Requirement of Mucosa-Associated Lymphoid Tissue Lymphoma Translocation Protein 1 Protease Activity for Fcγ Receptor-Induced Arthritis, but Not Fcγ Receptor-Mediated Platelet Elimination, in Mice

Kea Martin PhD,1, Ratiba Touil,1, Grozdan Cvijetic, Laura Israel PhD, Yeter Kolb, Sophie Sarret, Stéphanie Valeaux, Degl’Innocenti Elena, Le Meur Thomas, Nadja Caesar, Maureen Bardet, Christian Beerli PhD, Hans-Guenter Zerwes PhD, Jiri Kovarik PhD, Karen Beltz PhD†, Achim Schlapbach PhD, Jean Quancard PhD, Catherine H Régnier PhD, Marc Bigaud PhD, Tobias Junt PhD , Grazyna Wieczorek PhD, Isabelle Isnardi PhD, Amanda Littlewood-Evans PhD, Frédéric

Abstract

Objective. Fc receptors for IgGs (FcRs) play important roles both in protective and pathogenic immune responses. The assembly of the CARD9/BCL10/MALT1 signalosome is required for optimal FcR-induced canonical NF-B activation and pro-inflammatory cytokine release. Our goal was to clarify the relevance of the MALT1 protease activity in FcR-driven events and the therapeutic potential of selective MALT1 protease inhibitors in FcR-mediated diseases.
Methods. Using genetic and pharmacological disruption of MALT1 scaffolding and enzymatic activity, we assessed the relevance of MALT1 function in murine and human primary myeloid cells upon stimulation with immune complexes (IC) and using murine models of autoantibody-driven arthritis and immune thrombocytopenic purpura (ITP).
Results. MALT1 protease function is essential for optimal FcR-induced production of proinflammatory cytokines by various murine and human myeloid cells stimulated with ICs. In contrast, MALT1 protease inhibition did not affect the Syk-dependent, FcR-mediated reactive oxygen species (ROS) or leukotriene B4 (LTB4) production. Notably, pharmacological MALT1 protease inhibition in vivo reduced joint inflammation in the murine K/BxN serum-induced arthritis model (mean area under the curve [AUC] of paw swelling over time, 45.42 % in mice treated with a MALT1 protease inhibitor relative to control mice; p=0.0007) but did not affect platelet depletion in a passive model of immune thrombocytopenic purpura.
Conclusion. This work reveals the specific contribution of the protease activity of MALT1 to FcRmediated events and suggests therapeutic potential of MALT1 protease inhibitors for a subset of FcR-driven inflammatory disorders.

Introduction

Autoantibodies are important mediators of autoimmune and inflammatory diseases, such as rheumatoid arthritis (RA), lupus, and immune thrombocytopenic purpura (ITP) (1). Circulating immune complexes (IC) or antibody-coated cells activate immune cells via Fc receptors for IgGs (FcRs) leading to the release of pro-inflammatory mediators or the elimination of target cells.
Therefore, modulating FcR-mediated signaling events represents an attractive therapeutic strategy for multiple diseases (2). Signaling downstream of activating FcRs is mediated by ITAM domains present in FcRIIA or in the common FcR chain. Phosphorylation of ITAM domains by Src family kinases generates docking sites for the tyrosine kinase Syk, which in turn activates a signaling cascade leading to cellular activation and production of reactive oxygen species (ROS), leukotrienes, and proinflammatory cytokines (3). The first Syk inhibitor, fostamatinib, was recently approved for the treatment of ITP, highlighting the therapeutic relevance of this pathway (4).
Despite broad evidence for the role of MALT1 in signaling downstream of multiple ITAMcontaining receptors (5), its role in FcR signaling remains ill-defined (6-9). Triggering of ITAMcontaining receptors leads to protein kinase C (PKC)-dependent assembly of CBM signalosomes, comprised of a caspase recruitment domain (CARD) protein, BCL10, and MALT1, which
subsequently mediate canonical NF-B activation via activation of the inhibitor of NF-B kinase (IKK) complex (5). In addition to a scaffolding role, MALT1 has a proteolytic enzymatic function that modulates the intensity and persistence of CBM-driven signaling events via the cleavage of numerous substrates (5, 10). As a result, MALT1 protease inhibitors are being actively evaluated aiming at potential new therapies (5, 10).
Using myeloid lineage cells from MALT1 KO and MALT1 protease deficient (PD) mice, it was proposed that protease activity is required for FcR-induced cytokine release (8, 9). Moreover, platelet elimination in vivo was shown to be impaired in MALT1 PD animals in a passive mouse model of ITP (9). However, given that MALT1 PD animals develop a progressive T and B cellrelated multi-organ inflammatory disease driven by reduced regulatory T cells and characterized by drastically elevated serum IgG1 and IgE levels (11-14), the outcome of in vivo experiments using FcR-dependent models in these mice should be interpreted with caution.
To clarify the relevance of MALT1 protease in FcR-mediated events and in IC-mediated diseases such as RA and ITP, we used novel MALT1 protease inhibitors in addition to MALT1 KO and MALT1 PD mice. Using human and murine myeloid cells, we demonstrate that MALT1 protease function is critical for the optimal production of FcR-driven pro-inflammatory cytokines and chemokines, but not for LTB4 and ROS production. Pharmacological inhibition of the MALT1 protease in vivo was partially protective in a mouse model of autoantibody-driven arthritis. However, in contrast to previous studies (9), genetic or pharmacological abrogation of the MALT1 protease did not affect the elimination of IgG-coated platelets in an ITP model. We, therefore, propose that MALT1 protease inhibition may have therapeutic potential in diseases associated with FcR-driven pro-inflammatory cytokines such as RA but not in diseases driven by other FcR-mediated biological events such as platelet elimination in ITP.  

