Microglia and Chitotriosidase in Multiple Sclerosis

Initial biochemical investigations led to the unexpected discovery that plasma samples of patients with Gaucher disease (GD), a lysosomal storage disease, had a several hundred-fold elevated ability to hydrolyse chitin, a polymer of beta 1,4-linked N-acetylglucosamine that is largely diffuse in nature.¹ Later, the same investigators observed that lipid-laden activated macrophages accumulating in GD tissues were able to secrete exceedingly high levels of an enzyme able to cleave chitin and artificial chitotrioside substrates, therefore named chitotriosidase (Chit).^2,3 The measurement of plasma Chit activity has now also found an application in the clinical monitoring of patients with Fabry disease.⁴
Chit is a member of the mammalian chitinase family. It is mainly synthesised by activated macrophages and also by immature neutrophils⁵ as a 50kDa protein, proteolitically cleaved and predominantly secreted. Its natural substrate, chitin, is known for being an integral part of an arthropod’s exoskeleton.^6–9 Its content is variable within species and is reported to account for up to 20% of the fungal cell wall.^10,11 So far, no definitive mammalian chitin synthetase has been documented, although a pathogenic role of this oligosaccharide in vertebrates has been reported. Chitin is utilised in the enhancement of the inflammatory roles of macrophages and neutrophils^11,12 and, in such instances, may also modulate the cerebral microglial activation documented in neuroinflammatory and neurodegenerative diseases.^13,14
An important question arises when considering that chitin is absent in humans. However, the fact that the Chit gene is present and conserved in rodents and primates¹⁵ argues in favour of a biological role of this enzyme.

Chit, Chitin and the Central Nervous System

In pathological conditions Chit has achieved an increasingly important role as a marker of some neurological diseases,¹⁶ perhaps by virtue of the peculiar immune condition that characterises the brain. Nonetheless, physiological conditions such as ageing have also been found to correlate with a progressive increase of plasmatic Chit activity.^17,18 This might suggest that the innate immune system is involved in protecting the healthy human organism from cell damage that occurs during ageing, perhaps linked to oxidative processes.¹⁹
Given its peculiar immune ‘privilege’, healthy central nervous system (CNS) status is tightly regulated and maintained at a low functional level in order to prevent immune-mediated damage from occurring. Among brain cells, microglia and a few other cell types such as astrocytes represent components of the innate immune system that promptly activate a complex immune cascade soon after specific or unspecific stimuli – either ischaemic–oxidative or inflammatory – face the brain. Microglia are the resident macrophages in the CNS. They exert important functions such as phagocytosis of cellular debris and/or neuronal signal processing when activated through communications with neurons, immune cells and other glial cells.^20–22 Activation of microglia occurs in most pathological processes. This activation is accompanied by changes in morphology, upregulation of immune surface antigens and production of cytotoxic or neurotrophic molecules.^20,23,24
Despite the fact that chitin is absent in humans, Chit has been found to be elevated in some brain diseases, as described below. However, chitin-like substances have recently been found to accumulate within the brain in certain circumstances. Several lines of evidence exist that implicate impaired glucose metabolism in several CNS diseases.^25–27 It is known that brain hypoperfusion and several inflammatory conditions lead to increased glucose metabolism, which in turn strongly increases hexosamine pathway activity in endothelial cells, leading to synthesis of glucosamine polymers.^28,29 Glucosamine has various physiological properties: it inhibits pro-inflammatory cytokines from antigenpresenting cells (APCs), suppresses T-cell response by interfering with functions of APC and shows a direct inhibitory effect on CD3-driven T-cell proliferation.^30,31 Glucosamine administration in the animal model of multiple sclerosis (MS), known as experimental autoimmune encephalomyelitis (EAE), significantly reduces macrophage infiltration within the inflamed brain and reduces microglial activation and production of nitric oxide (NO) and inflammatory cytokines, such as interferon (IFN)-γ, interleukin (IL)-17 and tumour necrosis factor (TNF)-α, resulting in resistance to acute EAE.³¹ However, glucosamine polymers are also the building blocks of chitin and chitosaccharides.^28,32–34 Thus, the markedly augmented glucose metabolism that occurs during inflammation or hypoperfusion within the brain could induce increased formation of chitin-like substances.^33,34

Chit in Stroke

It is now very well established that after an acute brain ischaemia, early activation of microglia and endothelial cells and their transcription of TNF-α and IL-6 are able to induce a cascade of inflammatory pathways that transforms local endothelia into a pro-thrombotic state and allows peripheral mono- and poly-morphonuclear cytotoxic chemoattraction into the lesion site.^35,36 This early event correlates with the worsening of the cerebral damage, as a clear relationship between the extent of brain damage and the early increase of TNF-α plasma level is generally reported.³⁷ Besides the increase of innate immune cytokines, we and others^38,39 have also found increased Chit activity in stroke, suggesting a relevant role for infiltrated macrophages, or resident microglia activated by the ischaemic event, inside the brain. Our results also suggest that Chit, TNF-α and other pro-inflammatory cytokines are markers of microglia activation occurring during a stroke, which is independent of pre-existing inflammatory or infectious conditions of the patients.³⁸

