Dendritic Cell-Secreted Lipocalin2 Induces CD8+ T-Cell Apoptosis, Contributes to T-Cell Priming and Leads to a TH1 Phenotype

Lipocalin 2 (LCN2), which is highly expressed by dendritic cells (DCs) when treated with dexamethasone (Dex) and lipopolysaccharide (LPS), plays a key role in the defence against bacteria and is also involved in the autocrine apoptosis of T-cells. However, the function of LCN2 when secreted by DCs is unknown: this is a critical gap in our understanding of the regulation of innate and adaptive immune systems. Tolerance, stimulation and suppression are functions of DCs that facilitate the fine-tuning of the immune responses and which are possibly influenced by LCN2 secretion. We therefore examined the role of LCN2 in DC/T-cell interaction. WT or Lcn2−/− bone marrow-derived DCs were stimulated with LPS or LPS+IFN-γ with and without Dex and subsequently co-cultured with T-cells from ovalbumin-specific TCR transgenic (OT-I and OT-II) mice. We found that CD8+ T-cell apoptosis was highly reduced when Lcn2−/− DCs were compared with WT. An in vivo CTL assay, using LPS-treated DCs, showed diminished killing ability in mice that had received Lcn2−/− DCs compared with WT DCs. As a consequence, we analysed T-cell proliferation and found that LCN2 participates in T-cell-priming in a dose-dependent manner and promotes a TH1 microenvironment. DC-secreted LCN2, whose function has previously been unknown, may in fact have an important role in regulating the balance between TH1 and TH2. Our results yield insights into DC-secreted LCN2 activity, which could play a pivotal role in cellular immune therapy and in regulating immune responses.


Introduction
Dendritic cells (DCs) are professional antigen-presenting cells that coordinate innate and adaptive immune responses [1]. They are also the major source of cytokines, which can modulate effector cells. Because of these remarkable properties, DCs are valuable tools when developing vaccination strategies against tumours [2]. To improve their therapeutic use, it is important to understand their biology and how DCs regulate innate and adaptive immune responses in the tumour microenvironment [3,4]. The results of recent studies suggest that DCs follow a multistage regulation program after encountering danger signals, which facilitates the fine-tuning of the immune response: tolerance maintenance, immune-stimulation and immune-suppression are time-dependent-maturation-functions of DCs [5]. Immature DCs patrol the body against pathogens and are active in endocytosis and in maintaining tolerance [2,6]. However, when DCs encounter danger signals, such as pathogen-associated molecular patterns (PAMP) [7,8,9], they mature [10] with concomitantly increased expression of costimulatory molecules, cytokines [11,12] and up-regulation of the major histocompatibility complex (MHC class I and II), all of which are immune-stimulators. Finally, during the late phase of DC maturation, DCs switch to a suppressive phenotype, which is characterized by the expression of molecules such as indoleamine 2,3-dioxygenase (IDO) [13,14,15], the soluble IL-2 receptor alpha molecules (sIL2RA, sCD25) [16] and IL-10 [17,18], all immune-suppressors.
In earlier experiments, we identified high expression of LCN2 by DCs treated with dexamethasone (Dex) and LPS in a genomewide expression analysis [19]. Glucocorticoids (GC) are immunosuppressive and anti-inflammatory drugs widely used to treat autoimmune diseases or allergies and to enhance or inhibit targetgene transcription. When treated with GC, DCs acquire a tolerogenic phenotype [20].
We wanted to determine the role of LCN2 when secreted by DCs. LCN2 is a small glycoprotein involved in a number of biological processes such as acute phase responses (APR) [21,22], autocrine apoptosis of pro B-cells and IL-3-dependent bone marrow cells [23,24], tumourigenesis [25,26,27,28], and host defence against bacteria through the capture of iron-loaded siderophores [29,30]. Because LCN2 is highly secreted after treatment with Dex and during late maturation, our hypothesis was that LCN2 is involved in the immune suppressive phase of DCs by inducing T-cell depletion. We therefore treated bone marrow-derived DCs with LPS and LPS+IFN-c with and without Dex, to study the role of LCN2 in T-cell apoptosis in DC/T-cell interaction, using ovalbumin-specific TCR transgenic (OT-I and OT-II) mice. We then investigated in vivo T-cell suppression by means of a cytotoxic T-lymphocyte (CTL) killing assay using LPS treated Lcn2 2/2 or WT DC immunisation. Finally, we monitored the in vivo microenvironment that had evolved from DC inoculation. Our results identified LCN2 as a previously unrecognized molecule involved in the regulation of the balance between T H 1 and T H 2. This has important implications for cellular immune therapy against cancer, allergies, and diseases such as type-1 diabetes.

