The pseudo-kinase family of tribbles (TRIB) proteins has been linked to a variety of cell signalling pathways and appears to have functionally divergent roles with respect to intracellular protein degradation and the ability to regulate signal transduction pathways. In the arthritides, inflammation and a wide variety of pro-inflammatory pathways have been implicated to drive the cartilage destruction and consequent disability associated with both rheumatoid arthritis (RA) and osteoarthritis (OA). Despite burgeoning evidence linking the TRIB to inflammation-related pathologies such as diabetes, multiple sclerosis and cancer, very little is known about their roles in arthritis. The present review discusses current knowledge of the impact of TRIB on pro-inflammatory cellular mechanisms and pathways known to be important in the pathogenesis of RA and OA.
The mammalian TRIB proteins (TRIB1, TRIB2 and TRIB3) have functionally divergent roles including regulation of a variety of cell signalling pathways. This regulation may be manifested via their ability to negate the transcriptional influence of signalling moieties (either via proteosome-mediated degradation or acting as decoy kinases) or by directly regulating signal transduction cascades . TRIB are well documented to have roles in pathologies (e.g., diabetes, multiple sclerosis and leukaemia [2–4]) that involve inflammation.
In the arthritides, inflammation and pro-inflammatory cytokine-mediated signalling pathways culminate in cartilage destruction, contributing to the disability associated with both rheumatoid arthritis (RA) and osteoarthritis (OA). Although the chronic inflammation characteristic of autoimmune RA pathology is driven by adaptive immune mechanisms, it is likely that the innate immune system is important in establishing early disease via an environmentally-mediated adjuvant effect . In the context of OA, it is increasingly apparent that innate immune mechanisms are involved in the low-grade inflammation observed in a large proportion of cases .
An important role for ‘cytokine networks’ (especially when associated with inflammation) in such diseases has gained considerable support as evidenced by the number of anti-cytokine biologics now available (see  and references therein) to treat arthritis. Such networks and their signal transduction mechanisms appear to be important drivers of disease in both RA  and OA  and include G-protein-coupled receptors, chemokines and a myriad of pro-inflammatory cytokines [8,10,11].
The literature now supports the involvement of various signalling pathways and even specific protein kinases , in the ‘inflammation’ associated with the arthritides. It is becoming increasingly apparent that for a wide variety of these stimuli, including mechanical load , a diverse but overlapping set of signal transduction pathways drives the ensuing tissue damage. Several such pathways are convincingly linked to the aberrant expression of metalloproteinases, such as a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTSs) and matrix metalloproteinases (MMPs). These enzymes are known to degrade the key protein components of the cartilage extracellular matrix (aggrecan and collagen) (see  and references therein). A detailed understanding of pro-catabolic signalling has long been suggested to be the key to identifying new therapeutic targets for arthritis . However, much remains to be learnt about the complex mechanisms regulating processes such as MMP expression or endoplasmic reticulum (ER) stress, before this knowledge can be utilized clinically.
Below, we describe current literature pertaining to the impact of TRIB on a range of inflammatory and signalling processes considered important in arthritis pathogenesis and consider possible directions for future research.
TRIB1 has been suggested to be an important mediator of innate immunity. In RAW264.7 macrophages, activation by toll-like receptor (TLR)2 ligands, tumour necrosis factor α (TNFα) production and monocyte chemoattractant protein 1 (MCP-1)-dependent migration are all regulated by TRIB1 levels . TRIB1 silencing leads to an increase in TNFα and expression of CCAAT/enhancer-binding protein (C/EBP)β, which is an important transcription factor for metabolic disturbances such as ER stress and inflammation. C/EBPβ is expressed in synovial tissues and chondrocytes and, along with other C/EBP family members, mediates MMP expression and chondrocyte differentiation . Moreover, C/EBPβ also promotes expression of receptor activator of nuclear factor kappa B (NFκB) ligand (RANKL) and osteoclast formation in RA synovium and plays a key role in cartilage degradation during inflammatory arthritis (see  and references therein). TRIB1 deficiency has been reported to severely inhibit the differentiation of tissue-resident M2-like anti-inflammatory macrophages  leading to exacerbated metabolic dysfunction in obesity, a major risk factor for OA . Thus, TRIB1-mediated regulation of myeloid cells may be a significant factor in arthritis pathogenesis.
