Interleukin‐6 signalling mediates Galectin‐8 co‐stimulatory activity of antigen‐specific CD4 T‐cell response

Summary

Galectin‐8 (Gal‐8) is a mammalian lectin endowed with the ability to co‐stimulate antigen‐specific immune responses. We have previously demonstrated that bone‐marrow‐derived dendritic cells produce high levels of interleukin‐6 (IL‐6) in response to Gal‐8 stimulation. As IL‐6 is a pleiotropic cytokine that has a broad effect on cells of the immune system, we aimed to elucidate whether IL‐6 was involved in Gal‐8‐dependent co‐stimulatory signals during antigen recognition by specific CD4 T cells. With this aim, splenocytes from DO11.10 mice were incubated with a low dose of the cognate ovalbumin peptide in combination with Gal‐8. Interleukin‐6 was found significantly increased in cultures stimulated with Gal‐8 alone or Gal‐8 plus cognate peptide. Moreover, IL‐6 signalling was triggered during Gal‐8‐induced co‐stimulation, as determined by phosphorylation of signal transducer and activator of transcription 3. Interleukin‐6 blockade by neutralizing monoclonal antibody precluded Gal‐8 co‐stimulatory activity but did not affect the antigen‐specific T‐cell receptor activation. Different subsets of dendritic cells, as well as macrophages and B cells, were identified as the cellular source of IL‐6 during Gal‐8‐induced co‐stimulation. To confirm that IL‐6 mediated the Gal‐8 co‐stimulatory effect, antigen‐presenting cells from IL‐6‐deficient or wild‐type mice were co‐cultured with purified CD4 T cells from OTII mice in the presence of cognate peptide and Gal‐8. Notably, Gal‐8‐induced co‐stimulation, but not the antigen‐specific response, was significantly impaired in the presence of IL‐6‐deficient antigen‐presenting cells. In addition, exogenous IL‐6 fully restored Gal‐8‐induced co‐stimulation. Taken together, our results demonstrate that IL‐6 signalling mediates the Gal‐8 immune‐stimulatory effect.

Abbreviations

APC

antigen‐presenting cell

BMDC

bone marrow‐derived dendritic cell

DC

dendritic cell

FMDV

foot‐and‐mouth‐disease virus

Gal

galectin

IFN‐γ

interferon‐γ

IL‐6

interleukin‐6

IL‐6R

interleukin‐6 receptor

KO

knockout

MHCII

major histocompatibility complex class II

OVA

ovalbumin

pSTAT3

phosphorylated signal transducer and activator of transcription 3

TCR

T‐cell receptor

Tfh

follicular helper T cell

TLR

Toll‐like receptor

WT

wild‐type

Introduction

Galectins constitute a family of mammalian lectins characterized by the presence of conserved carbohydrate‐recognition domains that bind to N‐acetyl‐lactosamine‐containing glycans on target cells. Several studies positioned galectins as cue mediators of the immune response because they are implicated in many different processes such as tumour escape, autoimmune disorders, tolerance induction and host defence.1, 2 Our group previously demonstrated that galectin‐8 (Gal‐8) has a predominant activating role in the elicitation of primary adaptive immune response.3, 4, 5 We found that Gal‐8 co‐stimulates borderline antigen‐specific T‐cell responses by synergizing the T‐cell receptor (TCR) ‐specific signalling in the presence of a low dose of the antigen. From a mechanistic view, Gal‐8 strengthens TCR signalling through activation of the same pathways triggered by the antigen.4

Although CD4 T cells were identified as target cells of the Gal‐8 co‐stimulatory effect, the simultaneous presence of antigen‐presenting cells (APC) and the antigen is required to reach the galectin booster of the T‐cell activation.4 This observation unveils a role for Gal‐8 on the cooperative link between APC and CD4 T cells during antigen presentation. In line with this, we have recently reported that Gal‐8 induces full activation of mouse bone‐marrow‐derived dendritic cells (BMDC) and splenic dendritic cells (DC), which may represent one of the mechanisms involved in the elicitation of the adaptive immune response previously observed. Indeed, Gal‐8‐stimulated DC showed increased expression of major histocompatibility complex class II (MHCII), CD80 and CD86 molecules, an augmented capacity to activate antigen‐specific T‐cell responses, and a strong production of interleukin‐6 (IL‐6), among other inflammatory cytokines.5

