Introduction
Multiple sclerosis (MS) is an autoimmune, demyelinating disorder of the central nervous system (CNS), affecting approximately 2.5 million people worldwide [
1,
2]. The most common clinical form is relapsing-remitting MS, representing about 85% of the cases. After a period ranging from 10 to 20 years, approximately 60% of the patients enter a phase that is characterized by a steady decline of neurological functions (secondary progressive MS, SPMS). Primary progressive MS (PP-MS) affects a subset of patients (10–15%) that is characterized by continuous progression of the disease from its onset. Current treatments (i.e., disease-modifying drugs) reduce the number and severity of relapses in RR-MS, but are ineffective in halting disease progression. Hence, there is a strong need to develop new therapeutic strategies for MS.
Galectins are a conserved family of β-galactoside-binding lectins, implicated in a variety of biological processes, e.g., adhesion and motility, cell survival and proliferation, angiogenesis, and inflammation [
3]. Galectins contain either one or two carbohydrate recognition domains (CRDs) and can be found in both the cytosol and nucleus, as well as in the extracellular compartment, as they are secreted by a non-classical secretion process [
3]. A growing body of pieces of evidence suggests that galectins, including galectin-1 and galectin-3, are involved in the regulation of CNS homeostasis and neuroinflammation by controlling microglial activation, inhibiting neurodegeneration, and promoting neuroprotection [
4]. Galectin-1, galectin-3, galectin-4, galectin-8, and galectin-9 are expressed at high levels in the brain [
5]; however, most studies on CNS function and protection have been dedicated to galectin-1 and galectin-3 [
6].
There has been increasing interest in the role of galectin-3 (Lgals3) in neuroinflammation and neurodegeneration [
7]. In particular, it was shown that the deletion of the galectin-3 gene improves experimental allergic encephalomyelitis (EAE) in mice [
8] and was associated with a decreased Th17 and an increased regulatory T-cell response [
8]. In another model, Wallerian degeneration following the injury of the sciatic nerve in galectin-3-deficient mice was associated with a significant increase in the inflammatory cytokines, IL-1β and TNF-α, and with the upregulation of toll-like receptors (TLR)-2 and -4 [
9]. On the other hand,
Lgals3−/− mice, despite having a similar susceptibility to cuprizone-induced demyelination as wild-type mice, have an impaired efficiency of remyelination [
10]. However, in another study, during cuprizone-induced demyelination in
Lgals3−/− mice, oligodendrocyte precursor cell (OPC) maturation was not affected by the deficiency of galectin-3 [
11].
Interestingly, it has been reported the presence of auto-antibodies against galectin-3 in the sera from patients with secondary progressive multiple sclerosis (SPMS), which could be partly responsible for the progressive damage to the blood-brain barrier (BBB) associated with MS [
12]. Moreover, the inhibition of the expression of galectin-3 in the brain microvascular endothelial human cells (BMEC) triggered an increase in the expression of intracellular adhesion molecule 1 (ICAM-1), responsible for the extravasation of leukocytes to the CNS [
12]. Overall, the authors proposed that galectin-3 could drive a downregulation of ICAM-1 in BMEC, thus exerting a protective effect in MS [
12].
In the present study, we aimed to characterize the transcriptional levels of galectin-3, along with the expression levels of the other galectin gene family members, in the different cell populations that are involved in MS pathology.
Discussion
Galectin-3 is the unique representative of the chimera-type galectins, containing one CRD linked to the N-terminal domain, which allows its oligomerization and the formation of pentamers [
17]. Galectin-3 has a molecular weight of 31 kDa, and it was first identified in immune cells and subsequently found in a series of normal and cancer cells. The structure of galectin-3 consists of two structurally specific domains: the N-terminal domain and the C-terminal CRD. The N-terminal domain contains sites for phosphorylation involved in the regulation of secretion, while the C-terminal CRD is connected to a collagen-like sequence, assembled of proline, glycine, and tyrosine tandem repeats [
17]. There are several ligands for galectin-3, both in the intracellular and the extracellular space, such as Bcl-2, synexin, and β-catenin, inside the cells, and laminin, fibronectin, CD29, CD66, α1β1 integrin, and galectin-3-binding protein (Lgals3bp) in the extracellular space [
17].
In the present study, we have evaluated the expression levels of galectin-3 in cell populations that are involved in the pathogenesis of MS. We have also provided data on the modulation of the other members of the galectin gene family for comparison.
Our data show a significant increase in galectin-3 expression in encephalitogenic CD4+ T cells in a preclinical model of MS. Also, significantly higher levels of galectin-3 were found in microglial cells, astrocytes, and oligodendrocytes, both in MOG-induced EAE and in human MS-related white matter lesions.
Galectin-3 has been shown to be expressed in neuronal and glial cells [
18]. The nerve growth factor induces galectin-3 expression in mouse dorsal root ganglia neurons, thereafter promoting neurite outgrowth and neural cell adhesion [
19,
20]. Also, galectin-3 is expressed in astrocytes, microglia, macrophages, dendritic cells, and in activated T and B cells. Growing data are accumulating on the role of galectin-3 in neuroinflammation and neurodegeneration [
17]. Gal-3 has been associated with microglial activation in several neuropathological conditions, e.g., traumatic brain injury [
21], viral encephalitis [
22], ischemia [
23], and demyelination [
24]. Attenuated inflammatory responses were observed in
galectin-3 deficient mice suffering from EAE, associated with a marked reduction in the number of CNS-infiltrating leukocytes [
8]. Also, it was shown that the expression of
galectin-3 is upregulated before motor impairment and that its expression levels persisted in activated microglia throughout disease progression in a murine model of Huntington’s disease [
25]. Furthermore, serum galectin-3 is increased in Parkinson’s disease patients and correlates to the Hoehn-Yahr stage [
26].