Ethics Statement
All animal experiments were approved by the Institutional Review Board of the Medical University of Vienna and the Ministry of Sciences (BMWF-66.009/0323-11/3b/2012).

Mice
Pathogen-free 8-12 week-old female C57BL/6 wild-type (WT) and Lcn2 2/2 mice were used for experiments. Lcn2 2/2 mice [29] were kindly provided by Akira, University Osaka, and were backcrossed for ten generations to a C57BL/6 background. Lcn2 2/2 and C57BL/6 were bred and housed at the animal facility of the Medical University of Vienna. OT-I and OT-II transgenic mice were housed at the animal care unit of the Department of Pharmacology, Medical University of Vienna, Austria.
T-cells were stained with anti-CD3, anti-CD4, anti-CD8, anti-CD25 and anti-Va2 TCR for OT-I and OT-II mice (eBioscience). Cell viability was analysed with DAPI (Sigma-Aldrich). Apoptosis was measured with Annexin V (BD Pharmingen). CFSE (7 mM, Invitrogen) or Cell Proliferation Dye eFluor 670 (CPD, 5 mM, eBioscience) were used to detect proliferation. Flow cytometry was done on an LSR II (BD Pharmingen). Data were analysed by FlowJo (Version 9.6.2 Treestar). The difference in apoptosis induction was calculated using absolute cell number, determined with BD Trucount tubes.
This method is based on lyophilized pellet, containing a known number of fluorescent beads, which dissolves once the monoclonal antibody reagent is added. Absolute numbers (cells/ml) of positive cells in the sample are calculated following the equation: number of cell events/number of bead events x Trucount bead concentration.
CD8 + and CD4 + T lymphocyte isolation CD8 + and CD4 + T-cells were isolated from the spleens of OT-I or OT-II mice using MACS CD8 + or CD4 + Cell Isolation Kit (Miltenyi Biotech GmbH, Germany). T-cell purity of the enriched CD8 + or CD4 + was evaluated by flow cytometry.

RNA isolation
RNA was isolated using Trizol (Sigma-Aldrich) and reverse transcribed to cDNA using the High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Life technologies, Austria).

Quantitative Real Time PCR
Primers for Lcn2 (Mm 01324470_m1), Lcn2R (24p3receptor, Mm 00480680_m1), Megalin (Mm 01328171_m1) and b2microglobulin (Mm 00437762_m1, Applied Biosystems) were used for RT-PCR. Amplifications were done using a 7500 Fast Real-Time PCR system cycler according to the manufacturer's instructions. Gene expression was calculated following the equation 2 2DDCT = [(C T gene of interest -C T internal control) sample A -(C T gene of interest -C T internal control) sample B] [32].

Intracellular staining
Five replicates of splenocytes (10 5 ) were plated in 200 ml of medium supplemented with 0.05 mg/ml PMA (Sigma-Aldrich), 0.001 mg/ml ionomycin calcium salt (Sigma-Aldrich) and 0.2 ml BD GolgiPlug (brefeldin A, BD Pharmingen). As control, the cells were stimulated with PMA and Ionomycin without GolgiPlug. After 4 h of incubation at 37uC and 5% CO 2 , 100 ml of the supernatant was frozen at 220uC for ELISA. The cells were first stained with anti-CD8a and anti-CD4 before following the instruction for cytokines' intracellular staining with a mix of anti-IL-2, anti-IL-4, anti-IL-10, anti-IL-17, anti-IL-22 and anti-IFN-c, or intracellular staining for transcription factors with a mix of anti-Foxp3, anti-T-bet, anti-ROR-ct and anti-Gata-3 (eBioscience). The staining of intracellular cytokines and enzymes of CD8 + T-cells was done with a mix of anti-IL-2, anti-IFN-c, anti-granzyme B and anti-perforin (reagents and isotypes from eBioscience). For FACS analysis, cells were resuspended in 50 ml PBS (1x).