Material and methods Human samples

Buffy coats from healthy volunteers were provided under informed consent and collected through the Interregionale Blutspende SRK. Fresh human blood from healthy volunteers was provided under informed consent and collected through Santemed Switzerland (Novartis). Synthesis and characterization of MLT-748 and MLT-695 and Cpd11Cpd11 and MLT-748 synthesis was performed as previously described (15, 16) (WO2015181747). The synthesis of MLT-695 was performed as described in the supplementary information. MLT-695 compound characterization: 1H NMR (400 MHz, DMSO-d6) (major rotamer), δ: 8.54 (s, 1H), 7.80 (s, 1H), 7.19 (d, J=4.3 Hz, 2H), 6.94 (s, 1H), 6.92 (d, J=4.3 Hz, 2H), 6.43 (q, J=9.1 Hz, 1H), 5.29 (q, J=6.7 Hz, 1H), 3.06-3.29 (m, 5H), 3.17 (s, 3H), 2.91 (s, 3H), 1.93-2.16 (m, 4H), 1.60 (d, J=6.7 Hz, 3H). 13C NMR (101 MHz, DMSO) δ: 174.58, 151.36, 146.85, 146.32, 144.11, 140.95, 129.52, 125.20 (q), 123.65, 120.65, 114.80, 95.47, 72.06, 57.05, 55.14 (q), 49.54, 35.58, 30.83, 26.98, 17.28. HR-MS: [M+H]+ C24H28ClF3N5O4S calc: 574.1510, found: 574.1498.

Generation of human granulocytes and immature monocyte-derived dendritic cells (iMoDCs)

Granulocytes were isolated from heparinized peripheral blood as described earlier (17). Briefly, red blood cells were excluded by incubation with 1% dextran (Sigma) and granulocytes were isolated from the pellet after density gradient centrifugation using Histopaque 1077 (Sigma). iMoDCs were generated from PBMCs isolated from buffy coats as previously described (18). Enriched monocytes (Easy Sep Human Monocyte Enrichment kit, StemCell Technologies) were cultured in complete RPMI-1640 (Gibco) supplemented with 100 ng/ml GM-CSF (Novartis) and 80 ng/ml IL-4 (Novartis) for 7 days.
MALT1 KO and MALT1 PD mice on a C57BL/6 genetic background have been described previously (11). For experiments comparing MALT1 PD or KO animals to WT mice, WT littermate mice were used as a control. For experiments involving the use of MLT-695, C57BL/6J animals have been purchased from Charles River (France) and BALB/cByJrj from Janvier Labs (France). All animal studies were performed in accordance with the Swiss Federal laws and all experimental procedures were reviewed and approved by the Basel-Stadt Cantonal Veterinary Office.

Isolation of murine neutrophils and generation of BMDCs

Mouse BM cells were obtained by flushing femurs and tibiae with PBS. Neutrophils were enriched from BM cells using the EasySep Mouse Neutrophil enrichment kit (Stemcell). BMDCs were generated as previously described (18).

IC preparation and stimulation of DCs and neutrophils

Cells were plated at 1×105 or 1×106 cells/well, in a 96 well plate, pretreated for 1 hour with 1 M MLT-748, MLT-695, Cpd11 or DMSO and then stimulated with IC prepared the previous day by mixing 120 g/ml OVA (Enzo) with 800 g/ml anti-ovalbumin (OVA) polyclonal rabbit IgGs (Sigma C6534). This resulted in an IC concentration of 120 g/ml and ICs were used at 8 g/ml final concentration.

Cytokine, LTB4 and ROS measurement

Mouse paws were snap-frozen in liquid nitrogen and subsequently lysed and homogenized in cell lysis buffer (Cell Signaling Technology) with protease inhibitor cocktail (Complete Mini Tablet, Roche) using a Precellys24. Protein concentration (BCA Kit, Thermo Scientific) in homogenates was assessed and normalized. Cytokine concentrations in paw homogenates, mouse serum samples and human and mouse cell culture supernatants were assessed by V-Plex kits (MesoScale Discovery). LTB4 concentration was determined by ELISA (Cayman Chemical). ROS were assessed by flow cytometry using the dye Dihydrorhodamine 123 (Sigma). 1×105 cells were pre-treated with MLT-748 and Cpd11 (both 1 M) for 1 hour, followed by 30 min incubation at 37°C with OVA/anti-OVA IC (8 g/ml) or medium containing Dihydrorhodamine 123. After incubation, samples were placed on ice and rapidly acquired on a BD LSRFortessaTM.