Chit in Alzheimer’s Disease and Ageing

Amyloid plaques, senile plaques and amyloid angiopathy of Alzheimer’s disease (AD) brains have been described to co-localise with chitin-like glucosamine polymers. Such polymers are suggested to induce the formation of pathogenic amyloid fibrils.^33,34 Based on this original finding, we first conducted an association study of Chit in the plasma of people with AD,¹⁸ analysing Chit activity in AD patients and healthy individuals. Results were unexpectedly interesting and indicated that Chit values in the AD group were significantly higher than those obtained from individuals of comparable age. Moreover, plasma Chit activity level correlated with the individual’s age in the whole control group, confirming an age-related physiological Chit increase (see Figure 1).
It is thought that β-amyloid brain deposition is involved in the AD neurodegenerative cascade, perhaps through direct neurotoxicity or an immune network.^33,34,40,41 In this light, the high Chit expression – as demonstrated also by Di Rosa and collaborators⁴² – may have a dual role: it either represents a mere epiphenomenon of a strong macrophage– microglia activation due to β-amyloid deposition, or reflects a scavenging protective immune activity against potentially pathogenic chitin-like glucosamine polymers.³³ Thus, encouraged by these findings, we performed an immunocytochemistry study in order to explore the presence of chitin-like substances within AD, MS and healthy control (HC) brains. On the one hand, we could confirm the presence of abundant chitin-like substances in the brains of AD patients, which may well relate to the detection of peripheral Chit activity. Conversely, we failed to demonstrate the deposition of chitin-like substances in both MS and normal brains.⁴³

Chit in Multiple Sclerosis

MS is a chronic inflammatory/degenerative disease of the CNS. Most studies in humans, principally based on the EAE animal model, claim that an adaptive T- and B-cell antigen-specific response is pathophysiologically characteristic of this disease.⁴⁴ However, other studies indicate that the innate immune response predominates in most MS lesion subtypes; this response is initiated by both resident (microglia) and peripheral infiltrating macrophages.^45,46 TNF-α, IL-6, NO, reactive oxygen species (ROS) and other macrophage-derived products unfortunately show only a modest correlation with clinical activity in MS, and are as yet of very limited usefulness in clinical routine.⁴⁷ Through a case-control study, we demonstrated that Chit activity is increased in the blood and, particularly, in the cerebrospinal fluid (CSF) of MS patients,⁴⁸ and that this activity correlates well with the extent of CNS damage as scored by the Extended Disability Status Scale (EDSS) at follow-up⁴⁹ (see Figures 2 and 3).
Considering that with the normal immunoglobulin G (IgG) index found in these patients the Chit activity was of intrathecal rather than peripheral origin, we also calculated what we call the ‘Chit ratio’ (CSF Chit/ plasma Chit) and compared the EDSS score of the individual patients, obtained after a mean follow-up of five years (range two to eight years), with both oligoclonal bands (OCB) number and Chit ratio at the time of CSF withdrawal. While there was no correlation between EDSS and OCB, there was a significant correlation between EDSS and the Chit ratio (p<0.001). Higher disease severity (EDSS ≥5.5 at follow-up) was restricted to those patients with initial higher intrathecal Chit level and Chit ratio (see Table 1).
This study⁴⁸ not only confirms the important role of the innate immune response in MS phenomenology, but also allows Chit determination in CSF and plasma to be proposed as a diagnostic and monitoring method in an MS laboratory. As a result of our analysis, plasma and, to a larger extent, CSF Chit levels better correlate with the extent of CNS damage compared with the previously proposed macrophage-derived markers such as TNF-α, IL-6, IL-1, NO and ROS. As Chit CSF production is unrelated to plasma level and albumin quotient, we argued that Chit activity is compartmentalised in MS.
Despite the fact that this enzymatic activity likely is derived from infiltrating macrophages, we thought that resident microglia might be induced to produce Chit as these cells can gradually transform phenotypically into lipid-laden activated macrophages.⁵⁰ This intrathecal, possibly microglia-derived Chit activity found in MS patients could counterbalance the naturally occurring glucosamine aggregation and its transformation into possibly detrimental chitin-like polymers.

