Regulation of Th1 T Cell Differentiation by Iron via Upregulation of T Cell Immunoglobulin and Mucin Containing Protein-3 (TIM-3)

Iron plays an important role in host–pathogen interactions, in being an essential element for both pathogen and host metabolism, but also by impacting immune cell differentiation and anti-microbial effector pathways. Iron has been implicated to affect the differentiation of T lymphocytes during inflammation, however, so far the underlying mechanism remained elusive. In order to study the role of iron in T cell differentiation we here investigated how dietary iron supplementation affects T cell function and outcome in a model of chronic infection with the intracellular bacterium Salmonella enterica serovar typhimurium (S. Typhimurium). Iron loading prior to infection fostered bacterial burden and, unexpectedly, reduced differentiation of CD4+ T helper cells type 1 (Th1) and expression of interferon-gamma (IFNγ), a key cytokine to control infections with intracellular pathogens. This effect could be traced back to iron-mediated induction of the negative immune checkpoint regulator T cell immunoglobulin and mucin domain-containing protein 3 (TIM-3), expressed on the surface of this T cell subset. In vitro experiments demonstrated that iron supplementation specifically upregulated mRNA and protein expression of TIM-3 in naïve Th cells in a dose-depdendent manner and hindered priming of those T cells towards Th1 differentiation. Importantly, administration of TIM-3 blocking antibodies to iron-loaded mice infected with S. Typhimurium virtually restored Th1 cell differentiation and significantly improved bacterial control. Our data uncover a novel mechanism by which iron modulates CD4+ cell differentiation and functionality and hence impacts infection control with intracellular pathogens. Specifically, iron inhibits the differentiation of naive CD4+ T cells to protective IFNγ producing Th1 lymphocytes via stimulation of TIM-3 expression. Finally, TIM-3 may serve as a novel drug target for the treatment of chronic infections with intracellular pathogens, specifically in iron loading diseases.

Iron plays an important role in host-pathogen interactions, in being an essential element for both pathogen and host metabolism, but also by impacting immune cell differentiation and anti-microbial effector pathways. Iron has been implicated to affect the differentiation of T lymphocytes during inflammation, however, so far the underlying mechanism remained elusive. In order to study the role of iron in T cell differentiation we here investigated how dietary iron supplementation affects T cell function and outcome in a model of chronic infection with the intracellular bacterium Salmonella enterica serovar typhimurium (S. Typhimurium). Iron loading prior to infection fostered bacterial burden and, unexpectedly, reduced differentiation of CD4 + T helper cells type 1 (Th1) and expression of interferon-gamma (IFNg), a key cytokine to control infections with intracellular pathogens. This effect could be traced back to iron-mediated induction of the negative immune checkpoint regulator T cell immunoglobulin and mucin domaincontaining protein 3 (TIM-3), expressed on the surface of this T cell subset. In vitro experiments demonstrated that iron supplementation specifically upregulated mRNA and protein expression of TIM-3 in naïve Th cells in a dose-depdendent manner and hindered priming of those T cells towards Th1 differentiation. Importantly, administration of TIM-3 blocking antibodies to iron-loaded mice infected with S. Typhimurium virtually restored Th1 cell differentiation and significantly improved bacterial control. Our data uncover a novel mechanism by which iron modulates CD4 + cell differentiation and functionality and hence impacts infection control with intracellular pathogens. Specifically, iron inhibits the differentiation of naive CD4 + T cells to protective IFNg producing Th1 lymphocytes via stimulation of TIM-3 expression. Finally, TIM-3 may serve as a novel drug target for the treatment of chronic infections with intracellular pathogens, specifically in iron loading diseases.

