Introduction
Quinolinic acid (QUIN) is a downstream metabolite produced through the kynurenine pathway (KP) of tryptophan metabolism [
1],[
2]. In physiological conditions, QUIN is present in nanomolar concentrations and used as substrate by cells to synthesize the essential co-factor nicotinamide adenine dinucleotide (NAD
+). However, during neuroinflammation, the KP can be chronically or acutely activated through the induction of one of its initial enzymes, indoleamine 2,3-dioxygenase (IDO-1). QUIN is then produced in excess and can kill brain cells including neurons, astrocytes and oligodendrocytes by at least six different mechanisms [
2],[
3]. QUIN has been implicated in several neurological diseases, including multiple sclerosis (MS) [
4],[
5]. The involvement of the KP in neurological diseases is complex, as it revolves around the metabolic balance between excessive production of neurotoxic metabolites, such as QUIN, and neuroprotective compounds, such as kynurenic acid (KYNA) [
6],[
7].
At the cellular level, QUIN is cytotoxic for most of the brain cell types
in vitro at micromolar concentrations. Cammer [
8],[
9] showed that exposure to 1 mM of QUIN induces cell death in rat oligodendrocytes [
8],[
9]. Similar toxic effects are also observed in primary human astrocytes and neurons at pathophysiological concentrations of 150 nM [
10], and more recently in motor neurons at concentrations of 100 nM [
11]. Furthermore, this effect can be abolished by using antagonists of the N-methyl-D-aspartate (NMDA) receptor - such as memantine, MK801 and AP-V - implying excitotoxicity as the main mechanism inducing cell death [
10],[
11]. Current evidence suggests only monocytic lineage cells have the ability to produce QUIN [
12],[
13]. Brain cell types, including neurons, astrocytes, pericytes and endothelial cells are likely to uptake QUIN and catabolize it [
14]-[
17].
The function of the KP in oligodendrocytes remains to be investigated, although an earlier study demonstrated that IDO-1 and tryptophan 2,3-dioxygenase (TDO-2) are not expressed in human primary oligodendrocytes [
5]. This potentially has strong implications for MS pathology. The lack of these two KP regulatory enzymes in oligodendrocytes is associated with a higher cell susceptibility to allogenic T-cell challenge, since IDO-1 plays a crucial role in immune regulation - particularly in suppressing T cell proliferation [
18]. The KP profile has been shown to be altered in both MS patients and in experimental autoimmune encephalitis (EAE) mouse models [
19]-[
21]. Rejdak
et al. [
19] reported that the concentration of the neuroprotective KYNA in the cerebrospinal fluid (CSF) of MS patients is decreased during the remission stage. This indicates that the KP stays inclined towards neurodegeneration long after inflammation. We hypothesise that in MS the sustained activation of the KP in the CNS leads to an unbalanced production favoring neurotoxic over neuroprotective intermediates.
This study investigates the profile of the KP in oligodendrocytes and characterizes to what extent KP metabolism changes during immune challenge. We have also identified novel approaches to protect oligodendrocytes against pathophysiological concentrations of QUIN. The outcomes of this study could have significant implications for potential therapeutic strategies in MS and other neurodegenerative diseases that involve KP dysregulation with overproduction of QUIN.