Microglial Production of Chit

For this reason we have performed a new study with the aim of documenting a microglia-derived Chit production (Sotgiu et al., submitted). To this purpose, murine microglial cell line N9 was plated in 96-well tissue-culture plates with Royal Park Memorial Institute (RPMI) supplemented with 10% foetal bovine serum at 4×10⁴ per 100μl per well and incubated in the presence or absence of lipopolysaccharide (LPS) 100ng/ml and TNF-α 100U/ml, or in combination. Different conditions were performed in quadruplicate.
Chit was determined according to published methods^38,48 in the supernatant of microglial cell cultures: briefly, after 24 hours of culture, cell-free supernatants were collected; 30μl were incubated with 22μlmol/l 4-methylumelliferyl-β-d-N,N’,N’’triacetylchitotriose in 0.5M citrate-phosphate buffer at pH 5.2, and fluorescence was measured on a fluorimeter. Chit activity was calculated as nanomoles of substrate hydrolysed per ml per hour (nmol/ml/h). In Figure 4, Chit production from microglial cells is shown; this production is susceptible to modification by changing experimental conditions.

Conclusions

Until recently, Chit was known to be produced only in lysosomal storage diseases.¹ Thanks to the contribution of other groups, including ours, this enzyme has progressively acquired new associations with other previously unrelated diseases, particularly neurodegenerative and neuroinflammatory diseases (reviewed in reference 16). In such circumstances, and by analogy to storage diseases, it has long been thought that the increase of Chit activity in the plasma of neurological patients, as well as in normal ageing, likely depends on macrophage activation at a peripheral level, perhaps linked to the known scavenging role of macrophages.¹⁹ For the first time, we described that such enzymatic activity is also present in the CSF of patients with MS and that it is compartmentalised within the CNS, and therefore is of intrathecal origin.⁴⁸
The question remained open for a couple of years as to whether such Chit production was of macrophage or microglial origin. Macrophages are known to rapidly and diffusely infiltrate the MS lesion sites⁴⁷ through the blood–brain barrier and are also known to produce in loco a huge amount of inflammatory products.^45,46 However, microglia within the MS lesion are also known to give rise to continuous transformation into macrophages under facilitating circumstances such as digestion of myelin products.⁵⁰ Thus, the question as to whether intrathecal Chit production is autochthon (i.e. microglial-derived) or exogenous (from infiltrating macrophages) has remained unsolved for some time.
We now document that murine microglia can produce Chit under certain favourable in vitro circumstances, i.e. pro-inflammatory stimuli, and suggest that intrathecal Chit, recently found in MS patients,⁴⁸ is of microglial origin. It is unknown whether Chit activity has direct influences on the CNS or whether it simply reflects a relic of an archaic macrophage response to chitin-containing pathogens.^5,16,48 Thus, another important issue is whether Chit is determinant in the pathogenesis of MS. Perhaps an even more important issue involves the cerebral accumulation of chitin or chitin-like polymers and the role, if any, that they play. New histopathological studies on the AD brain^33,34,43 allow us to argue in favour of a protective role of Chit in the CNS by avoiding brain accumulation of chitin-like oligosaccharides.
Several contradictory reports in the literature document that β-amyloid plaques may be either protective or pathogenic elements in AD.^{13,14,27,51,57} If β-amyloid has a pathogenic and not only a pathognomonic role in AD, the role of chitin may be that of favouring and accelerating its detrimental deposition.³³ Recently, Castellani and colleagues³⁴ suggested that chitin-like polysaccharides within the AD brain could provide a scaffolding for β-amyloid deposition and that glucosamine may therefore facilitate the process of amyloidosis. This finding may support the hypothesis that an imbalance between glucosamine polymerisation and degradation in AD favours fibril formation. Thus, the different histochemical content of chitin-like substances recently found between MS and AD brain samples may be a reflection of different pathophysiological mechanisms.^33,34,43 The markedly augmented glucose metabolism occurring during inflammation in the MS brain could induce a spontaneous increase of glucosamine-derived chitin-like substances. In this context, the microglia-derived Chit activity⁴⁸ could counterbalance the naturally occurring glucosamine aggregation as well as its transformation into detrimental chitin, thus having a protective rather than a detrimental role within the CNS. In contrast, deposition of chitin-like substances in the AD brain could result from a reduced immune response that characterises the most severe clinical expression of disease,⁵⁸ or from an impaired chitin cleavage in relation to the advanced age of AD patients.
In conclusion, the role of Chit in MS is intriguing for its possible protective function. Much remains to be done in this area, but, even if it is still unclear whether Chit is a protagonist or a bystander, it could well represent a simple and sensible marker for monitoring MS, and as such deserves further attention. ■

Acknowledgements

We wish to warmly thank Mr Giuseppe Rapicavoli, Department of Pediatrics, University of Catania, for his skilful technical assistance in laboratory work.
Thanks to the patients who generously contributed to the studies reviewed and all collaborating colleagues in Sassari (Giannina Arru, Maria Laura Fois) and Catania (Rita Barone). Murine microglial cells were kindly provided by V Dalla Bianca, University of Verona.