Interleukin‐6 is a pleiotropic cytokine that provides key early signals to shape the innate immune response,6 CD4 and CD8 T‐cell effector responses, as well as memory formation.7, 8, 9, 10, 11, 12, 13 In particular, IL‐6 has been identified as a key driver of follicular helper T (Tfh) cell differentiation, which supports high‐affinity antibody production by germinal centre B cells.14, 15, 16 At a molecular level, IL‐6 positively regulates cathepsin S expression in DC, which leads to a more diverse repertoire of Tfh cells.17 Recently, Brahmakshatriya et al.18 demonstrated that high levels of IL‐6 directly delivered from Toll‐like receptor (TLR) ‐pre‐activated DC are a central factor for generating optimal helper T‐cell responses that drive an effective humoral immunity in aged mice. In this regard, we have recently demonstrated that a single dose of inactivated foot‐and‐mouth‐disease virus (FMDV) formulated with soluble Gal‐8 triggers a neutralizing humoral response that enhanced protection against homologous viral challenge. Remarkably, the Gal‐8‐induced anti‐FMDV response was preceded by a peak of IL‐6 and interferon‐γ (IFN‐γ) production at 48 hr post immunization, which returned to basal levels at day 5 after immunization, suggesting that Gal‐8 circumscribed its effects to the very beginning of the immune response induction.5 Considering that inactivated virus‐pulsed DC secrete IL‐6, which is crucial for the rapid anti‐FMDV antibody response,19 we postulated that Gal‐8‐induced DC‐derived IL‐6 could be involved in the stimulation of humoral response in the FMDV experimental model.

Despite the accumulating data regarding IL‐6 production during the immune‐stimulatory activity exerted by Gal‐8, the actual involvement of IL‐6 in the Gal‐8‐induced adaptive response still remains to be investigated.

Methods

Mice

C57BL/6J, C.Cg‐Tg(DO11.10)10Dlo/J (DO11.10), B6.Cg‐Tg(TcraTcrb)425Cbn/J (OTII) and B6.129S2‐Il6^tm1Kopf/J (IL‐6KO) breeding pairs were obtained from The Jackson Laboratory (Bar Harbor, ME) and bred in our facilities. The Ethics Committee Boards of the Universidad Nacional de San Martín and CIBICI‐CONICET approved all procedures involving animals.

Galectins

Recombinant Gal‐8 was obtained as described previously.5 Briefly, Gal‐8L mouse isoform (GenBank EF524570) was expressed in Escherichia coli BL‐21 and purified by Lactosyl‐Sepharose (Sigma‐Aldrich, St Louis, MO) followed by immobilized metal affinity chromatography. Lectin activity was tested by haemagglutination as reported elsewhere.20

Splenocytes and CD4 T‐cell purification or depletion

Mouse splenocytes were obtained as described previously.4 For CD4 T‐cell purification, a MojoSort Mouse CD4 Naive Cell Untouched Isolation kit was used, following the manufacturer's instructions (BioLegend, San Diego, CA). Cell purity (> 90%) was checked by flow cytometry. For CD4 T‐cell depletion, MiniMacs columns and anti‐CD4‐coupled paramagnetic particles (Miltenyi Biotec, Auburn, CA) were used, following the manufacturer's instructions. Depletion was confirmed by flow cytometry.