In addition to regulating the β-isoform of the C/EBP family, TRIB1 is reported to be essential for the degradation of C/EBPα via interaction with the E3 ubiquitin ligase constitutively photomorphogenic 1 (COP1), forming what has been referred to as a leukaemogenic oncoprotein complex . With regard to joint tissues, C/EBPα (like the β-isoform) is important in regulating osteoclastogenesis , such that TRIB1 levels could impact on skeletogenesis. Furthermore, C/EBPα expression has been observed in developing growth plate and articular cartilage and proposed to be a marker of early chondrogenic development  since it is absent from chondrocytes in the proliferative zone. TRIB1 might conceivably therefore contribute to the dysregulated recapitulation of developmental pathways that are increasingly thought to underpin aberrant tissue remodelling in joint diseases such as OA .
TRIB1 negatively regulates retinoic acid receptor (RAR) signalling by interacting with both RARs and the closely related retinoid X receptors (RXRs) , which function as lipid-activated transcription factors. Although RAR antagonism is an established therapeutic target in RA  and RAR activity also mediates loss of Sox9 and matrix gene expression in chondrocytes , diverse downstream consequences of specific RAR/RXR activation suggest that TRIB1 regulation of these factors in cell-specific contexts requires further elucidation to assess the full therapeutic potential.
Of specific relevance to RA, Foxp3 (forkhead box P3) is considered to be a master regulator of regulatory T-cells (Treg) which are key players in helping to prevent autoimmunity by maintaining self-tolerance. Recent evidence confirms the therapeutic potential of Foxp3 in inflammatory arthritis  and TRIB1 has been shown to bind Foxp3, a nuclear interaction that positively promotes Foxp3 expression . It is tempting to speculate that TRIB1 activity might therefore act to prevent the development of RA by maintaining an adequate anergic Treg population.
Similar to TRIB1, TRIB2 appears to play roles in innate immunity. In monocytes, TRIB2 appears to act as a constitutive brake, holding inflammatory activation in check prior to release, occurring via stimulus-dependent down-regulation of TRIB2 and subsequent interleukin (IL)-8 induction through specific mitogen-activated protein kinase (MAPK)-dependent pathways . Conversely, other groups have reported that the same stimulus (modified low-density lipoprotein) enhances TRIB2 and reduces IL-10 expression in monocyte-derived macrophages, although the mechanism of this effect was not investigated . In addition, TRIB2 has been reported to regulate TLR signalling, which may be a significant factor in OA pathology . It has been proposed that TRIB2 expression is induced by TLR5 activation as part of a negative feedback loop, moderating excessive pro-inflammatory NFκB activity by binding the p100 subunit and that this mechanism is ineffective during irritable bowel disease .
Again similar to TRIB1, TRIB2 may play important roles in acute leukaemias and, once more, interactions with C/EBPα appear to be important . In the context of articular cartilage and re-activation of developmental pathways during OA, it is also noteworthy that Notch1 mutations in T-cell acute lymphoblastic leukaemias have been associated with elevated expression of the target gene TRIB2 . This may well have relevance for arthritis since the Notch pathway is also proposed to contribute to chondrocyte dysfunction in OA . TRIB2 also regulates Wnt signalling in a cell-specific manner, whereby TRIB2 associated-ubiquitin E3 ligases β-transducin repeat-containing E3 ubiquitin protein ligase (βTrCP), COP1 and small ubiquitin regulatory factor (Smurf)1 reduce transcription factor 4 (TCF4)/β-catenin expression. In liver cancer cells this occurs via direct binding of TRIB2 to these E3 ligases, facilitating nuclear ubiquitination of TCF4 and β-catenin . Involvement of the canonical Wnt/β-catenin pathway in OA pathogenesis remains a debated topic (see  and references therein). In liver cancer, TRIB2 promotes protein stabilization of the Yes-associated protein (YAP) co-activator via interaction with βTrCP. It is suggested that TRIB2 functions as a hub to integrate Wnt/β-catenin, Hippo/YAP and C/EBPα pathways . Since YAP suppression serves to maintain chondrocyte phenotype and inhibit proliferation , elevated TRIB2 and YAP stabilization in chondrocytes could therefore be detrimental if such interactions extrapolate to OA pathology.
