Inflammatory cytokines regulate the expression of glycosyltransferases involved in

Transcription

1 Inflammatory cytokines regulate the expression of glycosyltransferases involved in the biosynthesis of tumor-associated sialylated glycans in pancreatic cancer Sònia Bassagañas M.S a, Helena Allende MD b, Lara Cobler PhD c, M. Rosa Ortiz MD d, Esther Llop PhD a, Carme de Bolós PhD b, Rosa Peracaula PhD a * a Biochemistry and Molecular Biology Unit, Department of Biology, University of Girona. Girona, Spain b Pathology Department. Vall d'hebron University Hospital. Barcelona, Spain. c Institut Hospital del Mar d Investigacions Mèdiques (IMIM). Barcelona, Spain. d Pathology Department. Dr. Josep Trueta University Hospital. Girona, Spain. * Corresponding author: Rosa Peracaula, Biology Department, Faculty of Sciences, University of Girona, Campus Montilivi, s/n Girona, Spain. Tel: ; fax: ; address: rosa.peracaula@udg.edu Short running title: Cytokines regulate glycosyltransferase expression in cancer.

2 Abstract Background: Pancreatic ductal adenocarcinoma (PDAC) is characterized by an abundant stroma containing several pro-inflammatory cytokines, which are described to modulate the expression of important genes related to tumor promotion and progression. In the present work we have investigated the potential role of these cytokines in the biosynthesis of tumorassociated carbohydrate antigens such as sialyl-lewis x (SLe x ) through the regulation of specific glycosyltransferase genes. Methods: Two human PDAC cell lines MDAPanc-3 and MDAPanc-28 were treated with proinflammatory cytokines IL-1β, TNFα, IL-6 or IL-8, and the content of tumor-associated carbohydrate antigens at the cell membrane was analyzed by flow cytometry. In addition, variation in the mrna expression of sialyltransferase (ST) and fucosyltransferase (FUT) genes, which codify for the ST and FucT enzymes involved in the carbohydrate antigens biosynthesis, was determined. The inflammatory microenvironment of PDAC tissues and the expression of Lewis-type antigens were analyzed by immunohistochemistry to find a possible correlation between inflammation status and the presence of tumor-associated carbohydrate antigens. Results: IL-1β stimuli increased SLe x and α2,6-sialic acid levels in MDAPanc-28 cells and enhanced the mrna levels of ST3GAL3-4 and FUT5-7, which codifiy for ST and FucT enzymes related to SLe x biosynthesis, and of ST6GAL1. IL-6 and TNFα treatments increased the levels of SLe x and Le y antigens in MDPanc-3 cells and, similarly, the mrna expression of ST3GAL3-4, FUT1-2 and FUT6, related to these Lewis-type antigens biosynthesis, were increased. Most PDAC tissues stained for SLe x and SLe a and tended to be expressed in the tumor samples with a higher presence of inflammatory immune cells. Conclusions: The inflammatory microenvironment can modulate the glycosylation pattern of PDAC cells, increasing the expression of tumor-associated sialylated antigens such as SLe x, which contributes to pancreatic tumor malignancy.

3 Keywords: α2,6-sialic acid, cytokines, fucosyltransferases, Lewis-type antigens, pancreatic adenocarcinoma, sialyltransferases.