Co‐stimulation assays

For co‐stimulatory assays, splenocytes (3 × 10⁵ cells) from DO11.10 mice were cultured for 48 hr at 37° in 5% CO₂ in flat‐bottom, 96‐well plates in 0·2 ml RPMI‐1640 medium (Thermo Fisher Scientific, Waltham, MA), in the presence of 10% fetal calf serum (Gibco, Thermo Fisher Scientific), 2 mm glutamine, and 5 mg/ml gentamicin (complete medium), in the presence of the cognate ovalbumin_323–339 (OVA_323–339) peptide (Genscript, Piscataway, NJ) at the indicated concentrations and 0·2 μm of recombinant Gal‐8. For IL‐6 neutralization, 2 μg/ml of IL‐6 neutralizing monoclonal antibody (Clone: MP5‐20F3) or matching isotype control (BioLegend) was used. Two different approaches were carried out to obtain IL‐6KO and C57BL/6J mice‐derived APC. For mitomycin APC, splenocytes were pretreated with 200 μg/ml mitomycin C (Sigma) in RPMI‐1640 medium for 1 hr on ice, and cells were washed three times with ice‐cold phosphate‐buffered saline. For CD4 T‐cell‐depleted APC, CD4 T cells were depleted from splenocytes as described above. Complete blocking of cell division was checked by inhibition of Concanavalin A‐activated proliferation. To assess proliferation, 1 μCi [³H]methylthymidine (New England Nuclear, Newton, MA) was added to each well 18 hr before harvesting. Thiodigalactoside (Sigma) was added 30 min before the stimuli. Recombinant mouse IL‐6 (BioLegend) was added together with the stimuli. Unstimulated cells' basal response ranged from 200 to 1000 counts/min and was subtracted in all experiments. Assays were performed in quadruplicate.

Cytokines quantification

Levels of IL‐6 and IFN‐γ were quantified in co‐stimulation‐assay supernatants by commercial enzyme‐linked immunosorbent assay (BioLegend) following the manufacturer's instructions, using recombinant cytokines–standard curves.

Intracellular IL‐6 expression

To analyse IL‐6 intracellular expression, co‐stimulation assays were performed as described before; and for the last 6 hr, Monensin and Brefeldin A (BioLegend) were added to the cultures together with 1 μg/ml of lipopolysaccharide (Sigma). To label surface antigens, 4 × 10⁶ cells were incubated in 100 μl of cold phosphate‐buffered saline‐azide plus anti‐Fcγ receptor monoclonal antibody (CD16/32, Clone: 93) for 20 min at 4°. Then anti‐CD11c (Clone: N418), anti‐CD11b (Clone: M1/70), anti‐B220 (Clone: RA3‐6B2), anti‐F4/80 (Clone: BM8), anti‐CD4 (Clone: GK1.5), or anti‐MHCII (Clone: M5/114.15.2) monoclonal antibodies conjugated to different fluorochromes, were added at the recommended concentrations. After 30 min, cells were washed and fixed with Cytofix/Cytoperm Fixation/Permeabilization Solution (BD, Franklin Lakes, NJ) for 20 min at 4°. Then, IL‐6 intracellular labelling was performed following the manufacturer's instructions, using anti‐IL‐6 (Clone: MP5‐20F3). All monoclonal antibodies and their isotype controls were from BioLegend. The IL‐6‐positive cells were determined by isotype and ‘fluorescence minus one’, negative controls.

Signal transducer and activator of transcription 3 phosphorylation

For Western blot analysis, co‐stimulation assays were performed for different time periods (2, 4 and 7 hr). Then, cells were washed with ice‐cold phosphate‐buffered saline containing 2 mm sodium orthovanadate and lysed in cracking buffer added with protease and phosphatase inhibitor cocktails from Sigma, followed by ultrasonication. Cell extracts were run in 10% SDS‐PAGE and then transferred to nitrocellulose membranes (GE Healthcare Limited, Little Chalfont, UK). Blots were probed with anti‐phosphorylated signal transducer and activator of transcription 3 (pSTAT3) antibody (Cell Signaling Technology, Beverly, MA), followed by goat anti‐rabbit IRDye800 CW secondary antibody (LiCor Biosciences, Lincoln, NE). Fluorescence emission was detected with an Odyssey clx infrared imaging system, and signal intensity was analysed using image studio lite software (LiCor Biosciences). Stripped blots were reprobed with anti‐STAT3 antibody (Cell Signaling Technology) and revealed as for pSTAT3.

