Fcμ receptor as a Costimulatory Molecule for T Cells

Article Fcmreceptor as a Costimulatory Molecule for T Cells Graphical Abstract Highlights dFcmR is expressed by T cells to ensure persistent IgM uptake dIntracellular accumulation of IgM enhances surface T cell receptor expression dT cell effector functions are boosted by FcmR-mediated accumulation of IgM dMethylation of FCMR gene is associated with decreased protein expression in old age Authors Andreas Meryk, Luca Pangrazzi, Magdalena Hagen, ..., Mikko Hurme, Klemens Trieb, Beatrix Grubeck-Loebenstein Correspondence andreas.mery[email protected] In Brief Meryk et al. demonstrate that uptake of IgM mediated by FcmR expressed on T cells increases the surface expression of TCR and costimulatory molecules to facilitate T cell activation, particularly when antigen concentrations are low. Consequently, FcmR increases TCR signaling, proliferation, and cytokine release. Meryk et al., 2019, Cell Reports 26, 2681–2691 March 5, 2019 ª2019 The Author(s). https://doi.org/10.1016/j.celrep.2019.02.024 Cell Reports Article Fcmreceptor as a Costimulatory Molecule for T Cells Andreas Meryk, 1,6,7, *Luca Pangrazzi, 1,6 Magdalena Hagen, 1,6 Florian Hatzmann, 1 Brigitte Jenewein, 1 Bojana Jakic, 2 Natascha Hermann-Kleiter, 2 Gottfried Baier, 2 Juulia Jylha ¨va ¨, 3 Mikko Hurme, 4 Klemens Trieb, 5 and Beatrix Grubeck-Loebenstein 1 1 Department of Immunology, Institute for Biomedical Aging Research, University of Innsbruck, 6020 Innsbruck, Austria 2 Division of Translational Cell Genetics, Medical University of Innsbruck, 6020 Innsbruck, Austria 3 Department of Medical Epidemiology and Biostatistics, Karolinska Institute, 17177 Stockholm, Sweden 4 Faculty of Medicine and Life Sciences, University of Tampere, Tampere 33014, Finland 5 Department of Orthopedic Surgery, Hospital Wels-Grieskirchen, 4600 Wels, Austria 6 These authors contributed equally 7 Lead Contact *Correspondence: andreas.m[email protected] https://doi.org/10.1016/j.celrep.2019.02.024 SUMMARY Fc receptor for IgM (FcmR)-deficient mice display dysregulated function of neutrophils, dendritic cells, and B cells. The relevance of FcmR to human T cells is still unknown. We show that FcmR is mostly stored inside the cell and that surface expression is tightly regulated. Decreased surface expression on T cells from elderly individuals is associated with alterations in the methylation pattern of the FCMR gene. Binding and internalization of IgM stimulate transport of FcmR to the cell surface to ensure sustained IgM uptake. Concurrently, IgM accumulates within the cell, and the surface expression of other receptors increases, among them the T cell receptor (TCR) and costimulatory molecules. This leads to enhanced TCR signaling, proliferation, and cytokine release, in response to low, but not high, doses of antigen. Our findings indicate that FcmR is an important regulator of T cell function and reveal an additional mode of interaction between B and T cells. INTRODUCTION The Fc receptor for IgM (FcmR) is a transmembrane protein initially referred to as ‘‘Fas apoptosis inhibitory molecule 3’’ (FAIM3) and TOSO (Hitoshi et al., 1998). IgM is bound with high avidity in a 1:1 stoichiometry of FcmRtoIgM(Kubagawa et al., 2009; Shima et al., 2010; Vire et al., 2011). This feature of the receptor led to the misleading assumption of a potent inhibition of Fas/CD95-induced apoptosis in early studies (Hitoshi et al., 1998; Nguyen et al., 2011), which has now been disproved (Honjo et al., 2012b; Kubagawa et al., 2009). We recently demonstrated in FcmR-deficient mice that dysregulated function of neutrophils increased susceptibility to bacterial infection (Lang et al., 2013). Defects in the maturation and differentiation of dendritic cells also impaired viral control (Lang et al., 2015). Regarding adaptive immune cells, micelackingtheFcmR had increased IgG autoantibodies and natural IgM (Honjo et al., 2012a; Ouchida et al., 2012), enhanced differentiation of B-1 cells, and dysregulated homeostasis of B-2 cells (Nguyen et al., 2017a). Increased surface expression of IgM-BCR in FcmR-deficient B cells was demonstrated, and it was concluded that FcmR downregulates surface expression of IgM-BCR (Nguyen et al., 2017a). All mice studies have been performed without analyzing the consequence of IgM binding to its cognate receptor. In contrast to the situation in mice, human FcmRexpressionis restricted to B and T lymphocytes and, to a lesser extent, natural killer (NK) cells, but is not expressed by other hematopoietic cells (Kubagawa et al., 2009; Murakami et al., 2012). Human FcmRisoverexpressedinBcelllymphomas(Vire et al., 2011) and has been linked to disease progression (Li et al., 2011; Pallasch et al., 2008). A functional characterization of FcmR in T cells is still missing, presumably because FcmR is absent on murine T cells (Nguyen et al., 2017a; Shima et al., 2010). T cell activation is a crucial checkpoint in adaptive immunity. Signaling downstream of the T cell receptor (TCR) following antigenic stimulation results in proliferation, differentiation, and effector cytokine release. Dysregulated TCR activation may support immunodeficiency or autoimmunity (Notarangelo, 2014; Theofilopoulos et al., 2017). Cellular signaling is strictly regulated and is known to exhibit