Preparation of MLT-695-loaded food pellets

The formulation of drugs in food was achieved using a custom-built machine to reform powdered mouse food containing MLT-695 into pellets. Dry powdered food (400 g) for rodents (Provimi Kliba SA, Kaiseraugst, Switzerland) was mixed well using a commercial kitchen mixer with MLT-695 powder at a final concentration of 0.1 or 0.3 g per kg of food and gradual addition of 200 ml sterile water. The MLT-695 food mixture (or control empty food) was loaded into a custom-built extruder, compressed and broken up into 2-4 cm pellets and dried at 35°C for 24 h in an airflow cabinet. Pellets were used within 3 weeks after preparation.

MLT-695 in vivo dose range-finding studies

C57BL/6J mice were fed for four days with empty food pellets containing no MLT-695 to acclimatize mice. They were then substituted with MLT-695-loaded food pellets at 0.1 or 0.3 g/kg (g compound per kg food). Blood samples were collected on days 3, 10 and 12 at 2 hours pre-dark period, and on day 4 and day 13 at 1-hour post dark period. Body weight was recorded on each of these days. K/BxN serum transfer-induced arthritis
Arthritis was induced on day 0 in WT, MALT1 KO or MALT1 PD mice by i.v. injection of 150 µl pooled serum obtained from K/BxN arthritic mice. The severity of paw swelling was scored as described earlier (19), and similarly, the area under the curve (AUC) was evaluated as described previously (20).
For experiments involving MLT-695 treatment, mice were allowed to acclimatize to empty food pellets. Two days prior to K/BxN serum administration, animals were either kept on empty food pellets (control group) or switched to MLT-695-loaded food pellets which were provided ad libitum until termination of the experiment.

Histology and Immunohistochemistry

Hind paws were fixed in 10% buffered formalin for 48 hours and decalcified over 6 days in Immunocal (ref 1440, Decal Chemical Corp) before paraffin embedding. Three μm thick sections were stained with hematoxylin and eosin (HE) and Safranin O (SafO). Histopathological changes were blindly scored on a scale of 0 (normal) to 3 (severe changes). The following parameters were assessed: inflammatory cell infiltrates, joint damage and proteoglycan loss (21).
Immunohistochemical staining for neutrophils/granulocytes with Ly6-2B or macrophages with F4/80 was performed on the Ventana Discovery XT stainer (Roche Diagnostics) using monoclonal rat antiLy-6B2 (clone 7/4, AbD Serotec) and rat anti-F4/80 (clone: Cl:A3-1 AbD Serotec), followed by goat anti-rat IgG (Jackson Immunoresearch Laboratories) and DapMap detection system (Roche Diagnostics).

Passive ITP model and platelet enumeration

Passive ITP in MALT1 KO, MALT1 PD and WT littermates (all C57BL/6) was induced by intraperitoneal injection of 4 µg of a rat IgG1 anti-mouse CD41 monoclonal antibody (clone MWReg30) in 200 µl PBS. Some animals received 2 g/kg bodyweight of IVIG (Privigen huIgG, CSL Behring AG; Swissmedic #58314) intraperitoneally 2 hours prior to injection of anti-CD41 antibody. For experiments involving MLT-695 treatment, WT BALB/cByJrj mice were injected intraperitoneally with 2 µg of the anti-CD41 antibody. IVIG (2 g/kg) and oral MLT-695 (20mg/kg) treatment was performed 2 hours prior to administration of the anti-CD41 antibody. Four hours after the anti-mouse CD41 antibody injection, animals were terminated and 450 µl of blood was collected in tubes containing Na-Citrate (3.2% solution). Circulating platelets were quantified by flow cytometry using counting beads. Platelets were identified using a PE-labelled antibody specific for CD61 (clone 2C9.G2) and eFluor660-labelled F(ab’)2 Goat anti-Rat IgG was used to verify the effective injection of anti-CD41 antibody for in vivo depletion of platelets.

DNP-KLH immunization

After empty food acclimatization, C57BL/6 animals were fed with MLT-965-loaded food pellets at different concentrations on day -2. Animals were immunized intraperitoneally with DNP-KLH (50 g/mouse, produced in-house) in Alu-Gel-S (Serva) on day 0 and serum anti-DNP IgG and anti-DNP IgM was measured on day 8 post-immunization by ELISA as described earlier (11).
Bar graphs in the figures represent average values + SEM unless indicated otherwise. Statistical significance between groups was calculated using a two-tailed unpaired Student t-test, ordinary oneway or two-way ANOVA using GraphPad Prism (v8) and is indicated in the graphs as follows: p<0.05, p<0.01, p<0.001, p<0.0001. Non-significant differences are not indicated. 

