Contact-dependent delivery of IL-2 by dendritic cells to CD4 T cells in the contraction phase promotes their long-term survival
- 141 Downloads
Common γ chain cytokines are important for immune memory formation. Among them, the role of IL-2 remains to be fully explored. It has been suggested that this cytokine is critically needed in the late phase of primary CD4 T cell activation. Lack of IL-2 at this stage sets for a diminished recall response in subsequent challenges. However, as IL-2 peak production is over at this point, the source and the exact mechanism that promotes its production remain elusive. We report here that resting, previously antigen-stimulated CD4 T cells maintain a minimalist response to dendritic cells after their peak activation in vitro. This subtle activation event may be induced by DCs without overt presence of antigen and appears to be stronger if IL-2 comes from the same dendritic cells. This encounter reactivates a miniature IL-2 production and leads a gene expression profile change in these previously activated CD4 T cells. The CD4 T cells so experienced show enhanced reactivation intensity upon secondary challenges later on. Although mostly relying on in vitro evidence, our work may implicate a subtle programing for CD4 T cell survival after primary activation in vivo.
Keywordsdendritic cell contact dependence long term survival T cell memory IL-2
In the linear model of T cell memory (Gasper et al., 2014), long lasting memory cells develop out of the contracting phase of activation. Cytokines, particularly IL-7 and IL-15, have been implicated in the formation, homeostasis and subtype differentiation of memory T cells (Gasper et al., 2014; Raeber et al., 2018). With all these advances in our understanding, exact determinant that leads to T cells’ commitment to memory is not clearly understood, particularly with regard to CD4 T cells (Bell and Westermann, 2008; MacLeod et al., 2009). Despite of different proposed mechanisms, a common question remains whether cell extrinsic factors can guide their long-term survival following primary infection.
Compared with IL-7 and IL-15, IL-2 signaling in memory T cell formation is less understood. For CD8+ T cells, IL-2 in the primary infection expands the cell number and renders them resistant to secondary challenges, mainly via anti-apoptotic signaling (Liao et al., 2013). In 2006 Bevan’s group reported that to establish CD8 T cell memory, the presence of CD25 on these cells during primary infection is critical. In that report, the authors ruled out autocrine IL-2, instead suggested an environmental source, such as CD4 T cells or dendritic cells (DCs) (Williams et al., 2006). This signaling was regarded as a strong one as it could be replaced only with IL-2/IL-2 antibody complex. A later report, however, suggested that CD8+ T cells could produce IL-2 for their own use in memory formation (Feau et al., 2011). Interestingly, it is generally agreed that a strong and sustained IL-2 signaling via CD25 leads to a terminal effector differentiation, detrimental to memory development (Kalia et al., 2010). While these findings may not be mutually exclusive, it does however, seem to indicate a discreet and managed availability of IL-2 at this stage of T cell activation. A similar analysis was performed for the role of IL-2 in CD4 T cell memory as well (McKinstry et al., 2014). IL-2 was found to be critical for the expression ratio of Blimp vs. Bcl-2, promoting cell survival and at the same time upregulate CD127 (IL-7Rα). Again, the intensity of IL-2 signaling at this stage was considered to be high. Because in the absence of autocrine IL-2 (IL-2 deficient T cells), the memory development required high levels of external IL-2 or in the form of IL-2/IL-2 antibody complex, and was minimally rescued by co-transfer of IL-2 sufficient primed CD4 T cells (McKinstry et al., 2014). However, it is not clear why IL-2 is belatedly needed in the primary stimulation, and more ominously whether CD4 T cells produce sufficient IL-2 this late to support their own transition to memory (Sojka et al., 2004).
In this short report, we describe an in vitro subtle stimulation event in activated T cells late in the primary phase of activation. In an experiment to study the CD69 expression profile in a typical in vivo T cell activation event, we noticed the re-expression of CD69 on activated CD4 T cells several days after its peak expression in vivo. In vitro, a similar activation can be driven by DCs in an antigen independent manner. In our settings, contact of previously activated/rested CD4 T cells with dendritic cells (DCs) sent a very mild activation signal to these T cells and led to a re-production of IL-2. T cells after this mild stimulation, upon transfer into naïve recipients, can better maintain their response to antigen in a secondary stimulation. Instead of signaling via JAK/STAT pathways (Ross and Cantrell, 2018), RNA-seq analysis indicated this activation event enhanced the expression of Bcl2 as well as Oasl2 and Isg15, two type I interferon-related genes (Zhao et al., 2013; Choi et al., 2015). These results provide suggestive evidence that optimal secondary response requires a brief encounter with DCs following antigen clearance and imply that DCs could be the source of the “trickling” IL-2 in the later phase of T cell activation, potentially pointing to a missing link in the overall picture of CD4 T cell survival after their primary activation.