Flow cytometry analysis

FlowMax cytometer PASIII (Partec, Münster, Germany) and flowjo software (FlowJo, Ashland, OR, USA) were used throughout this work.

Statistical analysis

Analysis of variance test was used, except for the analysis shown in Fig. 2, where Student's t‐test was used. P‐values < 0·05 were considered significant.

Figure 2.

Antigen‐presenting cells (APC) produce interleukin‐6 (IL‐6) during galectin‐8 (Gal‐8) ‐induced co‐stimulation. IL‐6 intracellular expression was determined by flow cytometry during Gal‐8‐induced co‐stimulation. Splenocytes (4 × 10⁶) from DO11.10 mice were cultured for 24 hr in the presence of 0·1 μg/ml of ovalbumin_323–339 (OVA _323–339) peptide and 0·2 μm of Gal‐8 (OVA + Gal‐8), or left untreated (Control). Cells were labelled with specific monoclonal antibodies for surface antigens (MHCII, CD11b, CD11c, F4/80, CD4 and B220) and for intracellular IL‐6. IL‐6⁺ population is box enfolded in the counter plots. Bars indicate percentage of IL‐6⁺ cells and the mean fluorescence intensity (MFI). FSC, forward side scatter. Depicted assays are representative of two independent experiments and were carried out, each time, with different recombinant protein preparations. *P < 0·05; **P < 0·01; ***P < 0·001.

Results and Discussion

When naive splenocytes from DO11.10 mice are incubated with low doses of OVA_323–339 cognate peptide in the presence of Gal‐8, a synergistic CD4 T‐cell response defined as Gal‐8‐induced co‐stimulation is observed.3, 4, 21, 22 In the present work, we used this established and well‐characterized in vitro model to assess whether IL‐6 was involved in the elicitation of antigen‐specific responses induced by Gal‐8. First, we analysed IL‐6 secretion during Gal‐8‐induced co‐stimulation. Hence, splenocytes from DO11.10 mice were cultured for 48 hr in the presence of OVA alone, or in combination with Gal‐8, and IL‐6 levels were determined in the supernatants by enzyme‐linked immunosorbent assay. As observed in Fig. 1(a), IL‐6 levels were significantly increased in those cultures stimulated with Gal‐8 alone or in combination with the antigen. However, no differences were recorded between antigen‐stimulated and control cells, indicating that in this limited‐antigen condition, IL‐6 secretion was triggered specifically in the presence of Gal‐8. Remarkably, the amount of IL‐6 was significantly higher in the supernatants from Gal‐8 plus OVA compared with cells stimulated with Gal‐8‐alone, indicating that cognate TCR activation potentiated IL‐6 secretion induced by Gal‐8. Furthermore, IL‐6 levels mirror the synergistic effect of Gal‐8 on TCR activation, evidenced by an augmented cell proliferation and IFN‐γ secretion (Fig. 1b), suggesting its involvement in the Gal‐8 co‐stimulatory effect. It should be stressed that IL‐6 level is not a mere reflection of cell proliferation rate, as it was also increased in Gal‐8‐alone‐treated cells, where proliferation was similar to basal level. Conversely, in proliferating OVA‐treated cells, the amount of secreted IL‐6 remained scarce (Fig. 1a,b). Pre‐incubation with thiodigalactoside, a Gal inhibitor, prevented the IL‐6 increment induced by Gal‐8 alone or in combination with the antigen, so highlighting both Gal‐8 specificity and the dependence of lectin–glycan interaction at the cell surface. It is well known that pathogen‐recognition or damage‐associated molecular patterns activate specific danger receptors like TLRs, to stimulate a range of signalling pathways including nuclear factor‐κB, which enhances IL‐6 production.23, 24, 25 In this regard, we have previously demonstrated that Gal‐8 induces nuclear factor‐κB activation and IL‐6 secretion in the human endothelial cell line HMEC‐1.26 Hence, it is plausible to consider that Gal‐8 could directly interact with TLRs to positively regulate IL‐6 transcription. In line with this, Gal‐3 was recently described as an endogenous ligand for TLR4 that promotes neuroinflammation.27 Considering that TLR4 activation leads to Gal‐8 secretion in endothelial cells, B cells and BMDC,5, 26, 28 the presence of a positive feedback among TLR‐signalling, Gal‐8 and IL‐6 that amplifies the inflammatory response, can be postulated.