Results

Murine myeloid cells rely on MALT1 protease for optimal FcR-induced cytokines

To dissect the relevance of MALT1 scaffolding versus protease functions for FcR-induced cytokine release, we established a functional in vitro assay using primary neutrophils isolated from bone marrow of WT, MALT1 KO, and MALT1 PD mice. Neutrophils were stimulated with soluble OVA/anti-OVA ICs. In line with previous findings (8), compared to WT, MALT1 KO neutrophils displayed a significantly reduced production of multiple pro-inflammatory cytokines including TNF-α and IL-6 (Figure 1A). This was recapitulated in the MALT1 PD neutrophils (Figure 1A), indicating that the enzymatic activity of MALT1 is the critical mediator of FcR-induced cytokine production. We extended our analyses to other myeloid cell types to demonstrate that MALT1 PD and MALT1 KO BMDCs also displayed reduced cytokine release upon IC stimulation (Figure 1B).
Canonical NF-B activation downstream of ITAM-containing receptors in myeloid cells involves a multistep signaling cascade requiring the kinase Syk, leading to the formation of the CARD9/BM signalosome and subsequent activation of the IKK complex (5). Using a previously reported low molecular weight inhibitor of Syk (Cpd11) (16), we could efficiently abrogate IC-driven cytokine release in both WT murine neutrophils (Figure 1C) and BMDCs (Figure 1D), substantiating the requirement of Syk enzymatic activity for FcR-induced pro-inflammatory cytokine release by murine myeloid cells. To extend the findings obtained with MALT1 PD cells, we next evaluated the impact of a novel MALT1 protease inhibitor (MLT-748) (15) displaying similar potency and selectivity to the recently reported inhibitor MLT-827 (Supplemental Figure 1A) (22, 23). In line with the observations using MALT1 PD cells, treatment with 1µM of MLT-748 efficiently inhibited ICdriven TNF-, IL-6 and KC/GRO production in WT neutrophils (Figure 1C) and BMDCs (Figure 1D). Overall, using genetic approaches and pharmacological inhibitors, our results strengthen and extend previous data (8, 9) to demonstrate that the protease activity of MALT1 is critical for the optimal production of FcR triggered cytokines.FcR-driven cytokine release but not ROS or LTB4 production in human myeloid cells is dependent on MALT1 protease activity
Next, we evaluated whether these findings translated to human myeloid cells. IC-based FcR stimulation of primary human granulocytes or iMoDCs efficiently elicited multiple cytokines (Figure 2A-B). In line with the mouse data, Syk inhibition by Cpd11 abrogated the production of all the cytokines. Inhibition of the MALT1 protease by MLT-748 efficiently suppressed TNF- and IL-6 in both granulocytes and iMoDCs (Figure 2A- B) and partially impacted IL-8 and IL-1 production. Of note, MLT-748 inhibited IC-driven cytokine release with a similar potency to that reported in other MALT1 protease-dependent assays (Supplementary Figure 1A-B). In addition, FcR-mediated TNF and IL-6 production by granulocytes and iMoDCs relied on CBM-driven canonical NF-B activation as pharmacological inhibition of IKK and MLT-748 suppressed cytokine release with similar potency (data not shown). This is in line with what was reported for other CARD9-dependent innate receptors (24, 25).
Triggering of FcRs by IC also leads to rapid cellular activation and release of ROS and leukotrienes such as LTB4 (3). IC-induced activation of human neutrophils resulted in the significant release of LTB4 and ROS, which was efficiently blocked by Syk inhibition (Figure 2C). In contrast, inhibition of the MALT1 protease by MLT-748 had no or only limited influence on LTB4 and ROS release (Figure 2C). We confirmed similar findings on ROS production by iMoDCs (Figure 2D). Together, these data demonstrated that MALT1 protease is required for optimal induction of several FcR-mediated pro-inflammatory cytokines in human myeloid cells but is dispensable for other effector functions such as ROS and LTB4 release.