4 Background Pancreatic adenocarcinoma (PDAC) is the fourth leading cause of cancer death and has a 5- year survival rate of less than 5% [1]. This extremely poor outcome is mainly due to the late diagnosis in advanced stages, aggressive invasion of surrounding tissues, early and extensive metastases, and an exceptional resistance to chemotherapy [2]. The characteristic intense desmoplasic reaction in PDAC leads to a stroma composed by several cell types including fibroblasts, pancreatic stellate cells, endothelial cells, immune cells and an altered extracellular matrix (ECM), constituting a microenvironment that promotes tumor growth and invasion [3, 4]. Inflammatory cells like macrophages and infiltrating mast cells produce pro-inflammatory cytokines as well as pro-angiogenic molecules and basic fibroblast growth factor, which are known to activate oncogenic signaling pathways on the tumor cells regulating the expression of genes involved in tumor promotion and progression [5, 6]. Focusing on pro-inflammatory cytokines, TNFα and IL-1β signaling is described to promote the activation of several transcription factors, including nuclear factor-κb (NF-κB) and activator protein-1 (AP-1), which act as transcription factors of a number of target genes [7, 8]. The expression of these two cytokines by pancreatic cells is thought to be a relatively early event in the activation of the cytokine cascade in acute pancreatitis [9]. In addition, serum IL-6 correlates with disease severity in humans [10] and is useful as an early diagnostic marker [11]. IL-8, in its turn, has also been detected in the pancreatic epithelial cells of clinical and pre-clinical pancreatitis tissue [12], suggesting a damaged pancreatic epithelium as a possible source of IL-8. Another important feature of PDAC is an altered glycosylation pattern on cell membrane and secreted proteins, such as the expression of sialyl-lewis x (SLe x ) and other tumor-associated carbohydrate epitopes [13-15]. One of the main roles of SLe x is to act as an E-selectin ligand mediating the arrest of tumor cells on the activated endothelial cells and thus facilitating their extravasation during the hematogenous metastasis [16]. The last steps of SLe x biosynthesis are catalyzed by specific fucosyltransferases (FucTs) and sialyltransferases (STs), whose

5 expression can be dysregulated in tumor tissues. Although the specific mechanisms that regulate FucT and ST expression are still not fully understood, epigenetic regulation, transcriptional and post-translational controls have been postulated as possible underlying mechanisms [17, 18]. In this regard, several reports have pointed out that pro-inflammatory cytokines can promote the overexpression of SLe x and related epitopes through the upregulation of specific FUT and ST genes in several types of cancer cells, such as hepatocellular carcinoma cells [19], prostate cancer cells [20], lung cancer cells [21] and bronchial mucose [22]. The cytokines assayed differed among these studies, as well as some of the genes that were up-regulated, giving further evidence that the regulation and expression of glycosyltransferases is tissue- and cell-type dependent. Taking into account these precedents, and given the importance of the inflammatory microenvironment and SLe x expression in PDAC, in this study we have investigated whether the main pro-inflammatory cytokines found in the PDAC stroma could modulate the expression of specific FUT and ST genes, which codify for the ST and FucT enzymes that lead to SLe x expression and thus promote tumor malignancy. To address this issue, we treated the two human pancreatic adenocarcinoma cell lines MDAPanc-28 and MDAPanc-3 with TNFα, IL1β, IL-6 or IL-8 cytokines, and afterwards determined SLe x and related antigens expression on the cells surface by flow cytometry. We have also quantified the mrna levels of the FUT and ST genes, which codify for the enzymes involved in the synthesis of the aforementioned epitopes. In addition, we have characterized the inflammatory component of human pancreatic adenocarcinoma tissues and have studied a possible correlation with Lewis-type antigens expression.

6 Material and methods Cell lines and culture The human pancreatic adenocarcinoma cell lines used were MDAPanc-3, established from a liver metastasis of a moderately differentiated adenocarcinoma of the head of the pancreas [23], and MDAPanc-28, established from a poorly differentiated adenocarcinoma of the body of the pancreas presenting local invasion [24]. Cells were routinely cultured in Dulbecco s Modified Eagle s Medium (DMEM) GlutaMAX-I containing 10% fetal bovine serum (FBS), 100 U/ml penicillin G, 100 µg/ml streptomycin and 0.25 µg/ml amphotericin B (all supplied by Gibco; Paisley, UK) and kept at 37 C in humidified atmosphere containing 5% CO 2. Cell growth and morphology were daily assessed under the field microscope. Cytokine treatments The cytokine treatment protocol was based on a previously described method [25], with minor modifications. Briefly, 1.8 x 10 5 MDAPanc-3 or 1.5 x 10 5 MDAPanc-28 cells were seeded in 25 cm 2 flasks (Nunc; Roskilde, Denmark) and cultured for 72 h in standard conditions. At subconfluence, cells were washed twice with sterile phosphate buffered saline (PBS). Cells were then serum starved for 4 h to bring cells to a basal activation state. Following this, the cells were treated with or without the presence of IL-1β, IL-6, IL-8 or TNFα pro-inflammatory cytokines (all of them from R&D Systems; Minneapolis, USA) at various concentrations within the pm range, in DMEM for 4, 12 or 24 h (for glycosyltransferase mrna evaluation) or 40 h (for the detection of carbohydrate determinants by flow cytometry). At least three independent experiments were performed for each treatment. RNA isolation and cdna synthesis RNA was isolated from the cells using RNeasy Mini Kit including on-column DNase digestion using the RNAse-Free DNase Set, according to the manufacturer s instructions (Qiagen GmbH; Hilden, Germany). Extraction yield and RNA purity were

