The role of ADAM17 in the T-cell response against bacterial pathogens

ADAM17 is a member of the A Disintegrin And Metalloproteinase family of proteases. It is ubiquitously expressed and causes the shedding of a broad spectrum of surface proteins such as adhesion molecules, cytokines and cytokine receptors. By controlled shedding of these proteins from leukocytes, ADAM17 is able to regulate immune responses. Several ADAM17 targets on T cells have been implicated in T-cell migration, differentiation and effector functions. However, the role of ADAM17 in T-cell responses is still unclear. To characterize the function of ADAM17 in T cells, we used Adam17fl/fl×CD4cre+ mice with a T-cell restricted inactivation of the Adam17 gene. Upon stimulation, ADAM17-deficient CD4+ and CD8+ T cells were impaired in shedding of CD62L, IL-6Rα, TNF-α, TNFRI and TNFRII. Surprisingly, we could not detect profound changes in the composition of major T-cell subsets in Adam17fl/fl×CD4cre+ mice. Following infection with Listeria monocytogenes, Adam17fl/fl×CD4cre+ mice mounted regular listeria-specific CD4+ TH1 and CD8+ T-cell responses and were able to control primary and secondary infections. In conclusion, our study indicates that ADAM17 is either not required in T cells under homoeostatic conditions and for control of listeria infection or can be effectively compensated by other mechanisms.


Introduction
Proteases of the ADAM (A Disintegrin And Metalloproteinase) family regulate various aspects of immune cell development and function. ADAM proteases cause the release of cytokines but also the ectodomain shedding of cell surface molecules including adhesion molecules and cytokine receptors [1,2]. For ADAM17, more than 70 target proteins have been identified [2]. ADAM17 is responsible for the TNF-α release and thus for the paracrine and systemic activity of the cytokine [3,4]. Inhibition of ADAM17 as well as global or myeloid cell-restricted deletion of ADAM17 in mice results in reduced systemic TNF-α levels following LPS treatment and protects mice from fatal endotoxemia [5,6,7]. Similarly, mice with defective iRhom2 (inactive Rhomboid protein 2), which facilitates trafficking of ADAM17 to the cell surface, are also less susceptible to LPS endotoxemia [8,9]. ADAM17 has also been identified as sheddase for TNFR1 and TNFR2 [5,10,11,12]. Loss of functional TNFRs reduces on one hand the a1111111111 a1111111111 a1111111111 a1111111111 a1111111111 sensitivity of cells towards TNF-α, on the other hand, soluble TNFRs sequester TNF-α and thereby limit its activity. Under inflammatory conditions, the IL-6Rα chain is also shed by ADAM17 [5]. In contrast to TNF-α, binding of IL-6 by soluble (s)IL-6Rα does not result in IL-6 neutralization, rather, the IL-6/sIL-6Rα can still interact with gp130 on the surface of cells and induce IL-6 signaling. Since gp130 is ubiquitously expressed, this IL-6 trans-signaling considerably widens the spectrum of IL-6 target cells [13]. Following various stimuli, L-selectin (CD62L) is rapidly shed from neutrophils and lymphocytes by ADAM17, and neutrophils deficient in ADAM17 show altered rolling and adhesion to endothelia and accelerated accumulation in the inflamed peritoneum [14,15,16].
Mouse T cells constitutively express Adam17 mRNA [17] and we could recently demonstrate impaired shedding of ADAM17 targets in T cells from hypomorphic ADAM17 mice with substantially reduced ADAM17 expression or in T cells cultured with ADAM17 inhibitors [5]. ADAM17 targets are involved in diverse T-cell functions including migration and differentiation to effector cells, therefore it has been postulated that ADAM17-mediated shedding of these proteins can regulate these processes [2,18]. Here, we use mice with a T-cell restricted ADAM17-deficiency (Adam17 fl/fl ×CD4cre + mice) to investigate the role of ADAM17 in T cells under homeostatic conditions and following infection of mice with L. monocytogenes. Surprisingly, deficiency of ADAM17 in T cells did not result in substantial alterations in the composition of peripheral T cells and in the protective T-cell response to L. monocytogenes.