thresholds (Das et al., 2009; van den Berg et al., 2013). The number of triggered TCRs is essential, and a reduction of surface TCR expression severely compromises the capacity to reach the activation threshold (Viola and Lanzavecchia, 1996). With age, an increased TCR activation threshold leads to a reduced signaling capacity of the ERK pathway which impairs signal strength and activation of individual T cells (Li et al., 2012). In addition, extracellular factors such as the dose of antigen and the duration and the strength of TCR signaling influence T cell activation (Constant et al., 1995; Huppa et al., 2003; Kalergis et al., 2001; van Panhuys et al., 2014). We designed a study to investigate the consequences of IgM binding to FcmR on human T cells. We demonstrate that with T cell Cell Reports 26, 2681–2691, March 5, 2019 ª2019 The Author(s). 2681 This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). differentiation, and following antigenic stimulation of the TCR, surface FcmR is strongly downregulated. FcmR expression is significantly reduced in naive and memory T cell populations from elderly people. Most FcmR is stored within the cell and traffics continuously to the cell membrane. Sustained FcmR expression on the cell surface leads to accumulation of the complex IgM:FcmR within the cell. Enrichment of IgM accelerates protein transport between the cell surface and the cell interior and thereby increases the expression of TCR and costimulatory molecules. Thus, FcmR-mediated binding of IgM increases TCR signaling, proliferation, and cytokine secretion of peripheral T cells. In contrast, bone marrow (BM) microenvironment downregulates surface FcmR expression, keeping the T cells in a resting state. RESULTS Differentiation and Activation of Human CD4 + T Cells Downregulate Surface FcmR Expression As demonstrated for B cells (Kubagawa et al., 2009; Nguyen et al., 2017a; Vire et al., 2011), IgM was also bound by the FcmR on CD4 + T cells, and the complex FcmR:IgM was internalized (Figure 1A). ImageStream analysis demonstrated a concentrationdependent binding and internalization of IgM, reaching a peak between 10 and 50 mg/mL of IgM (Figure 1B). Next we analyzed FcmR expression on naive, central memory (cMEM), and effector memory (eMEM) CD4 + T cells defined by the markers CCR7 and CD45RA (Figure 1C). With differentiation from naive to antigenexperienced T cells, FcmR expression decreased significantly AB DC EF G A B D C E F G Figure 1. FcmR Enables Human T Cells to Bind IgM but Is Downregulated by Differentiation and TCR-Mediated Activation (A) Confocal microscopy of CD4 + T cells (blue), incubated with rabbit a-FcmR and IgM-Alexa 488 (green) for 2 h at 37C in serum-free medium. Fixed and permeabilized cells were co-stained with secondary Cy3 a-rabbit (red) to visualize internalized FcmR. (B) PBMCs were incubated for 2 h with IgMAlexa 488 using concentrations between 0.7 and 47 mg/mL. IgM binding and internalization were quantified using ImageStream. (C) Flow cytometry of PBMCs gated on live CD3 + CD4 + T cells. A contour plot of naive (CCR7 + CD45RA + , black), cMEM (CCR7 + CD45RA  , blue), and eMEM (CCR7  CD45RA  ,red)CD4 + T cells is shown (left). Right: overlay histogram showing surface expression of FcmR protein on naive (black), cMEM (blue), and eMEM (red) CD4 + T cells. Gray represents fluorescence minus one (FMO). (D) Flow cytometry quantification of surface FcmR expression on CD4 + T cells (gated as in C) and presented as change in mean fluorescence intensity (MFI). (E and F) Flow cytometry analysis of IgM binding on naive, cMEM, and eMEM CD4 + T cells and presented as change in MFI (F). Overlay histogram (E) showing IgM binding on naive (black), cMEM (blue), and eMEM (red) CD4 + T cells. FMO control (without IgM-Alexa 488 incubation) is shown in gray. (G) Magnetic-activated cell sorting (MACS) sorted CD4 + T cells MACS were stimulated for 6 and 24 h with 1 mg/mL a-CD3 or 1 mg/mL a-CD3 plus 1mg/mL a-CD28 in serum-free medium. Surface expression of FcmR on CD4 + T cells was assessed using flow cytometry and presented as change in MFI. Samples of six to ten donors per group pooled from at least three independent experiments (B–G) or one representative picture of three stained samples (A). Data are shown as mean ±SEM. **p < 0.01 and ***p < 0.001 (Wilcoxon matched-pairs test). See also Figure S1. 2682 Cell Reports 26, 2681–2691, March 5, 2019 on peripheral blood (PB) CD4 + T cells (Figure 1D). Consequently, naive CD4 + T cells bound significantly more IgM than cMEM and eMEM T cells, as indicated by their higher mean fluorescence intensity of IgM (Figures 1E and 1F). The CD8 + T cell compartment also showed a downregulation of surface FcmR expression from naive to antigen-experienced cells (Figure S1A), but the surface FcmR was lower on naive and cMEM CD8 + T cells compared with their CD4 + T cell counterparts (Figure S1A). In line with the surface expression, CD8 + T cells bound and internalized lower amounts of IgM compared with CD4 + T cells (Figure S1B). A hallmark of T cells is their capacity to expand and differentiate after TCR-mediated activation. Thus, we investigated whether signaling via the TCR could influence FcmR expression. Six and 24 h after aCD3 or aCD3 plus aCD28-mediated TCR activation, FcmR protein was strongly