7 spectrophotometrically determined using a Nanodrop TM instrument (ND-1000, Thermo Fisher Scientific; Wilmington, DE, USA). 2,000 ng of total RNA were reverse transcribed to single stranded cdna using MultiScribe TM Reverse Transcriptase and random hexamer primers (Applied Biosystems Inc.; Foster City, CA, USA). Semiquantitative PCR The expression of ST6GAL1 and ST6GALNAC1-2 genes in MDAPanc-28 cells was examined by semiquantitative PCR using a MyCycler TM thermal cycler (BioRad Labs; Hercules, CA, USA). The primer sequences, annealing temperatures and products sizes were 5 -AAAAACCTTATCCCTAGGCTGC-3 (F) and 5 -TGGTAGTTTTTGTGCCCACA-3 (R), 60.8ºC, 379 bp for ST6GAL1 [26]; 5 -GGACTATGAGTGGCTGGAAGCA-3 (F) and 5 - CTGGTACAGCCGGATTATCCCT-3 (R), 65ºC, 421 bp for ST6GALNAC1 [27]; 5 - CTGCCAGTAAATTCAAGCTGC-3 (F) and 5 -TTGCTTGTGATGAATCCATAGG-3 (R), 60.8ºC, 184 bp for ST6GALNAC2 [28]; and 5 - ATCTGGCACCACACCTTCTACAATGAGCTGCG-3 (F) and 5 - CGTCATACTCCTGCTTGCTGATCCACATCTGC-3 (R), 68.2ºC, 838 bp for β-actin [29]. The intensity of the amplified cdna bands was measured using the Quantity-One software package (BioRad) and was normalized to β-actin. Data are expressed as mean ± standard deviation (SD) of the intensity values of three independent PCR assays. Quantitative real time PCR (qrt-pcr) Quantitative real time PCRs were performed in triplicate using RNA from three biological replicas. Primers and probe sequences for ST3GAL3 and β-actin genes were Custom Taqman Gene Expression Assays TM described by Pérez Garay et al. [30]. The other probes used were Hs _m1 for ST3GAL4, Hs _m1 for ST6GAL1, Hs _m1 for ST6GALNAC1, Hs _m1 for FUT1, Hs _s1 for FUT2, Hs _s1 for FUT3, Hs _s1 for FUT4, Hs _s1 for FUT5, Hs _s1 for FUT6, Hs _m1 for FUT 7 and Hs _m1 for TATA box binding protein (TBP) as