Mice and Listeria monocytogenes infection
Adam17 fl/fl mice [6] were kindly provided by Carl Blobel and back crossed with CD4cre + mice (B6.Cg-Tg(Cd4-cre)1Cwi/BfluJ; Jackson, Bar Harbor, ME). Due to CD4 expression in double positive thymocytes, Adam17 fl/fl ×CD4cre + mice show ADAM17 deletion in peripheral CD4 + and CD8 + T cells. All mice were genotyped by PCR. ADAM17 deletion was confirmed by FACS for loss of CD62L shedding from T cells after incubation of peripheral blood cells for 30 min with PMA and ionomycin. Age-and sex-matched Adam17 fl/fl ×CD4cre + and Adam17 fl/fl αCD4crecontrol mice of 8 to 12 weeks of age were used in all experiments. Mice were intravenously infected with wildtype Listeria monocytogenes strain EGD or a L. monocytogenes strain recombinant for ovalbumin [19]. Mice received 2α10 4 listeria in 200 μl PBS. Bacterial inocula were controlled by plating serial dilutions on PALCAM agar plates. For determination of bacterial burdens in infected mice, spleens and livers were homogenized in H 2 O, 0.5% Triton X-100 and serial dilutions of homogenates were plated on PALCAM agar. Colonies were counted after incubation at room temperature.
This study was carried out in strict accordance with the state guidelines. The protocol was approved by local ethics committee of the Behörde für Gesundheit und Verbraucherschutz of the City of Hamburg (Permit Number: 81/14). Mice were housed under specific pathogen free conditions in individually ventilated cages with standard food and water ad libitum. During infection experiments, mice were controlled daily and mice with signs of severe disease were euthanized to minimize suffering.
In vitro proliferation was measured by CFSE dilution assay. Spleen cells were incubated in PBS with 5μM CFSE for 15min at 37˚C. Cells were washed with PBS and 4 ×10 5 cells/well were cultured in culture medium in 96-well plates coated with anti-CD3 mAb in the presence of anti-CD28 mAb (clone 37.51, Biolegend). After three days, staining intensity of CFSE on CD4 + and CD8 + T cells was determined by flow cytometry. In parallel, cells were restimulated with 50 ng/ml PMA and 1 μM ionomycin for 4h. For the last 3.5h, 10 μg/ml brefeldin A (Sigma Aldrich) was added to the cultures to prevent cytokine secretion. Subsequently, CD40L and cytokine expression was determined by intracellular mAb staining and flow cytometry.

In vivo cytotoxicity assay
Spleen cells from C57BL/6 mice were incubated in culture medium with 10 -6 M of OVA 257-254 or LCMVgp 33-41 peptide (KAVYNFATM, JPT) at 37˚C. After 1h, cells were washed with PBS and incubated in PBS with 5μM or 0.5μM CFSE for 15min at 37˚C. Cells were washed with PBS and counted. CFSE low and CFSE high cells were mixed in a ratio of 1:1 and a total of 6×10 6 cells was i.v. injected into naive mice or mice which had been infected with LmOVA. After 3h, spleen and liver of recipients were analyzed for CFSE-positive cells. % killing was calculated: 100 − ((% relevant peptide-pulsed cells in immunized mice / % irrelevant peptide-pulsed cells in immunized mice) / (% relevant peptide-pulsed cells in control mice/% irrelevant peptidepulsed cells in control mice)) × 100

Flow cytometry
For surface staining, cells were incubated for 5 min with 10 μg/ml 2.4G2 (anti-FcγRII/III; BioXCell, West Lebanon, NH) and 1:100 rat serum in PBS to minimize unspecific antibody binding. Staining was performed on ice with fluorochrome-conjugated mAb according to standard methods. Dead cells were labelled with a fixable dead cell stain (Pacific Orange succinimidyl ester; Life Technologies, Waltham, MA). For measurement of intracellular cytokines, cells were incubated with mAb against surface proteins and with Pacific Orange succinimidyl ester. After washing with PBS, cells were fixed for 20 min with PBS, 2% paraformaldehyde at room temperature. Cells were washed with PBS, 0.2% BSA, permeabilized with PBS, 0.1% BSA, 0.3% saponin (Sigma, Aldrich), and incubated in this buffer with 1% rat serum. Cells were measured with a Canto II flow cytometer (BD Biosciences) and data were analyzed with the DIVA software (BD Bioscience) or FlowJo software (Treestar, Ashland, OR). Debris, doublets and Pacific Orange + dead cells were excluded from analysis.