Figure 1.

Galectin‐8 (Gal‐8) induces interleukin‐6 (IL‐6) secretion during co‐stimulation of antigen‐specific CD4 T‐cell response. (a) Quantification of IL‐6 by ELISA in supernatants from Gal‐8‐induced co‐stimulation cultures. (b) Quantification of interferon‐γ (IFN‐γ) by ELISA in supernatants (left) and cell proliferation (right) of Gal‐8‐induced co‐stimulation cultures. For all assays, splenocytes (3 × 10⁵ cells) from DO11.10 mice were cultured for 48 hr in the presence of 0·1 μg/ml of ovalbumin_323–339 peptide (OVA), and/or 0·2 μm of Gal‐8. Thiodigalactoside (TDG, 30 mm) was added 30 min before the stimulus. TDG had no effect on OVA response (not shown). ND, not detected. Depicted assays are representative of at least three independent experiments and were carried out, each time, with different recombinant protein preparations. **P < 0·01; ***P < 0·001; ****P < 0·0001.

Next, we asked which cells were actually responsible for IL‐6 production in response to Gal‐8 co‐stimulatory activity. For this purpose, splenocytes from DO11.10 mice were stimulated with Gal‐8 plus OVA for 24 hr and IL‐6 intracellular levels were determined in different cell subpopulations by flow cytometry. As shown in Fig. 2, CD11b⁻ and CD11b⁺ conventional DC (cDC) and plasmacytoid DC (pDC) as well as macrophages, B cells and CD4 T cells significantly increased IL‐6 production in response to Gal‐8‐induced co‐stimulation. As expected, a large proportion of cDC were induced to express IL‐6 upon Gal‐8 co‐stimulation (about 23–44%), being CD11b⁺ cDC the subpopulation that expressed higher levels of this cytokine (MFI > 10). A lower frequency (between 9 and 18%) of pDC, macrophages and B cells produced IL‐6 in response to Gal‐8 plus OVA, and among these subpopulations, pDC produced the highest level of this cytokine (MFI = 10). Cytokine production was almost marginal in both stimulated and unstimulated CD4 T cells. These findings are in agreement with our previous observations where both MHCII^int CD11b^high and MHC^high CD11b^int BMDC subpopulations, which resemble cDC and monocyte‐derived macrophages respectively, produce IL‐6 in response to Gal‐8 stimulation. Moreover, IL‐6 secretion was impaired in BMDC differentiated from Gal‐8 knockout (Gal‐8KO) mice, suggesting a physiological role for endogenous Gal‐8 in the regulation of IL‐6 production.5 Regarding B cells, and also in line with our findings, Tsai et al.28 have reported that despite both Gal‐8 and Gal‐1 playing important roles in the generation of plasma cells, only Gal‐8 induces IL‐6 expression on B cells. Taken together, our results indicate that Gal‐8 stimulates different splenic subpopulations of APC to produce and secrete IL‐6 during antigen‐specific CD4 T‐cell co‐stimulation.

To test our hypothesis by which IL‐6 signalling is involved in the Gal‐8‐dependent co‐stimulatory effect during antigen recognition by specific CD4 T cells, we repeated the assay depicted in Fig. 1(b), but in the presence of an IL‐6‐neutralizing monoclonal antibody (anti‐IL‐6). Notably, IL‐6 neutralization inhibited Gal‐8‐induced co‐stimulation, this reduction being specifically dependent on IL‐6 blockade as no significant differences were observed in the presence of the isotype control antibody (Fig. 3a). It should be noted that lymphocyte proliferation in response to OVA was independent of IL‐6, as no changes in T‐cell activation were recorded when anti‐IL‐6 was added to cells stimulated with antigen alone. This observation correlated with those results described in Fig. 1(a,b) where OVA‐stimulated cells proliferated and produced IFN‐γ in the absence of IL‐6. These results indicate that the co‐stimulation induced by Gal‐8, but not the antigen‐response itself, depends on IL‐6.

