High Expression of Galectin-3 in Patients with IgG4-Related Disease: A Proteomic Approach

Objectives Immunoglobulin G4-related disease (IgG4-RD) is a multiorgan condition manifesting itself in different forms. This study aimed to investigate protein expression profiles and to find the possible biomarker for IgG4-RD by liquid chromatography mass spectrometry (LC-MS) using tissue sections in IgG4-RD patients. Methods Protein expression profiles in five IgG4-related pancreatitis and three normal pancreatic samples were compared using LC-MS and were validated by quantitative real-time PCR (qRT-PCR), immunoblotting, and immunohistochemistry. ELISA was employed in the serum of 20 patients with systemic IgG4-RD before and during steroid treatment. Results LC-MS indicated that the levels of 17 proteins were significantly higher and 12 others were significantly lower in IgG4-related pancreatitis patients compared to controls. Among these proteins, galectin-3 levels were 13-fold higher in IgG4-related pancreatitis (P < 0.01). These results were confirmed by immunoblotting and qRT-PCR. The average number of galectin-3 + cells in various organs of IgG4-RD patients, including salivary glands, lungs, and lymph nodes, was higher than in controls. Galectin-3 was detectable in macrophages, dendritic cells, and stromal myofibroblast-like cells, but not in lymphocytes by immunofluorescence staining. Serum galectin-3 levels were higher in patients with IgG4-RD compared with healthy donors and remained high during steroid therapy. Conclusion Galectin-3 was overexpressed in IgG4-RD and the levels were indirectly related to clinical activity.


Introduction
Immunoglobulin G4-related disease (IgG4-RD) is an autoimmune multiorgan condition, characterized by a dense lymphoplasmacytic infiltration with a high number of IgG4positive plasma cells and high IgG4 levels in serum [1,2]. The disease has an indolent course and may only be detected after complications dependent on masses [1,2]. IgG4 is synthesized and secreted by plasma cells as a part of an immuneprotective mechanism and accounts for less than 5% of the total IgG in the serum of healthy individuals. IgG4 is generally considered as an anti-inflammatory immunoglobulin, because the ability to fix complement and bind activating Fc receptors is limited [3]. Thus, it is unclear if this immunoglobulin links to disease pathogenesis [2,4].
Chronic antigen exposure stimulates IgG4 production, leading to a shift in the IgG4 : IgG1 ratio. The mechanism driving this switch still remains unclear. However, Th2 interleukins such as IL-4, IL-5, and IL-13 can mediate the transition from IgG1 to IgG4 [4,5]. Th2s in CD4+ T cells and regulatory T cells play a role in the excessive production of IgG4 and stromal fibrosis in IgG4-RD [6][7][8][9]. However, more recent studies show an abundance of CD4+ cytotoxic T cells and a paucity of Th2 cells in IgG4-RD [10,11]. About 50% of IgG4-RD patients have also allergic diseases [2,12] and these conflicting results may come from IgG4-RD with or without a history of atopy [13]. Nevertheless, the etiology and pathogenesis of IgG4-RD are not well known.
Recent advancements in technical procedures and bioinformatic methods improved the sensitivity of mass spectrometry, making proteomic analysis on formalin-fixed paraffin embedded (FFPE) tissues possible [14]. In this paper, we aimed to investigate protein expression profiles in IgG4-RD and to identify possible biomarkers against this condition using liquid chromatography mass spectrometry (LC-MS).