8 internal control; all from TaqMan Pre-Designed Gene Expression AssaysTM (Applied Biosystems-Applera Hispania SA; Spain). Data collection was performed using an ABI Prism 7000 Sequence Detection System. For MDAPanc-28 cells, α1,3/4-fucts expression was quantified using QuantiTect SYBR green RT-PCR (Qiagen). The primers used for amplification of FUT3, FUT4, FUT6 and FUT7 were described by Higai et al. [31], and FUT5 primers were 5 -TGGGTGTGACCTCGGCGTGA-3 (F) and 5 - AAACCAGCCTGCACCATCGCC-3 (R). Hypoxanthineguanine phosphoribosyltransferase (HPRT) mrna (Gene-Cards database, NCBI36:X) was analyzed as internal control. Data collection was performed using an ABI Prism 7900HT system. The relative concentration of each genes were calculated by the comparative Ct method. Flow cytometry analysis Detection of carbohydrate determinants was performed by indirect fluorescence as previously described, with minor modifications [30]. Briefly, 2.5 x 10 5 viable cells were incubated in the presence or absence of the corresponding antibodies diluted 1/20 in PBS- 1% Bovine Serum Albumin (BSA): Le y (clone F3), Le x (clone P12), SLe a (clone KM231) from Calbiochem (EMD Chemicals Inc.; San Diego, CA, USA) or SLe x (clone CSLEX) from BD Pharmingen (Becton and Dickinson; NJ, USA); or biotinylated lectins Sambuccus nigra agglutinin (SNA) diluted 1/20 or Maackia amurensis (MAA) diluted 1/200, from Vector Laboratories Inc. (Burlingame, CA). After a wash, cells were incubated with the secondary antibody Alexa Fluor 488 goat anti-mouse IgG or Streptavidin Alexa Fluor 488 (Invitrogen Life technologies; Frederick, MD, USA). Mean fluorescence intensity (MFI) was calculated as the quotient between the positive and negative GeoMean for each cell line. Three independent assays for each sample were performed using FACSCalibur from BD Biosciences. No aggregation or loss of viability of the cells was detected after antibody or lectin incubations.

9 Human tissue samples Pancreatic adenocarcinoma tissue samples (n=20) were obtained from the paraffinembedded tissue bank Biobanc of the Institut Hospital del Mar d Investigacions Mèdiques (IMIM), and the study was approved by the institution s ethical committee (CEIC). Tissue samples were processed in 4 µm sections and haematoxylin-eosin stained to be used for diagnostic purposes and characterization of the inflammatory component. Immunohistochemistry For immunohistochemical analysis, the primary antibodies 57/27 against SLe a and 77/180 against Le y [32] were used as supernatant at a 1/2 dilution in PBS-1% BSA, and KM93 against SLe x (Chemicon Int.; CA, USA) was used following manufacturer s instructions. The indirect immunoperoxidase technique was performed as previously described [33] and sections were developed using DAB (Dako; Glostrup, Denmark). Statistical analysis Data are expressed as mean ± SD. Normality of data was tested using the Kolmogorov- Smirnov test and the homogeneity of variances was checked using the Levene s test. Mean scores were compared with the Student s t-test using SPSS statistical software for Windows (version 15.0, SPSS Inc.; Chicago, IL, USA). The criterion for significance was set at p<0.05. Results Two pancreatic adenocarcinoma cell lines, MDAPanc-3 and MDAPanc-28, with different pattern of cell surface glycans and different expression FUT and ST gene expression were chosen to address the influence of pro-inflammatory cytokines in remodeling their glycome, especially focusing on Lewis-type and α2,6-sialylated glycans. MDAPanc-28 cells display very high levels of α2,6-sialic acid in their cell membrane, and with regards to Lewis antigens

10 expression, they express SLe x while they lack the remaining Lewis type I and type II structures. MDAPanc-3 cells, in their turn, express high levels of several Lewis antigens, such as SLe a, SLe x, Le x, and Le y, whereas α2,6-sialic acid determinants are present at low levels [14]. SLe x and α2,6-sialic acid levels are enhanced in MDAPanc-28 cells after IL-1β treatments MDAPanc-28 cells were treated with IL-1β, IL-6, IL-8 or TNFα at pm range concentration and only IL1-β produced significant changes in the cell content of sialylated glycans. In particular, after IL-1β treatments (60 pm) for 40 h, flow cytometry analyses showed a medium of 52% (range 46-58%) enhanced expression of SLe x, which represents a 1.5-fold increase in the mean fluorescence intensity (MFI) value. Moreover, α2,6-sialic acid levels (recognized by SNA lectin), which were initially very high in this cell line, increased a 19% after the treatments (18-20%), giving 1.2-fold increase in the MFI value (Figure 1A). Changes in the levels of Le x, Le y, SLe a or other α2,3-sialic acid structures detected by MAA lectin, which is unable to bind to SLe x [34], could not be detected in IL-1β treated MDAPanc- 28 cells. IL-6 and TNFα treatments modify Lewis-type antigen expression in MDAPanc-3 cells IL-6 and TNFα were the pro-inflammatory cytokines that induced changes in the membrane glycosylation pattern of MDAPanc-3 cells. Treatments with IL-6 (50 pm) for 40 h caused a significant enhancement of type II structures, SLe x (34%, range of 32-36%) and Le y (22%, range of 19-25%), with 1.3 and 1.2-fold increases in the MFI values, respectively (Figure 1B). In TNFα treatments (60 pm) for 40 h, SLe x and Le y were also overexpressed, with 1.3 and 1.2-fold increases in the MFI values, respectively (Figure 1B). No expression changes of other Lewis-type or sialylated determinants in IL-6 or TNFα treatments were detected.