downregulated (Figure 1G). Together AB DC E Figure 2. Reduced FcmR Expression on CD4 + T Cells Is Associated with Changes in FCMR Methylation during Aging (A and B) PBMCs were obtained from young (<35 years) and elderly (>65 years) donors. (A) Flow cytometry quantification of surface FcmR expression on naive, cMEM, and eMEM CD4 + T cells, presented as change in MFI. (B) Naive, cMEM, and eMEM CD4 + T subpopulations were sorted using flow cytometry. Expression of FCMR mRNA, normalized to the control gene b-Actin, was measured. (C–E) FCMR transcript and methylation pattern of the FCMR gene were analyzed in PBMC samples obtained from the Vitality 90+ study. (C) Expression of FCMR mRNA and (D) agerelated differences in methylation sites in FCMR gene from young and elderly donors. (E) Correlations of the methylation sites cg05721773, cg22671342, cg22945467, and cg23088126 with FCMR mRNA. Five to 13 samples per group pooled from at least three independent experiments (A and B). Twenty-one young and 122 old individuals derived from the Vitality 90+ study (C–E). Data are shown as mean ±SEM. *p < 0.05, **p < 0.01, and ***p < 0.001 (unpaired Mann-Whitney test, A–D). Each dot point represents one individual. Spearman’s coefficient (r) and p value are shown in the graph (E). See also Figure S2 and Table S1. these data demonstrate that FcmRis highly expressed on the surface of naive PB CD4 + T cells, which enables them to bind high amounts of IgM. Following TCR-mediated activation and with differentiation, FcmR expression declines on T cells. Age-Related Alterations in the Methylation of the FCMR Gene Decrease mRNA and Protein Expression Aging leads to a progressive diminution of the naive and expansion of the eMEM T cell compartment (Grubeck-Loebenstein et al., 1998; Linton and Dorshkind, 2004). To assess whether this change was reflected by low FcmR expression, we analyzed the expression of FcmR on the subpopulations of CD4 + T cells from healthy young (<35 years) and elderly (>65 years) persons. FcmR expression was decreased on naive and memory CD4 + T cells obtained from elderly donors compared with their counterparts from young donors (Figure 2A). Consistently, IgM binding was strongly reduced on all CD4 + T cell subpopulations from elderly individuals (Figure S2). Because naive and antigen-experienced T cells were equally affected, the expansion of memory T cells was presumably not the only cause of decreased FcmR protein in CD4 + T cells from elderly persons. We were therefore interested in whether transcriptional and genomic regulatory mechanisms were involved. For this purpose, we purified CD4 + T cell Cell Reports 26, 2681–2691, March 5, 2019 2683 subpopulations from young and elderly persons. Consistent with surface protein expression, the FCMR mRNA expression level was highest in naive CD4 T cells obtained from young donors (Figure 2B). The mRNA transcript decreased with age and differentiation. To strengthen our observation that aging leads to decreased mRNA transcript, we analyzed FCMR mRNA and its association with the methylation levels in the FCMR gene in more than 100 PB mononuclear cell (PBMC) samples obtained from the Vitality 90+ study (Marttila et al., 2015). The expression of FCMR transcripts was significantly downregulated in elderly individuals (Figure 2C). Next we examined whether the 16 methylation sites in the FCMR gene that were present in the Illumina 450k array were affected by age. Six of these sites showed differential methylation between young and elderly individuals (Figure 2D). A significant correlation between FCMR transcript and methylation level was observed for 4 of these sites (Figure 2E; Table S1). DNA methylation is known to regulate gene expression, but the exact mechanisms are unknown. Promoter methylation, however, typically represses gene expression by blocking the binding of transcription factors. One of the 4 sites, cg22945467, is located upstream of a transcription start site (TSS1500); hence, the observed inverse correlation between the methylation level on this site and FCMR expression aligns with the scenario of hypermethylation leading to downregulated expression. This findings suggest that the aging-associated alterations in FCMR methylation could affect its expression in an age-dependent manner. Agerelated alterations of surface receptors may at least partially contribute to decreased T cell response in old age (Li et al., 2012), but a potential impact of FcmR on T cell function has not yet been investigated. Tissue Microenvironment Modulates Surface FcmR Expression Independent from an Intracellular Reservoir Confocal microscopy of the T cell zone from human tonsil indicated FcmR localization on the cell surface as well as inside the cell (Figure 3A). Thus, we examined FcmR in permeabilized and non-permeabilized PB CD4 + T cells. A strong intracellular presence of FcmR compared with the surface was visible (Figure 3B). For a better quantification, surface and intracellular expression was measured using flow cytometry. Just a small proportion of the total FcmR protein was detected on the cell surface (Figure 3C). Again, surface FcmR decreased with differentiation from naive to antigen-experienced cells, whereas the highest intracellular protein amount was present in cMEM T cells (Figure 3D). After