Previously activated CD4 T cells upregulate CD69 in response to DC binding
PA T encounter with DCs results in limited activation indicated by upregulating of Il2 and Bcl2
PT-A reactivation is associated with IL-2 production by DCs
Surface delivery of IL-2 from DCs to PA T
IL-2+DC contact leads to an altered gene expression profile
IL-2-positive DC coculture promotes activated CD4 T cells into a long lasting population in vivo
IL-2 production by DCs was first revealed by Granucci et al., with the advent of gene expression microarray (Granucci et al., 2001), and it was reported at that time to play a role in both CD4 and CD8 T cell proliferative responses to allogenic DCs. It was later reported by the same group that microbial products, including common PAMPs such as GpG, peptidoglycan and LPS, drove the IL-2 production. The production could be tuned by the CD40 signaling originated from activated T cells as well (Granucci et al., 2003). At least for Dectin-1-mediated activation, IL-2 transcription was controlled via NFAT in phagocytes (Goodridge et al., 2007). However, this notion of DC-originated IL-2 being critical for primary T cell activation was questioned by another report (Schartz et al., 2005). Of particular interest, in the latter report, adoptive transfer of antige-loaded IL-2-sufficient and deficient DCs did not result in any difference in CD8 memory response. The setup was meant to address IL-2 production by DCs at the primary activation of T cells. An inconspicuous oversight of that report, in light of our work here, was whether the IL-2 expression in recipient resident DCs was involved as an environmental and temporally discreet factor to support T cell memory beyond initial DC/T cell interaction (Schartz et al., 2005). Therefore, it remains premature to dismiss the impact of DC-intrinsic IL-2 production in the totality of T cell biology.
Regarding the role of IL-2 in T cell memory development, one unsettled issue is the timing of bioavailability. IL-2 production has been found to be rapid and transient (Sojka et al., 2004). This pattern is well explained from the negative feedback in the environmental milieu (Villarino et al., 2007). In human, IL-2 production peaks in day one and becomes undetectable on day three or four, to be followed by a more sustained T cell proliferation (Venuta et al., 1983). This certainly raises the issue of timing as the IL-2 suggested to be mandatory in late phase of primary activation is not available from the same T cells unless intervened otherwise. Our results seem to draw such a scenario: T cells expand in the presence of antigen with a robust IL-2 production to support the proliferation. The IL-2 production is rapidly reduced thereafter in preparation of the contraction from the peak T cell activation. At this moment, a percentage of DCs by coincidence bind to DCs in the absence of antigen. The IL-2 produced in these DCs triggers a secondary IL-2 production in these previously activated CD4 T cells to mediate a program of cell survival. Due to the transient nature of DC/T cell contact, another possibility is that T cells that have accumulated sufficient number of contacts with DCs venture to enter the second round of IL-2 production. The probabilistic nature of the DC/T cell contact may also limit a large number of T cells from entering memory programing, providing a crude selection filter at this moment. Regarding the observation that DC CD25 appears to contribute to the subtle T cell activation, one possible involvement is its role in bringing IL-2 to the DC/T cell contact interface, as T cells at this stage do not express this high affinity receptor. We have shown in this report that extrinsic addition of IL-2 was not as effective. In reality, it would be extraordinarily difficult to sustain IL-2 at high levels days after the primary activation. Low amounts of IL-2 would in theory require some degree of assistance for its efficient delivery to T cells. Previous work on several cytokines seem to point to a shared mechanism. In TH-17 programing, IL-6 is presented on DCs via IL-6Rα chain for highly efficient activation of T cells (Heink et al., 2017). In that report, one theory is the formation of some type of signaling complex at the contact site across two cells. However, the topography of combining cytokine and multiple receptor signaling chains in such a configuration has not been formally proposed. In Daclizumab treatment, a humanized monoclonal antibody used to treat inflammation, IL-2 was also presented by CD25 on DCs for robust signaling in T cells (Wuest et al., 2011). Our work again suggests the utilization of this special method of cytokine delivery. Why this way of cytokine delivery is superior to external loading is not clear. One possibility is the pre-existing CD25/IL-2 complex formation inside DCs, leading to a more sustained/stable presence on the DC surface. On the other side of DC/T cell pair, whether the subtle PA T reactivation induced gene profile changes are a consequence of this particular mode of cytokine/receptor arrangement or determined by other temporal factors or ligand density would require more sophisticated setups. IL-2 activates STAT5A/B and to some degree also signals via STAT1 and STAT3. The JAK-STAT pathway is critical for the differentiation of T cells subsets (Spolski et al., 2017). Whether a hard to detect, miniature signal in the contraction state of T cell activation also regulates the latter long term survival is a worthy follow-up study. Lastly, we cannot rule out epigenetic changes in DCs that are deficient in IL-2/CD25. Although these proteins are not known to directly regulate DC development or phenotype, strong evidence suggests that IL2/CD25 (via STATs and IRFs)-associated signaling is a part of DC programming (Tian et al., 2017). These differences themselves may also explain why the lack of these two factors leads to an altered T cell survival upon DC/T encounter.
