Galectin-3 in autoimmunity and autoimmune diseases

Systemic lupus erythematosus

Systemic lupus erythematosus (SLE) is a complex autoimmune multi-organ disorder in which genetic predisposition, environmental triggers and imbalanced immune response lead to autoantibody secretion and tissue injury.^33,34 Beside autoreactive B lymphocytes and autoantibody-secreting plasma cells, T_reg cells, NKT cells, mesenchymal stem cells, DCs and autoreactive T cells are also involved in the immunopathogenesis of SLE.^34–36

Mice susceptible to autoimmune diseases, including lupus, have abnormal germinal center reactions, which seem essential in triggering autoimmune signals. ³⁷ Spontaneously generated splenic germinal centers were associated with autoantibody production and atypical distribution of B220⁺ PNA⁺ (peanut agglutinin) GL-7⁺ (Ly77 or B-/T-cell activation marker) B cells and M1⁺ (Murine 1) follicular DCs. ³⁸ Splenic follicular gal-3⁺ cells have DC morphology, ³⁹ but it is not clear whether follicular macrophages also express gal-3 within lymphoid follicles. ²² In gal-3^−/− mice chronically infected by helminthes, such as Schistosoma mansoni, niches of CD138⁺ plasma cells and B220⁺ B lymphocytes were severely disturbed in the spleen and mesenteric lymph nodes, affecting the activation and survival of these cells. Moreover, the high number of follicular apoptotic bodies indicates that gal-3 is an important mediator of apoptosis and/or clearance in these lymphoid organs. ²² Notably, both mechanisms are involved in the pathogenesis of SLE and other autoimmune diseases. ⁴⁰

In the thymus, exogenous gal-3 plays a de-adhesive role favoring intrathymic migration of CD4/CD8 T cells during differentiation events. ⁴¹ By contrast, intracellular gal-3 suppresses T_H17 phenotype by inhibiting IL-23/IL-17-axis in DCs. ⁴²

In gal-3^−/− mice, the serum levels of IL-23, transforming growth factor (TGF)-β1 and IL-1β were significantly increased leading to a decrease in IL-12, interferon (IFN)-gamma and T_H1 cells. ⁴³ Since SLE progression is critically associated with pro-inflammatory T_H17 cells secreting IL-17, IL-1, IL-6 and IL-23, ⁴⁴ it is likely that gal-3 plays a regulatory role in the pathogenesis of SLE by suppressing T_H17 cell differentiation.

SLE patients with predominant type I IFN signature have a high rate of macrophage differentiation and autoantibody production leading to a more severe disease. ⁴⁵ Plasmacytoid DCs (pDCs) responsible for type I IFN signature are activated by self-nucleic acid complexes via CD32 and toll-like receptor (TLR)-9. ⁴⁶ Self-nucleic acid complexes are abundantly released by apoptotic cells and by neutrophils during neutrophil extracellular traps formation. ⁴⁷

On peripheral blood mononuclear cells, gal-3/CD32 (low-affinity IgG receptor—Fcgamma RII) interaction inhibits T_H2 cytokine responses and favors a type I IFN phenotype. ¹⁸ It is well known that the inhibition of type I IFN signals leads to a delayed symptoms in experimental SLE ⁴⁸ ; thus, gal-3 could promote SLE by increasing type I IFN expression.

Another mechanism potentially involved in gal-3 inducing SLE development consists in the interaction with transcription factors that induce B-/T-cell differentiation and hypergammaglobulinemia, such as Blimp-1 (B lymphocyte-induced maturation protein-1) which follows DC activation. ⁴⁹ In MRL.Fas^lpr mice, the absence of DCs leads to milder SLE symptoms, reduced T-/B-cell expansion and lower production of autoantibodies. ⁵⁰ Although poorly explored in SLE, polymorphisms in Blimp-1 have been proposed as a “risk factor” for SLE. ⁴⁹ Inhibition/deletion of gal-3 in vivo leads to up-regulation of Blimp-1 affecting plasma cell behavior, survival and synthesis of IgG and IgE.^3,24

IgG anti-galectins have been reported during the early stage and progression of SLE, especially anti-gal-2, −3, −7, −8 and −9. ⁵¹ In humans, anti-gal-3 autoantibody serum levels were significantly higher in SLE patients than in healthy donors or in patients with rheumatoid arthritis (RA), primary Sjögren's syndrome (pSS) and systemic sclerosis (SSc). ⁵²