ELISPOT
Elispot analysis was performed after 3 days of DC/T-cell coculture. After that CD4 + /OT-II T-cells were counted and plated again over night at 37uC in a 96-well/ELISpot-plate (Cellular Technology Limited, CTL, ImmunoSpot, Cleveland, USA), with 6 h-LPS-treated DC-OVA, just OVA or TRP as controls, and PMA+Ionomycin. To monitor unspecific signals and background, T-cells, DC-OVA, DC-TRP, Lcn2 2/2 DCs and WT DCs, were also co-cultured. We used ELISpot plates (Millipore, Massachusetts, USA) for IFN-c, IL-10 and IL-17A (from Mabtech, Sweden), following the ImmunoSpot protocol and recommendations. Spots were counted using Immunospot 5.0 software. Histograms are the results of the spots from T-cell+WT DC-OVA or T-cell+Lcn2 2/2 DC-OVA calculated after subtracting the spots of the controls.

Statistical Analysis
Data are expressed as mean 6 SEM. Two groups were compared using unpaired student's t-test and multiple comparisons using two-way ANOVA followed by Bonferroni post-tests. A p value ,0.05 was considered as statistically significant (*p,0.05; **p,0.01; ***p,0.001; ****p,0.0001). The obtained data were analysed by GraphPad Prism, version 5.0.