We must stress that results in this work are mostly suggestive pertaining to T cell memory. First, we used an in vitro model to simulate a potential in vivo process. Whether the CD69 upregulation seen in the late phase of T cell activation is related to our observation is unknown. Second, verifications will be needed to show that the in vivo CD69+ population has the survival advantage in real life T cell response. We therefore do not advocate a direct link between our findings and in vivo DC/T cell behavior, let alone the precise role of IL-2 from DCs in T cell memory formation. This report may be of some value in providing the missing link regarding the IL-2 availability during the late phase of primary CD4 activation that is conducive to memory formation. While this finding is intriguing, advanced animal work will provide more precise insight. For instance, a timed or activation stage-triggered IL-2 deletion in DCs, or a detailed analysis of accumulation DC/PA T binding will be highly informative. In addition to IL-2 production in the T cells, this event appears to regulate several groups of gene expression. Whether and how much these gene regulations individually and collectively contribute to IL-2 production or even to the memory development should present intriguing topics for future work.
Materials and methods
All mice were in C57BL/6 background. WT, OT-II, CD11a−/−, IL-2RA−/−, CD40−/− CD45.1 congenic mice were bred and housed at Tsinghua University Animal Facilities. IL-2RA−/− were made in house with CRISPR/Cas9 techniques and verified by genotyping. IL-2−/− mice were purchased from JAX. H2-IAb (H2AB) deficient mice (MHC class II−/− mice) were a gift from Dr. Hai Qi of Tsinghua University. The Animal Experiments Committee of Tsinghua University approved all of the experiments reported in this study.
Production of Cas9 mRNA and sgRNA
CRISPR/Cas9 vectors were gifts from Dr. Hai Qi of Tsinghua University. Vector pST1374-N-NLS-flag-linker-Cas9 expressing cas9 was digested with AgeI (NEB) and SpeI (NEB), and the linearized vector was gel purified, and used as the template for in vitro transcription using mMESSAGE mMACHINE T7 ULTRA kit (Life Technologies, AM1345). Vector pUC57kan-T7-gRNA expressing sgRNA was digested with DraI (NEB), and the linearized vector was gel purified. A pair of oligos for targeting Il2ra was annealed, phosphorylated, and ligated to the linearized vector, and used as the template for in vitro transcription using MEGAshortscript T7 Transcription kit (Life Technologies, AM1354). Both the Cas9 mRNA and the sgRNA were purified using MEGAclear kit (Life Technologies, AM1908) and eluted in RNase-free water, and stored in −80 °C. sgRNA-Il2ra: 5′-TAGGGAACCATAGTACCCAGTTGTC-3′; and 5′-AAACGACAACTGGGTACTATGGTTC-3′
Intracytoplasmic RNA microinjection
The intracytoplasmic RNA microinjection was performed at Tsinghua University Animal Facilities. Each embryo was microinjected with 25 ng/µL sgRNA and 50 ng/µL Cas9 mRNA into the cytoplasm. After microinjection, surviving embryos were implanted on the same day into oviduct of pseudo pregnant C57BL/6 female mice. Full-term pups were obtained by natural labor at 19.5-day post coitum (dpc).
