL-10 is required for regulatory T cells to mediate tolerance to alloantigens in vivo. J. Immunol. 166, 3789–3796 (2001).

661

© 2006 Nature Publishing Group http://www.nature.com/natureimmunology

ARTICLES 12. Karim, M., Kingsley, C.I., Bushell, A.R., Sawitzki, B.S. & Wood, K.J. Alloantigeninduced CD25+CD4+ regulatory T cells can develop in vivo from CD25–CD4+ precursors in a thymus-independent process. J. Immunol. 172, 923–928 (2004). 13. Cobbold, S.P. et al. Induction of Foxp3+ regulatory T cells in the periphery of T cell receptor transgenic mice tolerized to transplants. J. Immunol. 172, 6003–6010 (2004). 14. Abe, M., Wang, Z., de Creus, A. & Thomson, A.W. Plasmacytoid dendritic cell precursors induce allogeneic T-cell hyporesponsiveness and prolong heart graft survival. Am. J. Transplant. 5, 1808–1819 (2005). 15. Bai, Y. et al. L-selectin-dependent lymphoid occupancy is required to induce alloantigen-specific tolerance. J. Immunol. 168, 1579–1589 (2002). 16. Ochando, J.C. et al. Lymph node occupancy is required for the peripheral development of alloantigen-specific Foxp3+ regulatory T cells. J. Immunol. 174, 6993–7005 (2005). 17. Quezada, S.A. et al. Analysis of the underlying cellular mechanisms of anti-CD154induced graft tolerance: the interplay of clonal anergy and immune regulation. J. Immunol. 175, 771–779 (2005). 18. Murphy, D.B. et al. A novel MHC class II epitope expressed in thymic medulla but not cortex. Nature 338, 765–768 (1989). 19. Nakano, H., Yanagita, M. & Gunn, M.D. CD11c+B220+Gr-1+ cells in mouse lymph nodes and spleen display characteristics of plasmacytoid dendritic cells. J. Exp. Med. 194, 1171–1178 (2001). 20. Grouard, G. et al. The enigmatic plasmacytoid T cells develop into dendritic cells with interleukin (IL)-3 and CD40-ligand. J. Exp. Med. 185, 1101–1111 (1997). 21. Pachynski, R.K., Wu, S.W., Gunn, M.D. & Erle, D.J. Secondary lymphoid-tissue chemokine (SLC) stimulates integrin a4b7-mediated adhesion of lymphocytes to mucosal addressin cell adhesion molecule-1 (MAdCAM-1) under flow. J. Immunol. 161, 952–956 (1998). 22. Rosen, S.D. Endothelial ligands for L-selectin: from lymphocyte recirculation to allograft rejection. Am. J. Pathol. 155, 1013–1020 (1999). 23. Berg, E.L., McEvoy, L.M., Berlin, C., Bargatze, R.F. & Butcher, E.C. L-selectinmediated lymphocyte rolling on MAdCAM-1. Nature 366, 695–698 (1993). 24. Stamper, H.B., Jr & Woodruff, J.J. Lymphocyte homing into lymph nodes: in vitro demonstration of the selective affinity of recirculating lymphocytes for high-endothelial venules. J. Exp. Med. 144, 828–833 (1976). 25. Yoneyama, H. et al. Evidence for recruitment of plasmacytoid dendritic cell precursors to inflamed lymph nodes through high endothelial venules. Int. Immunol. 16, 915–928 (2004). 26. Bajenoff, M., Granjeaud, S. & Guerder, S. The strategy of T cell antigen-presenting cell encounter in antigen-draining lymph nodes revealed by imaging of initial T cell activation. J. Exp. Med. 198, 715–724 (2003). 27. Alferink, J. et al. Compartmentalized production of CCL17 in vivo: strong inducibility in peripheral dendritic cells contrasts selective absence from the spleen. J. Exp. Med. 197, 585–599 (2003). 28. Iellem, A. et al. Unique chemotactic response profile and specific expression of chemokine receptors CCR4 and CCR8 by CD4+CD25+ regulatory T cells. J. Exp. Med. 194, 847–853 (2001). 29. Shinohara, M. et al. SWAP-70 is a guanine-nucleotide-exchange factor that mediates signalling of membrane ruffling. Nature 416, 759–763 (2002). 30. Sivalenka, R.R. & Jessberger, R. SWAP-70 regulates c-kit-induced mast cell activation, cell-cell adhesion, and migration. Mol. Cell. Biol. 24, 10277–10288 (2004). 31. Asselin-Paturel, C., Brizard, G., Pin, J.J., Briere, F. & Trinchieri, G. Mouse strain differences in plasmacytoid dendritic cell frequency and function revealed by a novel monoclonal antibody. J. Immunol. 171, 6466–6477 (2003). 32. Cella, M., Facchetti, F., Lanzavecchia, A. & Colonna, M. Plasmacytoid dendritic cells activated by influenza virus and CD40L drive a potent TH1 polarization. Nat. Immunol. 1, 305–310 (2000). 33. Jahnsen, F.L. et al. Experimentally induced recruitment of plasmacytoid (CD123high) dendritic cells in human nasal allergy. J. Immunol. 165, 4062–4068 (2000). 34. Salio, M., Palmowski, M.J., Atzberger, A., Hermans, I.F. & Cerundolo, V. CpG-matured murine plasmacytoid dendritic cells are capable of in vivo priming of functional CD8