11 IL-6 and TNFα treatments modify the mrna expression of several FUT and ST genes in MDAPanc-28 and MDAPanc-3 cell lines Cell RNAs were isolated at different time points of treatment (4, 12 and 24 h) to determine time course changes in the pattern of expression of specific glycosyltransferase genes which codify for the α2,3-sts, α2,6-sts, α1,2-fucts and α1,3/4-fucts participating in the final steps of Lewis-type and α2,6-sialylated glycans. The α2,3-sts that can sialylate type II structures (Galβ1-4GlcNAc) giving rise to the α2,3-sialylated precursor of SLe x antigen are ST3Gal III, ST3Gal IV and ST3Gal VI [35]. The α1,3-fucts that can fucosylate the α2,3- sialylated type II structures to render SLe x antigen are FucT III to FucT VII, which possess different efficiencies and substrate specificities [36]. FucT III and FucT V can participate in the synthesis of a broad spectrum of Lewis antigens, acting on both type I and type II chains leading to Le a, Le b and SLe a, and Le x, Le y and SLe x, respectively. FucT IV acts mainly on neutral structures giving rise to Le x ; and FucT VI can act on neutral and especially on sialylated type II structures giving rise to Le x and SLe x, respectively. FucT VII can only synthesize the SLe x antigen from the sialylated type II acceptor. Potential enzymes that sialylate in α2,6-linkage and generate Sambuccus nigra lectin (SNA) detectable glycans are ST6Gal I and ST6Gal II (the latter is mainly expressed in brain tissues in human [37]), which act preferentially on type II chains; and ST6GalNAc I and ST6GalNAc II acting on O-glycans and leading to α2,6-sialylated GalNAc, including the sialyl-tn antigen [38]. MDAPanc-28 cells The α2,3-siayltransferase genes ST3GAL3 and ST3GAL4 were analyzed in MDAPanc-28 cells. This cell line expressed much lower levels of ST3GAL4 than ST3GAL3 transcripts and did not express ST3GAL6, as previously reported by our group [14]. At 4 h of IL-1β treatment, 1.8 and 8.5-fold increases were found for ST3GAL3 and ST3GAL4 mrna levels, respectively (p<0.001); and 1.5-fold increase was detected for ST3GAL3 at 24 h (p<0.001) (Figure 2, top).