Our data indicate that the dramatic increase in
galectin-3 expression in the spinal cord of EAE mice is associated with the concordant upregulation of
Lgals3 in all the different cell populations taking part in the MS evolution, including the extravasating T helper cells, microglia, oligodendrocytes, and astrocytes. To our knowledge, this is the first report showing the upregulated expression of
galectin-3 in astrocytes in a model of MS. It is known that loss of galectin-3 function inhibits gliogenesis, while galectin-3 overexpression increases it [
27]. Also, galectin-3 overexpression increases the percentage of striatal astrocytes generated by the subventricular zone (SVZ), but decreases the percentage of oligodendrocytes [
27]. Moreover, galectin-3 immunoreactivity was increased in the perinatal human SVZ and striatum after hypoxia/ischemia [
27]. Interestingly, we observed that
Lgals3 expression was not influenced
in vitro by the exposure of astrocytes to Th1- or Th17-conditioned medium (Supplementary Fig.
S2), advocating the involvement of other factors implicated in the crosstalk among the other cell populations within the CNS lesions.
We have to mention that caution should be used when interpreting data from a preclinical model of human disease. Rodent experimental autoimmune encephalomyelitis (EAE) is currently the most common animal model used for investigating MS etiopathogenesis and for the development and testing of novel therapies. Indeed, most of the FDA-approved MS disease-modifying treatments have previously shown efficacy in EAE, including interferon beta, glatiramer acetate, natalizumab, sphingosine 1-phosphate modulators, dimethyl fumarate, and B cell depletion therapies [
28]. However, despite being the model that most closely mimics human MS, EAE has some significant flaws. For instance, rodent EAE mostly affects the spinal cord and causes less brain lesions, while MS patients experience lesions in the brain and/or spinal cord. Moreover, the innate and adaptive immune systems of rodents differ from those of humans, which may limit the translatability of some discoveries [
28]. Furthermore, the development of EAE requires an external vaccination step, where adjuvants are frequently used to cause sensitization to myelin antigens. These adjuvants contain bacterial components that trigger the innate immune system through the engagement of pattern recognition receptors. In addition, unlike MS, which lacks a specific identifiable antigen, the inducing antigens in EAE are well identified, even though in some, but not most studies, myelin-derived peptides, including peptides of myelin basic protein (MBP), proteolipid protein (PLP), and myelin oligodendrocyte glycoprotein (MOG), have been proposed as a putative target in MS [
29]. Therefore, significant differences in the activation and priming of autoreactive T cells may occur between EAE and MS. Also, the CD8+-shift seen in MS lesions contrasts with the predominance of CD4+-T cells found in the majority of EAE models [
28].
It is hence clear that there is no single EAE model that is able to mirror the complexity of human diseases and pitfalls occur when preliminary laboratory results are translated into human prematurely.
Despite these limitations, the analysis of a publicly available transcriptomic dataset generated on MS lesions has produced data that are, at least partially, in accordance with those obtained with the EAE model employed in our study. In particular, in comparison to the normal white matter, MS lesions were found to express significantly higher levels of Lgals3, and the application of the co-expression modules analysis combined with the gene ontology strongly suggested that the modulation of Lgals3 occurred in parallel with the involvement of microglia and immune activation in the MS lesions. Further investigations need to be performed in order to ascertain the pattern of expression of Lgals3 in the different brain cell populations in MS patients and the effect of Lgals3 modulation in each of these cellular populations in disease activity and progression.
It is worth mentioning that, in the present study, we observed a similar pattern of transcriptional modulation between
Lgals3 and
Lgals3bp. Lgals3bp (also known as 90 K or Mac-2 BP) is a 90-kDa oligomeric glycoprotein that was originally identified as a ligand of the lactose-specific S-type lectin, galectin-3 (formerly Mac-2).
Lgals3bp expression is induced by viral infection and upon exposure to pro-inflammatory, such as IFN-α, IFN-β, IFN-γ, and TNF-α. Also, a prominent role for Lgals3bp in tumor progression has been revealed, as elevated expression levels of Lgals3bp in serum and tumor tissue positively correlate with poor survival or a more advanced and/or metastatic disease, in most of the human solid cancers (reviewed in [
30]). Hence, the combined inhibition of both galectin-3 and Lgals3bp for the treatment of MS could be envisaged.
The accumulating data regarding the involvement of
Lgals3 in MS suggest the potential use of galectin-3 inhibitors in this setting. Up to date, different inhibitors have been produced, such as 33DFTG, which has been shown to be able to reduce corneal neovascularization and fibrosis in preclinical models [
31], and TD139, which reduced concanavalin A-induced liver injury [
32]. Another galectin-3 inhibitor (a modified citrus pectin) has also been shown to prevent the breakdown of the blood–brain barrier and brain injury in a mouse model of subarachnoid hemorrhage [
33]. Given the multiple effects of galectin-3 exerted on the different CNS resident and invading cells, a better understanding of the involvement of galectin-3 during neuroinflammation and neurodegeneration will potentially define galectin-3 as a therapeutic target.
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.