Indoleamine 2,3-dioxygenase is a signaling protein in long-term tolerance by dendritic cells Maria T. Pallotta, 1 Ciriana Orabona, 1 Claudia Volpi, 1 Carmine Vacca, 1 Maria L. Belladonna, 1 Roberta Bianchi, 1 Giuseppe Servillo, 2 Cinzia Brunacci, 2 Mario Calvitti, 1 Silvio Bicciato, 3 Emilia M.C. Mazza, 3 Louis Boon, 4 Fabio Grassi, 5 Maria C. Fioretti, 1 Francesca Fallarino, 1* Paolo Puccetti, 1* Ursula Grohmann 1*† Departments of 1 Experimental Medicine and Biochemical Sciences and 2 Clinical and Experimental Medicine, University of Perugia, 06126 Perugia, Italy. 3 Department of Biomedical Sciences, University of Modena and Reggio Emilia, 41100 Modena, Italy. 4 Bioceros BV, Utrecht 3584 CM, Netherlands. 5 Institute for Research in Biomedicine, Bellinzona, 6500 Switzerland *These authors share senior authorship. Nature Immunology doi:10.1038/ni.2077
13
Embed
Indoleamine 2,3-dioxygenase is a signaling protein in long ... · Indoleamine 2,3-dioxygenase is a signaling protein in long-term tolerance by dendritic cells Maria T. Pallotta,1
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Indoleamine 2,3-dioxygenase is a signaling protein in long-term tolerance by dendritic cells
Maria T. Pallotta,1 Ciriana Orabona,1 Claudia Volpi,1 Carmine Vacca,1 Maria L.
Belladonna,1 Roberta Bianchi,1 Giuseppe Servillo,2 Cinzia Brunacci,2 Mario Calvitti,1
Silvio Bicciato,3 Emilia M.C. Mazza,3 Louis Boon,4 Fabio Grassi,5 Maria C. Fioretti,1
Francesca Fallarino,1* Paolo Puccetti,1* Ursula Grohmann1*†
Departments of 1Experimental Medicine and Biochemical Sciences and 2Clinical and
Experimental Medicine, University of Perugia, 06126 Perugia, Italy. 3Department of
Biomedical Sciences, University of Modena and Reggio Emilia, 41100 Modena, Italy. 4Bioceros BV, Utrecht 3584 CM, Netherlands. 5Institute for Research in Biomedicine,
Bellinzona, 6500 Switzerland
*These authors share senior authorship.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 1. Silencing of the Ido1 gene is both efficient and specific in pDCs. (a)
The kinetics of PCR analysis of Ido1 expression in pDCs treated with two distinct Ido1
siRNA (Ido1-1 and Ido1-2; see Methods for details) or Ido2 siRNA are shown. Control cells
were treated with negative control (nc) siRNA. Since no detectable transcripts for Ido2 in
pDCs were found (data not shown), efficiency and specificity of Ido1 and Ido2 gene silencing
was compared using liver cells incubated for 18 h with 1 µg/ml lipopolysaccharide. One
experiment is shown representative of five. (b) Expression of IDO protein was evaluated by
immunoblotting lysates from pDCs transfected with nc, Ido1-1, Ido1-2, or Ido2 siRNA at 48 h
post-transfection. One experiment of two. Because no anti-IDO2 antibody is currently
available, a control group for any effect of Ido1 siRNA on IDO2 protein expression could not
be introduced.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 2. pDC conditioning with IFN-γ but not TGF-β inhibits proliferation of
CD4+ T cells in an IDO-dependent fashion. CD4+CD25– T cells (3 × 105 per well) were
cocultured for 4 d with pDCs (6 × 105), the latter being preincubated with medium alone
(control), IFN-γ, or TGF-β for 24 h prior to extensive washing, in the presence of 4 µg/ml
anti-CD3 antibody. pDCs were used as such (control) or after transfection with negative
control (nc) or one of two distinct Ido1-specific (Ido1-1 or Ido1-2; see Methods for details)
siRNA. 1-MT (4 µM) was also present in some cocultures. Proliferation of CD4+ T cells in
the cocultures was evaluated in terms of incorporation of 5-ethynyl-2´-deoxyuridine (EdU)
into DNA in gated CD4+ cells. One experiment representative of four.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 3. pDC conditioning with IFN-γ but not TGF-β promotes apoptosis of