Statistical analysis
Statistical analyses were performed with Prism software (GraphPad Software Inc., La Jolla, CA). Results were analyzed with the tests indicated in the figure legends. A p-value of <0.05 was considered significant ( Ã : p<0.05; ÃÃ : p<0.01; ÃÃÃ : p<0.001; ns: not significant).

Results
Impaired shedding of ADAM17 substrates by T cells from Adam17 fl/fl × CD4cre + mice In the first set of experiments, we tested whether ADAM17-deficent T cells were able to shed verified ADAM17 substrates. Spleen cells from naive Adam17 fl/fl ×CD4crecontrol mice and Adam17 fl/fl ×CD4cre + mice were incubated with PMA and ionomycin. At different time points, activation was stopped by placing the samples on ice. Cells were surface stained for CD62L, TNF-α, TNFR1, TNFR2 and IL-6Rα, and analyzed by flow cytometry (Fig 1). Following stimulation, CD4 + and CD8 + T cells from control mice rapidly lost surface expression of CD62L. In contrast, T cells from Adam17 fl/fl ×CD4cre + mice completely failed to shed CD62L. TNF-α was not visible on the surface of T cells from control mice throughout the stimulation period. However, we detected surface TNF-α after two and four hours of stimulation on subpopulations of CD4 + and CD8 + T cells from Adam17 fl/fl ×CD4cre + mice. TNFR1 was expressed at low levels on the surface of T cells from control mice. Comparison of MFI values revealed that control T cells rapidly lost TNFR1 surface expression following stimulation. We also observed a decrease of surface TNFR1 on CD4 + and CD8 + T cells from Adam17 fl/fl ×CD4cre + mice, however, decrease was delayed when compared to control T cells. TNFR2 was expressed on subpopulations of CD4 + and CD8 + T cells. T cells from control mice rapidly lost this expression. In contrast, there was only marginal reduction of surface TNRF2 from ADAM17-deficient T cells. Likewise, shedding of IL-6Rα was delayed and far less pronounced from T cells of Adam17 fl/fl ×CD4cre + mice when compared to control T cells.
Spleen cells were also incubated in plates coated with anti-CD3 mAb and surface expression of CD62L, IL-6Rα and TNFR1 was determined (Fig 2). In comparison to PMA and ionomycin treatment, shedding of these proteins was less pronounced and delayed after anti-CD3 mAb treatment. However, we also observed impaired shedding of CD62L, IL-6Rα and TNFR1 from T cells from Adam17 fl/fl ×CD4cre + mice when compared to T cells from control mice.
In conclusion, deficiency of ADAM17 completely prevented shedding of CD62L indicating the efficient deletion of Adam17 in CD4 + and CD8 + T cells in Adam17 fl/fl ×CD4cre + mice. This result also suggests that reduction of other ADAM17 substrates from the surface of activated T cells from Adam17 fl/fl ×CD4cre + mice was most likely due to ADAM17-independent mechanisms, such as shedding by other proteases or altered surface turnover.