662

T cell responses to endogenous but not exogenous antigens. J. Exp. Med. 199, 567–579 (2004). 35. Dzionek, A. et al. BDCA-2, a novel plasmacytoid dendritic cell-specific type II C-type lectin, mediates antigen capture and is a potent inhibitor of interferon a/b induction. J. Exp. Med. 194, 1823–1834 (2001). 36. Lakkis, F.G., Arakelov, A., Konieczny, B.T. & Inoue, Y. Immunologic ‘ignorance’ of vascularized organ transplants in the absence of secondary lymphoid tissue. Nat. Med. 6, 686–688 (2000). 37. Chen, Y., Demir, Y., Valujskikh, A. & Heeger, P.S. The male minor transplantation antigen preferentially activates recipient CD4+ T cells through the indirect presentation pathway in vivo. J. Immunol. 171, 6510–6518 (2003). 38. Reed, A.J. et al. Alloreactive CD4 T cell activation in vivo: an autonomous function of the indirect pathway of alloantigen presentation. J. Immunol. 171, 6502–6509 (2003). 39. Wagers, A.J. & Kansas, G.S. Potent induction of a(1,3)-fucosyltransferase VII in activated CD4+ T cells by TGF-b1 through a p38 mitogen-activated protein kinasedependent pathway. J. Immunol. 165, 5011–5016 (2000). 40. Chen, W. et al. Conversion of peripheral CD4+CD25– naive T cells to CD4+CD25+ regulatory T cells by TGF-b induction of transcription factor Foxp3. J. Exp. Med. 198, 1875–1886 (2003). 41. Corbascio, M. et al. CTLA4Ig combined with anti-LFA-1 prolongs cardiac allograft survival indefinitely. Transpl. Immunol. 10, 55–61 (2002). 42. Kandula, S. & Abraham, C. LFA-1 on CD4+ T cells is required for optimal antigendependent activation in vivo. J. Immunol. 173, 4443–4451 (2004). 43. Brandt, M., Steinmann, J., Steinhoff, G. & Haverich, A. Treatment with monoclonal antibodies to ICAM-1 and LFA-1 in rat heart allograft rejection. Transpl. Int. 10, 141–144 (1997). 44. Liu, Y.J., Oldfield, S. & MacLennan, I.C. Memory B cells in T cell-dependent antibody responses colonize the splenic marginal zones. Eur. J. Immunol. 18, 355–362 (1988). 45. Ellyard, J.I. et al. Antigen-selected, immunoglobulin-secreting cells persist in human spleen and bone marrow. Blood 103, 3805–3812 (2004). 46. Shapiro-Shelef, M. & Calame, K. Regulation of plasma-cell development. Nat. Rev. Immunol. 5, 230–242 (2005). 47. Crowley, M., Inaba, K. & Steinman, R.M. Dendritic cells are the principal cells in mouse spleen bearing immunogenic fragments of foreign proteins. J. Exp. Med. 172, 383–386 (1990). 48. Josien, R., Heslan, M., Brouard, S., Soulillou, J.P. & Cuturi, M.C. Critical requirement for graft passenger leukocytes in allograft tolerance induced by donor blood transfusion. Blood 92, 4539–4544 (1998). 49. Baldwin, W.M., III, Rhoton, K. & Sanfilippo, F. IgM and IgG alloantibody production by splenocytes and deposition in rat renal allografts are decreased by donor-specific blood transfusion. Transplantation 51, 481–485 (1991). 50. Han, S. et al. Cellular interaction in germinal centers. Roles of CD40 ligand and B7–2 in established germinal centers. J. Immunol. 155, 556–567 (1995). 51. de Heer, H.J. et al. Essential role of lung plasmacytoid dendritic cells in preventing asthmatic reactions to harmless inhaled antigen. J. Exp. Med. 200, 89–98 (2004). 52. Itano, A.A. et al. Distinct dendritic cell populations sequentially present antigen to CD4 T cells and stimulate different aspects of cell-mediated immunity. Immunity 19, 47– 57 (2003). 53. Moseman, E.A. et al. Human plasmacytoid dendritic cells activated by CpG oligodeoxynucleotides induce the generation of CD4+CD25+ regulatory T cells. J. Immunol. 173, 4433–4442 (2004). 54. Kretschmer, K. et al. Inducing and expanding regulatory T cell populations by foreign antigen. Nat. Immunol. 6, 1219–1227 (2005). 55. Fontenot, J.D. et al. Regulatory T cell lineage specification by the forkhead transcription factor Foxp3. Immunity 22, 329–341 (2005). 56. Corry, R.J., Winn, H.J. & Russell, P.S. Primarily vascularized allografts of hearts in mice. The role of H-2D, H-2K, and non-H-2 antigens in rejection. Transplantation 16, 343–350 (1973).

VOLUME 7

NUMBER 6

JUNE 2006

NATURE IMMUNOLOGY