Genotyping of mice
All mice were genotyped with the tail DNA. Two pairs of primers were used to identify the IL-2RA−/− mice. Primer Il2ra-WT (5′-CATTAACCATAGTACCCAGTT-3′) and common-R (5′-GCAAAGCCAAACCATCCCTG-3′) can detected the wild type anneal. The product size was 305 bp. Primer Il2ra-M (5′-CATTAACCATAGTACCCAGTG-3′) and common-R can detected the mutant anneal. The product of PCR was 304bp. The PCR reaction conditions were as follows: 95 °C, 5 min; 95 °C, 30 s, 58.5 °C, 30 s, 72 °C, 30 s, 30 cycles; 72 °C, 10 min.
Antibodies and reagents
Antibodies for FACS, Vα2 TCR chain, CD69, CD4, CD25, CD127, CXCR5, CCR7, CCR9, CD62L, CD83, CD45.1 and CD103, and ELISA kits were purchased from eBioscience. Others are as follows, anti-mouse CD3e (eBioscience, 85-16-0032-85), anti-mouse CD28 (eBioscience, 85-16-0281-85), murine GM-CSF (Peprotech, 315-03-1000), murine IL-4 (Peprotech, 214-14-1000), human IL-2 (Peprotech, 200-02-1000), anti-mouse CD4 APC (eBioscience,17-0041-82), rat IgG2b K isotype control PE (eBioscience, 12-4031-82), rat IgG2b k isotype control APC (eBioscience, 17-4031-82), rat IgG2b k isotype control FITC (eBioscience, 11-4031-82), anti-mouse V alpha 2 TCR eFluor 450 (eBioscience, 48-5812-82), anti-mouse CD69 APC (eBioscience, 17-0691-82), anti-mouse V alpha 2 TCR PE (eBioscience, 12-5812-82), anti-mouse CD103 (Integrin alpha E) APC (eBioscience, 17-1031-82), anti-mouse CD197 (CCR7) PE (eBioscience, 85-12-1971-80), anti-mouse CD199 (CCR9) PE (eBioscience, 85-12-1991-80), anti-mouse CD83 APC (eBioscience, 17-0839-41), anti-mouse CD45.1 PE (eBioscience, 12-0453-82), anti-mouse CD62L FITC (eBioscience, 85-11-0621-82), PerCP/Cy5.5 anti-mouse CD127 (BioLegend, 135021), PE anti-mouse CD25 (BioLegend, 102007), PE anti-mouse CD185 (CXCR5) (BioLegend, 145503), mouse IL-2 ELISA Set (eBioscience, 555148), mouse IFN γ ELISA Ready-SET-Go (eBioscience, 85-88-7314-77), 1× TMB solution (eBioscience, 00-4201-56), cell lines and primary cell culture DC1940 (a gift from Dr. Li Wu of Tsinghua University), A20, 3T3 and MEF were cultured in complete RPMI1640 (GIBCO) supplemented with 10% FBS (GIBCO), 100 U/mL penicillin, and 100 mg/mL streptomycin,10 mmol/L HEPES and 50 µmol/L β-mercaptoethanol. Bone marrows were differentiated for 6 days with 20 ng/mL GM-CSF and 10 ng/mL IL-4 in complete RPMI medium to produce BMDCs respectively. On day 6, BMDCs which were grown in 24-well plate were used directly for experiments. 293FT cells were cultured in DMEM (GIBCO) supplemented with 10% FBS. All cells were cultured at 37 with 5% CO2.
Isolation of primary cells
Primary CD4 T cells were isolated from OT-II or B6 mice splenocytes by EasySep Mouse CD4+ T cell Isolation Kits (Stem Cell, 19852) and sometimes sorted by FACS with an anti-Vα2 TCR antibody. Primary CD11c DC cells were isolated from B6 or different gene knockout mice splenocytes by EasySep Mouse CD11c positive selection kit II (Stem Cell, 18780). Primary cells were obtained according to the protocol of the kits.