T-cell development and peripheral T-cell composition in
Due to the impaired shedding of proteins involved in T-cell function, we wondered whether ADAM17-deficiency causes alterations in maturation and composition of peripheral CD4 + and CD8 + T cells. Therefore, the cellular composition of thymus, spleen, inguinal lymph nodes and liver of Adam17 fl/fl ×CD4cre + mice was analyzed by flow cytometry. We observed a similar distribution of CD4 -CD8 -(double negative), CD4 + CD8 + (double positive), and CD4 + CD8and CD4 -CD8 + (single positive) cells in the thymus of Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4crecontrol mice (Fig 3A, S1 Fig). Likewise, Adam17 fl/fl ×CD4cre + mice showed a regular composition of conventional CD4 + and CD8 + T cells in spleen, inguinal lymph nodes and liver as well as similar total numbers of CD4 + and CD8 + T cells in the spleen (Fig 3B). In some experiments, we observed slightly reduced CD4 + and increased CD8 + T-cell frequencies, however, these alterations were not consistent in all experiments. T cells from spleen, lymph nodes and liver were stained for CD44 and CD62L to determine the distribution of naive (CD44 -CD62L + ), effector/effector memory (CD44 + CD62L -) and central memory T cells (CD44 + CD62L + ). There was a similar distribution of these T-cell subpopulations in Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4crecontrol mice ( Fig 3C).
IL-6 plays a fundamental role in the differentiation of T H17 cells and prevents the formation of peripheral T reg cells. Since impaired shedding of IL-6Rα might influence these differentiation pathways, we determined the frequencies of FoxP3 + CD4 + T reg cells ( Fig 3D) as well as T H17  (Fig 3E). T reg cells were detected by intracellular FoxP3 staining (S1 Fig). For determination of T H1 and T H17 cells, spleen cells were incubated with PMA/ionomycin or without stimulation. After 4h, intracellular expression of CD40L, which is upregulated on all conventional CD4 + T H cells following TCR stimulation, as well as IFN-γ, TNF-α and IL-17A was determined by flow cytometry. We observed similar frequencies of T reg cells, and after stimulation equal frequencies of CD40L + IFN-γ + TNF-α + T H1 cells and CD40L + IL-17A + T H17 cells among CD4 + T cells from spleens of Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice.
In conclusion, deficiency of ADAM17 in CD4 + and CD8 + T cells did not cause profound changes in the composition, distribution and differentiation status of conventional T cells.

In vitro responses of T cells from Adam17 fl/fl ×CD4cre + mice
In the next set of experiments, we tested the function of T cells in vitro. Spleen cells from Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice were labelled with CFSE and cultured with anti-CD3 mAb and anti-CD28 mAb. After 3 days, proliferation was determined by loss of CFSE staining (S2 Fig). We detected similar CFSE staining in CD4 + T cells from both mouse strains but somewhat higher CFSE staining in CD8 + T cells from Adam17 fl/fl ×CD4crecontrol mice, indicating slightly less extensive proliferation of these cells. After 3 days, T cells were, in addition, re-stimulated for 4h with PMA and ionomycin and cytokine expression was determined by intracellular staining and flow cytometry (S2 Fig). The vast majority of CD4 + T cells from both mouse strains responded with upregulation of CD40L. A larger fraction of ADAM17-deficient CD4 + T cells were able to produce TNF-α and IFN-γ. Stimulation resulted in induction of TNF-α and IFN-γ in most CD8 + T cells from both mouse strains. Here, we detected slightly lower frequencies of TNF-α + CD8 + T cells in cultures of Adam17 fl/fl ×CD4cre + spleen cells.
In summary, ADAM17 deficiency did not substantially alter the in vitro T-cell response to polyclonal stimulation. In some assays, ADAM17-deficient T cells showed stronger responses, however, differences were in all cases only very modest.

T-cell responses to Listeria monocytogenes infection in
To test the role of ADAM17 in the generation and function of a T-cell response in vivo, we applied the Listeria monocytogenes infection model. L. monocytogenes infection induces strong CD4 + T H1 and CD8 + T-cell responses, which are essential for the clearance of the bacteria. Furthermore, control of L. monocytogenes highly depends on TNF-α [24,25,26,27].