MALT1 protease inhibition ameliorates autoantibody-induced arthritis in mice

Based on our in vitro results, we next wanted to study the relevance of MALT1 in an in vivo model of autoantibody-driven disease, in which inflammation is strongly dependent on FcRs. Genetic deficiency of CARD9 and specific FcRs has been reported to suppress K/BxN autoantibodyinduced arthritis (8, 26). To assess the relevance of MALT1 scaffolding and protease function, we evaluated the induction of K/BxN serum-induced arthritis in MALT1 KO, MALT1 PD and WT littermate controls (Figure 3A-B). Consistent with the data reported in CARD9 deficient animals and the relevance of MALT1 in FcR-induced inflammation, MALT1 KO mice displayed significantly reduced paw swelling (Figure 3A-B). Histological analysis of the joints 11 days after serum transfer revealed reduced joint damage and proteoglycan loss, as well as diminished macrophage and neutrophil infiltration in MALT1 KO animals (Figure 3C and Supplementary Figure 2). Consistent with these findings and our in vitro data we observed lower levels of pro-inflammatory cytokines in the paw homogenates and serum of MALT1 KO mice (Figure 3D-E). In contrast, MALT1 PD mice were not protected in the K/BxN arthritis model and displayed variable results in the paw swelling in independent experiments (Figure 3A-B). We assessed the degree of swelling (area under the curve) relative to WT across different experiments (Figure 3B) and found a significant improvement in disease score in MALT1 KO, while MALT1 PD animals displayed a similar degree of disease as compared to WT. The lack of protection and variable swelling observed in MALT1 PD mice may indicate a limited relevance of the MALT1 protease function for FcR-dependent effects in vivo. Alternatively, it may be a reflection of the multiple immunological alterations occurring in these animals (11). Indeed, MALT1 PD mice suffer from hypergammaglobulinemia (IgG1 and IgE) (11), which may alter the threshold of FcR signaling in vivo, in spite of the reduced cytokine production observed in our in vitro system. Moreover, additional changes associated with the underlying inflammatory pathology occurring in MALT1 PD mice may contribute to the lack of protection in this model.
To evaluate the relevance of the MALT1 protease function independent of the effects of the MALT1 PD mouse pathology, we tested the impact of a novel MALT1 protease inhibitor MLT-695 in WT mice injected with K/BxN serum. Compared to MLT-748, MLT-695 displays equal selectivity and similar in vitro potency (Supplementary Figure 1A-B), but improved solubility and permeability which results in an overall more favorable pharmacokinetic profile for in vivo applications. For continuous in vivo administration, we used food pellets loaded with MLT-695. This approach led to continuously high drug exposure throughout the day and night period (Supplementary Figure 1C-D). To demonstrate that this dosing regimen efficiently blocked MALT1 protease function in vivo, we used immunizations with DNP-KLH and confirmed inhibition of DNP-specific IgG and IgM in the serum of MLT-695 treated animals (Supplementary Figure 1E-F), consistent with the defective antibody response to DNP-KLH reported in MALT1 PD mice (11, 14). As opposed to the observations in MALT1 PD animals, pharmacological inhibition of MALT1 protease using MLT-695 in WT mice resulted in partial protection from K/BxN serum-induced arthritis (Figure 4A-B), histopathological parameters (Figure 4C-D), and paw and serum cytokine levels (Figure 4E-F).
In summary, our data demonstrate that pharmacological inhibition of the MALT1 protease ameliorates autoantibody-mediated, cytokine-dependent FcR-driven joint inflammation in vivo. Of note, results in MALT1 PD mice were not predictive of those obtained with pharmacological MALT1 protease inhibition. Progressive disease in MALT1 PD animals likely masked the effect of MALT1 protease inhibition on arthritis development.

Elimination of IgG-coated platelets in a passive ITP model occurs independently of MALT1 scaffolding or protease activity

We next evaluated MLT-695 in a passive ITP model in Balb/c mice in which rapid onset of thrombocytopenia is induced by administration of an anti-CD41 antibody. In this model, disease was caused by FcR-mediated phagocytosis of platelets. Consistent with the relevance of FcRs in this model, FcR blockade via the systemic administration of intravenous immunoglobulins (IVIG) efficiently prevented platelet depletion (Figure 5A). In contrast, treatment with 20 mg/kg of MLT-695 via oral gavage 2 hours prior to the administration of the anti-platelet antibody, did not affect platelet elimination (Figure 5A). The lack of protection in this model could not be explained by limited exposure of the compound in vivo, as MLT-695 blood levels achieved by oral gavage were in the same range as, or higher than, those observed in the K/BxN arthritis model (data not shown).
To extend these findings, we performed the ITP model in MALT1 KO and MALT1 PD mice on a C57BL/6 background. While IVIG pre-treatment efficiently prevented platelet elimination, antiCD41 antibody-mediated platelet depletion occurred normally in both MALT1 PD and MALT1 KO mice (Figure 5B), confirming the data obtained with the MALT1 protease inhibitor in Balb/c mice. This indicates that abrogation of the MALT1 scaffolding or protease activities has no impact on FcRmediated platelet depletion in vivo. Together with our in vitro data and the results from the KBxN model of arthritis, these data suggest that MALT1 protease function plays a role in a subset of FcRmediated diseases as it is important for cytokine release downstream of FcRs, but not for other cellular effector functions.

