Toll-like receptors hit calcium

Abstract

Mitochondrial Ca²⁺ uptake is a multifarious signal that controls both the activity of matrix dehydrogenases and the sensitivity to apoptotic and necrotic challenges. Recent evidence indicates that mitochondria also play a role in triggering inflammation, as mitochondrial DNA, when released by the cell, is an important damage-associated molecular pattern (DAMP). Now, Toll-like receptors (TLRs) are shown to close the loop, by affecting in turn mitochondrial activity. Two studies by Shintani and colleagues, one in this issue of EMBO reports ^{1, 2}, identify a new TLR transduction mechanism that impinges directly on mitochondrial function. Upon binding of CpG oligodeoxynucleotides, TLR9—which in non-immune cells is retained in the ER—inhibits SERCA2, thus reducing Ca²⁺ transfer to the mitochondria and aerobic metabolism.

A few years ago, putting together inflammation and mitochondrial Ca²⁺ handling would have been quite a bizarre idea. The complex signaling network downstream of cytokine receptors was accepted to lead to the nucleus and minimally affect mitochondria. In addition, during inflammatory responses, more attention was placed on oxygen consumption by NADPH oxidases than on the housekeeping machinery of aerobic respiration. Then, some surprising, novel information gradually set the stage for the heterodox association.

Indeed, Ca²⁺ accumulation by energized mitochondria, an old notion of bioenergetics, has entered a glittering phase. Numerous examples now highlight the notion that cellular Ca²⁺ signals, evoked by a variety of physiological or pathological challenges, are decoded within mitochondria into effects as diverse as increased ATP production, release of apoptotic cofactors or bioenergetic collapse in necrosis. Moreover, altered mitochondrial Ca²⁺ handling plays a role in the pathogenesis of a variety of human diseases, ranging from neurodegenerative and metabolic disorders to cancer ³. Then, mitochondria directly stepped into the mechanisms of inflammation, as they were shown not only to be a target of toxic and/or immune damage, but also to directly promote the initiation and/or potentiation of inflammatory reactions by triggering TLR signaling. TLRs are a family of receptors, initially identified in immune cells, that includes 10 and 12 paralogues in humans and mice, respectively. Upon binding of specific ligands of bacterial, viral or fungal source (pathogen-associated molecular patterns, PAMPs), a signaling cascade is activated that culminates in the transcription of genes for inflammatory mediators, such as TNF-α and IL-6. In addition to microbial PAMPs, TLRs can also sense endogenous molecules released from infected or stressed cells (DAMPs). These ligands include nuclear structural components (such as HMG-B1), heat-shock proteins (HSP60 and HSP70) and also components of mitochondria (such as mtDNA) ⁴. The latter is released extracellularly upon tissue damage and is rich in unmethylated CpGs.

Finally, the paradigm that TLRs are invariably associated with pro-inflammatory effects has been recently amended by the evidence that small doses of PAMPs may result in an attenuated inflammatory response to subsequent larger doses of PAMPs or to injury. This phenomenon is thought to be due to the transcription of genes coding for inhibitors of the TLR-NFκB signaling pathway ⁵. Moreover, evidence that TLRs are not exclusively expressed in immune cells but also in several other types of cells, including neurons and cardiomyocytes ⁵, suggested that this anti-inflammatory mechanism might operate directly on the potential targets of the inflammatory damage. Among the TLR ligands able to trigger an anti-inflammatory response, unmethylated CpG-oligodeoxynucleotide (CpG-ODN) ligands of TLR-9 were shown to be very potent. Indeed, their administration, which is well tolerated clinically, attenuates the acute inflammatory cardiac dysfunction induced by both LPS and ischemia–reperfusion, by inhibiting the NFκB pathway in ventricular myocytes ⁶.

“…altered mitochondrial Ca²⁺ handling plays a role in the pathogenesis of a variety of human diseases, ranging from neurodegenerative and metabolic disorders to cancer”