Liquid Chromatography Mass Spectrometry (LC-MS).
Proteins for LC-MS analysis were extracted from FFPE tissue samples (five IgG4-related pancreatitis and three normal pancreas tissue) using the Liquid Tissue MS Protein Prep kit (Expression Pathology Inc, Rockville, MD, USA) [15]. All cases analyzed for LC-MS were the partial resection for pancreas. For normal pancreas tissues, negative surgical margins taken at resection for pancreatic cancer were selected. Briefly, after deparaffinization, three 0.4 m thick tissue sections (10 × 10 mm) were dissected using a needle and solubilized in 20 uL of Liquid Tissue buffer and protein digestion was performed with trypsin (Promega Corp, Madison, WI, USA) for 18 h at 37 ∘ C. Samples were dried and solubilized in 0.1% formic acid (Wako, Osaka, Japan) and 1 g aliquots for each sample were separated by nanoflow reversed-phase LC (NanoLC-Ultra 2D-Plus, Eksigent, Dublin, CA, USA) equipped with cHiPLC Nanoflex (Eksigent). Eluted peptides were analyzed by a quadrupole-time-of-flight hybrid mass spectrometer (Triple TOF5600+ system, AB SCIEX, Framingham, MA, USA).

Peptides Identification and Quantification.
Mass spectra were searched against the Uniprot/Swissprot human proteomic database (2014-6 release) from the European Bioinformatics Institute using ProteinPilot version 4.5 (AB SCIEX). False discovery rates (FDRs) were determined after peptide/protein identification using Proteomic System Performance Evaluation Pipeline provided as a part of Protein-Pilot (AB SCIEX). Label-free quantification of peptides was performed using Progenesis QI for Proteomics (Nonlinear Dynamics, Newcastle upon Tyne, UK). Proteins identified by at least two distinct peptides were used for further analysis.

Real-Time Quantitative PCR (qRT-PCR).
Total RNA was extracted from FFPE samples using the NucleoSpin total RNA FFPE kit (Macherey-Nagel, Düren, Germany). cDNA synthesis was performed using the ReverTra Ace qPCR RT kit (TOYOBO, Osaka, Japan). Amplifications were carried out in triplicate on a 96-well plate in a 10 L volume per well using Fast SYBR Green Master Mix (Applied Biosystems, Life Technologies, Carlsbad, CA, USA). These reactions were performed with an Applied Biosystems 7500 Fast real-time PCR machine. Expression was normalized to GAPDH, using the 2 −ΔΔCt method. The primer sequences were LGALS3 (forward: 5 -GCCTCGCATGCTGATAACAA-3 , reverse: 5 -CGTGGGTTAAAGTGGAAGGC-3 ) and GAPDH (forward: 5 -GGTATCGTGGAAGGACTCATGAC-3 , reverse: 5 -ATGCCAGTGAGCTTCCCGTTCAGC-3 ).   to the principles expressed in the Declaration of Helsinki. The Ethics Committee of Kyoto University approved this study (R0305).

Statistical
Analysis. Data were analyzed with Graph-Pad Prism 6 (mdf, Tokyo, Japan) or R version 3.2.0. To compare continuous variables, unpaired Student's -test or Mann-Whitney test and two-way analysis of variance (ANOVA) followed by Bonferroni's multiple comparisons test were used. Differences with a value smaller than 0.05 were considered statistically significant.

Protein Identification and Quantification. LC-MS
revealed that 956 proteins were differentially expressed between IgG4-related pancreatitis and normal pancreatic samples (Figures 1(a), 1(b), 1(c), and 1(d)). Among these, 352 were significantly upregulated and 604 were downregulated in IgG4-related pancreatitis samples compared with normal pancreatic tissues (Figure 1(e)). Proteins with high confidence values ( value < 0.05) and showing a twofold difference in their expression between IgG4-related pancreatitis and normal pancreatic samples were selected for further analysis; the levels of 17 proteins were higher and those of 12 proteins were lower in IgG4-related pancreatitis tissues than in healthy pancreas (Table 1). We selected galectin-3, as it was one of the highly upregulated proteins in IgG4-related pancreatitis (13-fold, value 0.013).

