Essential role of IκBNS for in vivo CD4+ T‐cell activation, proliferation, and Th1‐cell differentiation during Listeria monocytogenes infection in mice

Abstract Acquisition of effector functions in T cells is guided by transcription factors, including NF‐κB, that itself is tightly controlled by inhibitory proteins. The atypical NF‐κB inhibitor, IκBNS, is involved in the development of Th1, Th17, and regulatory T (Treg) cells. However, it remained unclear to which extend IκBNS contributed to the acquisition of effector function in T cells specifically responding to a pathogen during in vivo infection. Tracking of adoptively transferred T cells in Listeria monocytogenes infected mice antigen‐specific activation of CD4+ T cells following in vivo pathogen encounter to strongly rely on IκBNS. While IκBNS was largely dispensable for the acquisition of cytotoxic effector function in CD8+ T cells, IκBNS‐deficient Th1 effector cells exhibited significantly reduced proliferation, marked changes in the pattern of activation marker expression, and reduced production of the Th1‐cell cytokines IFN‐γ, IL‐2, and TNF‐α. Complementary in vitro analyses using cells from novel reporter and inducible knockout mice revealed that IκBNS predominantly affects the early phase of Th1‐cell differentiation while its function in terminally differentiated cells appears to be negligible. Our data suggest IκBNS as a potential target to modulate specifically CD4+ T‐cell responses.


Introduction
The development and function of immune cells are regulated by a variety of transcription factors including NF-κB. NF-κB acts as a molecular switch that regulates many immunological processes including cell proliferation, activation of immune cells, and regulation of inflammation [1]. The activation of NF-κB is controlled by inhibitory proteins, such as the classical NF-κB inhibitor IκBα, which retains NF-κB in the cytoplasm by masking the nuclearlocalization sequence [1]. Next to the cytoplasmic inhibitory proteins, a second class of atypical NF-κB inhibitors comprising Bcl3, IκBζ, IκBη, and IκB NS exists in the nucleus, which either have activating or suppressing functions on NF-κB-mediated gene expression [2]. The NF-κB family of transcription factors contributes to Th1 differentiation. For instance, mice expressing a nondegradable IκBα mutant exhibit reduced IFN-ɣ production and Th1 responses [3,4]. With respect to atypical NF-κB inhibitors, Bcl-3deficient mice exhibit impaired Th1 responses toward intracellular pathogens [5,6]. In contrast, T-cell-specific deletion of IκBζ results in increased IFN-ɣ expression [7]. IκBζ counteracts RelA/p65 activity at the Ifng locus and TGF-β-induced IκBζ represses Ifng promoter activity by reducing acetylation of histones associated with the Ifng locus [7]. Similar to IκBζ, which interacts with chromatin-modifying enzymes [8,9], Bcl-3 acts as a bridge to nuclear coregulators [10,11].
IκB NS was first described in thymocytes in the context of negative selection [12]. IκB NS is also expressed in different T-cell subsets such as Th1 cells, regulatory T-cell precursors, and Th17 cells [13][14][15]. Of note, IκB NS −/− mice exhibit reduced numbers of Tregs, because IκB NS acts in concert with c-Rel and p50 to regulate Foxp3 expression, thereby, mediating the transition of Treg precursors into mature Treg cells [14]. In terms of T-cell development and function, both IκB NS-deficient CD4 + and CD8 + T cells, exhibit a proliferation defect upon in vitro TCR stimulation [13,16]. Moreover, IκB NS -deficient T cells show decreased secretion of IL-2 following in vitro stimulation and IκB NS −/− Th1 cells produce less IFN-γ [13,15]. Furthermore, IκB NS is critical for the development of effector functions in Th17 cells both in vitro and in vivo [15]. While together these data indicate a crucial role of IκB NS in the development and function of different T-cell subsets, no data are available on how IκB NS -deficiency affects the activation, proliferation, and effector function of T cells specifically responding to a pathogen-derived antigen during in vivo infection.

IκB NS fosters CD4 + T-cell activation and Th1 cytokine induction during Listeria monocytogenes infection
To specify the role of IκB NS in CD4 + T-cell activation following in vivo pathogen recognition, we combined systemic infection with ovalbumin-expressing Listeria monocytogenes (LM-OVA) and adoptive transfer of IκB NS sufficient or deficient TCR-transgenic OT-II CD4 + T cells (Fig. 1A). Analysis of reisolated cells revealed the first genotype-specific differences in the spleen on day 3 post infection (Fig. 1B). Of note, by day 5, LM-OVA-specific IκB NS +/+ CD4 + T cells underwent extensive proliferation, which was impaired in CD4 + T cells lacking IκB NS (Fig. 1B). Analyzing the proliferated CD4 + T cells for the expression of activation markers and Th1-related effector cytokines revealed striking differences between the genotypes (Fig. 2). Here, lack of IκB NS resulted in reduced frequency of LM-OVA-specific CD4 + T cells expressing the activation markers CD44 and PD-1, as well as the Th1-effector cytokines IFN-γ, IL-2, and TNF-α. We conclude that IκB NS is required for CD4 + T-cell activation and expansion in in vivo infectious settings and is critically involved in Th1-cell differentiation.