Figure 3.

Galectin‐8 (Gal‐8) ‐induced CD4 T‐cell co‐stimulation is dependent on interleukin‐6 (IL‐6). (a) IL‐6 neutralization. Splenocytes (3 × 10⁵ cells) from DO11.10 mice were cultured in the presence of 0·1 μg/ml of ovalbumin_323–339 peptide (OVA), and/or 0·2 μm of Gal‐8, in the presence or absence of IL‐6 neutralizing monoclonal antibody (anti‐IL‐6) or isotype control (Iso). After 48 hr, proliferation was assessed. (b) Signal transducer and activator of transcription 3 (STAT3) activation. Splenocytes (9 × 10⁶ cells) from DO11.10 mice were stimulated for 2 or 7 hr with 0·1 μg/ml of OVA and/or 0·2 μm of Gal‐8, in the presence or absence of IL‐6 neutralizing monoclonal antibody (anti‐IL‐6) or isotype control (Iso). For positive control, cells were incubated for 15 min with 100 ng/ml of recombinant IL‐6 (rIL‐6). Protein expression levels of pSTAT3 and total STAT3 were detected by Western blot analysis. Bars depict pSTAT/STAT3 fluorescence signal ratio. (c) Gal‐8‐induced co‐stimulation in the presence of IL‐6‐deficient antigen‐presenting cells (APC). IL‐6KO and C57BL/6J mice‐derived APC (2 × 10⁵): Mitomycin APC (upper panels) or CD4‐depleted APC (lower panels) were co‐cultured with CD4 T cells (1 × 10⁵) purified from OTII mice, in the presence of 0·05–0·1 μg/ml of OVA and/or 0·2 μm of Gal‐8. After 48 hr, cell proliferation (right) and quantification of interferon‐γ (IFN‐γ) by ELISA in supernatants (left) were assessed. (d) Same as (c) with Mitomycin APC, and with addition of 5 ng/ml of rIL‐6 to the co‐cultures containing IL‐6KO APC. Fold increase was calculated as the counts/min (OVA+Gal‐8)/counts/min (OVA) ratio. Depicted assays are representative of three (a) and two (b–d) independent experiments and were carried out, each time, with different recombinant protein preparations. *P < 0·05; ***P < 0·001.

The multiple functions of IL‐6 are initiated upon its binding to the IL‐6 receptor (IL‐6R) system, which comprises the IL‐6Rα and the gp130 signal‐transducing chain. Cytokine binding induces the homodimerization of IL‐6R, which in turn, triggers the activation of the downstream Janus‐activated kinase–STAT3 pathway.29 Various sets of IL‐6‐responsive genes are induced by the activation of STAT3.25 Having demonstrated that IL‐6 is increased during Gal‐8‐induced co‐stimulation of the CD4 T‐cell response (Fig. 1a), we next asked whether IL‐6 signalling was actually triggered in the presence of Gal‐8. For this purpose, DO11.10 mouse splenocytes were stimulated with OVA, Gal‐8 or OVA plus Gal‐8 for different time‐points, and then, the phosphorylation of STAT3 (pSTAT3) was determined by Western blot. As shown in Fig. 3(b), Gal‐8 stimulation triggered STAT3 phosphorylation independently of the presence of the antigen after 2 hr of treatment. Notably, a synergistic effect on the transcription factor activation was observed in the presence of both Gal‐8 and OVA. Similar results were obtained after 4 hr (data not shown) and 7 hr of treatment (Fig. 3b, right). Gal‐8‐induced pSTAT3 was inhibited in the presence of the anti‐IL‐6, indicating a tight dependence on interleukin signalling. Conversely, antigen‐induced TCR activation was independent of IL‐6 signalling, as no differences in transcription factor phosphorylation were observed in the presence of anti‐IL‐6 or the isotype control. These observations are in agreement with the absence of IL‐6 secretion and signalling in cells treated with OVA alone (results depicted in Figs 1a and 3a, respectively). Our results strongly suggest that Gal‐8 induces APC to produce and secrete IL‐6, which in turns signals through IL‐6R in antigen‐specific CD4 T cells to promote Gal‐8 co‐stimulatory activity.