Suppression by TRIB2 of the activity of FOXO (forkhead box O) proteins has been reported . Like Foxp3, these belong to the forkhead family of transcription factors. ‘O’ family members are regulated by phosphoinositide 3-kinase (PI3K)/Akt signalling and are tumour suppressors that are inactivated in many human cancers due to hyperactivation of the PI3K/Akt pathway. FOXO3a activity is reported to underpin the persistence of neutrophils within joints during inflammatory arthritis via suppression of Fas ligand . FOXO3a has also been shown to promote tumour metastasis through the induction of MMPs . In chondrocytes, however, adenosine monophosphate-activated kinase activation induces FOXO3a; this limits oxidative stress by increasing superoxide dismutase (SOD)2 expression, thereby serving to inhibit cartilage damage . Furthermore, FOXO3 is reported to support oxidative stress resistance in chondrocytes . Since Akt signalling (expected to down-regulate FOXO3a) is an important mediator of cytokine-induced MMP expression in chondrocytes , TRIB2 might well influence FOXO activity to different outcomes in different disease contexts.
TRIB3 plays a role in forkhead transcription factor regulation. In tumorigenesis for example, TRIB3 acts as an inhibitor of Akt, whereas TRIB3 loss leads to enhanced Akt phosphorylation and the subsequent hyperphosphorylation (and inactivation) of FOXO3 . As suggested above, the impact of this relationship (if applicable) in arthritis pathology remains unclear. However, in mast cells, TRIB3 has been shown to be a negative regulator of cytokine/chemokine production, effectively controlling the extent of the inflammatory response . Among developmental pathways and again like TRIB2, TRIB3 is reported to be a key regulator of Notch signalling [via transforming growth factor β (TGFβ) signalling and extracellular signal-regulated kinase (ERK); , whereas TRIB3 silencing has been shown to reduce Notch1 expression . It has already been suggested that inhibition of Notch signalling would be therapeutically beneficial for both inflammatory arthritis  and OA .
A number of other TRIB3-interacting proteins have been identified, most of which are nuclear with roles in signal transduction (see  and references therein). For example, the E3 ligase seven in absentia homologue 1 (SIAH1) can bind TRIB3 and target it for proteasomal degradation. This interaction was shown to be sufficient to reduce TGFβ-mediated gene expression . Cartilage contains latent TGFβ which is thought to help maintain tissue homoeostasis following a ‘pro-inflammatory’ stimulus via activin-like kinase (ALK)5 and Smad2/3 signalling . Interestingly, TRIB3 interacts with Smad3 and a positive feedback loop exists whereby Smad3 enhances TRIB3 promoter activity. TRIB3 appears to initiate the degradation of Smurf2, which reduces Smad2 degradation, enhancing Smad3 phosphorylation and promoting nuclear localization . In OA, it is thought that Wnt signalling ultimately skews TGFβ signalling to favour ALK1 and Smad1/5/8 (with concomitant expression of MMP-13) , which might be expected to reduce TRIB3 levels. However, the elevated TRIB3 expression reported for OA chondrocytes  may therefore be a consequence of TGFβ and Smad2/3 signalling whereas the aforementioned Akt-inhibitor mode of TRIB3 action has been suggested to result in reduced responsiveness to insulin-like growth factor 1 (IGF-1)  and consequent cell death. Of relevance here, both IGF-1 and TGFβ have been shown to be protective against cytokine-induced cartilage breakdown .