Mice were i.v. infected with 2×10 4 bacteria of a L. monocytogenes strain recombinant for ovalbumin (LmOVA) [19] and the T-cell response in spleen and liver was analyzed at different time points post-infection. For C57BL/6 mice, immunodominant CD8 + T-cell epitopes from L. monocytogenes are currently not known. Since ovalbumin contains a strong CD8 + T-cell epitope (OVA 257-264 ), application of LmOVA allows determination of ovalbumin-specific CD8 + T-cell responses generated during L. monocytogenes infection of C57BL/6 mice.
Compared to naive mice (Fig 3), we observed a shift in frequencies and numbers from CD44 -CD62L + naive to CD44 + CD62Leffector CD4 + and CD8 + T cells in spleen (Fig 4A and  4C) and liver (Fig 4B and 4D) at 8 days post L. monocytogenes infection. However, changes in T-cell compositions were similar in Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice. We following stimulation of spleen cells for 4 h with PMA (50 ng/ml) and ionomycin (1 μM). Scatter plots give results for individual mice and the means +/-SD from representative experiments. Experiments were repeated at least 2 times. Groups were compared with student's t test. A p-value of <0.05 was considered significant. https://doi.org/10.1371/journal.pone.0184320.g003 The role of ADAM17 in the T-cell response against bacterial pathogens PLOS ONE | https://doi.org/10.1371/journal.pone.0184320 September 6, 2017 also determined the expression levels of CD62L and IL-6Rα on activated CD44 + CD4 + and CD44 + CD8 + T cells (S3 Fig). Surface expression of both proteins was reduced on CD44 + T cells following infection. However, T cells from Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice did not differ in CD62L and IL-6Rα expression levels.
To determine the formation of specific CD8 + T cells, cells isolated from spleen and liver were incubated with OVA 257-264 peptide and the production of IFN-γ and TNF-α were measured by intracellular cytokine staining and flow cytometry. At day 8 and 15 post infection, we observed high frequencies and numbers of OVA 257-264 -specific CD8 + T cells in spleens, The role of ADAM17 in the T-cell response against bacterial pathogens however, frequencies and numbers were similar in Adam17 fl/fl ×CD4cre + and Adam17 fl/fl × CD4cremice (Fig 5A and 5B). In some experiments, we observed slightly higher frequencies of cytokine producing CD8 + T cells in the liver of Adam17 fl/fl ×CD4cre + mice, however, these differences were not consistent in all experiments (Fig 5C).
Finally, the cytolytic activity of CD8 + T cells was tested (Fig 7). Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice were infected with LmOVA. Eight days later, mice received a 1:1 mixture of spleen cells loaded either with OVA 257-264 or with an irrelevant peptide. After further 3h, frequencies of both peptide-loaded cell populations were determined in spleen and liver. Compared to non-infected recipients, we observed a strong reduction of OVA 257-264loaded spleen cells in spleens and livers of infected mice, indicating that LmOVA infection had induced OVA-specific cytotoxic CD8 + T cells. However, we did not observe a difference in the cytotoxic activity between Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice.
In conclusion, ADAM17 deficiency in CD4 + and CD8 + T cells did not substantially alter the T-cell response to L. monocytogenes in terms of frequencies of responding cells, expression profile of the analyzed cytokines and cytotoxicity.
Control of Listeria monocytogenes infection in Adam17 fl/fl ×CD4cre + mice TNF-α is fundamental for the immune response to L. monocytogenes. Thus, impaired shedding of TNF-α, TNFR1 and TNFR2 from T cells could result in impaired control of listeria. Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice were i.v. infected with 2×10 4 listeria and 5 days later, listeria titers in spleens and livers were determined (Fig 8A). There was no significant difference in the bacterial titers of spleen and liver between Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice, indicating that shedding of TNF-α or of other surface proteins from T cells is not required for the control of L. monocytogenes.