To further confirm our hypothesis, in which APC‐derived IL‐6 is responsible for the co‐stimulatory activity, antigen presentation assays were performed using either Mitomycin C‐treated or CD4 T‐cell‐depleted splenocytes from IL‐6KO or wild‐type (WT) mice as APC source, co‐cultured with purified CD4 T cells from OTII mice. As observed in Fig. 3(c) (upper panels), Gal‐8‐induced co‐stimulation was significantly impaired when mitomycin APC were unable to produce IL‐6, evidenced by cell proliferation (right) and IFN‐γ secretion (left). Similar results were obtained when CD4 T‐cell‐depleted splenocytes from IL‐6KO mice were used as APC (Fig. 3c, lower panels). Once again, antigen‐specific CD4 T‐cell stimulation was not affected in the absence of IL‐6, since no differences in T‐cell proliferation or IFN‐γ secretion were observed in cells stimulated with OVA alone in the presence of WT or IL‐6KO APC. Additionally, this experimental design allowed the demonstration that antigen‐specific CD4 T cells are a target for IL‐6‐mediated effects upon Gal‐8 stimulation. Finally, to confirm that the reduction in the Gal‐8 co‐stimulatory effect was dependent on IL‐6 deficiency, the assay depicted in Fig. 3(c) was performed with the exogenous addition of recombinant IL‐6. Proliferation fold increase was calculated as the ratio: counts/min (OVA + Gal‐8)/counts/min (OVA) of co‐cultures using IL‐6KO or WT APC. As depicted in Fig. 3(d), exogenous IL‐6 restored the proliferation rate of IL‐6KO APC : CD4 T‐cell co‐cultures, demonstrating that the presence of IL‐6 was sufficient to rescue Gal‐8‐induced co‐stimulation. Altogether, our findings demonstrate that Gal‐8 stimulates APC to produce IL‐6, which signalling synergizes the TCR activation during antigen recognition on CD4 T cells, resulting in the Gal‐8 co‐stimulatory activity.

In summary, the in vitro model used throughout this work allowed us to show the participation of IL‐6 signalling in Gal‐8‐induced co‐stimulation during the elicitation of the antigen‐specific CD4 T‐cell response. As IL‐6 is a key cytokine involved in the differentiation of Tfh cells and germinal centre formation, our findings support a plausible mechanism by which Gal‐8 stimulates the antigen‐specific protective humoral response previously observed in a viral vaccine model.5 Interestingly, aged CD4 T cells require higher levels of IL‐6 produced during the CD4 T cell : APC cognate interaction compared with young CD4 T cells, to effectively induce IL‐6 signalling, read out as IgG antibody production.10, 18 Gal‐8, by activating IL‐6 production from APC during cognate interaction with naive CD4 T cells, could circumvent the insufficient response of aged cells, so becoming an attractive candidate to add to vaccine formulations for compromised populations.

Disclosures

Authors declare no conflict of interest.

Author contributions

JC, CAP, OC and MVT conceived the project and designed experiments. JC, CAP and LMS performed the experiments. JC, CAP, LMS, MPA, OC and MVT analysed data, discussed results and wrote the manuscript. OC and MVT supervised work. JC and LMS are fellows and MPA, OC and MVT are researchers from Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET; Argentina); CAP is fellow from Agencia Nacional de Promoción Científica y Tecnológica (ANPCyT, Argentina).