Discussion
The present report showed that null mutations of the mouse Sharpin gene did not affect the steady-state distribution of splenic DC subsets nor the development and phenotype of BMDC. However, loss of SHARPIN significantly diminished the capacity of BMDC to secrete inflammatory cytokines and nitric oxide. The attenuated cytokine production was not due to the presence of anti-inflammatory inhibitors, and can be largely explained by selective inactivation of NF-κB signaling. Stimulated cpdm BMDC exhibited Th2-biased T cell-polarizing capabilities, consistent with the Th2 cytokine-dominant phenotype in cpdm mice. Together, these results indicate an indispensible role of SHARPIN in regulating DC immunological functions, disruption of which may contribute to the development of immune diseases.
Since WT and cpdm mice are both specific pathogen-free, the nature of the trigger of the severe inflammation in cpdm mice is not obvious. One such initiating factor could be endogenous apoptotic and/or necrotic cells that can release danger-associated molecular patterns (DAMPs) to launch and amplify an inflammatory response [32]. Such ‘sterile’ inflammation could be initiated and take place in all organs and tissues affected in cpdm mice, thus causing multi-organ inflammatory disorders. This hypothesis is supported by several recent studies. Fibroblasts of cpdm mice are highly sensitive to TNFα-induced cell death and the cpdm phenotype can be partially rescued by deletion of TNF [7], [8], suggesting that deficiency of SHARPIN compromises the anti-apoptotic mechanisms in cpdm mice resulting in cell death-induced inflammatory disease. Apoptosis of keratinocytes is a prominent feature of the skin lesions in cpdm mice [33] and this is mediated by caspase-dependent mitochondrial pathways [34]. These induced and/or intrinsic apoptotic cells can release various types of DAMPs that exert their pro-inflammatory properties by activating DC through pattern recognition receptors [32], [35], such as HMGB1 recognized by TLR2/4 [36], [37].
Consistent with the impaired Th1 immune response in cpdm mice [5], stimulated BMDC weakly polarized Th1 differentiation, but strongly supported the development of Th2 effector cells. Combined with the dramatic effect of IL12 treatment on the phenotype of these mice [5], this suggests that the Th2-biased systemic inflammation in cpdm mice is caused by reduced IL12P70 production from DC. The importance of IL12P70 in regulating Th1 and Th2 immune responses in mice was clearly demonstrated in IL12P35- and IL12P40-deficient mice [38], [39] and through clinical studies in human patients [40]–[42].
NF-κB, TBK/IRF3, and MAPK signaling are important pathways activated by LPS or poly I:C and disruption of either pathway may lead to decreased cytokine expression. NF-κB activation was selectively inhibited in LPS and poly I:C stimulated cpdm BMDC compared to wild type, while activation of TBK1/IRF3, ERK1/2, or p38 was not, indicating that disrupted NF-κB signaling in cpdm DC is responsible for the defective cytokine expression. These results seem to contradict a recent study that found increased levels of NF-κB activation and IL1 transcription in cpdm mice [43]. This difference can probably be attributed to the different cells and tissues used in these studies, and may point to cell- and tissue- type specific functions of SHARPIN. This is consistent with recent reports of tissue-specific effects of NF-κB signaling [44]. Ubiquitous activation of NF-κB by removing inhibitors such as A20 and ITCH through genetic manipulation results in widespread inflammation, consistent with the role of NF-κB in the production of pro-inflammatory mediators [45], [46]. However, selective inhibition of NF-κB activation in parenchymal cells of the skin, liver, and intestine results in chronic inflammation driven by NF-κB competent leukocytes [47]–[50]. This indicates that a balance between the pro-inflammatory role of NF-κB in leukocytes and the anti-inflammatory role in parenchymal cells is critical in the maintenance of tissue homeostasis. Experiments with mice with cell- and tissue-specific deletion of Sharpin, currently under way, will help to elucidate the role of SHARPIN in inflammation.
Despite defective NF-κB activation in the absence of Sharpin expression, there was no significant change in splenic DC populations or expression of co-stimulatory molecules on BMDC. Different NF-κB subunits involved in the canonical and non-canonical branches of the NF-κB signaling pathway have distinct functions to control specific aspects of DC development and function [51], [52]. The non-canonical pathway (p100 processing to produce p52) appears to be intact in SHARPIN-deficient cells [7], [9] suggesting that the NF-κB heterodimer p52/RELB is sufficient to maintain the normal regulation of DC homeostasis and maturation.
The molecular basis by which the Sharpin mutation causes reduced NF-κB activation in BMDC remains to be determined. LPS and poly I:C used here are well-defined ligands that specifically engage TLR4 and TLR3, respectively. The expression profile of surface TLR4 complexes is similar between WT and cpdm BMDC suggesting that defective NF-κB activation is not a result of differential TLR expression on target cells. LPS engages TLR4 to activate MYD88-dependent and TRIF-dependent pathways, whereas TLR3 stimulated by poly I:C only triggers TRIF-dependent signaling [53]. The defective NF-κB activation by both stimuli suggests that the Sharpin mutation interferes with the protein adaptors or kinases shared by both signaling pathways, such as RIP1 and TRAF6 [18]. Recent studies demonstrated that SHARPIN interacts with HOIP to form LUBAC that exerts its linear-ubiquitin-chain-ligase activity on NF-κB signaling players RIP1 and NEMO [7], an essential step for intact TNFα-stimulated NF-κB activation. The Sharpin null mutation disrupts the ubiquitylation process and abrogates the TNFα-induced NF-κB signaling pathway. Since TNFR and TLR partially share their downstream signaling cascades, a similar ubiquitin-mediated regulation may hold true for SHARPIN in LPS- and poly I:C-induced NF-κB activation.
In summary, the present study identified an indispensible role of SHARPIN in the production of pro-inflammatory mediators and TLR-induced NF-κB signaling. The impaired Th1-stimulating ability of Sharpin-deficient DC may account for the Th2-dominant inflammatory phenotype of cpdm mice. The balance between Th1 and Th2 differentiation is critical for immune homeostasis. A better understanding of how such balance is maintained will help design cytokine treatment for human diseases with Th2-biased cytokine secretion similar to the mouse cpdm, such as allergies and hypereosinophilic syndromes.