Discussion

Inhibition or disruption of the MALT1 protease has been reported to influence signaling events downstream of multiple immune receptors (5, 22, 27). Despite this, its contribution to FcR-mediated events has so far not been studied in detail. The data presented here consistently demonstrate that the protease function of MALT1 is critically required for the optimal production of canonical NF-B induced pro-inflammatory cytokines upon activation of FcRs in both murine and human myeloid cells. In contrast, the abrogation of MALT1’s function did not influence other FcR-mediated cellular effector functions such as ROS and leukotriene release (8, 9). Similarly, in mast cells activated through the IgE receptor FcR, the pro-inflammatory cytokine release is dependent on MALT1 and Bcl10, while degranulation and leukotriene synthesis occur via independent pathways (28). In sharp contrast to the selective impact of MALT1 protease on FcR-mediated cytokine production, Syk inhibition affected all tested FcR-mediated effector functions (3). Consistently, Syk inhibition reduced disease development in the K/BxN and the ITP model (29, 30). Therefore, our data help position MALT1 protease inhibitors for inflammatory diseases in which autoantibodies lead to cytokine production.
Consistent with the critical role of the MALT1 protease for optimal production of FcR-induced proinflammatory cytokines, we demonstrated that MALT1 deficiency/pharmacological inhibition can ameliorate joint inflammation and recruitment of inflammatory cells in an in vivo model of autoantibody-driven arthritis. In contrast, MALT1 PD animals did not show reduced disease progression. This is a likely consequence of the complex inflammatory pathology occurring in these mice consequent to a reduced and impaired regulatory T cell compartment (11-14, 31-34). Of note, MALT1 PD mice display a progressive elevation in serum IgG1 and IgE (11), which may alter the threshold of FcR signaling in vivo, and display signs of chronic mast cell degranulation (unpublished observation, T. Calzascia), which may directly influence the mast cell-dependent joint inflammation process once initiated by the autoantibodies in the K/BxN serum (35). Using a novel MALT1 protease inhibitor in vivo (MLT-695) we were able to evaluate the role of MALT1 protease function in the development of auto-antibody induced arthritis in absence of the established immune alterations occurring in MALT1 PD animals. Importantly, the inconsistency between the data generated in WT mice treated with the MALT1 protease inhibitor and the results obtained in MALT1 PD mice highlights how the use of genetically modified animals may at times mislead drug discovery and how critical it is to generate appropriate pharmacological inhibitors.
Consistent with the results presented here, previous work highlighted the importance of the CARD9/BM signalosome in the autoantibody-dependent K/BxN arthritis model. CARD9-deficiency abrogated the capacity of macrophages and dendritic cells (DCs) to produce pro-inflammatory cytokines upon stimulation of multiple FcRs (24, 25, 36, 37), and the selective deletion of CARD9 in neutrophils was protective in the K/BxN arthritis model (8). In the T cell-dependent collagen-induced arthritis model, MALT1 deletion in T cells only affected disease onset but not the course of the disease (38). This suggests that the role of MALT1 in early arthritis, during the formation of autoantibodies is most prominent in the adaptive immune system, while MALT1 and the CBM complex are most important in the myeloid compartment in established disease, once the autoantibodies are formed (38). Given the multiple biological pathways and cell types contributing to arthritis, MALT1 protease represents an attractive therapeutic target. However, the severe immune alterations and regulatory T cell impairment observed in mice affected by systemic or T cell subsetspecific MALT1 protease deficiency also raised potential safety concerns associated with chronic MALT1 inhibition (11-14, 31-34). As the potency and selectivity of currently used MALT1 protease inhibitors such as Mepazine and MI-2 are limited (39-41), MLT-695 may provide a novel pharmacological tool to help evaluate the contribution of MALT1’s enzymatic function to different biological processes in vivo, as well as the potential safety liabilities associated with chronic MALT1 inhibition.
Genetic or pharmacological abrogation of MALT1 activity was ineffective at preventing antibody-dependent platelet depletion in a passive ITP model. While multiple mechanisms contribute to platelet destruction in ITP (42), thrombocytopenia induction in passive ITP murine models was shown to rely on platelet elimination by phagocytic cells (43). Our findings in the ITP model are consistent with multiple reports suggesting that genetic abrogation of PKC, or CBM components does not prevent phagocytosis mediated by FcR-mediated or other CARD9- and MALT1-dependent ITAM-containing receptors (6, 7, 24, 44, 45). In contrast to our findings, a single study suggested that BCL10 silencing impacts the phagocytosis of IgG-coated red blood cells (6), and one other study using a novel MALT1 PD mouse line recently proposed that the MALT1 protease function is essential for platelet elimination in a passive mouse model of ITP (9). A potential explanation for the discrepancy may lie in the novel BALB/c MALT1 PD line generated by Nakamura et al., which other than the more established C57BL/6 MALT1 PD lines (11-14), is not yet thoroughly characterized with respect to extensive and progressive immune alterations (9). Based on published data and our own observation, the severity of the spontaneous pathology affecting MALT1 PD lines varies significantly depending on the genetic background and housing conditions. Of note, homozygous MALT1 PD mice on a BALB/c background displayed a severe purulent eye inflammation before 7 weeks of age which led us to stop all further activities on such genetic background (unpublished observation, T. Calzascia).
In conclusion, our findings offer critical insights into the relevance of the MALT1 function in FcRmediated events. MALT1 protease function exclusively governs FcR-mediated cytokine production, while other FcR effector functions are MALT1 independent. A novel MALT1 antagonist, MLT-695, rather than MALT1 PD mice, was essential to obtain this knowledge, and to position MALT1 antagonists in indications such as RA, where FcR activation leads to cytokine production. While drug discovery always relies on functional knowledge of the molecular target, our study shows that the reverse can be true as well: Novel compounds can advance target biology and, ultimately, translational research towards clinical application.