INTRODUCTION
Because of its high redox activity, iron is a key component of several enzymatic processes. Virtually every cell of the body requires iron for metabolism and proliferation. In mammals most iron for the daily needs is deliverd by macrophages which ingest aged or damaged red blood cells (1). After phagocytosis iron is extracted from erythrocyte heme and exported from the macrophage via the iron-exporter ferroportin-1 to the circulation, where it is bound to transferrin, and taken up by cells via transferrin receptor-1 (TfR-1) (2). This uptake is thus of high relevance for the differentiation of most cells including lymphocytes (3,4). Since iron is crucial for both microbes and mammalian cells, iron homeostasis undergoes subtle changes during infection and inflammatory processes which is mediated by various mechanisms (5). Moreover, iron availability affects lymphocyte proliferation and function (6)(7)(8). In human T lymphocytes, iron induces refractoriness to IFNg/STAT1 (Signal transducer and activator of transcription 1) signaling via involvement of the TfR-1 (9). The TfR-1 regulates IFNg signaling in activated T cells by interacting with the T cell receptor (TCR) after being recruited to the immunological synapse in response to TCR activation (10). Activation of the TCR causes naïve CD4 + T cells to proliferate and differentiate into different subsets [Th type 1 (Th1) cells, Th2, Th17, Th9, Th22, and regulatory T cells (Treg)] (11), depending on the specific cytokine milieu.
As a consequence of persisting antigenic stimulation and inflammation, like in chronic infections and cancer, T cell exhaustion develops as a state of cellular and immunological dysfunction (12). Exhausted T cells are characterized by reduced cytokine production and over-expression of a distinct set of inhibitory receptors like programmed cell death protein 1 (PD-1), cytotoxic T-lymphocyte-associated Protein 4 (CTLA-4), Lymphocyte-activation gene 3 (Lag-3), and T cell immunoglobulin and mucin containing protein-3 (TIM-3) (12). Several studies demonstrated a key role of TIM-3 in T cell dysfunction and exhaustion (13)(14)(15) and in activating signaling cascades leading to the regulation of immune cell functions (16)(17)(18)(19)(20).
Whereas inhibitory receptors on lymphocytes such as PD-1, CTLA-4, Lag-3, T cell immunoreceptor with Ig and ITIM domains (TIGIT-1), and TIM-3 have been identified as crucial components mediating T cell exhaustion in cancer and viral infections (21), the role in chronic bacterial infections is largely elusive. However, upon infection with Mycobacterium tuberculosis Tim-3 knockout mice show an improved survival (22), pointing to the importance of TIM-3 in bacterial infections.
Salmonella enterica serovar Typhimurium (S. Typhimurium) is a Gram negative, facultative intracellular bacterium which causes typhoid fever like infections in mice. Many studies with attenuated bacterial strains have emphasized an essential role for Salmonella-specific CD4 + Th1 cells as important host response mechanisms to intracellular pathogens (23,24). Of note, Salmonella are siderophilic bacteria and iron loading is associated with increased bacterial multiplication and impaired immune control of infection (25).
Here, we provide evidence that detrimental effects of high systemic iron on the integrity of bacterial immune defense in a model of chronic S. Typhimurium infection can be traced back to a strong iron-mediated upregulation of TIM-3 expression in helper T cells and an impaired differentiation into protective IFNg producing Th1 lymphocytes. Administration of TIM-3 blocking antibodies to iron-loaded animals restored the Th1 cell expansion and greatly reduced bacterial burden, indicating that the negative checkpoint modulator acts as a crucial, irondependent regulator of T cell immune response to bacterial pathogens.

Infection of Mice
Wildype Salmonella enterica serovar typhimurium (S. Typhimurium) strain ATCC14028 was used for experiments and grown under sterile conditions in LB broth (Sigma-Aldrich) to late-logarithmic phase. For the infection model male mice were fed with different iron diets for two weeks and during the course of the infection. Low iron diet had an iron content of ≤9 mg iron/kg diet (26), high ion diet contained 5 g iron/kg diet (both diets from Altromin). Male mice were used at 8-12 weeks of age and infected intraperitoneally with 500 CFU of S. Typhimurium in 200 ml PBS (27)(28)(29). After 14 days of infection mice were euthanized by cervical dislocation, spleens and livers were isolated, erythrocytes lysed, and flow cytometry was performed. The bacterial load of organs was determined by plating serial dilutions of organ homogenates on LB agar (Sigma-Aldrich) under sterile conditions and the number of bacteria calculated per gram of tissue.

Flow Cytometry Analysis
Spleens were homogenized through 100 µm nylon cell strainer (Falcon) and red blood cells were lysed by incubation in ACK buffer (150 mM NH 4 Cl, 10 mM KHCO 3 , 0.1 mM Na 2 EDTA) for 2 min at room temperature.