IκB NS affects the early phase of Th1-cell differentiation
To investigate in more detail IκB NS dependency of Th1-cell differentiation, we used the newly generated reporter mouse Nfkbid lacZ that contains a LacZ cassette within the IκB NS -encoding Nfkbid gene and expresses ß-galactosidase under the control of the Nfkbid promoter. After confirmation that the Nfkbid lacZ mouse represents a faithful reporter to quantify Nfkbid gene expression (Supporting Information Fig. 1), we analyzed the kinetics of Nfkbid promoter activity under Th1-polarizing conditions. Promoter activity increased until day 3 following T-cell activation (Fig. 3A), suggesting that IκB NS is especially important for the early phase of Th1-cell differentiation. To further confirm this, we utilized Nfkbid FL/FL × Rosa CreERT2 mice, which allow for targeted deletion of Nfkbid at time of interest following T-cell stimulation. After confirming complete loss of IκB NS in CD4 + T cells within 48 h following stimulation (Fig. 3B), we evaluated at which phase of Th1-cell differentiation IκB NS is required. Both, IFN-γ and CD44 expression were reduced in IκB NS −/− CD4 + T cells when IκB NS -deficiency was induced early (day 2) during Th1-cell differentiation ( Fig. 3C) but were not affected by adding tamoxifen on day 4 ( Fig. 3C). Notably, the expression of T-bet, which is crucial for Th1-cell differentiation [17] was not affected in the absence of IκB NS (Fig. 3C), suggesting that other early factors of Th1-cell differentiation are affected. Since IFN-ɣ amplifies Th1cell differentiation, we suspect that the reduced IFN-ɣ expression in the absence of IκB NS is responsible for reduced Th1 responses. As IκBζ and IκB NS have opposing activities on IL-6 expression in macrophages [16,18], it is tempting to speculate that a similar counteracting function operates in Th1 cells with respect to IFN-ɣ expression.
Another hint regarding a potential mechanism underlying the impact of IκB NS on Th1-cell differentiation comes from data published by Touma et al., who showed that the F-box protein family member Fbxo17 is highly overexpressed in resting and activated IκB NS −/− T cells [13], a finding we confirmed in OT-II CD4 + T cells (data not shown www.eji-journal.eu CD4 + T cells from Thy1.1 + OT-II x WT and OT-II x IκB NS −/− were transferred into C57BL/6 mice. One day post transfer recipients were infected with 5 × 10 3 LM-OVA. CD4 + T cells from spleen and liver were analyzed for CFSE loss at the indicated time post infection by flow cytometry. (B, C) Flow cytometry data are representative for two (day 3) or three (day 5) independent experiments with similar outcome with n = 4-5 individually analyzed mice/group and data were constrained to alive singlet Thy1.1 + CD4 + T cells and are shown in columns side-by-side in a concatenated qualitative dot plot in which each column represents data of an individual mouse. The summary plots are depicted as mean ± SEM of 4-5 individually analyzed mice/group and indicate the percentages of proliferated (CFSE low ) transferred CD4 + T cells in spleen and liver samples. Statistics were performed using two-tailed unpaired student's t-test. **p < 0.01, ****p < 0.0001.
involved in regulation of cell cycle and proliferation [19]. Fbxo17 regulates proteasomal degradation of glycogen synthase kinase-3β (GSK3β) [20]. Since GSK3 isoforms are crucial for Th1 differentiation [21], increased Fbxo17 expression in IκB NS −/− T cells might explain impaired Th1 differentiation. The exact mechanism of IκB NS and/or Fbxo17 in Th1 differentiation remains, however, unresolved and will be the focus of future studies.
Listeria monocytogenes induces a Th1-dominated response with almost no Th17 cells and Th2-cell responses even actively being   suppressed by pathogen-derived factors [22,23]. In line with data obtained in the in vitro Th1 differentiation, IκB NS -deficient CD4 + T cells failed to upregulate CD44 and to produce the Th1-effector cytokines IFN-γ, IL-2, and TNF-α following in vivo pathogen encounter (Fig. 2). IκB NS -dependency of IL-2 and IFN-γ production in CD4 + T cells has been described before [13] and we recently showed that IκB NS -deficiency impairs CD4 + T-cell proliferation during in vitro Th-cell differentiation and is critically involved in the development of Th1 and Th17 cells [15]. However, the impact of IκB NS on T-cell differentiation appears to be context dependent.