As far as clinical manifestations are concerned, Shi et al. found an association between anti-gal-3 antibodies and SLE cutaneous vasculitis. Indeed, SLE patients bearing skin lesions have high serum levels of anti-gal-3 autoantibodies, which are localized on the vessel walls. Notably, the injection of anti-human gal-3 antibodies in BALB/c mice evoked a dermal vasculitis 36 h post injection. ⁵³

Gal-3 and anti-gal-3 antibodies might also be involved in development of lupus nephritis, which affects approximately 50–60% of SLE patients. ⁵⁴ During the early phases of lupus nephritis, autoimmune T cells respond to tissue injuries reacting against chromatin or nucleosomes derived from cell death, including apoptosis, necrosis, necroptosis, autophagy and possible disturbances in the clearance of dying cells, leading to differentiation of autoimmune lymphocytes.^55,56

Interestingly, SLE patients affected with lupus nephritis were seen to display higher serum levels of anti-gal-3 antibodies compared with controls; additionally, gal-3 stained in glomeruli in the majority of patients, while it was absent in controls. ³⁸ Notably, gal-3 was expressed in distal tubules of normal kidneys. ³⁸ Gal-3 expression scores correlated with anti-double stranded (ds) DNA antibody titers, renal histological activity indexes and complement consumption in SLE patients. ³⁸

In rats, gal-3 modulates mesangial cell proliferation and matrix synthesis during experimental glomerulonephritis. Moreover, it was demonstrated that gal-3 is significantly secreted by infiltrating macrophages and distal tubular epithelium. ⁵⁷ Recently, Nielsen et al. ⁵⁸ revealed an increased population of gal-3-binding protein (G3BP)-positive microparticles either in sera or in electron-dense deposits from kidney biopsies of SLE patients affected with lupus nephritis, suggesting that gal-3 co-localizes in immune complex deposits. These observations support the potential role of gal-3 as a putative autoantigen in development of specific lupus manifestations.

Taking into account these data, whether gal-3 is involved in fueling or dampening SLE pathways is currently unclear (Table 1).

Rheumatoid arthritis

RA is an autoimmune disease characterized by an aberrant immune response against citrullinated self-peptides and progressive joint destruction. ⁵⁹ Failure in CD4⁺ T_reg-mediated self-tolerance and increase in T_H17 cells seem to be critical in the immunopathogenesis of RA. ⁶⁰ Indeed, in RA synovitis, the inflammatory process is orchestrated by activated monocytes (secreting IL-1β, IL-6, IL-7 and tumor necrosis factor-alpha (TNF)-α), fibroblast-like synoviocytes (synthesizing metalloproteases, chemokines and cytokines, mainly IL-15), and dysregulated osteoclasts (due to high levels of TNF-α and IL-17). ⁶¹

Gal-3 seems to play a pro-inflammatory role in RA ⁶² and high serum levels of gal-3 have been found in mice during collagen-induced arthritis. ⁶³ In humans, gal-3 mRNA and protein are overexpressed in the synovial membrane close to joint destruction and are highly correlated with C-reactive protein (CRP) serum levels and IL-6, granulocyte–macrophage colony stimulating factor (GM–CSF), chemokine (C-X-C) motif ligand 8 (CXCL8), metalloproteinase-3 (MMP-3), TNF-α, chemokine (C-C motif) ligand 2 (CCL2), CCL3 and CCL5 secretion by synovial fibroblasts. ⁶⁴ The lack of gal-3 prevented antigen-induced arthritis in mice which showed lower levels of antigen-specific IgG, IL-6, TNF-α and IL-17. ⁶²

Genetic polymorphisms in gal-3 gene seem to be involved in RA susceptibility. Indeed, it has recently been shown that gal-3 gene (LGALS3) + 292C allele is associated with RA in Taiwanese population. ⁶⁵ Notably, recombinant gal-3 in gal-3^−/− mice can restore RA symptoms and cytokine production. ⁶² Finally, gal-3 short hairpin RNA (shRNA) injected in inflamed joints reduced collagen-induced arthritis in rats. ⁶³ Thus, gal-3 is likely to play a critical role in RA development by regulating synovial fibroblast functions (Table 1).