TRIB3 has been suggested to mediate insulin resistance in obesity caused by dietary excess of saturated fatty acids . Obesity has implications for autoimmunity (including RA ) and is a major risk factor for OA, where it is now thought ‘inflamed’ adipose tissue and dyslipidaemia play key roles in pathogenesis . Saturated fatty acids have also been proposed to mediate TRIB3 expression via induction of ER stress, the phenomenon whereby malfolded proteins accumulate in the ER, initiating inter-organelle signalling (ER to nucleus) known as the unfolded protein response (UPR). TRIB3 is part of the autophagy system  and is induced by ER stress via activation of activating TCF4 , a transcription factor that TRIB3 negatively regulates in a self-repressing feedback loop . Similarly, the UPR has been shown to increase TRIB3 expression  and cartilage pathology can be induced by ER stress  by triggering the UPR . This causes chondrocyte apoptosis via C/EBP homologous protein (CHOP), which serves to constitutively resolve the UPR and CHOP expression increases with OA progression . Autophagy protects against mitochondrial dysfunction, which is elevated in OA cartilage due to reduced SOD2 levels . In normal cartilage, autophagy is constitutively active, but becomes compromised with aging and precedes chondrocyte death and cartilage damage . UPR can be induced by biomechanical injury and IL-1 in chondrocytes  and TRIB3 may be implicated further, since CHOP facilitates the catabolic and apoptotic responses to IL-1 .
Conversely, it has been suggested that TRIB3 is associated with cell survival under prolonged hypoxic stress, in a hypoxia-inducible factor 1α-dependent manner . Since hypoxia is a situation articular cartilage is well adapted to , this could provide an alternative explanation for increased TRIB3 in OA cartilage  as a consequence of severe hypoxia in diseased tissue. TRIB3 has also been shown to be both up- and down-regulated by the steroid dexamethasone , which is used to treat many inflammatory and autoimmune conditions including RA. Indeed, the ability of dexamethasone to modulate TRIB3 expression may underpin its effectiveness, as it is known that TNFα positively regulates cortisol metabolism in inflammatory arthritis .
The roles of TRIB proteins in both cancer and immune regulation are of significant relevance when we consider the potential implications of these regulatory pseudokinases in arthritis. Firstly, many cancers produce MMPs as a means of growth expansion, enabling tumours to invade new tissue by remodelling the tissue matrix. Many of the MMPs expressed are the same enzymes that are ultimately up-regulated in the arthritides to drive cartilage destruction. Furthermore, the inflamed synovium or pannus has been likened to a cancer due to its hyperplastic appearance and the propensity for pannus tissue to ‘invade’ cartilage. Indeed, some therapies originally developed as anti-cancer agents are now being evaluated for RA. Secondly, inflammation clearly drives RA progression and increasing evidence suggests that low level inflammation is at least a contributory factor in OA. Therefore, TRIB-mediated dysregulation of immune responses may play a significant role in these diseases.
TRIB1 appears to have opposing effects on various cell-types, making any therapy that seeks to modulate TRIB1 levels potentially problematic. Overall, TRIB1 appears to be ‘anti-inflammatory’, regulating the levels of TNFα and key immune cells that serve to help maintain appropriate immune surveillance.
TRIB2 also appears to be important for immunity, with alterations in TRIB2 levels leading to the expression of a variety of immune modulators (both pro- and anti-inflammatory).
TRIB3 is a negative regulator of cytokine/chemokine production and appears to promote a ‘pro-inflammatory’ phenotype. Moreover, its role in processes reported to be important in OA pathogenesis such as ER stress, the UPR, autophagy and obesity all point towards TRIB3 being a potential player in the signal transduction pathways that underpin OA pathology including MMP expression.
Thus, there are many potential links between these intriguing pseudokinases and both RA and OA (summarized in Figure 1) which undoubtedly have significant roles to play in arthritis pathology. However, further research is required before we can definitively identify the pathways and cell types in which they exert their influence.
TRIB and how they may influence the inflammatory/catabolic index in arthritis
We thank Arthritis Research UK (grant 19485) and the Oliver Bird Rheumatism Programme for their generous funding.
β-transducin repeat-containing E3 ubiquitin protein ligase
C/EBP homologous protein
constitutively photomorphogenic 1
forkhead box O3
Forkhead box P3
nuclear factor kappa B
retinoic acid receptor
retinoid X receptor
small ubiquitin regulatory factor
transcription factor 4
transforming growth factor β
tumour necrosis factor α
unfolded protein response
Tribbles Pseudokinases on the Crossroads of Metabolism, Cancer, Immunity and Development: Held at The Aquincum Hotel, Budapest, 22–24 April 2015