antigen clearance, antigen-specific T cells migrate to the BM, where they reside as long-lived memory cells (Pangrazzi et al., 2017; Tokoyoda et al., 2009). To check whether this affects either surface or intracellular FcmR expression, we isolated CD4 + T cells from BM and PB of the same donor. In every donor, the expression of FcmR was significantly lower on BM CD4 + T cells compared with the corresponding PB CD4 + T cell subpopulations (Figures 3E, 3F, and S3A). BM CD4 + T cells still stored large amounts of FcmR within the cell but significantly less compared with PB CD4 + T cell subpopulations (Figures 3E and 3F). We speculated whether the BM environment might affect the surface expression of FcmR. Indeed, 20 h culture in serum-free medium raised the surface expression of FcmRon BM CD4 + T cells (Figure 3G). Typical BM T cell survival cytokines might be responsible for decreased FcmR surface expression. Therefore, we treated BMMCs for 20 h with IL-6, IL-7, and IL-15 or combinations of these cytokines. With the exception of IL-6 alone, these cytokines downregulated the surface expression of FcmR on BMMCs (Figure 3H) and even more on PB CD4 + T cells (Figure S3B) during 20 h culture. Our data indicate a tissue-specific regulation of surface FcmR expression mediated by the microenvironment. The FcmR is stored within T cells. Presumably, fast re-expression on the surface is therefore possible when BM T cells migrate from the BM back to the blood. Intracellular FcmR Traffics Continuously to the Cell Surface Leading to IgM Enrichment Inside the Cell We wondered why T cells store large amounts of FcmR. Thus we investigated the possibility of whether intracellular FcmR traffics continuously to the plasma membrane and back. To address this point, T cells were incubated with a-FcmR antibody (Ab) at 4C for 30 min, washed, and shifted to 37C for 20 h. During this second incubation step, FcmR disappeared from the cell surface but could still be found in the cytoplasm (Figure 4A, top). In contrast, strong intracellular and surface FcmR staining was visible when cells were incubated in the presence of a-FcmRAbat37 C for 20 h (Figure 4A, bottom). To provide further evidence of possible FcmR recycling between the cell interior and the plasma membrane, we treated CD4 + T cells with actinomycin D, brefeldin A, MG132, and IgM for 20 h in the presence of a-FcmR Ab. As expected, membrane trafficking was totally inhibited by brefeldin A but was not influenced by actinomycin D and MG132, suggesting that neither de novo synthesis nor proteasomal degradation of FcmR played a major role (Figure 4B). Surprisingly, even in the presence of IgM, an increased amount of FcmR was expressed on the cell surface (Figure 4B), indicating that even following IgM-mediated internalization, receptor recycling took place. To investigate the functional impact of FcmR recycling, we incubated CD4 + T cells with IgM and a-FcmR Ab for 2 and 20 h and documented IgM uptake by confocal microscopy. IgM internalization increased, and IgM and FcmR were highly enriched within the cells after 20 h compared with 2 h (Figure 4C). Flow cytometry confirmed the observed IgM enrichment during 20 h culture (Figure 4D). To study whether endocytosis might be involved in IgM uptake, we stained early endosomes. After 2 h, internalized IgM co-localized with EEA1 (early) endosomes (Figure 4E). As control for specific co-localization of IgM and EEA1 and to exclude coincidental overlay, we stained for Rab7 (late) and Rab11 (recycling). Both endosomal markers did not co-localize with IgM. These data show that FcmR traffics continuously to the plasma membrane, aiming to catch as much IgM as possible. The internalized IgM accumulates within the cell rather than being degraded. FcmR Acts as a Costimulatory Molecule Enhancing TCR Signaling, Proliferation, and Cytokine Production The function of FcmR and the consequences of IgM internalization and enrichment in T cells have not yet been investigated. Because IgM accumulates in T cells over time, we analyzed T cell proliferation, administrating IgM prior to and 2684 Cell Reports 26, 2681–2691, March 5, 2019 simultaneously with TCR-mediated activation. Twenty hours of preincubation with IgM significantly enhanced CD4 + and CD8 + T cell proliferation (Figures 5A, 5B, and S4A). This effect was absent following simultaneous administration of IgM and TCR stimuli (Figures 5A and 5B) or following preincubation with IgM for a period shorter than 20 h (Figure S4B). In contrast, two recent studies showed an inhibitory effect of IgM on T cell proliferation but did not provide an explanation for the underlying mechanism (Colucci et al., 2015; Lloyd et al., 2017). Experiments in both studies were performed in RPMI-1640 supplemented with 10% fetal bovine serum (FBS) using high concentrations of TCR stimuli (a-CD3/a-CD28-coated beads or PMA) and IgM purchased from Sigma. Sigma IgM contains sodium azide, whereas Jackson IgM (purchased from Jackson Immunoresearch), used in our experiments, is free of any preservative. We therefore AB CD FE GH Figure 3. Tissue-Specific Regulation of Surface and Intracellular FcmR Expression (A) Confocal microscopy of a paraffin-embedded tonsil stained with a-CD3 (green), DAPI (blue), and a-FcmR (red). (B) Confocal microscopy of non-permeabilized (-perm; top) and permeabilized (+perm; bottom) CD4 + T cells, stained with a-CD4 (green), DAPI (blue), and a-FcmR (red). (C and D) Flow cytometry analysis of surface and intracellular expression of FcmR protein on naive, cMEM, and eMEM CD4 + T cells (D). Overlay histogram (C) showing surface (left) and intracellular (right) expression of FcmR protein (red) and control staining (FMO; black). (E and F) Surface and intracellular expression of FcmR protein on cMEM and eMEM T cells from PB and BM of the same donor (F). Overlay histogram (E) showing surface and intracellular expression of FcmR protein on cMEM CD4 + T cells. PB surface (red, solid), BM surface (blue, solid), PB intracellular (red, dashed), and BM intracellular (blue, dashed) are shown. (G) Surface expression of FcmR on BM CD4 + T cells was measured directly after isolation of BM mononuclear cells (BMMCs) (0 h) and after 20 h culture at 37C in serum-free medium. (H) BMMCs were incubated for 20 h with 100 ng/mL of IL-6, IL-7, IL-15, or a combination of 100 ng/mL IL-6 and IL-15 or IL-7 and IL-15 in serum-free medium. Surface expression of FcmR on BM CD4 + T cells was measured using flow cytometry, and the fold change of untreated samples was calculated. Eight to ten samples per group pooled from at least three independent experiments (C–H) or one representative picture of three stained samples and tissues (A and B). Data are shown as mean ± SEM. *p < 0.05 and **p < 0.01 (Wilcoxon matchedpairs test). See also Figure S3. analyzed whether this might explain the discrepancy between the results (Figures S4C and S4D). Adding sodium azide to Jackson IgM results in reduced proliferation (Figures S4E and S4F), whereas the dialyzation of Sigma IgM diminished the inhibitory effect (Figure S4G). Using the standard T cell culture medium RPMI plus 10% fetal calf serum (FCS), we could not observe any difference in T cell proliferation in response to additional IgM (Figure S4H). FCS contains large amounts of natural IgM, and we cannot exclude that the human FcmR interacts with calf IgM. Therefore, we performed our experiments in serum-free medium optimized for T cell culture. To provide further evidence of a costimulatory function of FcmR, we stimulated PBMCs with different concentrations of aCD3 plus aCD28. IgM increased T cell proliferation when TCR stimuli were limited, and a high concentration of aCD3 plus aCD28 (10 mg/mL) overrode the need for FcmR-mediated costimulation (Figure 5C). To determine whether IgM binding to FcmR on CD4 + T cells is responsible for increased proliferation Cell Reports 26, 2681–2691, March 5, 2019 2685 or whether this effect is mediated by other cells, we purified CD4 + T cells from PB. CD4 + T cells stimulated with aCD3, or aCD3 plus aCD28 proliferated faster, and the frequency of CD25 + cells was strongly increased by IgM preincubation (Figures 5D–5F). Next, we were interested in whether the continuous presence of IgM influences T cell activation after several stimulation cycles, or whether it induces exhaustion. First PBMCs were stimulated with aCD3, then with IL-2, and finally with different combinations of aCD3 and IL-2 (Figure S5A). CD4 + T cells continuously stimulated in the presence of IgM kept proliferation increased 3 and 4 days after the last activation cycle (Figures S5B and S5C). Next, we quantified the production of IFNgover a time period of 48 h. IgM preincubation led to an increased IFNgsecretion of CD4 + T cells stimulated with aCD3 or aCD3 plus aCD28 at all time points (Figure 5G). To understand how IgM acts on the distal effector functions of TCR and CD28 signaling, including proliferation and cytokine production, we investigated signaling events following TCR and CD28 engagement. We preincubated purified CD4 + T cells with IgM for 20 h or left them untreated before TCR stimulation. Phosphorylation of ERK1/2 was altered by IgM preincubation and strongly enhanced after 30 min of activation with aCD3 plus aCD28 (Figures 5H and S5D). In contrast activation of the classical NFkB pathway was not altered, as measured by phosphorylation of IkBa(Figures 5H and S5E). Along this line, proximal TCR signaling investigated by analysis of Zap70 (Tyr319) phosphorylation was not significantly altered 1, 3, or 5 min after activation of IgM-treated or untreated cells (data not shown). Costimulatory function of FcmR after activation therefore does not influence early proximal TCR signaling but converges later, specifically at the Erk1/2 signaling pathway. Our findings show that preincubation with IgM boosts TCR signaling, proliferation, and cytokine production when antigen exposure is relatively low. AB DC E Figure 4. FcmR Trafficking between Cell Surface and Cell Interior Results in IgM Enrichment (A) PBMCs were incubated for 30 min at 4C with rabbit a-FcmR Ab, then washed and shifted to 37C for 20 h (top) in serum-free medium. Bottom: cells were incubated for 20 h at 37C in the presence of a-FcmR Ab in serum-free medium. After surface staining with a-CD4 (blue), cells were fixed, permeabilized, and stained with secondary a-rabbit Cy3 (red). (B) PBMCs were incubated either for 30 min at 4C with rabbit a-FcmR Ab or for 20 h at 37Cin the presence of a-FcmR Ab in serum-free medium. In addition, cells were pretreated for 0.5 h with 10 mg/mL brefeldin A (BFA), 20 ng/mL