Systemic sclerosis

SSc is a heterogeneous autoimmune disease characterized by progressive fibrosis of skin and internal organs due to an abnormal and chronic activation of fibroblasts. ⁶⁶ Gal-3 has been suggested to play a key role in inducing fibrosis in different tissues. ⁵ Gal-3 serum levels are substantially increased in SSc patients where they seem to be correlated with the fibrotic process; in fact, gal-3 seems to regulate fibroblast activation, one of the major pathways responsible for fibrosis in SSc. ⁶⁷ Indeed, gal-3 serum levels were not increased in the early stages of diffuse SSc but they progressively increased over disease course. ⁶⁸ Thus, gal-3 represents an interesting molecule, potentially involved in the pathogenesis of SSc, which can help us to better understand the mechanisms of fibrogenesis in experimental models and humans (Table 1). It might also represent a therapeutic target in SSc.

Polymyositis/dermatomyositis

Polymyositis and dermatomyositis are considered the two major subsets of idiopathic inflammatory myopathies. ⁶⁹ Polymyositis is characterized by a CD8⁺ T-cell response against muscle antigens presented by MHC Class I molecules. On the other hand, dermatomyositis is marked by a complement-mediated microangiopathy, typically mediated by type I IFN phenotype and autoantibodies with distinct specificities^70,71; however, the exact mechanism involved in complement activation remains unclear. ⁷² Current biological treatments in polymyositis/dermatomyositis include blockade of cells and cytokines, including B-cells by rituximab, ⁷³ T cells by abatacept and cytokines by anti-IL-1, anti-IL-6 or anti-IFNα monoclonal antibodies. ⁷⁴

Autoantibodies to gal-3 were found more frequently in the sera of patients with polymyositis/dermatomyositis than in those with other autoimmune diseases, including RA, pSS, SSc and SLE. ⁵² However, the analysis of sensitivity and specificity of these autoantibodies in large cohorts of patients is critical in order to determine whether anti-gal-3 could be used as diagnostic marker in patients with polymyositis/dermatomyositis.

Primary Sjogren syndrome

pSS is an autoimmune disease well defined by mononuclear cell infiltrate in the exocrine glands, primarily lacrimal and salivary glands. ⁷⁵ The inflammatory infiltrate leads to acinar atrophy, which is correlated with ectopic follicles (tertiary lymphoid tissues) composed by lymphocytes, plasma cells and stromal fibrosis. ⁷⁶ In the follicles, autoantigens are presented to T and B cells and there are germinal centers where autoantibodies can be generated. ⁷⁷ Gal-3 serum levels are significantly increased in pSS patients where they are correlated with serum levels of plasma CRP, fibrinogen, IgG and IL-17. ⁷⁸ However, the cellular and molecular mechanisms of gal-3 in pSS have been poorly investigated so far (Table 1).

IgA nephropaty

Glomerulonephritis refers to an immune inflammatory process within renal glomeruli, which is sometimes associated with autoantibodies, eventually leading to end-stage renal disease. ⁵⁵ The first step in the development of glomerulonephritis is the activation of TLRs⁺ and nod-like receptors (NLRs)⁺ cells by PAMPs and DAMPs. ⁷⁹ A combined T_H1- and T_H2-cytokine response induces CD4⁺ T-cell differentiation into TGF-β⁺IL-10⁺CTLA4⁺ T_reg cells via TGF-β induction, INF-γ⁺TNF-α⁺ T_H1 cells via IL-12 induction, IL-4⁺IL-5⁺IL13⁺ T_H2 cells via IL-2, IL-4 and IL-13 induction, or IL-17⁺IL-21⁺ T_H17 cells via TGF-β, IL-6 and IL-17, all of them involved in glomerular injury. ⁸⁰

The most prevalent glomerulonephritis in humans is the immunoglobulin A (IgA) nephropathy, which is characterized by huge glomerular deposition of IgA immune complexes leading to renal injury. ⁸¹ IgA nephropathy is hallmarked by high levels of IgA in the serum and poorly galactosylated polymeric IgA1 is widely deposited in the mesangium, following secretion by plasma cells in mucosa-associated lymphoid tissues (MALT). ⁸² Since gal-3 was shown to control mesangial cell proliferation and extracellular matrix compounds in a rat model of glomerulonephritis induced by passive transfer of anti-Thy1 antibodies, ⁵⁷ it is possible that gal-3 is involved in the pathogenesis of the IgA nephropathy.

Gal-3 protects renal tubules from chronic injury by limiting apoptosis. ⁸³ Consistently, gal-3^−/− mice develop pronounced changes in renal function and structure in age-related renal disease ⁸⁴ ; however, it is not clear whether and how gal-3 regulates renal function. Furthermore, no relationship between gal-3 and renal disease associated with IgA deposition has been demonstrated yet. Recently, Oliveira et al. ⁸⁵ demonstrated that IgA expression and secretion are up-regulated in serum and peritoneal fluid in the absence of gal-3, suggesting that IgA-differentiating plasma cells and IgA secretion are at least in part regulated by gal-3. Although gal-3 and IgA nephropathy are not correlated, these observations deserve further studies.

