MALT1 regulates innate immunity but how this occurs remains ill-defined. We used the selective MALT1 paracaspase inhibitor MLT-827 to unravel the contribution of MALT1 to innate signaling downstream of Toll-like or C-type lectin-like receptors, demonstrating that MALT1 regulates the production of myeloid cytokines, and downstream of C-type lectin-like receptors, selectively.
Besides its function in lymphoid cells, which has been addressed by numerous studies, the paracaspase MALT1 also plays an important role in innate cells downstream of pattern recognition receptors. Best studied are the Dectin-1 and Dectin-2 members of the C-type lectin-like receptor family that induce a SYK- and CARD9-dependent signaling cascade leading to NF-κB activation, in a MALT1-dependent manner. By contrast, Toll-like receptors (TLR), such as TLR-4, propagate NF-κB activation but signal via an MYD88/IRAK-dependent cascade. Nonetheless, whether MALT1 might contribute to TLR-4 signaling has remained unclear. Recent evidence with MLT-827, a potent and selective inhibitor of MALT1 paracaspase activity, indicates that TNF- production downstream of TLR-4 in human myeloid cells is independent of MALT1, as opposed to TNF- production downstream of Dectin-1, which is MALT1 dependent. Here, we addressed the selective involvement of MALT1 in pattern recognition sensing further, using a variety of human and mouse cellular preparations, and stimulation of Dectin-1, MINCLE or TLR-4 pathways. We also provided additional insights by exploring cytokines beyond TNF-, and by comparing MLT-827 to a SYK inhibitor (Cpd11) and to an IKK inhibitor (AFN700). Collectively, the data provided further evidence for the MALT1-dependency of C-type lectin-like receptor —signaling by contrast to TLR-signaling.
The paracaspase activity of MALT1 (Mucosa-associated lymphoid tissue lymphoma translocation protein 1) was revealed in 20081,2. Since then, a number of studies have reported its critical contribution to antigen receptor responses in lymphocytes. Genetic models in the mouse as well as pharmacology data support a key role in T cells, in T-cell dependent autoimmunity and in B-cell lymphoma settings3,4. In lymphocytes, MALT1 paracaspase activation occurs upon assembly of a CARD11-BCL10-MALT1 complex5, which is triggered by antigen-receptor proximal signaling downstream of the T- or B- cell receptor. There is also ample evidence that a similar CARD9-BCL10-MALT1 complex is important for propagating signals downstream of C-type lectin-like receptors (CLLR), e.g., Dectin-1, Dectin-2 and MINCLE in myeloid cells6,7. Dectin-1 has been particularly well studied because this pathway is critical for host defense against fungal infections8,9. Implication of MALT1 in Toll-like receptor (TLR) pathways, however, has remained controversial10. Recent evidence in human myeloid cells ruled out a direct role for MALT1 paracaspase activity in the regulation of TNF- production downstream of TLR-411.
In the present work, we used various experimental settings and stimulatory conditions in human and mouse myeloid cells to probe innate signaling pathways, relying on specific pharmacological tool inhibitors and measurement of cytokine production.
Experiments were conducted according to the guidelines and standards of the Novartis Human Research Ethics Committee.
1. Preparation of Peripheral Blood Mononuclear Cells (PBMCs) from Human Buffy Coats
NOTE: We received buffy coats from healthy volunteers one day after collection, in 50 mL bags. They were provided under informed consent and collected through the Interregionale Blutspende Schweizeriches Rotes Kreuz. We handled them using the procedure below, at room temperature unless specified otherwise.
2. Preparation of Monocytes from PBMCs
3. PBMCs and Monocytes Treatments and Stimulatory Conditions
4. Monocytes Differentiation into iMoDCs and Stimulatory Conditions
5. Monocytes and PBMCs Freezing/Thawing Procedures
6. Mouse Spleen Cells Preparation and Treatment
NOTE: We conducted animal sacrifices according to the guidelines and standards of the Novartis Animal Welfare Organization. Studies were approved by the Ethics Committee of the regional governmental authority (Kantonales Veterinäramt der Stadt Basel). We sacrificed animals by isoflurane over-exposure, with all efforts made to minimize suffering.