actinomycin D (ActD), or 1 mM MG132 prior to a-FcmR Ab incubation. Alternatively, 47 mg/ml IgM was added together with a-FcmR Ab. After surface staining of CD3 + CD4 + T cells, cells were fixed, permeabilized, and stained with secondary a-rabbit Alexa 647. (C) Confocal microscopy of T cells, incubated for 2and20hat37 C with rabbit a-FcmR AB and 47 mg/ml IgM-Alexa 488 (green) in serum-free medium. After washing, cells were stained with a-CD4 (blue) and then fixed and permeabilized. Cells were co-stained with secondary a-rabbit Cy3 (red) to visualize surface and internalized FcmR. (D) PBMCs were incubated for 2 h at 37Cor20h at 4C and 37C with 47 mg/mL IgM-Alexa 488. Flow cytometry analysis of IgM binding on CD4 + T cells presented as change in MFI. (E) PBMCs were incubated for 2 h at 37C with 47 mg/mL IgM-Alexa 488 (green). After surface staining with a-CD4 (blue), cells were fixed, permeabilized, and stained with a-EEA1 (top), a-Rab7 (middle), or a-Rab11 (bottom) plus secondary anti-rabbit-Cy3 (red). Eight or nine samples per group pooled from at least three independent experiments (B and D) or one representative picture of three stained samples and tissues (A, C, and E). Data are shown as mean ±SEM. **p < 0.01 (Wilcoxon matched-pairs test). 2686 Cell Reports 26, 2681–2691, March 5, 2019 Intracellular Accumulation of IgM Increases Surface Expression of TCR and CD28 by Regulating Protein Transport to the Cell Surface We demonstrated that T cells ensured high IgM uptake, which then led to increased TCR signaling. Because of the increased phosphorylation of ERK1/2, we focused on upstream molecules of the TCR pathway, which might be modulated by IgM enrichment. IgM preincubation enhanced surface expression of CD3 and CD28 on CD4 + T cells (Figures 6A and 6B). Consistent with the FcmR expression profile on CD4 + and CD8 + T cell subpopulations (Figures 1D and S1A), the percentage increase of CD3 and CD28 was higher on naive CD4 + and CD8 + T cells compared with antigen-experienced T cells (Figures 6C and S6A–S6C). The increased surface expression of the molecules induced by IgM was not due to alterations in gene transcription (Figure 6D). Generally, TCR expression is dependent on a balance of de novo synthesis, recycling, and degradation (Geisler, 2004). To investigate which of these pathways were most affected, we incubated PBMCs with IgM in the presence of actinomycin D, cycloheximide, or MG132. TCR expression was strongly enhanced by IgM regardless of inhibition of mRNA transcription, protein synthesis, or proteasomal AB D C E F HG Figure 5. Intracellular Accumulation of IgM Enhances T Cell Proliferation, Cytokine Production, and Signaling (A and B) IgM (47 mg/mL) was administered to division tracking-labeled PBMCs, 20 h prior to or simultaneously with TCR stimulation using 1mg/mL a-CD3 or 1 mg/mL a-CD3 plus 1 mg/mL a-CD28 in serum-free medium. As control, PBMCs were kept in serum-free medium without IgM. (A) Representative histograms of CD4 + T cells without IgM treatment (left), simultaneous IgM treatment and TCR engagement (middle), or preincubated with IgM (right). (B) Proliferation of CD4 + T cells with or without IgM administration was measured using flow cytometry at day 4 after stimulation. (C) IgM (47 mg/mL) was administered to division tracking-labeled PBMCs, 20 h prior to stimulation (+IgM). As control, PBMCs were kept in serumfree medium without IgM (IgM). PBMCs were stimulated with the indicated concentrations of a-CD3 plus a-CD28 in serum-free medium. Proliferation of CD4 + T cells was quantified using flow cytometry at day 4 after stimulation. (D–F) MACS-sorted CD4 + T cells were preincubated for 20 h with 47 mg/mL IgM (+IgM) or kept in serum-free medium (IgM). After preincubation, T cells were transferred to plates coated with 1 mg/mL a-CD3, and 1 mg/mL soluble a-CD28 was added to some wells. (D) Dot plots show CD25 expression and proliferation of control (IgM, left) and IgM-preincubated (+IgM, right) CD4 + T cells following activation with a-CD3 plus a-CD28 at day 4. Red numbers indicate the frequency of cells in each quadrant. (E and F) Proliferation (E) and expression of CD25 (F) were quantified using flow cytometry at days 3 and 4 after stimulation. (G) MACS-sorted CD4 + T cells were preincubated for 20 h with 47 mg/mL IgM (+IgM) or kept in serum-free medium (IgM). Supernatants were collected after 6, 12, 24, and 48 h after transfer to plates coated with 1 mg/mL a-CD3, and 1 mg/mL soluble a-CD28 was added to some wells. ELISAs for IFNgof a-CD3 (left) or a-CD3 plus a-CD28 (right) activated cells were performed. (H) MACS-sorted CD4 + T cells were preincubated for 20 h with 47 mg/mL IgM (+IgM) or kept in serum-free medium (IgM). After preincubation, T cells were transferred to plates coated with 1 mg/mL a-CD3, and 1 mg/mL soluble a-CD28 was added. Cells were harvested at indicated time points. Immunoblotting analysis of total and phosphorylated ERK1/2 and IkBawas performed. Six to 19 samples per group pooled from at least three independent experiments (A–G) or one representative gel out of three (H). Data are shown as mean ± SEM. *p < 0.05 and **p < 0.01 (Wilcoxon matched-pairs test in B, C, E, and F or two-way ANOVA in G). See also Figures S4 and S5. Cell Reports 26, 2681–2691, March 5, 2019 2687 degradation (Figure 6E). Therefore, we