Psoriasis

Psoriasis is characterized by red and scaly patches in the skin, ridged fingernails, hair loss, painful joints and systemic findings. ⁸⁶

Myeloid DCs (mDCs) drives the immunopathogenesis of this immune-mediated inflammatory disease by releasing IL-23 and IL-12, ⁸⁷ which are responsible for the activation of T_H17, T_H1 and T_H22 cells which produce the “psoriatic cytokines” IL-17, IFN-γ, TNF and IL-22 affecting keratinocyte functions. ⁸⁸ In contrast, T_reg cells control immune system by preventing autoimmune responses against self-antigens. ⁸⁹

The role of gal-3 in the pathogenesis of psoriasis is still poorly understood; however, it has been shown that it is up-regulated in dermal endothelial cells and down-modulated in the epidermis ⁹⁰ and Langerhans cells. ⁹¹ The mechanisms involved in these modifications of gal-3 expression have not been studied yet.

Atopic dermatitis

Atopic dermatitis (AD) is an autoimmune disorder characterized by defective skin barrier, which becomes permissive to penetrating allergens with systemic sensitization ⁹² and IgE and IgG anti-nuclear autoantibody production. ⁹³ It is first driven by a T_H2-cytokine phenotype which is then switched to a mix T_H1/T_H2 phenotype. ⁹⁴ In mouse models, gal-3 polarizes immune responses towards a T_H2 phenotype. ⁴³ AD-ovalbumin (OVA)-induced mice had high levels of serum gal-3 whereas OVA-sensitized gal-3^−/− mice developed a T_H1-polarized immune response associated with a decrease in inflammatory markers, ⁹⁵ suggesting that gal-3 may act as an immunomodulatory molecule during early stage of AD.

Multiple sclerosis

Multiple sclerosis (MS) is the most prevalent autoimmune disease of the central nervous system (CNS) and can be mimicked by experimental autoimmune encephalomyelitis in animal models. The pathogenesis of MS is marked by a potent T_H17 immune response against demyelinating degenerative cells. ⁹⁶ Apoptotic oligodendrocytes are engulfed by microglia in damaged tissues, which are infiltrated by macrophages, T/B lymphocytes, NKT cells and DCs.^97,98

The involvement of gal-3 in the pathogenesis of MS may be coupled with oligodendrocyte differentiation, since gal-3 contributes to myelin integrity and function. ⁹⁹ In gal-3^−/− mice, myelinated axons are numerically reduced and the myelin is severely disorganized around axons. ⁹⁹ However, during wallerian degeneration, sciatic nerve was rapidly regenerated by Schwann cell activities and the clearance of the myelin debris was significantly efficient in the absence of gal-3.^100,101 Although the reason is still unclear, gal-3 may play a regulatory role in the development of MS, at least in part by affecting phagocytosis of myelin debris.

Experimental autoimmune encephalomyelitis

Experimental autoimmune encephalomyelitis (EAE) partially reproduces human MS in experimental animals and is classified as a neuroinflammatory disease orchestrated by self-antigens recognized by autoreactive T cells. T_H17 and T_H9 cells have currently been considered pivotal in the pathogenesis of EAE, ¹⁰² besides the removal of degenerated myelin via phagocytosis mediated by microglia. ¹⁰³ In EAE mice, gal-3 is up-regulated by activated microglia during myelin degeneration and phagocytosis in spinal cords and optic nerves. ¹⁰⁴ In these activated cells, gal-3 activates PI-3 kinase pathway via rat sarcoma protein (Ras), phagocytosis mediated by complement receptor-3 and promotes a high-affinity interaction between degenerated myelin (myelin debris) and phagocytosis-receptors during tissue repair.^105,106

In EAE, gal-3 prevents apoptosis and IL-10 production on the one hand and induces T_H17 phenotype and IFN-γ synthesis on the other hand. In gal-3^−/− mice, EAE was marked by a reduction in monocyte/macrophage infiltration and an increase in apoptotic cells. ¹⁰⁷ Gal-3 regulates T-cell function forming complexes in TCR with Mgat5 (beta1,6 N-acetylglucosaminyltransferase V), an enzyme which exposes N-acetylgalactosamine domain, the binding site of gal-3. ¹⁰⁸ Gal-3 inhibition phenocopies Mgat5^−/− TCR clustering and increases the susceptibility to EAE. Thus, gal-3 can be involved in EAE development.