Iron Measurement
Serum iron concentrations are measured with a colorimetric iron quantification kit (QuantiChrom Iron Assay Kit, BioAssay Systems) following the manufacturer's instructions. Tissue iron wa s qu antifi ed using a colorimetric method with bathophenanthroline disulfonic acid (35). In brief, organ lysates were hydrolyzed with acid for 24 h at 65°C, mixed with a colorimetric solution containing sodium acetate, bathophenanthroline disulfonic acid and l-ascorbic acid and absorbance at 539 nm was measured. The iron content of the organ was calculated from a standard curve and normalized to the protein content of the lysate determined by the Bradford method.

BrdU Labeling of Splenocytes
Splenocytes were cultured as described before and pulsed with 10 µM BrdU (Sigma-Aldrich) 4 h before harvesting. Intracellular staining for BrDU with surface co-staining for CD3, CD4 and CD8 was performed with BrdU Flow Kit (BD) according to the manufacturers`instructions and cells were analyzed with flow cytometry. Iron sources ferric chloride FeCl 3 , ferric sulfate Fe 2 (SO 4 ) 3 , ferric citrate FeC 6 H 5 O 7 were added at indicated concentrations.

Statistics
Statistical analysis was generated using Prism GraphPad software (Version 7). Significance was determined by unpaired two-tailed t test to assess data, when only two groups were compared. For multiple comparisons Analysis of variance (ANOVA) combined with Tukey`s post test was performed. P values less than 0.05 were considered as statistically significant in any test.

Iron Supplementation Inhibits T Cell Priming and Expansion and Upregulates the Immune Checkpoint Regulator TIM-3
Based on previous observations that iron loading affects Th1 and Th2 cell differentiation (36,37) and the regulatory effects of TIM proteins on T cell priming (38) we hypothesized a possible interconnection between those pathways.
First, iron supplementation of anti-CD3-primed splenocytes isolated from naive mice induced a slight decrease in the proliferation and dramatically slowed cell cycle progression of  T cells as demonstrated by CFSE dilution ( Figure 1A) and BrdU pulsing ( Figure 1B) assays even at concentrations as low as 2.5 to 5 µM (39). Both effects were independent of the iron source (chloride, sulfate and citrate) and, hence, independent of iron source-inherent differences in solubility, dissociation properties, and cellular iron availability. In parallel, we found a strong upregulation of both transcript and surface protein levels of TIM-3 in the total CD4 + T cell fraction primed with anti-CD3/  CD28 and supplemented with iron, as compared to controltreated cultures ( Figures 1C, D). Again, this upregulation was observed with different iron supplements and occurred in a dosedependent manner. Notably, the same mode of regulation of Tim-3 mRNA and protein levels was observed in naive Th cells differentiated into Th1 lymphocytes in presence of iron ( Figures  1E, F). Of note, surface abundance of another checkpoint regulator, PD-1, was comparable in control-and ironstimulated Th1 differentiation cultures ( Figure 1G).