Control of secondary L. monocytogenes infection strongly relies on the acquired T-cell response and might therefore be affected in Adam17 fl/fl ×CD4cre + mice. To test the efficacy of these T-cells, Adam17 fl/fl ×CD4creand Adam17 fl/fl ×CD4cre + mice were infected with L. monocytogenes and after 40 days reinfected with a lethal listeria dose. Two days later, titers in spleen and liver were determined (Fig 8B). None of the reinfected mice succumbed to the high dose infection and Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice harbored comparable L. monocytogenes titers in spleen and liver. This result implies that Adam17 fl/fl ×CD4cre + mice can generate listeria-specific memory T cells which are effective in protection upon secondary listeria infection. In conclusion, Adam17 fl/fl ×CD4cre + were not impaired in their control of primary and secondary L. monocytogenes infection.

Discussion
ADAM17 is constitutively expressed in T cells and expression only marginally changes following T-cell activation [17]. Functional characterization of ADAM17-deficient T cells in vitro showed altered shedding of surface proteins such as CD62L, TNF-α, TNFR1, TNFR2 and IL-6Rα, all considered relevant for CD4 + and CD8 + T-cell function, as well as modest changes in proliferation and cytokine production. Surprisingly, ADAM17-deficiency apparently did not influence the T-cell function in vivo. We did not observe significant changes in maturation, migration, differentiation patterns and function of CD4 + and CD8 + T cells in naive and listeria-infected mice. The composition of conventional T-cell subsets was unchanged, there was no shift in T reg or T H17 -cell differentiation, and mice were able to mount an effective CD8 + and T H1 response to L. monocytogenes. We detected similar listeria titers in spleen and liver of Adam17 fl/fl ×CD4cre + and Adam17 fl/fl ×CD4cremice at day 5 of primary and day 2 of secondary infection indicating that ADAM17-deficient T cells were not impaired in controlling L. monocytogenes. During this study, we infected a total of 96 Adam17 fl/fl ×CD4creand 97 Adam17 fl/fl ×CD4cre + mice and followed these mice for up to 40 days. We lost 3 Adam17 fl/fl ×CD4cremice (3.1%) and 8 Adam17 fl/fl ×CD4cre + mice (8.2%) due to infection, which further confirms that ADAM17-deficiency does not substantially weakens the T-cell response to L. monocytogenes.
Naive and memory T cells require the addressin CD62L for entering the T-cell zone of secondary lymphoid tissues at high endothelial venules. Subsequent activation in the T-cell zone causes rapid loss of CD62L surface expression. Our in vitro experiments demonstrate that CD62L shedding from T cells is strictly dependent on ADAM17. However, we did not observe substantial changes in T-cell development and distribution as well as in the T-cell response to L. monocytogenes. Our results confirm studies demonstrating normal homeostatic T-cell Scatter plots give frequencies of CD40L + TNF-α + IFN-γ + (D), CD40L + IL-17A + (E) and CD40L + IL-17A + IFN-γ + (F) in spleen and liver. Results for individual mice and the means+/-SD from a representative experiment are shown. Experiments were repeated 2 times. Groups were compared with student's t test. A p-value of <0.05 was considered significant.
https://doi.org/10.1371/journal.pone.0184320.g006 distribution in mice with a protease resistant CD62L protein [28,29]. Thus, rapid CD62L shedding from activated T cells appears to be not essential for their function. mRNA expression analyses indicate that activated CD8 + T cells downregulate Sell mRNA (coding for CD62L) during listeria infection [17], which might compensate for defective shedding. We observed impaired shedding of TNF-α and TNFRI, and delayed shedding of TNFRII from T cells of Adam17 fl/fl ×CD4cre + mice. These defects did neither interfere with T-cell differentiation under homeostatic conditions nor with formation of a T-cell response against L. monocytogenes. Shedding of TNF-α from T cells was also not required for listeria control during primary and secondary infection. In the case of TNFRI and TNFRII other protease might compensate for the ADAM17-deficiency in T cells. Several studies could demonstrate that membrane-bound TNF-α is active [30,31,32,33,34] and capable of providing partial protection against L. monocytogenes infection [27,31,33]. In addition, protection against listeria mainly relies on TNF-α produced by myeloid cells and only under conditions with high titers, T-cell derived TNF-α becomes important [26]. Finally, control of secondary listeria infection does not require TNF-α [26,35]. Together, these results could well explain why a defect in shedding of TNF-α in Adam17 fl/fl ×CD4cre + mice does not substantially impair the control of L. monocytogenes.