Discussion
The KP profile has been extensively characterized in various human cell types including neurons, astrocytes, microglia, endothelial cells and several other immune cells [
14]-[
17]. However, the literature regarding the KP in oligodendrocytes is very limited [
5]. More importantly the biological function of the KP in oligodendrocytes remains to be delineated. In this study, we showed that murine oligodendrocytic cell lines express all the KP enzymes tested and have a fully functional KP unlike astrocytes, which lack the mid-stream enzyme KMO that converts kynurenine to 3-hydroxykynurenine [
15]. However, these data conflict with our previous findings showing that human primary oligodendrocytes lack expression of both IDO-1 and TDO-2 [
5]. These discrepancies are likely explained by the fact that KP expression: 1) varies between different species of the same cell type [
29]-[
31] and; 2) is significantly different between primary cells and their respective cell lines [
5]. It is also possible that the conditional transfection of the oligodendrocytes with simian virus 40 T antigens may alter the KP expression [
17]. Furthermore, in this study we showed that KP metabolism is different between two cell lines of the same cell type arising from the same species. N20.1 cells display a more active KP mRNA enzyme expression when compared to N19 (Figure
1). The degree of gene up-regulation by IFN-γ was also different: most of the KP enzymes were induced by IFN-γ in N20.1 cells but only limited to TDO-2, KAT-2 and KYNU in N19 cells. The increase of K/T ratio as a result of TDO-2 up-regulation in these 2 mice oligodendrocytic cell lines was also reported by our group in the human oligodendrocytic cell line, MO3.13 [
5]. It was also demonstrated in a recent study that TDO-2 is highly up-regulated in immortalized glial cell lines [
32].
The expression of KATs and secretion of KYNA by both cell lines suggests that the oligodendrocytes are likely to have neuroprotective functions, especially during neuroinflammatory events. This is reflected by higher expression of KAT-II in the presence of IFN-γ (see Additional file
1). This concurs with current literature pertaining to the mice model where KAT-2 is demonstrated to be the main enzyme associated with KYNA production in the brain [
33]. Expression of QPRT corresponding with catabolic rather than anabolic activity of QUIN further supports the neuroprotective role of oligodendrocytes. In addition, this suggests that oligodendrocytes are capable of taking up exogenous QUIN present in the extracellular matrix. They could, therefore, be involved in the detoxification of pathophysiological concentrations of exogenous QUIN, secreted by activated immune cells. Using immunocytostaining, we further showed a rapid uptake of QUIN by oligodendrocytes occurring within 30 minutes then gradually increasing with time (Figure
2). These results are in agreement with the KP metabolic profile found in human primary oligodendrocytes showing an uptake of exogenous QUIN from media and production of neuroprotective KYNA and PIC [
5]. We previously showed that QUIN is cytotoxic to both primary human neurons and astrocytes at concentrations as low as 150 nM and 350 nM respectively [
10].
As described above, N20.1 cells display a more active KP compared to N19 cells. This suggests that N20.1 cells are likely to have a higher tolerance to QUIN toxicity, that is a higher capacity to catabolize QUIN. Indeed, this was reflected by the LD
50 concentration to QUIN toxicity in N20.1 cells with a difference 1-fold higher compared to N19 (Figure
3A). Surprisingly, this was not reflected in the PRE-QUIN treatment group where cells are pre-exposed to QUIN followed by treatment with anti-QUIN mAb (Figure
3D and E). We expected N20.1 cells to have a higher capacity to catabolize QUIN. However, the cell death assay revealed otherwise. It is important to note that the QUIN concentrations based on LD
50 used in both cell lines are different. It may imply that QUIN at approximately 1 μM concentration range may exert a more permanent and lasting excitotoxic effect. This is regardless of the cells capacity to take up and catabolize QUIN. An important aspect to further dissect is the short-term and long-term effects of QUIN toxicity on oligodendrocytes. This may have important implications for neurodegeneration and repair in situations where the KP is activated for sustained and prolonged periods of time, such as progressive MS. We have previously shown that low micro molar concentrations of QUIN could exert excitotoxicity chronically in human neurons over a period of 5 weeks [
34]. Here, our data support that from a previous study by Cammer
et al. [
8] showing acute QUIN toxicity in primary rat oligodendrocytes at micro molar concentrations.
It is interesting to speculate that differences in susceptibility to QUIN-mediated toxicity observed between the two cell lines may also reflect their different maturation states and the associated expression of NMDA receptors. The relatively immature N19 cells, with a lower expression of NMDA receptors, experience reduced excitotoxicity in response to QUIN. Furthermore, it is also possible that there are differences in expression of QUIN transporter molecules and, therefore, varying levels of QUIN uptake possible between cell lines. The characterization of NMDA receptors and the capacity of the QUIN uptake transporter certainly warrant further investigation in oligodendrocyte development.