In our study, we treated OT-II mice splenocytes with 200 µg/mL OVA about 48 h, and isolated activated CD4+CD69+ T cells by FACS with anti-mouse Vα2 TCR and anti-mouse CD69 antibodies. Then the PA T cells were harvested. Sometimes we treated B6 splenocytes with anti-mouse CD3e and anti-mouse CD28 for about 48 h. PA T cells were cultured in complete RPMI1640 supplemented with 10% FBS, 200 U/mL penicillin, and 200 mg/mL streptomycin, 10 mmol/L HEPES and 50 µmol/L β-mercaptoethanol. After resting for 48 h, PA T cells were co-cultured with different kinds of DCs or cell-lines. In 24-well-plate co-culture assay in vitro, the ratio of PA T and DCs is 2 to 1 (2 × 105 and 1 × 105 cells). Co-culture was carried out for 24 h, PA T cells or DCs were harvested to detect surface markers. The supernatants were stored in −20 °C for testing IL-2 or IFN-γ secretion by ELISA.
The transwell assay
HTS Transwell-24-well permeable support with 0.4 µm Pore Polycarbonate Membrane and 6.5 mm Inserts, Sterile (Corning, 3396) were used in the transwell assay. In the transwell co-culture system, PA T cells were seeded in the upper layer, and DCs in the substrate. PA T cells were cultured at a 2:1 ratio with DCs for 24 h.
Gene knockdown in DC1940 cell-lines
Lentiviral plasmids encoding targeting Il2 shRNA or non-targeting shRNA (SHC002) were purchased from Sigma and the Il2 shRNA TRC numbers were as follows: TRCN0000067198, TRCN0000067200, TRCN0000067202. The lentivirus was produced by co-transfection of 293FT cells with lentivirus expression vector, pCMV-VSV-G and pCMV-dR8.91 using Lipofectamine 2000 (Invitrogen, 11668027). DC1940 cells were infected with lentivirus at 500 ×g, 32 °C for 1.5 h. DC1940 cells were selected with puromycin (1 µg/mL) 72 h post-infection. The knockdown efficiency was analyzed by real-time PCR.
RNA samples were prepared using TRIzol reagent (Invitrogen, 15596018) and first strand cDNA was synthesized with RrimeScriptTM RT-PCR Kit (TAKARA, RR047A). Real-time PCR was performed using 2 X RealStar Power SYBR mixture (GenStar, A311-10) in a BIO-RAD CFX96 Real-time System. GAPDH was used for normalization. The primer sequences were as follows: Il2, 5′-TGAGCAGGATGGAGAATTACAGG-3′ and 5′-ATGTGTTGTCAGAGCCCTTTAG-3′; Oasl2, 5′-AAACAGCTGAAGGGAGACCG-3′ and 5′-CTCGCTGCTGTACATTCCA-3′; Isg15, 5′-AGCAATGGCCTGGGACCTAAAG-3′ and 5′-TAAGACCGTCCTGGAGCACTG-3′; IL-7 receptor (IL-7r), 5′-ACAAGAACAACAATCCCACAGAG-3′ and 5′-TCGCTCCAGAAGCCTTTGAAG-3′; Bcl2, 5′-TGTGTGGAGAGCGTCAACAG-3′ and 5′-CAGACATGCACCTACCCAGC-3′; GAPDH, 5′- ATCAAGAAGGTGGTGAAGCA-3′ and 5′-AGACAACCTGGTCCTCAGTGT-3′.
Sampling and RNA extraction for RNA-seq
We treated OT-II mice splenocytes with 200 µg/mL OVA about 48 h, and isolated activated CD4+CD69+ T cells by FACS with anti-mouse Vα2 TCR and anti-mouse CD69 antibodies. Then PA T cells were harvested, and cultured in culture media. After resting 48 h, PA T cells were co-cultured with B6 or IL-2−/− CD11c+ DCs, which had been isolated from B6 or IL-2−/− splenocytes by EasySep Mouse CD11c Positive Selection Kit II. PA T cells were cultured in complete RPMI1640 alone as a control. After 24 h, these T cells were harvested by FACS with Vα2 TCR+CD11c−. The total RNA of each sample was isolated using TRIzol reagent (Invitrogen). RNA quality and concentration were determined using NanoDrop 2000 (Thermo Scientific). The RNA samples were stored in −80 °C, and later sent to BGI (China) for RNA sequencing.
Analysis of RNA sequencing
Sequencing was performed on a BGISEQ-500 RS unit using single-end sequencing, and averagely 23,656,293 raw sequencing reads were generated. Then we removed reads 1) with adaptors; 2) in which unknown bases were more than 10%; 3) having more than 50% bases in low quality (quality is no more than 5). A total of 23,593,592 clean reads were obtained. The fastq files were aligned to hg19 with HISAT2, with the following parameters: –phred64 -sensitive -I1 -X 1000.