focused on TCR recycling between the cytoplasmic pool and the cell surface, assuming that protein transport might play a role. Control and IgM-preincubated cells were treated with PDBu (phorbol 12,13-dibutyrate) for 1 h, which resulted in 20% TCR internalization (Figure 6F). The time course of surface CD3 recovery after PDBu removal was monitored using flow cytometry. IgM-preincubated cells recovered surface CD3 within 30 min, while control cells did not reach basal level in the observation time (Figure 6F). Next, PBMCs were incubated with IgM for 20 h or left untreated, and internalization of CD3 following TCR engagement was measured using flow cytometry. Preincubation with IgM enhanced internalization of CD3 (Figure 6G). To evaluate whether FcmR acts specifically on TCR recycling or generally on protein transport, we measured the kinetics of the de novo synthesized protein CD69 on the cell surface and within the cell. Two hours following TCR activation, the amount of intracellular CD69 was similar in IgM-preincubated and control cells, but surface expression was significantly increased (Figure 6H). Consistent with a general effect of FcmR on protein transport, the expression of CCR7 and CD45RA was also increased by IgM preincubation (Figures S6D and S6E). These data suggest that FcmR controls and regulates the speed of protein transport between the cell interior and the cell surface and therefore also the surface expression of the TCR and of costimulatory molecules. AB CD EF GH Figure 6. IgM Accumulation within CD4 + T Cells Accelerates Protein Transport (A–C) PBMCs were incubated for 20 h with 47 mg/mL IgM (+IgM) or kept in serum-free medium (IgM). Expression of CD3 and CD28 was measured using flow cytometry. (A) Overlay histograms showing surface expression of CD3 (left) and CD28 (right) on IgM (blue) and +IgM (red) CD4 + T cells. (B) MFI of CD3 (left) and CD28 (right) on CD4 + T cells. (C) MFI of CD3 was quantified on naive, cMEM, and eMEM CD4 + T cells. Numbers indicate the percentage increase as average of all samples, calculated as change in percentage (% = MFI of CD3 on IgM preincubated cells/MFI of CD3 on control cells 3100). (D) MACS-sorted CD4 + T cells were incubated for 6, 12, and 20 h with 47 mg/mL IgM or kept in serum-free medium, and the expression of CD3 (left) and CD28 (right) mRNA, normalized to the control gene b-Actin, was measured. Fold changes between IgM preincubated and control cells are shown. (E) PBMCs were incubated for 20 h with 10 mg/mL cycloheximide, 20 ng/mL actinomycin D, or 1 mM MG132 or kept untreated in serumfree medium in the presence (+IgM) or absence (IgM) of 47 mg/mL IgM. Expression of CD3 was quantified using flow cytometry. Percentage change in surface CD3 expression normalized to control cells without IgM administration is shown. (F) Relative levels of CD3 in IgM preincubated (+IgM) and control (IgM) cells, which were untreated or treated with PDBu for 1 h. After PDBu treatment, cells were washed and incubated at 37C for the indicated times. Percentage change in surface CD3 expression normalized to untreated cells. (G) PBMCs were incubated for 20 h with 47 mg/mL IgM (+IgM) or kept in serum-free medium (IgM). After 20 h, cells were activated with mouse a-CD3 at 37C for the indicated times and then shifted to ice and stained with a-mouse fluorescein isothiocyanate (FITC). Percentage change in surface CD3 expression normalized to unstimulated cells (0 min). (H) PBMCs were preincubated with 47 mg/mL IgM (+IgM) or left untreated (IgM), then cells were stimulated for 2 and 4 h using 1 mg/mL a-CD3 plus 1 mg/mL a-CD28 in the presence or absence of 10 mg/mL BFA in serum-free medium. After washing, CD4 + T cells were stained for surface and intracellular CD69 expression. MFI of CD69 was calculated as change in MFI (DMFI = MFI of stimulated cells MFI of non-stimulated cells). Six to ten samples per group pooled from at least three independent experiments (A–H). Data are shown as mean ±SEM. *p < 0.05 and **p < 0.01 (Wilcoxon matched pairs test in (B–F and H or two-way ANOVA in G). See also Figure S6. 2688 Cell Reports 26, 2681–2691, March 5, 2019 Ab. As control PBMCs were incubated for 30min at 4C with rabbit a-FcmR Ab. After surface staining of CD3 + CD4 + T cells, cells were fixed, permeabilized and stained with secondary a-rabbit Alexa647. Cell Sorting T cells were isolated from fresh PBMCs by magnetic cell sorting using the MACS human Pan T cell isolation kit (Miltenyi Biotech) following manufacturer’s protocol. Purified T cells were then stained with a-CD3, a-CD4, a-CD8, a-CD45RA, and a-CCR7. Labeled cells were sorted with a FACSAria II flow cytometer (BD Biosciences). Cells were sorted into CD4 + CCR7 + CD45RA + , CD4 + CCR7 + CD45RA - and CD4 + CCR7 - CD45RA - T cells. The cells were collected into RPMI 1640 medium containing 10% FCS and washed once prior RNA isolation. The purity of each subset was > 95% as determined by flow cytometry. ImageStream PBMCs were incubated in the presence or absence of IgM-Alexa488 using concentrations between 0.7mg/ml and 47mg/ml. After washing, cells were stained with a-CD4 PE and measured using ImageStream Mark II (MKII) Imaging Flow Cytometer system (Amnis). Cells were gated on aspect ratio versus bright field area to include only single cells, using the gradient