7. Cytokines and Viability Measurements
8. Compound Preparation
9. Stimuli Preparation
In myeloid cells, MALT1 relays activation signals downstream of several C-type lectin-like receptors, such as Dectin-1, Dectin-2 and MINCLE6. These pathways rely on (hem)ITAM motif-containing receptors (e.g., Dectin-1) or ITAM motif-containing co-receptors (e.g., FcRγ, for Dectin-2 and MINCLE) that recruit and activate the SYK kinase (Figure 1). This leads to activation of a protein kinase C isoform, namely PKCδ, which phosphorylates CARD9, thereby triggering CARD9/BCL10/MALT1 complex formation and recruitment of TRAF6 for downstream NF-κB activation12. By contrast, the TLR-4 pathway recruits TRAF6 in a MALT1-independent but MYD88/IRAK-dependent manner for NF-κB activation (Figure 1). Evidence for this differential involvement of MALT1 was obtained using genetic models of MALT1 deficiency as well as pharmacological treatment with the peptidic active-site inhibitor z-VRPR-fmk11,13,14.
We used the recently reported potent and selective MALT1 inhibitor MLT-82715 and asked if this compound would regulate TNF- production downstream of C type lectin-like and Toll-like receptors, respectively. Human PBMCs and mouse spleen cells were stimulated with depleted zymosan (DZ, a known agonist of Dectin-1) or lipopolysaccharide (LPS, a known agonist of TLR-4) and we measured TNF- release in the culture supernatant after 20 h. In both the human and the mouse assays, MLT-827 selectively blocked TNF- production driven by the Dectin-1 pathway, but not by the TLR-4 pathway (Figure 2). We obtained similar data upon incubation with the z-VRPR-fmk compound (Supplementary Figure 1).
To gain pathway insights, we conducted further experiments in human monocytes and in immature monocytes-derived dendritic cells (iMoDCs), comparing the effect of MLT-827 to that of the SYK inhibitor Cpd1116 and to that of the IKK inhibitor AFN70015. In monocytes stimulated with LPS, production of TNF-α was almost completely abrogated by AFN700 but was not sensitive to Cpd11 (Figure 3A), which is consistent with the dependency/independency of the TLR-4 pathway on NF-κB/SYK activity, respectively (see Figure 1). By contrast, TNF-α production driven by Dectin-1 in iMoDCs displayed sensitivity to Cpd11 in addition to sensitivity to MLT-827 and AFN700 (Figure 3B, Supplementary Figure 2), providing further evidence for involvement of a SYK/CBM signaling cascade in the Dectin-1 pathway (Figure 1). Noteworthy, production of IL-1, IL-6 and IL-23 upon Dectin-1 stimulation was also sensitive to the three inhibitors, thereby indicating regulatory mechanisms similar to TNF-. However, a limited effect of the three compounds on IL-8 production suggested a distinct regulatory mechanism for this cytokine (Figure 3B, Supplementary Figure 2).
In addition to Dectin-1, other CLLRs, such as Dectin-2 and MINCLE, function via stimulation of a CARD9 signalosome7. We therefore tested MLT-827 in iMoDCs challenged with the MINCLE agonist Trehalose-6,6-dibehenate (TBD). Raising TBD concentrations above 50 µg/mL led to production of TNF-, IL-6 and IL-1, which relied on MALT1 paracaspase activity as seen from the blocking effect of MLT-827 (Figure 4A). Consistent results were obtained when challenging iMoDCs with increasing concentrations of DZ to stimulate Dectin-1 (Figure 4B).
Figure 1: NF-κB signaling downstream of Dectin-1, MINCLE and TLR-4. The cartoon depicts the key features of canonical NF-κB activation pathways downstream of Dectin-1, Dectin-2, MINCLE or TLR-4 in myeloid cells. The hemITAM-containing Dectin-1 receptor17 can directly engage SYK to stimulate CBM (CARD9/BCL10/MALT1) complex formation, leading to TRAF6 dependent NF-κB activation. Other C-type Lectin-like receptors such as Dectin-2 or MINCLE need to recruit an ITAM-containing FcRγ chain to engage a CBM and activate NF-κB. TLR-4 receptors use another mechanism for NF-κB activation, relying on MYD88 and IRAK1/IRAK4 kinases upstream of TRAF6.