12 Regarding FUT expression, MDAPanc-28 cells showed high expression of FUT5, moderate expression of FUT7 and low expression of FUT6, whereas FUT3 and FUT4 transcripts were expressed at very low levels, in agreement with the data reported by Pérez-Garay et al. [39]. The expression of FUT5, FUT6 and FUT7 increased significantly at 4 and 24 h of IL-1β treatment (2.1 and 1.5-fold for FUT5; 1.7 and 3.9-fold for FUT6; 1.3 and 2-fold for FUT7) (p<0.001) (Figure 2 bottom), which together with the increase in the expression of ST3GAL3 and ST3GAL4 may explain the increase in SLe x membrane levels in IL-1β treated cells. In MDAPanc-28 cells, only transcript levels of ST6GAL1 were detected and these were doubled at 24 h in IL-1β treated cells (p<0.001), which may explain the corresponding raise in α2,6-sialic acid levels. MDAPanc-3 cells The mrna levels of the glycosyltransferases involved in the synthesis of SLe x and Le y antigens were analyzed in the IL-6 or TNFα treated MDAPanc-3 cells. MDAPanc-3 cells showed similar levels of ST3GAL3 and ST3GAL4 and no expression of ST3GAL6. With regards to the pattern of fucosyltransferase expression, MDAPanc-3 cells had higher expression of FUT1 than FUT2, and FUT4 was the highest expressed FUT that codifies for α1,3/4-fuct. Low levels of FUT3, FUT5, FUT6 and FUT7 mrna were detected. In IL-6 treatments, ST3GAL3 mrna levels increased significantly at 4, 12 and 24 h of treatment, although descended along time (2.2, 1.6 and 1.3-fold respectively) (p<0.01); ST3GAL4 mrna levels were enhanced at the three time points (1.6, 1.5 and 2.1-fold) (p<0.001) (Figure 3 top). At 4 h, there was a transcript decrease for the FUT genes that codify for α1,3/4-fucts, FUT3, FUT5 and FUT7, and then at 12 and 24 h the mrna levels increased until values that were similar to those of the untreated cells. FUT4 mrna levels showed a different pattern of expression, decreasing at 24 h compared to untreated cells (p<0.001) (Figure 3 middle). In contrast, FUT6 levels decreased considerably at 4 h of treatment (5.0-fold) (p<0.001) but doubled expression with respect to the untreated cells at

13 12 h and 24 h (p<0.001) (Figure 3 bottom). FUT1, which codifies for the α1,2-fuct that acts preferentially on type II structures and thus leads to the Le y epitope, showed increased mrna levels after stimulation at 4, 12 and 24 h (1.7, 2.0 and 1.7-fold) (p<0.001) (Figure 4 top); whereas the mrna levels of FUT2, which codifies for the α1,2-fuct that can act on both type I and II structures, initially decreased dramatically and then increased until reaching values that were similar to those of the untreated cells (Figure 3 middle). TNFα treatment of MDAPanc-3 cells led to enhancement of both ST3GAL3 and ST3GAL4 mrna levels at the three time points evaluated (1.5, 2.0 and 1.3-fold for ST3GAL3; 2.6, 3.3 and 2.9-fold for ST3GAL4) (p<0.001) (Figure 4 top). Similarly to IL-6 treatments, the genes that codify for α1,3/4-fucts followed heterogeneous patterns of expression. In particular, FUT3 was expressed constantly at low levels with respect to the untreated cells; FUT4 and FUT5 diminished their expression at 4 and 24 h; and FUT7 showed a significantly diminished expression at 4 and 12 h (p<0.001), and slightly increased expression at 24 h (1.2-fold) (Figure 4 middle and bottom). In contrast, FUT6 expression was again enhanced in the treated cells, especially at 12 and 24 h (1.8 and 1.4-fold) (p<0.001). Taken together these results suggest that the augment in FUT6, ST3GAL3 and ST3GAL4 transcript levels are the main contributors to the increase of SLe x in both IL-6 and TNFα treated MDAPanc-3 cells. With regards to FUT genes that codify for α1,2-fucts, FUT1 mrna levels were moderately increased at 4, 12 and 24 h of TNFα treatment (about 1.5-fold), whereas FUT2 was enhanced at 24 h of stimulation (1.8-fold) (p<0.001). In this case, both genes seem to contribute to the higher Le y expression after TNFα stimuli. Lewis antigen expression and inflammation grade in human PDAC samples Twenty pancreatic adenocarcinoma tissues with different level of differentiation (three well differentiated (WD); ten moderately differentiated (MD); five poorly differentiated (PD) and two without information about differentiation) were assayed for Lewis-type antigens expression, in particular for SLe a, SLe x and Le y. The TNM grade of the pancreatic cancer