In Vivo Dietary Iron Supplementation Impairs Pathogen Control, Upregulates TIM-3 and Reduces Th1 Differentiation
We then investigated if the strong inhibitory effects of iron on T cell priming and the stimulation of the checkpoint regulator TIM-3 can be recapitulated in vivo and if this affects antibacterial defense in a model of chronic bacterial infection.
To this end we used C57BL/6 mice with transgenic expression of a functional natural resistance associated macrophage protein 1 (NRAMP1 or SlC11A1), which results in an improved host resistance to infections with S. Typhimurium and thus prolonged bacterial infection allowing to study T cell responses over time (Figure 2, Supplementary Figure 1).
Mice were fed with low (<9 mg elementary Fe/kg) (26) or high iron (5 g/kg) diets for two weeks and during the course of the infection. After intraperitoneal infection with 500 CFU of S. Typhimurium (27-29) mice were followed up for 14 days. Since the canonical high iron feeding regime used in the short-term Salmonella infection model (25 g/kg) led to strong weight reduction and pre-term mortality, we decided to induce iron overload with chow containing less iron (5 g Fe/kg), which was sufficient to significantly increase the levels of the metal in the canonical storage organs, liver and spleen, as compared to mice on a low-iron diet (Figures 2A, B). This was accompanied by  (40), was progressively increasing over time ( Figure 3C). Of note, neither spleen cellularity ( Figures 3A, 4A), nor general abundance of this CD4 + T cell subset was changed by dietary iron supplementation ( Figure  3B). The IFNg expressing Th1 subset plays a central role in mounting effective immune protection against S. Typhimurium (40). Dietary iron loading strongly upregulated surface TIM-3 levels in that subset ( Figure 4A, Supplementary Figure 1D) and dramatically reduced frequencies of Th1 cells within splenic helper CD4 + and CD3 + pan-T cells on day 14 post bacteria challenge. This suggested reduced priming and differentiation of Th1 cells and/or their exhaustion by high iron availability ( Figure 4A, Supplementary Figure 1C). In line with the reduced Th1 differentiation, we found significantly decreased percentages of helper T cells expressing T-bet, the key transcription factor orchestrating development of the Th1 phenotype ( Figure 4A). Consequently, at the systemic level, the diminished Th1 cell expansion culminated in reduced circulating concentration of IFNg in iron-fed mice ( Figure  4B). Of note, we did not observe any substantial effects of dietary iron overload on the residual, self-sustaining pool of Th1, Th2, Th17 and Treg lymphocytes in the spleen (Supplementary Figure 4).
Analogically to the in vitro setting, iron loading in vivo had no impact on the levels of another immune checkpoint protein, PD-1, on the surface of helper T lymphocytes (Supplementary Figure 1E), which suggests that the inhibitory effects of the metal are limited to TIM-3 regulation. Importantly, the effects of Means ± SEM are shown in the plots. Statistical significance was assessed by two-tailed T test. *p < 0.05, **p < 0.01. (A) isotype control n = 12; Tim-3 blocking Ab n = 7. (B) spleen isotype control n = 6, Tim-3 blocking Ab n = 7; liver isotype control n = 7, Tim-3 blocking Ab n = 8.
iron were highly Th1 cell specific, as neither the IL-4 + Th2, nor the IL-17A + Th17, nor the regulatory FOXP3 + T cell subsets displayed numerical alterations upon high-iron diet ( Figure 4C). In addition, serum levels of IL-2, the key cytokine for T cell expansion were not changed after iron loading of S. Typhimurium infected mice (Supplementary Figure 2A). Of note, the levels of major inducers of TIM-3 expression, namely IL-12 (31) and IL-27 (32), were not significantly affected by iron challenges (Supplementary Figures 2B, C), suggesting that iron controls TIM-3 mRNA and protein expression rather directly via a cell-intrinsic mechanism and not via modulating the expression of those two TIM-3 inducing cytokines.
So far, our data clearly showed the negative impact of iron overload on Th1 cell differentiation, function, and microbial control. However, in theory these effects could also be attributed to the particularly effective T cell response in the low iron setting or to the extreme difference in iron content between the diets. To account for that, we performed the S. Typhimurium infection experiments in animals fed the low iron (<9 mg/kg), standard (166 mg/kg) and high iron chow (5 g/kg). As shown in Supplementary Figure 5, there was no significant difference neither in Th1 (CD4 + IFNg + ) differentiation or expression of surface TIM-3 in Th1 cells (A, B), nor in abundance of Th2, Treg and Th17 cells (C-E) between the low-and standard iron chow. Conversely, the Th1 cell percentages were significantly higher and their TIM-3 expression was significantly lower in the animals fed standard diet than in iron overloaded mice (Supplementary Figures 5A, B).

Dietary Iron Supplementation Has No Effect on Numbers of Non-Lymphoid Immune Cells in Response to S. Typhimurium Infection
Because TIM-3 is not only expressed on lymphocytes but also on myeloid cells such as monocytes and macrophages (41), we reasoned that innate immune cell numbers may also be affected by iron challenge. As shown in Supplementary Figure  3, we could not find any significant differences in relative frequencies of any of the investigated myeloid populations (neutrophils, macrophages, Ly6C hi classical monocytes and Ly6C lo resident monocytes) attributed to the iron content of the diets after 14 days of infection with S. Typhimurium. These results point towards comparable levels of systemic inflammation in both experimental groups. The percentages of TIM-3-expressing cells in the myeloid leukocytes were found, in general, to be less than half of the expression of TIM-3 in the Th1 lymphocytes and, in addition, were not significantly affected by dietary iron loading (Supplementary Figure 3, Figure 4A).