It is apparent that therapeutic strategies targeting QUIN could prevent excitotoxicity and ultimately significantly attenuate neurodegeneration. In this study, we demonstrated that neutralizing QUIN toxicity using an anti-QUIN mAb could prevent oligodendrocyte cell death
in vitro. There is still a limited number of studies targeting QUIN toxicity with a neutralizing antibody, despite extensive evidence showing QUIN as a key excitotoxin involved in several neuropathological diseases. This study is the first to propose the use of a mAb to neutralize
in vitro QUIN toxicity. Based on our findings, the application of neutralizing QUIN with antibodies could be further developed into monoclonal antibody therapy for neurodegenerative diseases such as MS or amyotrophic lateral sclerosis [
35]. Monoclonal antibody therapy has gained popularity over the years including the recently FDA approved natalizumab used in MS treatment [
36],[
37].
As described above, chronic inflammatory responses can easily trigger cumulative production of pathophysiological concentrations of QUIN by activated monocytic cells such as infiltrating macrophages and microglia. This likely creates an environment in which neuronal cells are highly susceptible to excitotoxicity. Thus, we investigated further how manipulation of the KP during inflammation could affect the survival of oligodendrocytes towards QUIN-induced toxicity. We attempted to limit the production of QUIN from the BV2 microglial cell line by inhibiting IDO-1, since this enzyme is highly inducible by IFN-γ and will be of pathophysiological relevance during inflammation. Our data show that the use of IDO-1 inhibitors (1-MT and berberine) were able to significantly decrease QUIN production by BV-2 cells and subsequently lead to complete abolishment of oligodendrocyte cell death at a concentration of 4 μM (Figure
5). These results represent another relevant therapeutic strategy for MS using KP inhibitors. An earlier study using Ro 61-8048, a KMO inhibitor, on EAE rats showed a significant alleviation of disease progression [
21]. In this same study the authors demonstrated that the advantage of KMO inhibition, in comparison with IDO-1, is that they could obtain both a decrease of QUIN synthesis and an increased production of the neuroprotective KYNA in the brain and spinal cord [
21].
The use of IDO-1 inhibitors as a treatment for EAE and MS requires further investigation. The positive outcomes of this study are based on IDO-1 inhibition specifically in activated microglial cells. However, there is no IDO-1 inhibitor able to selectively target cell types. For example, inhibiting IDO-1 in astrocytes, the main producers of KYNA [
15] would likely be deleterious. Furthermore, some studies have demonstrated that systemic inhibition of IDO-1 in EAE exacerbates the disease [
38],[
39]. It is likely that the timing of intervention is important in this context: normal KP activation certainly suppresses aberrant immune responses and inhibiting this effect is unwanted in autoimmune disease. However, excessive KP activation produces neurotoxic metabolites, so preventing this dysregulated activation of the KP is clearly neuroprotective. Some KP metabolites have critical roles in the regulation of the T-cell activity responsible for autoimmune disorders [
39]-[
41]. This further supports the notion that the homeostatic activation of IDO-1 in EAE does have beneficial outcomes. However, its continued inhibition could also have significant negative effects in progressive disease.
It is technically challenging to deliver IDO-1 inhibitors specifically to activated monocytic cells. However, cell-specific targeting might be possible using nanoparticles [
42]. Another possibility will be to deliver multiple targets of KP modulators that can limit QUIN production whilst increasing KYNA and PIC. Platten
et al. [
43] demonstrated that tranilast (Rizaben), a synthetic analog of the KP intermediate 3-HAA, was able to fully abolish EAE by skewing the immunological profile from pro-inflammatory to immunosuppressive [
43]. Combination therapy targeting KP enzymes and/or metabolites appears to be a promising therapeutic for EAE and MS but requires further investigation.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.