References

Ben Mkaddem S, Benhamou M, Monteiro RC. Understanding Fc Receptor Involvement in Inflammatory Diseases: From Mechanisms to New Therapeutic Tools. Front Immunol. 2019;10:811.
Hogarth PM, Pietersz GA. Fc receptor-targeted therapies for the treatment of inflammation, cancer and beyond. Nat Rev Drug Discov. 2012;11(4):311-31.
Nimmerjahn F, Ravetch JV. Fc gamma receptors as regulators of immune responses. Nature Reviews Immunology. 2008;8(1):34-47.
Bussel J, Arnold DM, Grossbard E, Mayer J, Trelinski J, Homenda W, et al. Fostamatinib for the treatment of adult persistent and chronic immune thrombocytopenia: Results of two phase 3, randomized, placebo-controlled trials. Am J Hematol. 2018;93(7):921-30.
Jaworski M, Thome M. The paracaspase MALT1: biological function and potential for therapeutic inhibition. Cell Mol Life Sci. 2016;73(3):459-73.
Rueda D, Gaide O, Ho L, Lewkowicz E, Niedergang F, Hailfinger S, et al. Bcl10 controls TCR- and FcgammaR-induced actin polymerization. J Immunol. 2007;178(7):4373-84.
Garcia-Gomez S, Alvarez Doforno R, Martinez-Barricarte R, Torres JM, Ferreira Cerdan A, Davila M, et al. Actin polymerisation after FCgammaR stimulation of human fibroblasts is BCL10 independent. Clin Immunol. 2016;163:120-2.
Nemeth T, Futosi K, Sitaru C, Ruland J, Mocsai A. Neutrophil-specific deletion of the CARD9 gene expression regulator suppresses autoantibody-induced inflammation in vivo. Nature communications. 2016;7:11004.
Nakamura Y, Yokoyama K, Igaki K, Tsuchimori N. Role of Malt1 protease activity in pathogenesis of inflammatory disorders mediated by FcgammaR signaling. Int Immunopharmacol. 2018;56:193-6.
Demeyer A, Staal J, Beyaert R. Targeting MALT1 Proteolytic Activity in Immunity, Inflammation and Disease: Good or Bad? Trends Mol Med. 2016;22(2):135-50.
Bornancin F, Renner F, Touil R, Sic H, Kolb Y, Touil-Allaoui I, et al. Deficiency of MALT1 paracaspase activity results in unbalanced regulatory and effector T and B cell responses leading to multiorgan inflammation. J Immunol. 2015;194(8):3723-34.
Gewies A, Gorka O, Bergmann H, Pechloff K, Petermann F, Jeltsch KM, et al. Uncoupling malt1 threshold function from paracaspase activity results in destructive autoimmune inflammation. Cell reports. 2014;9(4):1292-305.
Jaworski M, Marsland BJ, Gehrig J, Held W, Favre S, Luther SA, et al. Malt1 protease inactivation efficiently dampens immune responses but causes spontaneous autoimmunity. The EMBO journal. 2014;33(23):2765-81.
Yu JW, Hoffman S, Beal AM, Dykon A, Ringenberg MA, Hughes AC, et al. MALT1 Protease Activity Is Required for Innate and Adaptive Immune Responses. PloS one. 2015;10(5):e0127083.
Quancard J, Klein T, Fung SY, Renatus M, Hughes N, Israel L, et al. An allosteric MALT1 inhibitor is a molecular corrector rescuing function in an immunodeficient patient. Nat Chem Biol. 2019;15(3):304-13.
Thoma G, Smith AB, van Eis MJ, Vangrevelinghe E, Blanz J, Aichholz R, et al. Discovery and profiling of a selective and efficacious Syk inhibitor. J Med Chem. 2015;58(4):1950-63.
Quach A, Ferrante A. The Application of Dextran Sedimentation as an Initial Step in Neutrophil Purification Promotes Their Stimulation, due to the Presence of Monocytes. J Immunol Res. 2017;2017:1254792.
Martin K, Muller P, Schreiner J, Prince SS, Lardinois D, Heinzelmann-Schwarz VA, et al. The microtubule-depolymerizing agent ansamitocin P3 programs dendritic cells toward enhanced antitumor immunity. Cancer Immunol Immunother. 2014;63(9):925-38.
Koziczak-Holbro M, Littlewood-Evans A, Pollinger B, Kovarik J, Dawson J, Zenke G, et al. The critical role of kinase activity of interleukin-1 receptor-associated kinase 4 in animal models of joint inflammation. Arthritis Rheum. 2009;60(6):1661-71.
Littlewood-Evans A, Sarret S, Apfel V, Loesle P, Dawson J, Zhang J, et al. GPR91 senses extracellular succinate released from inflammatory macrophages and exacerbates rheumatoid arthritis. Journal of Experimental Medicine. 2016;213(9):1655-62.
Koenders MI, Lubberts E, Oppers-Walgreen B, van den Bersselaar L, Helsen MM, Kolls JK, et al. Induction of cartilage damage by overexpression of T cell interleukin-17A in experimental arthritis in mice deficient in interleukin-1. Arthritis Rheum. 2005;52(3):975-83.
Bardet M, Unterreiner A, Malinverni C, Lafossas F, Vedrine C, Boesch D, et al. The T-cell fingerprint of MALT1 paracaspase revealed by selective inhibition. Immunol Cell Biol. 2018;96(1):81-99.