Galectin-3 Is Highly Expressed in Different Organs with
IgG4-RD. To measure the expression of galectin-3 in different organs of IgG4-RD patients, FFPE samples from various organs were stained with anti-galectin-3 antibodies. Since we focused on immune cells, but not on epithelial cells, we counted galectin-3-positive cells in the stroma, including immune and fibroblast-like cells but excluding galectin-3positive epithelial cells, in high-power fields (HPF) from each specimen. The average number of galectin-3-positive stromal cells was higher in organ samples from IgG4-RD patients compared with healthy pancreas or lymph node samples. The differences between healthy pancreas and IgG4-related pancreatitis samples and between healthy lymph node and IgG4 lymphadenopathy were statistically significant ( value < 0.001 for both comparisons) (Figure 2(f)).

Galectin-3 Levels in Serum of Patients with Systemic
IgG4-RD. Galectin-3 levels in serum of IgG4-RD patients were 7.8 ± 0.58 ng/mL and 10.7 ± 0.58 ng/mL before and during PSL treatment, respectively, that is, 2-and 2.5-fold higher than 3.8 ± 0.46 ng/mL found in healthy donors (versus donor value < 0.01 for both comparisons; Figure 4(a)). We monitored the serum levels of IgG4 and galectin-3 during PSL tapering in five patients for an average of 24 months. While IgG4 serum levels markedly decreased after prednisolone (PSL) therapy, galectin-3 levels in the serum increased. Inversely, lower PSL doses were accompanied by a slight decrease in galectin-3 levels in serum (Figures 4(a)  and 4(b)). treatment; range 0.3-2 months; last samples from an average of 24 months after treatment; range, 5-59 months).
Galectin-3 levels in serum were on average 1.39 ± 0.058fold higher immediately after the start of the PSL therapy (on average 1 month) when compared to galectin-3 levels before the treatment. Galectin-3 levels in serum were on average 1.37 ± 0.15-fold higher after the last follow-up treatment, on average 24 months after the start of the therapy, compared to relative to the levels before the treatment (Figure 4(c)).