LCN2 is not affecting DC maturation
LCN2 is widely expressed by epithelial cells, fibroblasts, macrophages, T-cells [23] and DCs [19], and it is also present in the uterus during involution, a period of extensive apoptosis [22]. To find out whether LCN2, when secreted by DCs, participates in tolerance induction by T-cell clonal deletion, we first monitored and compared WT and Lcn2 2/2 DC maturation and then confirmed LCN2 mRNA and protein expression in bone marrow-derived DCs [31] by using quantitative real-time PCR and ELISA. Immature DCs were treated with LPS and LPS+IFNc, with or without Dex, and maturation was monitored by flow cytometry. Costimulatory molecules and the MHC class I and II increased over time after LPS stimulation. We observed a delayed up-regulation of MHC class II during the first 6 hours in the Lcn2 2/2 DCs, which reached a similar level in WT and Lcn2 2/2 DCs after 24 hours ( figure 1A). Thus, LCN2 does not disturb DC maturation and antigen presentation. Maturation was reduced by pre-treatment with Dex (figure 1B) as shown by Matyszak et al. [35]. IFN-c was used to mimic human monocyte-derived DCs that had been stimulated with LPS+IFN-c to increase the priming of T-cells [11,36]. Interestingly, when Dex was added prior to LPS+ IFN-c, human monocyte-derived DCs did not fully mature (figure S1). This is noteworthy because DC maturation delay is the first step in immune suppression, and GC, like Dex, are used in clinical trials for the treatment of cancer, allergies and autoimmune diseases. In mice, as expected [19], Lcn2 was highly transcribed by cells that had been pre-treated with Dex ( figure 1C). The increased expression after Dex+LPS stimulation has been confirmed in other cell types, as, e.g, in deactivated macrophages from the lung [37]. Secreted LCN2 levels steadily rose over time (figure 1C). LCN2 was also found when DCs were treated only with LPS or LPS+ IFN-c (figure 1C). After 6 hours of treatment, we monitored the increase in secreted levels of the proinflammatory cytokines IL-1a, IL-6, IL-12, TNF-a and IFN-c. Although differences were observed between WT and Lcn2 2/2 DC in the secreted levels of these cytokines (figure 1D), they were not statistically relevant. This result proved that Lcn2 2/2 DC function and secretion are comparable to WT. Our data show that LCN2 is expressed by bone marrow-derived DCs; it is highly secreted when Dex is added prior to other stimulants and the secreted level increases over time. As with mouse DCs, human monocyte-derived DCs did not undergo full maturation with Dex. To explain this observation, we hypothesize that LCN2, which is highly expressed during the late phase of DC maturation, promotes the suppressive phenotype of DCs.
The absence of LCN2 results in reduced CD8 + T-cell apoptosis LCN2 induces autocrine apoptosis in activated T-cells and in pro B-cells when IL-3 is withdrawn from the culture medium [23]. In order to investigate the induction of T-cell apoptosis mediated by the LCN2 secreted by DCs, we set up co-cultures with WT or Lcn2 2/2 DCs and T-cells from OT-I or OT-II transgenic mice.   OVA-loaded DCs were differently stimulated for 6 hours in order to obtain distinct amounts of secreted LCN2. The DC/T-cell cultures were analysed after 48 and 72 hours. T-cells were stained with CD3 + , CD4 + or CD8 + antibodies, and analysed for apoptosis using Annexin V and DAPI to identify dead cells ( figure 2A). During the first 48 hours, CD8 + T-cells underwent apoptosis when the DCs were treated with Dex, which resulted in an increase in secreted LCN2 ( figure 1C). However, T-cell apoptosis was highly reduced in co-cultures with Lcn2 2/2 DCs. After 48 hours we observed higher apoptosis in the presence of LCN2 for OT-I, while the difference was not significant for OT-II. CD4 + T-cells did not appear to be affected by apoptosis in the Dex+LPS (DL) condition. After 72 hours the effect of Dex was less evident (figure 2B). Our results indicate that induction