IL-6 supports the formation of T H17 cells and suppresses the induction of peripheral T reg cells [36,37,38,39]. Furthermore, there is controversial data on the role of IL-6 in T H1 responses [40,41]. Based on these studies, one could postulate that IL-6Rα shedding restricts the response of activated CD4 + T cells to IL-6 and thereby regulates CD4 + T-cell differentiation. In our in vitro study, we observe impaired IL-6Rα shedding from T cells of Adam17 fl/fl ×CD4cre + mice which is consistent with our results from experiments with T cells of hypomorphic ADAM17 mice or with metalloprotease inhibitors [5]. However, ADAM17-deficiency affected neither the composition of T-cell subsets under homeostatic conditions nor the formation of the T-cell response against L. monocytogenes. Listeria infection of mice is considered to be a T H1 model and thus provides only very limited information on T H17 -cell differentiation. In addition, it could recently be shown that T H17 differentiation requires IL-6 presentation by IL-6Rα + dendritic cells and is largely independent from IL-6Rα expression on T cells [38]. However, Adam17 fl/fl ×CD4cre + mice also showed normal frequencies of T reg cells and generated a regular T H1 and CD8 + T-cell response to L. monocytogenes. Thus, ADAM17-mediated shedding of IL-6Rα is either not important for these processes or compensated by other mechanisms such as shedding by other proteases, receptor internalization or down-regulation of Il6Ra mRNA expression in activated T cells [5,17].
In conclusion, our study indicates that although T cells bear several potential ADAM17 targets on their surface, ADAM17-deficiency does not result in profound changes of peripheral T-cell composition under homeostatic conditions and does not impair the T H1 and CD8 + Tcell response against L. monocytogenes. ADAM17-mediated shedding might simply be not required in these processes. Alternatively, ADAM17 targets could be shed by other proteases, removed from the cell surface by internalization or downregulated on the mRNA expression level. Our analysis was restricted to the T-cell response against L. monocytogenes. However, it is well conceivable that under different circumstances, such as responses to other pathogens or to tumors or in autoimmunity, T cells might require ADAM17 for proper function. Adam17 fl/fl ×CD4creand Adam17 fl/fl ×CD4cre + mice were labelled with CFSE and stimulated with anti-CD3 mAb and anti-CD28 mAb. After 3 days, CFSE expression was determined on CD4 + and CD8 + T cells. At this time point, T cells were also re-stimulated for further 4h with PMA and ionomycin or were left without stimulation (none). Subsequently, expression of CD40L, TNF-α and IFN-γ was determined by intracellular staining and flow cytometry. (A) Representative histograms for CFSE expression of CD4 + and CD8 + T cells from Adam17 fl/fl × CD4cre -(light grey) and Adam17 fl/fl ×CD4cre + mice (dark grey). Charts give the MFI for individually analyzed samples. (B) Frequencies of CD40L + , CD40L + TNF-α + and CD40L + IFN-γ + CD4 + T cells (left), as well as of TNF-α + and IFN-γ + CD8 + T cells (right). Charts give the mean +/-SD of triplicate cultures for each mouse strain. Groups were compared with student's t test. A p-value of <0.05 was considered significant. Data are representative for two independent experiments with similar outcome. (PDF) S3 Fig. CD62L and IL-6Rα surface expression on CD4 + and CD8 + T cells from Adam17 fl/fl × CD4cre + mice. Adam17 fl/fl ×CD4creand Adam17 fl/fl ×CD4cre + mice were infected with 2×10 4 LmOVA. Spleen cells from naive mice (A) and mice infected for 8 (B) and 15 days (C) were analyzed for surface expression by flow cytometry. Scatter plots give MFI (mean fluorescence intensity) for CD62L and IL-6Rα on CD44 + CD4 + and CD44 + CD8 + T cells. Results for individually analyzed mice and mean +/-SD are presented. Experiments were repeated 2 times. (PDF)