Figure 2: Dectin-1 signals via MALT1 for production of TNF-α in human and mouse cells. (A) Human PBMCs data as in Unterreiner et al., 2017 (Figure 2A)11. Human PBMCs were stimulated with 1 ng/mL of LPS (TLR-4 agonist) or 100 µg/mL DZ (Dectin-1 agonist) for 20 h in the presence of graded concentrations of MLT-827. TNF- released in the supernatant was quantified by HTRF. (B) Mouse spleen cells were treated with a concentration range of MLT-827 for 30 min and subsequently stimulated with 30 µg/mL DZ or 1 µg/mL LPS + 10 ng/mL IFN- for 18 h. TNF-α in the cell culture supernatant was measured by ELISA. One of two experiments with similar results is shown, as means ± SEM of three measurements.
Figure 3: IKK- and/or SYK-dependency of cytokine production downstream of TLR-4 and Dectin-1. (A) Human monocytes were pre-treated for 1 h with MLT-827 (1 µM), Cpd11 (1 µM), or AFN700 (3 µM) or vehicle (DMSO). Cells were stimulated with 10 ng/mL LPS for 20 h and TNF- in the supernatant was quantified by HTRF. (B) TNF-α, IL-1β, IL-6, IL-23 and IL-8 production by human monocytes-derived dendritic cells (iMoDCs) stimulated for 24 h with DZ (100 µg/mL) after 1 h pre-incubation with MLT-827, Cpd11, AFN700 (all at 1 µM) or DMSO. Cytokine levels in DMSO-treated samples were set at 100%. Data are means ± SD of three measurements, and are representative of three independent experiments. *P < 0.05; **P < 0.01; ***P < 0.001, unpaired two-tailed Student's t test.
Figure 4: C type lectin-like-dependent cytokine production by iMoDCs. TNF-α, IL-1β, and IL-6 production by iMoDCs stimulated for 24 h with the MINCLE agonist Trehalose-6,6-dibehenate (TDB, 100 µg/mL) (A) or with the Dectin-1 agonist DZ (100 µg/mL) (B) after 1 h pre-incubation with MLT-827 (1 µM) or DMSO. Data are means ± SD of three measurements and are representative of three independent experiments.
Supplemental File 1. Please click here to download this file.
Supplemental File 2. Please click here to download this file.
Supplemental File 3. Please click here to download this file.
In this work, we used simple experimental settings to study signaling pathways in human and mouse innate cells, and interrogate their dependency on MALT1 proteolytic function. Expanding on previous work11, our study showed that MALT1 paracaspase activity controls C-type lectin-like receptor induced cytokine production, including TNF-α. In contrast, TLR-4-induced TNF-α was independent of MALT1 in both species. Collectively, these data corroborated the key and selective contribution of the MALT1/CBM signalosome downstream of C-type lectin-like receptors, which was unveiled by earlier studies6,12,18.
Whether the clear independency of TLR-4 signaling on MALT1 in myeloid cells applies to other cell types remains to be explored. For instance, in B-lymphocytes, TLR signaling was previously shown to contribute to B-cell activation downstream of the B-cell antigen receptor19. In fact, we have unpublished evidence that TLR-4 stimulated human and mouse B cells display sensitivity to MLT-827. Therefore, further mechanistic insights downstream of the B-cell receptor will be valuable. In this context, a recent study in B cell lymphoma provided evidence for clustering of the signaling pathways downstream of the B-cell receptor and the TLR9 receptor20. TRAF6, which acts as a mediator for NF-κB activation in both the B-cell receptor and the TLR pathways, might be a point of crosstalk, which could explain the sensitivity of both pathways to MALT1 protease inhibition. Conversely, TRAF6 is also a common downstream player of CLLRs and TLRs for induction of NF-κB but these two pathways do not appear to crosstalk in a MALT1 paracaspase-dependent manner in myeloid cells.
This work focused on cytokine production, which provides an easy readout for signaling pathways and can be implemented readily for compound profiling. It highlighted the value of selective and potent inhibitors of MALT1 for unravelling MALT1 biology. Obtaining further mechanistic insights will require additional work and development of more proximal assays, e.g., to characterize the substrates of MALT1 involved in innate signaling regulation.
The authors have nothing to disclose.
We thank Elsevier for their authorization (license number 4334770630127) to reproduce here Figure 2A from Unterreiner et al. (2017).