of apoptosis is reduced when T-cells are co-cultured with Lcn2 2/2 , compared with WT DCs, and the response differs between OT-I and OT-II T-cells. We then measured the amount of secreted LCN2 in the co-cultures with WT and Lcn2 2/2 DCs ( figure 2C, figure S2). We found that in OT-I, LCN2 expression was less than in OT-II, and that CD8 + T-cells seemed to take up higher amounts of LCN2 (figure 2C). We therefore analysed the expression of Lcn2 receptors in OT-I and OT-II cells by qRT-PCR, Slc22A17 (known as Lcn2R) and Megalin (not found in our system). We found that CD8 + T-cells expressed Lcn2R but CD4 + T-cells did not ( figure 2D). We then decided to analyse if the pro-apoptotic protein Bim [24] was differently expressed, and hence stimulated CD8 + and CD4 + T-cells with recombinant LCN2 and dynabeads (CD3CD28). We found no difference in Bim expression between CD8 + and CD4 + T-cells (data not shown).
Taken together, there is a correlation between high LCN2 expression and elevated apoptosis for CD8 + T-cells but not for CD4 + T-cells. We concluded that the apoptosis response is T-cellspecific, and one possible explanation for this is the different receptor expression levels in T-cells.

LCN2 DCs do not impair the CD8 T-cell response
To test whether LCN2 affects the in vivo CD8 + T-cell response, we set up a cytotoxic T-lymphocyte killing assay (CTL), using DC immunisation. OVA-loaded DCs were stimulated for 6 hours with LPS and delivered s.c. to syngeneic WT mice. Dex was not used because, based on our protocol, it prevents the up-regulation of CCR7 cytokine, which is responsible for the DC migration to lymph nodes [19]; this is in contrast to Anderson et al. and Volchenkov et al. who used another treatment condition [38,39]. Three days after immunisation, mice received splenocyte-target cells loaded with distinct OVA-peptides in combination with different concentrations of fluorochrome. The draining lymph nodes were harvested after 6 and 18 hours and CFSE positive cells were quantified (figure 3A). Three peaks were detected in the control mice. Mice that had received DC immunisation showed a significant reduction of the SIINFEKL peak (Lcn2 2/2 DCs 6.4%, WT DCs 2.4% remaining cells, figure 3B). We observed a reduced killing response in mice receiving Lcn2 2/2 DCs (87% 62.0%) compared with WT DCs (96% 60.5%, p,0.01). The WT DC immunisation induced a higher cytotoxic response. Based on these results, we hypothesized that immunisation with Lcn2 2/2 DCs resulted in either improper priming of T-cells, or the development of CD4 + regulatory T-cells. To establish which hypothesis was correct we analysed T-cell proliferation and microenvironment ( figure 4 and figure 5).
When the experiment was repeated in Lcn2 2/2 mice, the killing response was reduced considerably as compared to WT mice, going from 87% to 40% (62.8) with Lcn2 2/2 DCs, and from 96% to 55% (68.1) with WT DCs ( figure 3B bottom). Although the trends were the same as for WT mice, the functionality and lytic ability were significantly reduced in Lcn2 2/2 mice (p, 0.0001 figure S3). In vitro intracellular staining for granzyme B and perforin of OT-I T-cells co-cultured with WT or Lcn2 2/2 DCs showed a trend not statistically relevant (figure 3C), indicating that LCN2 probably does not affect the cytotoxic ability of CD8 + Tcells. To test this conclusion for the in vivo condition, we used SIINFEKL 257-264 as peptide together with the adjuvant CpG [40,41] to immunise WT and Lcn2 2/2 mice for 7 days before inoculating target cells and 18 hours later measuring the killing effect. No difference was found between the WT and Lcn2 2/2 responses, antigen presentation and cytotoxic activity were not damaged by the absence of LCN2 ( figure 3D). In summary, LCN2 does not affect the killing function of CD8 + T-cells. The large difference between WT and Lcn2 2/2 recipient mice led us to speculate that other cell types mediate the response we observed with DC immunisation.