Type I diabetes (T1D)

The autoimmune type 1 diabetes (T1D) is caused by genetic predisposition and immune dysregulation leading to abnormal autoreactive T lymphocyte differentiation, pancreatic β cell apoptosis and deficient insulin production. ¹⁰⁹ According to genome-wide association studies, pro-apoptotic genes contribute to insulitis and β-cell destruction. ¹¹⁰ Gal-3 protects pancreatic β cells against IL-1-mediated apoptosis. ¹¹¹ On the other hand, inhibition or deletion of galectin-3 renders pancreatic β-cells more resistant to TNF-α/IFN-γ/IL-1β apoptosis. ¹¹² Probably, extracellular and intracellular gal-3 have a different effect in T1D pathogenesis. Approximately 50% of pancreatic cells express intracellular gal-3 in healthy tissues. ¹¹³

Gal-3^−/− mice are resistant to multiple low-dose streptozotocin-induced diabetes (MLD-STZ) showing modest mononuclear cell recruitment to pancreatic islets, low expression of IFN-γ and inducible nitric oxide synthesis (iNOS) and absence of expression of TNF-α and IL-17 in comparison with wild type mice. ¹¹⁴ The ablation of gal-3 was associated with up-regulation of “anti-apoptotic” genes, suggesting that gal-3 induces apoptosis of islet β-cells upon pro-inflammatory condition and exposition of autoantigens driving autoimmune responses during early T1D. The mechanism involved in the control of apoptosis on pancreatic β-cells by gal-3 has not been fully elucidated yet. However, gal-3 seems to play a role in the development of T1D and it could become a promising therapeutic target in T1D patients.

Autoimmune hepatitis (AiH)

Autoimmune hepatitis (AiH) is a chronic disease in which apoptotic hepatocytes and portal inflammatory reaction induce a collapse of hepatic lobules. ¹¹⁵ The pathogenesis of AiH is initially driven by a T_H1/T_H17 phenotype marked by infiltrating CD8⁺ T cell,^116,117 and is subsequently switched to T_H2 signature which promotes clonal expansion of autoantibody-secreting plasma cells and inhibits T_H1/T_H17 functions. ¹¹⁸ During the dampening of tissue inflammation, T_reg cells attenuate the immune response ¹¹⁹ and NKT cells drive a fibrotic response. ¹²⁰ Gal-3 is an interesting molecule due to its ability to induce hepatocyte apoptosis ¹²¹ and liver fibrosis, ¹²² both orchestrating the activation and differentiation of hepatic stellate cells (HSCs) into myofibroblasts. ¹²³

Concanavalin-A (Con A)-induced hepatitis is an experimental model that mimics pathological changes observed in AiH patients. ¹²⁴ The induction of hepatitis by Con A in gal-3^−/− mice resulted in a weak liver injury, marked by low levels of pro-inflammatory cytokines secreted by hypoactive lymphocytes and DCs and a high number of annexin V⁺ propidium-idodide⁺ later apoptotic cells. ³² Moreover, gal-3 induces myofibroblast activation ¹²² and favors phagocytosis by HSCs during liver fibrosis, ¹²⁵ indicating that gal-3 plays a critical role in the pathogenesis of Con A-induced hepatitis and probably in AiH.

Crohn's disease

Crohn's disease is an inflammatory bowel disease of autoimmune origin in which genetic predisposition and environmental factors results in abnormal T-cell immune response. ¹²⁶ Gal-3 secreted by resident macrophages and colonic epithelial cells activates atypical colonic fibroblasts in lamina propria inducing the production of IL-8 via NF-κB and driving fistula formation. ¹²⁷ During the evolution of the disease, the expression of gal-3 is decreased in the colon interfering with fibroblast migration into fistulae. ¹²⁸

Epithelium breakdown was significantly associated with reduced expression of gal-3 and high levels of TNF-α, ¹²⁹ suggesting that gal-3 regulates epithelial cell–cell interactions in the colon. In patients affected with Crohn's disease, high serum levels of IgG anti-gal-3 autoantibodies were inversely correlated with clinical symptoms. ¹³⁰ By contrast, high levels of gal-3 in the sera and a high number of CD14⁺ cells in the blood of patients with Crohn's disease were found during the active stages of the disease, ¹³¹ Altogether these data suggest that galactin-3 can exert a pro-inflammatory effect in Crohn's disease and may have a role in the disease development.