Shintani and colleagues identify an alternative TLR9 signaling pathway that, in addition to the canonical TLR-NFκB axis, accounts for the activation of an anti-inflammatory mechanism within the parenchymal cells of an inflamed tissue ^{1, 2}. The alternative route stems from a different intracellular sorting of TLR9 in immune and non-immune cells. In immune cells, the chaperone Unc93b1 shuttles TLR9 from the ER to the endo/lysosomal compartment, where processing of the receptor and binding to CpG-ODN initiates the canonical MyD88-dependent pro-inflammatory signaling pathway ⁷. In neurons or cardiomyocytes, which are at high risk of permanent damage by inflammation due to their poor regenerative capacity, Unc93b1 is expressed at low levels ⁸, and TLR9 is mainly retained in the ER ¹. There, the engagement by CpG-ODN triggers a different, hitherto unknown signaling route ². Through biochemical studies, Shintani and colleagues identify SERCA2 (isoform 2 of the sarco-endoplasmic reticulum Ca²⁺ ATPase) as a protein directly interacting with TLR9. They show that in cardiomyocytes (but not in cardiac fibroblasts), upon interaction with CpG-ODN, TLR9 binds the Ca²⁺ pump, reducing its activity and lowering [Ca²⁺] in the ER lumen. As to the downstream consequences, the authors appropriately draw their attention to the emerging link between mitochondrial [Ca²⁺] and pro-survival mechanisms, such as autophagy. Genetic ablation of the inositol 1,4,5 trisphosphate receptor (IP₃R), which is the Ca²⁺ release channel of the ER, was previously shown to dramatically increase autophagic rates by impairing Ca²⁺ transfer to mitochondria and Ca²⁺-dependent stimulation of aerobic metabolism. Hence, ATP levels were decreased and AMPK signaling to autophagy stimulated ⁹. Accordingly, Shintani et al ^{1, 2} also observe decreased mitochondrial [Ca²⁺] levels and ATP production, as well as increased AMPK signaling, in DAMP-challenged cells (Fig 1).

Figure 1. Effect of TLR9 on mitochondrial Ca²⁺ signaling in non-immune cells.

(A) In the absence of TLR9 engagement, Ca²⁺ transfer from the ER to mitochondria stimulates the Ca²⁺-sensitive matrix dehydrogenases (DH) of the Krebs cycle, thus increasing electron feeding to the respiratory chain and ATP production. The increase in the ATP/AMP ratio decreases AMPK activity and suppresses autophagy. In addition, matrix [Ca²⁺] favors the opening of the permeability transition pore (PTP), thus triggering the mitochondrial morphological and functional alterations that occur in apoptosis and necrosis. (B) Upon engagement by CpG-ODN, TLR9 binds to SERCA2 and inhibits its activity, thus reducing both DH stimulation (and hence autophagy suppression) and sensitivity to apoptotic/necrotic challenges.

Although the authors did not directly address this issue, decreased mitochondrial [Ca²⁺] loading is expected to correlate with lower sensitivity to apoptotic and necrotic challenges, thereby increasing cell resistance in the inflamed area. Indeed, numerous examples of pathology-related changes of mitochondrial Ca²⁺ homeostasis are available and provide a coherent picture ². Oncogenes reduce ER Ca²⁺ levels and cancer-related miRNAs reduce the expression of the mitochondrial Ca²⁺ uniporter, thus reducing sensitivity to apoptosis, whereas tumor suppressor genes and viral proteins have the opposite effect. Thus, the observation that TLR9 protects parenchymal cells from death, operating through a different component of the Ca²⁺ signaling machinery, is an important addition to a well-established conceptual framework.

The involvement of mitochondrial function as target of TLR activity, however, also opens the possibility that other processes—different from ATP production but still strictly dependent on mitochondrial bioenergetics—are involved in signal transduction. In particular, increased feeding of electrons to the respiratory chain due to the stimulation of Ca²⁺-dependent matrix dehydrogenases is expected to increase the production of ROS. Interestingly, the latter are known to contribute to the stabilization of HIF-1, a transcription factor that activates the expression of a number of genes involved in the adaptation of tissues to hypoxia ¹⁰. A reduction of mitochondrial Ca²⁺ loading should be paralleled by a decrease in ROS-dependent signaling. Thus, the cellular effects downstream of mitochondrial involvement could be very complex and include both rapid changes in sensitivity to cell death pathways and a global change of the proteomic profile and would be worth analyzing in detail.

Overall, the papers by Shintani and colleagues tie together the role of mitochondria in the initiation of inflammation and in the regulation of cell sensitivity to the inflammatory environment, by placing the focus on the Ca²⁺-mediated signaling liaison between the ER and the mitochondria. It is tempting to speculate that the recent explosive advance in the molecular understanding of mitochondrial Ca²⁺ transport ² will allow not only to rapidly expand these novel concepts, but also to develop new therapeutic approaches in the broad area of inflammatory diseases.

Conflict of interest

The authors declare that they have no conflict of interest.