LCN2 expressing DCs promote T-cell proliferation
LCN2 is known to be an iron-binding protein that can deliver iron to cells when LCN2, bound to the siderophore-iron complex (holo-Lcn2), is internalized. When the iron complex is lacking (apo-Lcn2), LCN2 is internalized, chelates iron and reduces its intracellular concentration [24]. Since iron concentration is important for cell proliferation, we sought to verify the ability of Lcn2 2/2 DCs to stimulate T-cells. We observed that WT DCs were able to activate both CD8 + and CD4 + T-cells, whereas Lcn2 2/2 DCs showed a significantly impaired T-cell division at 1:1 but not 1:10 DC/T-cell ratio ( figure 4A). This result might seem incongruous, but it is consistent with T-cells playing an active role in their own proliferation by producing IL-2 (data not shown). Independently of the cell ratio, apoptosis was always reduced when Lcn2 2/2 DCs were used (figure 4B), but when recombinant LCN2 was added to Lcn2 2/2 DC/T-cell co-culture, apoptosis increased again (figure 4C). A similar result was obtained using conditioned media from the WT DCs ( figure S4). To find out whether T-cell proliferation depends on iron-transportation, we used a chelating agent, deferoxamine to remove iron from the coculture and found that proliferation was highly reduced, while apoptosis increased, especially in the Lcn2 2/2 DC/T-cell coculture ( figure 4D). We conclude that LCN2 contributes to the proliferation of T-cells, which could explain the reduced killing ability of the Lcn2 2/2 in the in vivo CTL assay. selected based on the CFSE staining. On the left, the control group received just PBS, in the middle mice immunised with 6 h-LPS-treated Lcn2 2/2 DCs and on the right, mice immunised with 6 h-LPS-treated WT DCs. Histograms show the specific killing effects using Lcn2 2/2 or WT DC immunisation. These data show the most representative of pooled mice (5 mice/group) performed 3 independent times and shown as mean 6 SEM, p value is **p,0.01. (C) FACS analysis of perforin and granzyme B intracellular staining in OT-I cells after 72 h of WT or Lcn2 2/2 DC/T-cell co-culture. DCs were pre-treated for 6 h with LPS. Histograms show the percentage of positive cells in the co-culture. (D) Comparison of in vivo CTL assays in WT and Lcn2 2/2 recipient mice immunized with SIINFEKL 257-264 and CpG peptides for 7 days. These data show the most representative experiment of pooled mice (5 mice/group) performed 3 independent times. doi:10.1371/journal.pone.0101881.g003 CD4 + T-cell differentiation depends on the stimulus, the cytokine milieu [42] and the strength of stimulation [43]. Therefore, we monitored the microenvironment after 3 days of DC immunisation and a further 6 and 18 hours of target-cell inoculation. The splenocytes were stained to detect transcription factors for regulatory T-cells (T Reg ), T helper 1 (T H 1) and T helper 17 (T H 17), T helper 2 (T H 2) and the expression of various cytokines. WT mice receiving Lcn2 2/2 DCs showed slightly elevated levels of FoxP3 expression compared with mice receiving WT DCs (p,0.05), whereas T-bet, ROR-ct and Gata-3 showed no statistical change (figure 5A, figure S5). IL-2, IL-4, IL-5, IL-10, IL-12p70, as well as IFN-c, were measured upon restimulation of the splenocytes with PMA and ionomycin (figure 5B). The mice receiving WT DC immunisation produced more IL-12p70 and IFN-c after 6 and 18 hours of target-cell inoculation, indicating a tendency to induce a T H 1 response. The IFN-c secretion was also calculated using an Elispot assay to quantify the amount produced by the responding cells ( figure 5C). Mice receiving Lcn2 2/2 DCs showed increased expression of IL-2, IL-4, IL-5 and IL-10 (after 18 hours, figure 5B). Intracellular levels of IL-17 and IL-22 showed no statistical change after 6 hours (figure S6) but increased at 18 hours especially in mice that received Lcn2 2/2 DCs, indicating a delayed response. In addition, we calculated the ratio of cytokine expression between WT and Lcn2 2/2 DC immunisation (see figure 5D legend). The ranking order supports the observation that T-cell reprogramming was occurring, because the DCs induce two distinct environments: a T H 1 environment in the case of WT DC inoculation and a T H 2 milieu in the case of Lcn2 2/2 DC immunisation. We concluded that, amongst other molecules, LCN2 promotes a T H 1 response as opposed to the T H 2 response that is favoured in its absence.