root-mean-square feature to include cells in focus. Apoptotic cells were excluded from the analysis based on morphology. Cells in focus were gated for the expression of CD4 and IgM internalization was analyzed using IDEAS software. Immunofluorescence analysis of tonsil biopsies Formalin-fixed, paraffin-embedded 4-mm tonsil sections were deparaffinised in xylene and re-hydrated in ethanol. The slides were boiled in 0.01 M citrate buffer (pH 6) for 16 min in the microwave for epitope retrieval and allowed to cool for about 1 h at room temperature. After cooling the slides were washed twice with Aqua dest. and with TBS for 5min. Slides were blocked with 3% skim milk in TBS/Tween for 20 min at room temperature. Rabbit a-FcmR (Sigma Aldrich) and mouse a-CD3 were administered in TBS and incubated overnight at 4C. After washing three times with TBS/Tween, the slides were incubated with a-mouse Alexa488, a-rabbit Cy3 and dapi for 1h at 4C. The slides were washed three times with TBS/Tween and twice with TBS for 5min. Cell Voyager CV1000 microscope (Yokogawa-Bruker) were used to analyze stained slides. Immunofluorescence confocal microscopy FcmR mediated IgM uptake: Purified CD4+ T cells or PBMCs were incubated with 47mg/ml IgM-Alexa488 in presence of rabbit a-FcmR (Sigma Aldrich). Surface were stained with a-CD4-BV421 and cells fixed with BD cytoperm/cytofix buffer and washed with Perm/Wash buffer (BD). The cells were washed twice with PermWash buffer and PBS Cells and stained with a-rabbit Cy3 to visualize FcmR. IgM colocalization in endosomes: PBMCs were incubated with 47mg/ml IgM Alexa488.The surface was stained with a-CD4-BV421 and cells were fixed with BD cytoperm/cytofix buffer and washed with Perm/Wash buffer (BD). Cells were stained with rabbit a-EEA1, a-Rab7 or a-Rab11 (Cell Signaling). After washing, a-rabbit Cy3 (Abcam) was used to visualize endosome marker. FcmR surface trafficking: PBMCs were incubated for 30min at 4C in the presence of rabbit a-FcmR (Sigma Aldrich), then washed and shifted for 20 hours to 37C. Additionally cells were incubated for twenty hours at 37C in the presence of a-FcmR Ab. The surface was stained with a-CD4-BV421 and cells were fixed with BD cytoperm/cytofix buffer, washed with Perm/Wash buffer and stained with a-rabbit Cy3 to visualize FcmR which had been expressed on the surface during incubation time in the presence of rabbit a-FcmR Ab. After the last incubation step, cells were resuspended in 100ml PBS, transferred to coverslips by cytospin and mounted with fluorescent mounting medium (Dako). Images were acquired on Cell Voyager CV1000 (Yokogawa-Bruker). ELISA assay Purified CD4 + T cells were seeded at a density of 1 310 6 cells and incubated with 47mg/ml IgM for 20h or kept untreated in serum-free medium. Supernatants were collected after six, twelve, twenty-four and forty-eight hours after transfer to plates coated with 1mg/ml a-CD3 and 1mg/ml soluble a-CD28 was added to some wells. IFNgwas assessed in supernatants using Elisa kits (Biolegend) Western blot CD4 + T cells were isolated from PBMCs by negative selection with MACS and preincubated for 20 hours with 47mg/ml IgM or kept in serum-free medium. After preincubation, T cells were transferred to plates coated with 1mg/ml a-CD3 and 1mg/ml soluble a-CD28 was added. Cells were lysed in ice-cold lysis buffer [5mM NaP2P, 5mM NaF, 5mM Na3VO4, 5mM EDTA, 150mM NaCl, 50mM Tris (pH 7.3), 1% NP-40, aprotinin and leupeptin (50 mg/ml each)] and centrifuged at 15000xg for 15 min at 4C. Cell lysates were electrophoresed on a NuPAGE gel (Invitrogen) and transferred onto a polyvinylidene difluoride (PVDF) membrane (Millipore) by semi-dry blotting (100mA, 90 min). The primary antibodies were diluted in tris-buffered saline containing 0.5% Tween-20 and either with 5% nonfat dry milk or with 5% bovine serum albumin (BSA) for phospho antibodies. Peroxidase-conjungated antibodies (Pierce) served as secondary reagents (1:5000). Enhanced chemiluminescence was used for antigen detection (Super Signal, Thermo Fischer). As a loading control for whole cell lysates b-actin (Santa Cruz Biotechnology,sc-1615) was applied. Protein lysates were subjected to immunoblotting with antibodies against phospo-IkBa(Ser32), clone 14D4 (Cell signaling, #2859), pan IkBa(Cell Cell Reports 26, 2681–2691.e1–e5, March 5, 2019 e4 signaling, #9242), phospho-ERK1/2 (Thr202/Tyr204) (Cell signaling, #9101) and pan ERK1/2 (Cell signaling, #9102). Blot pixel densities were quantified using ImageJ (NIH) and normalized to actin (Schindelin et al., 2012). QUANTIFICATION AND STATISTICAL ANALYSIS All data are shown as mean ±standard error of the mean (SEM). Statistical analysis was performed using GraphPad Prism software version 5.0 (GraphPad Software). To determine the significance of differences between two groups, the Wilcoxon matched pairs test, the unpaired Mann-Whitney test and two way ANOVA test were used, as indicated in the figure legends. For correlations, statistical significance was assessed by Spearman correlation analysis. A p value less than 0.05 was considered significant, except in the assessment of the age-related differences in the FCMR methylation sites between young and elderly, a more stringent p value threshold (p < 0.001) was used. e5 Cell Reports 26, 2681–2691.e1–e5, March 5, 2019