100 µm nylon cell strainer | Sigma | CLS431752 | |
14 ml Falcon tube | BD Falcon | 352057 | |
15 mL Falcon tube | Falcon | 352090 | |
50 mL Falcon tube | Falcon | 352070 | |
6 well plates | Costar | 3516 | |
96 well flat-bottom plate, with low evaporation lid | Costar | 3595 | |
96 well V-bottom plate | Costar | 734-1798 | |
Ammonium Chloride – NH4Cl | Sigma | A9434 | |
Assay diluent RD1-W ELISA | R&D | 895038 | Assay diluent |
Cell culture microplate, 384 well, black | Greiner | 781986 | |
Depleted Zymosan | Invivogen | tlrl-dzn | now: tlrl-zyd |
Dimethyl sulfoxide | Sigma | D2650 | DMSO |
EDTA-Na2 | Sigma | E5134 | Ethylenediaminetetraacetic acid disodium salt dihydrate |
ELISA muTNF-α | R&D | SMTA00 | |
Ficoll-Paque Plus | GE Healthcare | 17-1440-03 | |
gentleMACS C tubes | MACS Miltenyi Biotec | 130-096-334 | |
gentleMACS dissociator | MACS Miltenyi Biotec | 130-093-235 | |
GM-CSF | Novartis | – | |
Heat-inactivated Fetal bovine serum | Gibco | 10082 | FBS |
HTRF hu IL-23 | CisBio | 62HIL23PEG | |
HTRF hu TNF-α | CisBio | 62TNFPEC | |
HTRF reconstitution buffer | CisBio | 62RB3RDE | 50mM Phosphate buffer, pH 7.0, 0.8M KF, 0.2% BSA |
IFN-γ | R&D | L4516 | |
IL-4 | Novartis | – | |
Isoflurane | Abbott | Forene | |
Lipopolysaccharides (LPS) | Sigma | L4391 | LPS used in human samples |
Lipopolysaccharides | Sigma | L4516 | LPS used in murine samples |
Lysis buffer | Self-made | – | 155 mM NH4Cl, 10 mM KHCO3, 1 mM EDTA, pH 7.4 |
Magnet | Stemcell | 18001 | |
Microplate, 384 well white | Greiner | 784075 | |
Monocytes enrichment kit | Stemcell | 19059 | |
Nalgene Mr. Frosty Cryo 1 °C Freezing Container | Nalgene | 5100-0001 | cooling device (containing Propanol-2) |
PBS 1x pH 7.4 [-] CaCl2 [-] MgCl2 | Gibco | 10010 | Phosphate-buffered saline |
Penicillin/Streptomycin | Gibco | 15140 | Pen/Strep |
Potassium bicarbonate – KHCO3 | Sigma | P9144 | |
PrestoBlue | Invitrogen | A13262 | Resazurin solution for viability assessment |
Propanol-2 | Merck | 1.09634 | |
Read buffer | MesoScale Discovery | R92TC-3 | Tris-based buffer containing tripropylamine |
Recovery cell culture freezing medium | Gibco | 12648-010 | freezing medium |
Roswell Park Memorial Institute Medium (RPMI) with Glutamax | Gibco | 61870 | + 10% FBS for iMoDCs + 10% FBS + 1 mM Sodium Pyruvate + 100 U/mL Pen/Strep + 5 µM β-mercaptoethanol for human PBMCs and monocytes + 10% FBS + Pen/Strep + 5 µM β-mercaptoethanol for murine splenocytes |
Separation buffer | Self-made | – | PBS pH 7.4 + 2% FBS + 1 mM EDTA pH 8.0 |
Sodium Pyruvate | Gibco | 11360 | |
Trehalose-6,6-dibehenate | Invivogen | tlrl-tdb | TDB |
Tween 20 | Sigma | P7949 | Polysorbate 20 |
UltraPure 0.5 M EDTA pH 8.0 | Invitrogen | 15675 | Ethylenediaminetetraacetic acid |
Viewseal sealer | Greiner BioOne | 676070 | |
V-PLEX Proinflammatory Panel 1 Human Kit | MesoScale Discovery | K15049D | electrochemiluminescent multiplex assay (IL-1β, TNF-α, IL-6, IL-8) |
β-Mercaptoethanol | Gibco | 31350 |