Administration of TIM-3 Blocking Antibodies Restores Th1 Cell Expansion and Improves Pathogen Control in Iron-Supplemented Mice
We next tested whether neutralization of TIM-3 affects iron induced Th1 cell exhaustion and the impaired immune control of chronic S. Typhimurium infection.
To test that, we administered isotype and blocking antibodies against TIM-3 (17) to iron loaded animals infected with S. Typhimurium. In this setting, blocking TIM-3 significantly increased the numbers of IFNg-producing Th1 lymphocytes within the CD4 + T cell compartment as compared with animals receiving the isotype antibody ( Figure 5A). Moreover, it virtually restored the Th1 differentiation to the levels observed in Salmonella-infected mice fed an iron-low chow ( Figure 4A). Importantly, this intervention had no effect on the total cellularity of the spleens ( Figure 4A). Furthermore, the TIM-3 blockade in vivo improved bacterial control as demonstrated by a significant reduction of bacterial burden in the liver and spleen ( Figure 5B). Collectively, our findings demonstrate that TIM-3 links the negative effects of iron on IFNg production by Th1 lymphocytes to impaired host defense against the intracellular pathogen S. Typhimurium.

Th1 Cells Express the Highest Levels of Cellular Iron Importers Among the Th Subsets
Finally, we sought to investigate the mechanism behind the highly specific TIM-3-mediated impact of iron on Th1 lymphocytes but not on the other CD4 + T cell subsets. The most obvious explanation is a potential difference in uptake of iron by various Th cell classes. To test that, we investigated levels of cellular iron importers in Th1, Th2, Th17, and Treg lymphocytes. Iron can be ingested by the cell in two forms: as transferrin-bound iron (TBI), being the physiological form of the element, and as chemically reactive, potentially toxic nontransferrin-bound iron (NTBI), as for example upon iron overload conditions. Interestingly, among the Th subsets differentiated ex vivo, Th1 cells demonstrated the highest mRNA and surface protein expression of both the TBI importer TfR1 (transferrin receptor 1) and NTBI uptake proteins DMT1 (di-valent metal transporter 1), as well as ZIP14 (Zrt-And Irt-Like Protein 14) (Supplementary Figure  6). This may be related to the superior iron uptake capabilities of Th1 cells and the higher sensitivity to iron compared with other CD4 + T cells.