CD4+ T cells in an IDO-dependent fashion. CD4+CD25– T cells (3 × 105 per well) were
cocultured for 24 h with pDCs (6 × 105), the latter preincubated with medium alone (control),
IFN-γ, or TGF-β for 24 h prior to extensive washing, in the presence of 4 µg/ml anti-CD3
antibody. pDCs were used as such (control) or after transfection with negative control (nc) or
one of two distinct Ido1-specific (Ido1-1 or Ido1-2; see Methods for details) siRNA. 1-MT (4
µM) was also present in some cocultures. Apoptosis of CD4+ T cells in the cocultures was
evaluated by cytofluorimetric analysis of PI+Annexin V+ cells in gated CD4+ cells. One
experiment representative of three.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 4. pDC conditioning with TGF-β but not IFN-γ increases numbers of
CD4+Foxp3+ T cells in an IDO-dependent fashion. CD4+CD25– T cells (3 × 105 per well)
were cocultured for 4 d with pDCs (6 × 105), the latter preincubated with medium alone
(control), IFN-γ, or TGF-β for 24 h prior to extensive washing, in the presence of 4 µg/ml
anti-CD3 antibody. pDCs were used as such (control) or after transfection with negative
control (nc) or one of two distinct Ido1-specific (Ido1-1 or Ido1-2; see Methods for details)
siRNA. 1-MT (4 µM) was also present in some cocultures. Expression of Foxp3 in the
cocultures was evaluated by cytofluorimetric analysis of co-expression of surface CD4 and
intracellular Foxp3. Control samples included CD4+CD25– cells and pDCs cultured alone.
One experiment representative of three.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 5. TGF-β induces the expression of Tgfb1 transcripts in pDCs in an
IDO-dependent fashion. Real-Time PCR kinetic analysis of Tgfb1 transcript levels in TGF-β–
conditioned pDCs from C57BL/6 (WT) or Ido1–/–. Gapdh (encoding glyceraldehyde 3-
phosphate dehydrogenase) was used for normalization. Data (means ± s.d. from three
experiments) are presented as fold change in normalized transcript expression in cytokine-
treated pDCs relative to untreated counterparts (in which fold change = 1; dotted line).
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 6. Percentages of CD4+Foxp3+ cells increase in popliteal lymph nodes
(PLN) harvested from peptide-challenged footpads of mice sensitized with a combination of
peptide-pulsed CD8– DCs and TGF-β– but not IFN-γ–treated pDCs. Co-expression of CD4
and Foxp3 was measured by cytofluorimetric analysis in PLN leukocytes harvested either
from peptide-challenged left (PLNL) or vehicle-injected right (PLNR) hind legs of mice (n = 6
per group) sensitized with peptide-pulsed DCs 15 d earlier as in Fig. 2a and challenged
intrafootpad with the IGRP peptide 24 h before. One representative experiment of three is
shown.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 7. Percentages of CD4+Foxp3+ cells do not increase in spleen and
mesenteric lymph nodes (MLN) of mice sensitized with a combination of peptide-pulsed
CD8– DCs and TGF-β–treated pDCs. Expression of Foxp3 was measured as percent of
CD4+Foxp3+ cells by cytofluorimetric analysis of spleen cells or leukocytes harvested from
MLN of mice (n = 6 per group) sensitized with peptide-pulsed DCs 15 d before as in Fig. 2a
and challenged intrafootpad with the IGRP peptide 24 h before.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 8. The polyclonal anti-pITIM2 antibody is specific for the tyrosine
phosphorylated form of the peptide sequence encompassing IDO ITIM2. (a) ELISA:
Phosphorylated (pITIM2) and nonphosphorylated (ITIM2) peptides were covalently coupled
to N-oxysuccinimide binding plates and probed with various dilutions of affinity-purified
anti-pITIM2. PBS alone (i.e., without peptide) was used as a control. Values are means ± s.d.