Unterreiner A, Touil R, Anastasi D, Dubois N, Niwa S, Calzascia T, et al. Myeloid Innate Signaling Pathway Regulation by MALT1 Paracaspase Activity. J Vis Exp. 2019(143).
Gross O, Gewies A, Finger K, Schafer M, Sparwasser T, Peschel C, et al. Card9 controls a non-TLR signalling pathway for innate anti-fungal immunity. Nature. 2006;442(7103):651-6.
Hara H, Ishihara C, Takeuchi A, Imanishi T, Xue L, Morris SW, et al. The adaptor protein CARD9 is essential for the activation of myeloid cells through ITAM-associated and Toll-like receptors. Nat Immunol. 2007;8(6):619-29.
Corr M, Crain B. The role of FcgammaR signaling in the K/B x N serum transfer model of arthritis. J Immunol. 2002;169(11):6604-9.
Gringhuis SI, Wevers BA, Kaptein TM, van Capel TM, Theelen B, Boekhout T, et al. Selective C-Rel activation via Malt1 controls anti-fungal T(H)-17 immunity by dectin-1 and dectin-2. PLoSPathog. 2011;7(1):e1001259.
Klemm S, Gutermuth J, Hultner L, Sparwasser T, Behrendt H, Peschel C, et al. The Bcl10Malt1 complex segregates Fc epsilon RI-mediated nuclear factor kappa B activation and cytokine production from mast cell degranulation. J Exp Med. 2006;203(2):337-47.
Jakus Z, Simon E, Balazs B, Mocsai A. Genetic deficiency of Syk protects mice from autoantibody-induced arthritis. Arthritis Rheum. 2010;62(7):1899-910.
Podolanczuk A, Lazarus AH, Crow AR, Grossbard E, Bussel JB. Of mice and men: an openlabel pilot study for treatment of immune thrombocytopenic purpura by an inhibitor of Syk. Blood. 2009;113(14):3154-60.
Cheng L, Deng N, Yang N, Zhao X, Lin X. Malt1 Protease Is Critical in Maintaining Function of Regulatory T Cells and May Be a Therapeutic Target for Antitumor Immunity. J Immunol. 2019;202(10):3008-19.
Demeyer A, Skordos I, Driege Y, Kreike M, Hochepied T, Baens M, et al. MALT1 Proteolytic Activity Suppresses Autoimmunity in a T Cell Intrinsic Manner. Frontiers in Immunology. 2019;10(1898).
Di Pilato M, Kim EY, Cadilha BL, Prussmann JN, Nasrallah MN, Seruggia D, et al. Targeting the CBM complex causes Treg cells to prime tumours for immune checkpoint therapy. Nature. 2019;570(7759):112-6.
Rosenbaum M, Gewies A, Pechloff K, Heuser C, Engleitner T, Gehring T, et al. Bcl10controlled Malt1 paracaspase activity is key for the immune suppressive function of regulatory T cells. Nature communications. 2019;10(1):2352.
Nigrovic PA, Lee DM. Mast cells in inflammatory arthritis. Arthritis Res Ther. 2005;7(1):1-11.
Rhoads JP, Lukens JR, Wilhelm AJ, Moore JL, Mendez-Fernandez Y, Kanneganti TD, et al. Oxidized Low-Density Lipoprotein Immune Complex Priming of the Nlrp3 Inflammasome Involves TLR and FcgammaR Cooperation and Is Dependent on CARD9. J Immunol. 2017;198(5):2105-14.
Phongsisay V, Iizasa E, Hara H, Yoshida H. Evidence for TLR4 and FcRgamma-CARD9 activation by cholera toxin B subunit and its direct bindings to TREM2 and LMIR5 receptors. Mol Immunol. 2015;66(2):463-71.
Gilis E, Gaublomme D, Staal J, Venken K, Dhaenens M, Lambrecht S, et al. MALT1-deletion in T-cells protects against the development of auto-immune arthritis, but causes spontaneous osteoporosis. Arthritis Rheumatol. 2019.
Unterreiner A, Stoehr N, Huppertz C, Calzascia T, Farady CJ, Bornancin F. Selective MALT1 paracaspase inhibition does not block TNF-alpha production downstream of TLR4 in myeloid cells. Immunol Lett. 2017;192:48-51.
Fontan L, Melnick A. Targeting lymphomas through MALT1 inhibition. Oncotarget. 2012;3(12):1493-4.
Meloni L, Verstrepen L, Kreike M, Staal J, Driege Y, Afonina IS, et al. Mepazine Inhibits
RANK-Induced Osteoclastogenesis Independent of Its MALT1 Inhibitor Function. Molecules. 2018;23(12).
Swinkels M, Rijkers M, Voorberg J, Vidarsson G, Leebeek FWG, Jansen AJG. Emerging Concepts in Immune Thrombocytopenia. Front Immunol. 2018;9:880.
Neschadim A, Branch DR. Mouse models of autoimmune diseases: immune thrombocytopenia. Curr Pharm Des. 2015;21(18):2487-97.
Strasser D, Neumann K, Bergmann H, Marakalala MJ, Guler R, Rojowska A, et al. Syk kinase-coupled C-type lectin receptors engage protein kinase C-sigma to elicit Card9 adaptormediated innate immunity. Immunity. 2012;36(1):32-42.
Li X, Cullere X, Nishi H, Saggu G, Durand E, Mansour MK, et al. PKC-delta activation in neutrophils promotes fungal clearance. J Leukoc Biol. 2016;100(3):581-8.