Discussion
Professional antigen-presenting cells are pivotal in the regulation of innate and adaptive immune responses [1,11,12]. Specifically for this characteristic, DCs have been exploited as vaccines against cancer [44,45]. The aim of DC vaccination is to induce tumour-specific effector T-cells to reduce tumour mass or induce prolonged and increased inflammation against tumour antigens. In this context, Nava et al., 2012, investigated the effect of DC vaccination against glioblastoma [46]. LCN2, an APR protein, may have a major role in DC/T-cell processes, given its abundant secretion by DCs after being treated with Dex and, subsequently, with LPS [19]. This secretion is not restricted to LPS stimulation but is also found with Dex in combination with IFN-c. In our study, cells were treated with IFN-c to mimic the human situation, where IFN-c is used to trigger DC maturation, and we showed that LCN2 is highly expressed when Dex is added prior to other stimuli.
In a tumour milieu LCN2 is potentiated by proinflammatory cytokines Il-1b, TNF-a, and IL-17 [47] and is involved in the epithelial to mesenchymal transition (EMT) [48,49]. LCN2, however, does have controversial roles: it is an APR [21], it plays a role in apoptosis [23] and it is bacteria-static by sequestering iron through its siderophores [29]. For all these reasons, it is important to understand the role of LCN2 when it is highly secreted by DCs.
We therefore monitored the kinetics of LCN2 secretion by the bone marrow-derived DCs. We found that DCs express LCN2 independently of subsets or culture conditions (GM-CSF alone or GM-CSF in combination with IL-4) [19], whereas mouse and human DCs efficiently secrete IL-12 (which is important for T-cell priming and T H 1 response) only when previously exposed to IL-4 [50,51]. The amount of secreted LCN2 is greatly increased by DCs incubated with Dex (DEX+LPS and DEX+LPS+IFN-c). By contrast, IFN-c does not seem to play a role in LCN2 release and, when used in combination with LPS, IFN-c seems to reduce LCN2 secretion. This observation is consistent with the observed increased expression of LCN2 by macrophages infected with Salmonella Typhimurium when incubated with IFN-c [52] but not when incubated with LPS. We thus conclude that LCN2 has a cell-and content-dependent function, which has also been welldocumented in the mouse intestinal tumour model APCmin [53].
Before using WT and Lcn2 2/2 DCs for in vivo experiments, we monitored in vitro the maturation and the expression of proinflammatory cytokines. The results showed that Lcn2 2/2 DCs are as functional as WT DCs and that their secretion of proinflammatory cytokines is higher without Dex treatment. We then studied the role of LCN2 secreted by DCs in the induction of Tcell apoptosis. We showed that LCN2, when highly secreted by DCs, plays a role in promoting the apoptosis of T-cells in a manner that is CD8 + T-cell-specific. This might be due to the expression of the Lcn2 receptor on the T-cell surface [54]. LCN2 binds its Lcn2R and is internalized to deliver iron, which is important for the survival of T-cells. CD8 + T-cells expressed Lcn2R but CD4 + T-cells did not. We speculate that, together with iron-homeostasis, apoptosis occurs when there is an unequal distribution of LCN2 or Lcn2R. Most likely other molecules participate in apoptosis in the presence of Dex. Consistent with this proposed role of LCN2, we found that supplementing the Lcn2 2/2 DC/T-cell co-culture with recombinant LCN2 resulted in increased T-cell apoptosis.
From our immunisation studies with DCs, we uncovered a subtle role of LCN2 in inducing both T-cell apoptosis and proliferation, and found that Lcn2 2/2 DCs impaired T-cell priming ability. The cytokine microenvironment -as compared to Lcn2 2/2 DC immunisation -also showed a distinct milieu when generated by WT DC. The WT DC inoculation induced a T H 1 response through the enhanced secretion of IFN-c and IL-12p70; conversely, Lcn2 2/2 DCs triggered a T H 2 response by releasing more IL-2, IL-4, IL-5 (which increased 10.000 fold within 12 hours), and IL-10 cytokines. Our data demonstrated that LCN2 plays a role in CD8 + T-cell apoptosis mediated by DCs, especially when pretreated with Dex+LPS. LCN2 was involved in T-cell priming after LPS stimulation in a ratio-dependent manner. DC immunisation was highly efficient: when using WT DCs, LCN2 contributed to an increase in the DCs' anti-tumour effectiveness by (A) Histograms represent CD8 + T-cell divisions. Activated T-cells were stained and gated on cells positive for CD8 + CD25 + . Generation 0 was the undivided population initially labelled with CPD and all other generations were the result of the sequentially halved CPD fluorescence. The results were analysed by the proliferation tool of FlowJo. (B) Analysis of the T-cell apoptosis at 1:1 (10 5 :10 5 ), 1:5 (2610 4 :10 5 ) and 1:10 (10 4 :10 5 ) DC/T-cell ratios. T-cells (OT-I) were co-cultured with 6 h-LPS-treated Lcn2 2/2 DCs (black bars) or 6 h-LPS-treated WT DCs (grey bars). Five independent experiments were performed in triplicate. Mean 6 SEM, *p,0.05, ****p,0.0001. (C) Analysis of T-cell apoptosis after adding Recombinant Lcn2 (Rec Lcn2) in DC/T-cell co-culture. Lcn2 2/2 DCs, pre-treated with LPS for 6 h, were co-cultured 72 h with T-cells in medium supplemented with Rec Lcn2. Five independent experiments were performed in triplicate. Mean 6 SEM, *p,0.05, ***p,0.001. (D) Proliferation and apoptosis of T-cells (OT-I) was determined after co-culture in the absence or presence of desferrioxamine (1:1, 1:5 and 1:10 ratio) for 72 h with DCs that had been previously incubated for 6 h with LPS. The two groups were analysed by ANOVA and shown as mean 6 SEM, *p,0.05, ***p,0.001. doi:10.1371/journal.pone.0101881.g004 inducing a T H 1 microenvironment phenotype. Physiologically, this tended to develop towards a cell-mediated, rather than a humoral or antibody-mediated response. Clearly, it would be of great benefit to the cellular immune therapy treatment of cancer and diseases such as multiple sclerosis, type-1 diabetes, asthma and allergies to regulate the balance between T H 1 and T H 2 [43,55]. While our data may be descriptive, they nonetheless demonstrate the importance of LCN2 in the DC/T-cell interaction. Further study using tetramer staining as well as on the LCN2 molecular mechanism should be done to unravel the dual role of LCN2 in Tcell commitment.