of triplicate determinations of the optical density of the ELISA products. (b) Immunoblot
analysis: Lysates from P1 tumor cells, either untransfected (P1 mock) or transfected with a
construct expressing IDO (P1.IDO) or an IDO mutant lacking the phosphorylable tyrosine
present in ITIM2 (P1.Y253F), were sequentially blotted with anti–pITIM2, anti–IDO, and anti–
β-tubulin antibodies. Cells were used as such or after treatment with the phosphatase inhibitor
Na3VO4 for 10 min. (a,b) one experiment representative of five.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 9. Silencing of Ptpn6 or Ptpn11 genes is both efficient and specific in
pDCs. (a) The kinetics of PCR analysis of Ptpn6 and Ptpn11 expression in pDCs treated with
Ptpn6 or Ptpn11 siRNA are shown. Control cells were treated with negative control (nc)
siRNA. One experiment is shown representative of three. (b) Expression of SHP-1 and SHP-2
proteins was evaluated by immunoblotting lysates from pDCs transfected with nc, Ptpn6, or
Ptpn11 siRNA at 48 h post-transfection. One experiment representative of two is shown.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 10. Expression of SHP-1/2, IKKα and type I IFN receptor is required
for the induction of CD4+Foxp3+ T cells by TGF-β–treated pDCs in vitro. CD4+CD25– T cells
(BALB/c or 129/Sv) were cocultured for 4 d with pDCs (BALB/c or Ifnar–/–, respectively),
the latter preincubated with TGF-β for 24 h prior to extensive washing, in the presence of 4
µg/ml anti-CD3 antibody. pDCs were used as such (no treatment) or after transfection with
negative control (nc), Chuk, Ikbkb, or a mixture of Ptpn6 and Ptpn11 (Ptpn6/Ptpn11) siRNA.
Expression of Foxp3 in the cocultures was evaluated by cytofluorimetric analysis of co-
expression of surface CD4 and intracellular Foxp3. Control samples included CD4+CD25–
cells and pDCs cultured alone (not shown). One experiment representative of three.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 11. Lack of expression of SHP-1/2, IKKα, or type I IFN receptor does
not impair the effects of IFN-γ–treated pDCs. (a) In vitro, CD4+CD25– T cells (BALB/c or
129/Sv) were cocultured for 4 d (for measuring proliferation) or 24 h (apoptosis) with pDCs
(BALB/c or either 129/Sv or Ifnar–/–, respectively), the latter preincubated with medium alone
or IFN-γ for 24 h prior to extensive washing, in the presence of 4 µg/ml anti-CD3. pDCs were
used as such or after transfection with negative control (nc), Chuk, Ikbkb, or a mixture of
Ptpn6 and Ptpn11 (Ptpn6/Ptpn11) siRNA. Proliferation of CD4+ T cells in the cocultures was
evaluated in terms of incorporation of EdU into DNA in gated CD4+ cells, whereas apoptosis
was evaluated by cytofluorimetric analysis of PI+Annexin V+ cells in gated CD4+ cells. One
experiment representative of two. (b) In vivo, a skin test assay was performed as in Fig. 2a.
BALB/c pDCs were used as such or after transfection with nc, Ptpn6 plus Ptpn11, Chuk, or
Ikbkb siRNA. Untransfected pDCs purified from wild type (129) or Ifnar–/– 129/Sv mice were
also used in combination with the HY peptide. (a,b) *P < 0.01 and **P < 0.001.
Nature Immunology doi:10.1038/ni.2077
Supplementary Fig. 12. Percentages of CD4+Foxp3+ cells are still high in the popliteal
lymph nodes (PLN) at three months of sensitization with TGF-β–conditioned pDCs.
Combinations of IGRP-pulsed (BALB/c) or HY-pulsed (C57BL/6 and Ido1–/–) CD8– DCs and
pDCs, the latter pretreated with either IFN-γ or TGF-β for 24 h, were administered to either
BALB/c (for BALB/c pDCs) or C57BL/6 (for both C57BL/6 and Ido1–/– pDCs) recipients.
Selected groups of recipient mice also received slow-release 1-MT or placebo pellets or anti-
TGF–β at the time of sensitization. Expression of Foxp3 was measured as percentage of
CD4+Foxp3+ cells by cytofluorimetric analysis in leukocytes harvested from PLNs (draining
left hind footpads) of mice (n = 6 per group) sensitized with peptide-pulsed DCs 3 months
earlier and challenged intrafootpad with the respective peptide (IGRP for BALB/c or HY for
C57BL/6 and Ido1–/–) 24 h before. One experiment out of three is shown.