DISCUSSION
Here we report a novel mechanism by which the increased availability of iron leads to an unfavorable outcome of infections with the intracellular bacterium Salmonella enterica serovar Typhimurium. Iron specifically acts on IFNg-producing CD4 + T cells and stimulates the expression of the negative regulatory surface receptor or 'immune checkpoint' inhibitor TIM-3, thus reducing the functionality of these cells. This is in line with earlier studies indicating that Th1 T cells are exquisitely sensitive to iron perturbations because intracellular iron depletion affects Th1 cells more than the Th2 cell subset (8,36,37). In our model of chronic bacterial infection, the effects of iron were highly specific for Th1 cells as neither the number of Th2, nor Th17 or Treg subsets displayed differences after high iron diet. Interestingly, our previous report provided evidence that the function of the CD8 + IFNg-producing T cell subset in murine mammary carcinoma was strongly hampered by intravenous iron supplementation (39).
Principally, iron can be ingested by cells as transferrin-bound iron, being the physiological form of the element, which is mediated by the TfR-1. Additionally, iron can be taken up as chemically reactive and potentially toxic non-transferrin-bound iron, mediated by the receptors DMT1 and ZIP14. This form of iron ingestion happens e.g. upon iron overload conditions. Our in vitro results showed a preferential expression of iron uptake receptors, both of the transferrin-and non-transferrin bound iron forms, in the Th1 subset which is likely to explain their particular sensitivity to iron overload. Appropriately, the CD4 + T cell subset was described to be the main effector subset in the defense of Salmonella infections (20,40,42).
Importantly, in chronic infections T cell exhaustion is described to develop as a state of cellular and immunological dysfunction as a consequence of persisting antigenic stimulation (12), leading to reduced cytokine production and over-expression of inhibitory receptors. TIM-3, an inhibitory receptor widely expressed on Th1 cells, is described as an important player in T cell dysfunction and exhaustion (12)(13)(14)(15). This supports our finding that iron specifically impacts on TIM-3 expression in Th1 cells during chronic infection. Indeed, we showed that dietary iron loading strongly upregulated surface TIM-3 levels on Th1 cells. A pathophysiological role of this interaction is strongly supported by the finding that anti-TIM-3 treatment improved infection control in iron loaded mice to levels observed in animals on an ironbalanced diet. This is in line with recently published studies, demonstrating the crucial role of TIM-3 over-expression on T cells in the control of infections with intracellular bacteria such as Mycobacterium tuberculosis, where IFNg mediated immune effector pathways play a decisive role (23,43,44). Accordingly, in M. tuberculosis infected mice administration of a TIM-3 fusion protein, acting as a molecular sink for TIM-3 ligands, reduced the bacterial burden (22). In our model, similar to the M. tuberculosis infection model, blocking TIM-3 overcame the detrimental effect of iron on IFNg-producing Th1 cells. Furthermore, it rescued Th1 differentiation, and reduced bacterial burden in livers and spleens of Salmonella infected mice This highlights the role of TIM-3 as a crucial negative regulator of Th1 cell expansion and Th1-mediated bacterial host defense acting downstream of iron in chronic Salmonella infection.
Our findings raise the question how iron impacts Th1 immunity by influencing the expression of TIM-3. According to our in vitro data ( Figure 1) iron dose-dependently increased TIM-3 expression, and blocked cell cycle progression and differentiation of Th0 to Th1 cells. Hence, two non-exclusive mechanisms may be proposed: first, interference of iron with the Th1-specific transcription factor network and second, signaling induced by reactive oxygen species, as postulated in our previous report on iron and anti-tumor CD8 + T cells (39). While iron did not affect the expression of TIM-3-inducing cytokines, IL-12 and IL-27 in vivo, it still could affect signals mediated by those cytokines via the IL-12 receptor or via T-bet, the master switch transcription factor of Th1 cells. This is supported by our data showing similar effects of iron on T-bet and on IFNg expression in Th1 cells. This would be in a line with the description of T-bet as important regulator of TIM-3 on Th1 cells (45). In addition, iron may act on transcription factors other than T-bet such as c-Jun N-terminal kinases (JNK) (46) which, in turn, can activate several down-stream factors including c-Jun and SMAD Family Member 4 (Smad4) which are known to trans-activate TIM-3 expression (47,48).
Our recently published data (39) indicate that high cellular iron content in CD8 + T cells induces accumulation of reactive oxygen species (ROS) in mitochondria and that treatment with a mitochondria-specific ROS scavenger could restore T cell priming even in the high iron setting. Whether a ROSmediated mechanism including ferroptosis (49) or mitochondrial dysfunction (50) as a consequence of iron loading underly the TIM-3 up-regulation and functional impairment of Th1 cells in high iron-fed Salmonella-infected animals remains to be investigated.
Importantly, the increased expression of TIM-3 in response to iron overload is pathophysiologically relevant as we were able to restore host defense against Salmonella by a TIM-3 blocking antibody. As microbial resistance against major classes of antibiotics continues to increase, immune-modulatory therapies that strengthen host defence mechanisms and promote bacterial killing will become more and more important. Therefore, the selective acquired immune deficit that iron overload exerts on Th1 immunity could be partly overcome by a monoclonal antibody against TIM-3. This may be specifically relevant in subjects with iron overload on the basis of repeated transfusions in the course of hematological diseases or hemoglobinopathies. Hence, our study provides a novel rational for an immune-modulatory therapy by blocking the checkpoint molecule TIM-3.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The animal study was reviewed and approved by BMWFW Austria.

AUTHOR CONTRIBUTIONS
PT participated in the study design, data collection and analysis, and drafted the manuscript. GK, SE, and NB participated in the data collection, MN participated in the data collection and revised the manuscript. AS and ED revised the manuscript. GW and CP-O participated in the study design, data collection, analysis, data interpretation, and manuscript preparation. All authors contributed to the article and approved the submitted version.