Discussion
We chose the protein galectin-3, identified by LC-MS with high confidence and differential expression in FFPE samples of IgG4-related pancreatitis tissues. Galectin-3, encoded by LGALS3, is a member of the galectin family and expressed in numerous cells, including epithelium cells of gastrointestinal and respiratory tracts and renal tubules, fibroblasts, chondrocytes, osteocytes, and endothelial cells. This protein has 8 Pathology Research International pro-and antiapoptotic activity [16][17][18] and regulates human monocyte and macrophage, participating in the autophagy pathway [19][20][21]. Galectin-3 is also expressed in cells involved in immune responses, such as neutrophils, eosinophils, basophils, mast cells, Langerhans cells, dendritic cells, and monocytes and macrophages [18]. Notably, in resting lymphocytes and several lymphoid cell lines, galectin-3 is not normally expressed, but it can be induced by various stimuli [18,22,23]. High galectin-3 expression is observed in many forms of cancer, including thyroid, pancreatic, and colon cancers, and has been linked to cancer progression and metastasis [24,25].
This study demonstrated an increase in galectin-3 expression in IgG4-RD patients. Galectin-3 was expressed in cells involved in immune response activity, including macrophages, dendritic cells, myofibroblasts, and epithelial cells. Galectin-3 signal was found in stromal deposition. Dendritic cells are specialized in antigen presentation cells and connect the innate and adaptive immune system [30]. They have the distinctive ability to stimulate and polarize naive T cells. For example, the expressions of IL-12 and IL-10 stimulate T helper-1 (Th1) and Th2 activation, respectively [30]. Th2 response controls the production of IgG4 [18]. Plasmacytoid dendritic cells participate in autoimmune diseases, such as IgG4-related pancreatitis, lupus erythematosus, psoriasis, and contact dermatitis, likely contributing to IFNproduction [31,32]. Galectin-3 secreted by dendritic cells mediates the binding of these cells to lymphocytes and might play a role in cell-cell interactions [33,34]. Galectin-3 binds integrins, the main membrane molecules involved in cell adhesion, including those present on the surface of macrophages such as 1 1 integrins and the subunit of M 1 integrin (CD11b/18, Mac-1 Ag), and these integrins are used as receptors [18,21]. Galectin-3 is also essential for effective phagocytosis by macrophages to remove apoptotic cells and therefore preventing autoimmune reactions [35]. In addition, endogenous galectin-3 has been found to drive a Th2 response in both dendritic cells and T cells, while galectin-3 deficiency resulted in the development of a Th1 response [29]. Recent studies show clonally expanded CD4+ cytotoxic T cells that express granzyme B, perforin, IFN-gamma, and TGF-beta1 are prominent in IgG4RD [10,11] Rituximabmediated B cell depletion is associated with a reduction in numbers of disease-associated CD4+ cytotoxic T cells, suggesting that B cells play an important role in the maintenance of disease-associated T cell clone [10]. Extracellular galectin-3 directly induces T cell apoptosis in a carbohydrate dependent manner by binding to its cell surface receptors, CD7 and CD29 [36]. Inhibition of galectin-3 in vivo skewed the balance toward plasma cell differentiation and increased immunoglobulin production through the up-regulation of Blimp-1, a transcription factor responsible for B cell apoptosis and essential for plasma cell commitment, in Trypanosoma cruzi infection model [23]. Additionally, galectin-3 acts as a negative regulator of the differentiation of B-1 lymphocytes into plasma cells [37]. Likewise, IL-4-mediated B cells activation, resulting in their differentiation into memory cells, is accompanied by a significant increase in galectin-3 expression [23]. Therefore, galectin-3 may act as a protective factor against the progression of IgG4-RD since galectin-3 blocks plasma cell differentiation in B cells and apoptosis of T cells [23,36,37], although this protein has other biological roles [18].
In our study, galectin-3 serum levels in patients with IgG4-RD before and during PSL treatment were significantly higher compared with those in healthy donors. IgG4 levels in serum decreased dramatically after PSL treatment, while galectin-3 levels in serum remained high throughout the treatment. Galectin-3 serum levels increased with PSL treatment and gradually decreased with the tapering of PSL. This implies that galectin-3 cannot serve as a realistic biomarker for IgG4-RD in routine clinical practice. Glucocorticoid therapy could be a bias, because galectin-3 expression increases significantly in macrophages and lung Clara cells after glucocorticoid treatment [38]. Similarly, mineralocorticoid hormones increase galectin-3 expression in vascular smooth muscle cells [39]. Therefore, the induction of galectin-3 expression after treatment with PSL, a synthetic glucocorticoid, may be in agreement with these previous reports. Galectin-3 expression is higher in SLE and systemic sclerosis patients with active diseases than those with an inactive one. However, galectin-3 expression is not directly correlated with the disease activity and severity indexes in each SLE and systemic sclerosis patient, nor with the duration of the disease [28]. In our study, galectin-3 was present in the stroma and myofibroblasts in IgG4-RD patients. Immune cells subside immediately after steroid therapy, while tissue fibrosis may remain even after treatment [40]. It is difficult to assist tissue fibrosis because follow-up biopsy is uncommon after the treatment of IgG4-RD. The only available data come from a recent work by Della-Torre et al. [41], who found a reduction in the total number of myofibroblasts after rituximab treatment in one patient with cutaneous IgG4-RD. The presence of extracellular galectin-3 can be a determining factor for prolonged galectin-3 levels in serum after PSL treatment. It is possible that galectin-3 seen in patients with IgG4-RD represents the response against inflammation. To understand the significance of galectin-3 in relation to the pathogenesis of IgG4-RD, the decreased expression of galectin-3 in the affected tissues after successful corticosteroid therapy would be important; however no tissues after therapy were available.
In conclusion, galectin-3 was overexpressed in IgG4-RD and the levels were indirectly related to clinical activity and treatment. Further studies will need to explore the exact mechanism behind the function of galectin-3 in IgG4-RD and other autoimmune diseases.