After alignment, the read counts for each gene were calculated by RSEM, and the expression values of each gene were normalized using FPKM.
KEGG was used to perform pathway enrichment analysis of DEGs, which was automated by clusterProfiler. The calculated P-values were then corrected by Bonferroni’s method and a threshold of 0.05 was set. Heatmap (log2 scaled) was plotted with pheatmap, and protein-protein associations were analyzed by STRING (von Mering et al., 2003).
Mouse multiplex cytokine assay
PA T cells were co-cultured at a 2:1 ratio with DC1940 cells 24 h, then CD69+ and CD69− cells were obtained by FACS with anti-mouse Vα2 TCR and anti-mouse CD69 antibodies. These sorted cells were cultured 1 × 105 per well in 24-well-plate in complete RPMI1640 respectively. The supernatants were collected at different time points (15 h and 24 h) and stored in −20 °C for mouse multiplex cytokine detection. According to the manufacture’s protocol (Millipore, MCYTOMAG-70K, MCYTOMAG-70K-PMX, MCYTMAG-70K-PX32), the secretion of cytokines was detected by Luminex.
Listeria monocytogens infection
Listeria monocyte expressing ovalbumin (LM-OVA) were a gift from Dr. Chen Dong of Tsinghua University. Frozen stocks of LM-OVA were thawed and then diluted in fresh brain heart infusion (BHI) medium (BD) to reach mid-log growth phase. 8-week-old mice were intravenously injected via the lateral tail vein with sub-lethal (5 × 104 CFUs) dose of LM-OVA that was suspended in 200 µL PBS per mouse. After 4 h, spleens and lymph nodes were collected as day 0 samples. At day 1–10 days after infection, spleens and LNs were collected and dissociated in PBS containing 0.05% Triton X-100, and bacterial CFUs were determined by plating on BHI agar plates. The CD4+ T cells in spleens and LNs were analyzed for CD69 expression by FACS.
Adoptive transfers and re-stimulation clonal expansion experiments
After resting 48 h, PA T cells were co-cultured at a 2:1 ratio with B6 or IL-2−/− CD11c+ DCs. PA T cells were cultured in complete RPMI1640 alone as a control. After 24 h, these T cells were harvested by FACS with Vα2 TCR+CD11c−. These PA T cells were transferred 1 × 106 per mice into naïve C57BL/6 mice via intravenously. After one month, spleens and LNs were collected, 1 × 106 or 2 × 106 cells were seeded with 2 × 106 feeder cells and treated with 100 µg/mL OVA for 48 h to 96 h in 24-well-plate. The feeder cells were pre-treated X-Ray with 3,000 rad radiations (RS 2000 Pro). The culture supernatants were collected and stored −80 °C for IFN-γ and IL-2 secretion analysis by ELISA.
Number of experimental repeats are shown in the figure legend. All bar graphs are means with SEM. Statistical analysis was performed with Student’s t test in GraphPad Prism software. P value < 0.05 was considered significant. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. ns, not significant.
We thank Drs. Li Wu, Hai Qi, and Chen Dong for providing mice and reagents and Dr. Fei Shu for help in Listeria monocytogens infection and adoptive transfer experiments. Y.S. is supported by the joint Peking-Tsinghua Center for Life Sciences and the National Natural Science Foundation of China grants 81621002, 31630023, 31370878 and 20171312479. X.H. is supported by Ministry of Science and Technology of China National Key Research Projects 2015CB943201, National Natural Science Foundation of China grants 31821003, 31725010, 81661130161, 91642115 and 81571580.
D. T performed all the experiments with assistance from L. Z, F. N and Y. X. Y.S. and X.H conceptualized the project. Y.S. wrote the manuscript with inputs from X.H
COMPLIANCE WITH ETHICS GUIDELINES
Dan Tong, Li Zhang, Fei Ning, Ying Xu, and Xiaoyu Hu declare that they have no conflict of interest. Yan Shi has received a research grant from Boehringer Ingelheim. All institutional and national guidelines for the care and use of laboratory animals were followed.
- Schartz NE, Chaput N, Taieb J, Bonnaventure P, Trebeden-Negre H, Terme M, Menard C, Lebbe C, Schimpl A, Ardouin P et al (2005) IL-2 production by dendritic cells is not critical for the activation of cognate and innate effectors in draining lymph nodes. Eur J Immunol 35:2840–2850CrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.