While EAE induction in IκB NS
−/− mice did not affect the Th1cell pool [24], the DSS colitis model uncovered increased IFN-γ www.eji-journal.eu expression in IκB NS −/− CD4 + T cells [15,16]. In contrast, intestinal infection of IκB NS -deficient mice with Citrobacter rodentium resulted in reduced frequencies of IFN-γ-producing CD4 + T cells in spleen but not in colon and local lymph nodes [15]. Of note, the present study is fundamentally different since we specifically track T cells responding to pathogen-derived antigen in mice exhibiting an IκB NS -sufficient immune system. In line with published data obtained with polyclonal T cells [13], we also found OT-I CD8 + IκB NS −/− T-cell proliferation to be impaired following in vitro antibody-induced TCR stimulation (data not shown) while CD8 + T-cell proliferation induced by antigen-specific activation following in vivo pathogen encounter does not depend on IκB NS (Supporting Information Fig. 2). Moreover, with the exception of TNF-α, IκB NS -deficiency had no or only transient effect on the expression of all other markers analyzed (Supporting Information Fig. 3). Data from adoptive transfers suggest that IκB NS might affect cytotoxic function of CD8 + T cells during the early phase of infection, which is likely to be compensated by the WT adaptive immunity in the later infection phase (Supporting Information Fig. 4A). This was supported by LM-OVA infections in WT and conventional IκB NS −/− mice that neither revealed differences in bacterial elimination (Supporting Information Fig. 4B) nor uncovered defects in the establishment of cytotoxic T-cell responses from the polyclonal TCR pool (Supporting Information Fig. 4C). This at least in part excludes that the observed effects in IκB NS −/− CD8 + T cells are due to the transgenic OT-I TCR specific for the model antigen expressed by LM-OVA. We hypothesize that the effects of IκB NS -deficiency in CD8 + T cells that are observed in vitro are compensated in vivo by the presence of host-derived proinflammatory cytokines that are produced during infection. This is in line with our observation that genotype-dependent differences in the activation pattern of CD8 + T cells are only transient. Interestingly, the percentage of TNFα-producing LM-OVA-specific IκB NS −/− CD8 + T cells was significantly higher compared to IκB NS +/+ CD8 + T cells. There is evidence that IκB NS may function as suppressor of TNF-α [25].
To rule out the possibility that the fraction of TNF-α + CD8 + T cell is per se higher in IκB NS −/− mice, we analyzed its expression at 1 day post infection revealing even more TNF-α + CD8 + T cells in the IκB NS +/+ group (data not shown). Nevertheless, despite distinct changes in the activation pattern and kinetics, IκB NS −/− CD8 + T cells are fully capable of producing IFN-γ and mice with IκB NS -deficient T cells are competent in developing cytotoxic T-cell responses against LM-OVA (Supporting Information Fig. 4).

Concluding remarks
While being largely dispensable for the in vivo acquisition of effector function in CD8 + T cells, we unveiled IκB NS to play a pivotal role in the activation, expansion, and Th1-cell differentiation of CD4 + T cells following in vivo pathogen encounter. In terms of timing, IκB NS affects the early phase of Th1-cell differentiation but is dispensable in terminally differentiated Th1 cells.

Bacterial infection
OVA-expressing L. monocytogenes (LM-OVA, strain 10403S) were grown overnight (37°C, 180 rpm) in BHI broth (BD Biosciences). A 1:5 dilution was prepared with fresh BHI and after 3 h bacteria were harvested and diluted in sterile PBS to establish an infection dose of 5 × 10 3 CFU/mouse. To determine CFU in spleen and livers organs were homogenized in 0.2% IGEPAL CA-630 (Sigma-Aldrich) lysis buffer and serial dilutions were plated on BHI agar plates to quantify colonies after incubation at 37°C for 24 h.

Cell preparation
Spleens and livers were squeezed through 100 μm cell strainers (Falcon), washed with PBS (300 × g, 10 min, 4°C) followed by erythrocyte lysis. Splenocytes were passed through a 30 μm cell strainer and resuspended in PBS. Liver cells were separated by density centrifugation in a 35% mixture of Easycoll (Biochrom) and PBS. After centrifugation (360 × g, 20 min, RT) cells were washed and resuspended in PBS.

Flow cytometric analyses
Flow cytometric analyses were performed in adherence to the "Guidelines for the use of flow cytometry and cell sorting in immunological studies" [29]. Cells were stained with anti-CD16/32 (Fc-block) (BioLegend) and with Fixable Viability Dye-eFluor 780 (eBioscience) to exclude dead cells. After washing, cells were incubated with the antibody mixture containing CD8a-BV510 (53-6.7) or CD4-BV510 (GK1. . IκB NS -dependency of cytotoxic function in CD8 + T cells was determined as described before [30]. Cells were analyzed with a BD FACS Canto II, BD LSR II (BD Biosciences) or an Attune NxT Acoustic Focusing Cytometer (Thermo Fisher Scientific).

Statistics
Results are expressed as mean ± SEM. Statistical analyses were performed with the GraphPad Prism 5.4 Software (La Jolla, CA, USA) and a p-value below 0.05 was considered significant.