14 tissues were: one stage IB (pt2n0), three stage IIA (pt3n0), twelve stage IIB (pt1-3n1), one stage IV (pt3nxm1), one pt3 and two samples had no stage information. The analysis of the tissue inflammation pattern was performed according to the presence of polymorphonuclear cells (PMN), lymphocytes and plasma cells in the tumor areas, which was scored in four levels (3, 2, 1 and 0) corresponding to high, moderate, slight and no expression. The tissue inflammation grade corresponded to the maximum score obtained by any of the inflammatory cells. One tissue showed grade 3 inflammation, especially due to PMN, seven tissues were grade 2 and eleven were grade 1. Only one tissue did not show any inflammatory component. Most of the PDAC samples stained positively for SLe a (19 out of 20) and SLe x (18 out of 20), and 15 out of 20 stained for Le y (Table 1). SLe a was highly expressed in 17 tumor tissues (one inflammation grade 3, seven grade 2, eight grade 1 and one grade 0) (Figure 5). Two tumor tissues showed low SLe a expression and corresponded to grade 1 inflamed tissues with a predominant PMN component. The tumor tissue with no SLe a expression was a low grade inflamed tissue. The SLe x stained tissues could be classified into high SLe x expressing (13 tissues) and low SLe x expressing (5 tissues). High SLe x expressing tissues were classified into inflammation grade 3 (1 tissue), grade 2 (5 tissues) and grade 1 (7 tissues), and the five tissues with low SLe x expression were inflammation grade 2 (2 tissues), grade 1 (2 tissues) and grade 0 (1 tissue) (Figure 5). The two SLe x negative tissues were grade 1 inflamed tissues. Le y also showed positivity in tissues of all inflamed grades. The eleven highly expressing Le y tissues were grade 2 (3 tissues), grade 1 (7 tissues), and grade 0 (1 tissue) (Figure 5). Four low expressing Le y tissues were grade 3 (1 tissue), grade 2 (1 tissue) and grade 1 (2 tissues), and the five tissues that did not stain for Le y corresponded to three tissues of grade 2 and two of grade 1.

15 Although no significant correlation could be found between inflammation grade and Lewistype antigens expression, the immunohistochemical results suggest a tendency towards inflammation being found in SLe x and SLe a positive tissues, and that these sialylated Lewis antigens tend to decrease slightly in low inflamed tissues. On the contrary, Le y is not so widely expressed and seems to predominate in low inflamed conditions (grade 1 tissues). Discussion As early as 1863, Virchow hypothesized that the origin of cancer was at sites of chronic inflammation, based on morphological observations. Nowadays it is broadly accepted that a pro-inflammatory environment influences in the progression of tumorigenesis, and in that sense tumors can be considered as wounds that fail to heal [40]. Several studies describe an increased incidence of PDAC in patients with hereditary or sporadic forms of chronic pancreatitis, and inflammatory infiltrates are frequently observed in PDAC tissues [41, 42]. The influence of inflammation in cancer can cover a wide variety of signaling pathways regulating gene expression. The present work focuses on the influence of pro-inflammatory cytokines in the expression of specific fucosyltransferase and sialyltransferase genes which codify for enzymes involved in the biosynthesis of Lewis-type and sialylated glycans in PDAC. Pro-inflammatory cytokine treatments enhance cell membrane SLe x and Le y levels Both MDAPanc-28 and MDAPanc-3 cells showed enhanced SLe x levels after treatment with pro-inflammatory cytokines. In all cases these treatments entail an increase in the ST3GAL3 and ST3GAL4 mrna levels, as well as in some FUT gene expression levels, in particular those that codify for specific α1,3/4-fuct that lead to SLe x biosynthesis. In MDAPanc-28 cells, IL-1β treatment increased ST3GAL3, ST3GAL4, FUT5, FUT6 and FUT7 mrna levels at 4 and 24 h, except for ST3GAL4, whose increase was detected only at

16 4 h. FUT3 and FUT4 mrna levels, which are expressed at very low levels, did not change along the treatment. The highest transcript increases were found for ST3GAL4 and FUT6 genes. In agreement with these results, IL-1β stimulation on hepatocellular carcinoma cells also caused increased levels of SLe x antigen and showed enhanced ST3GAL4 (at 4 h of treatment) and FUT6 (late in time) gene transcription [19, 43]. Likewise, IL-6 and TNFα caused an increase of SLe x in MDAPanc-3 cells likely explained by the enhanced expression of ST3GAL3 and ST3GAL4 genes along time course. With regards to the FUT genes, which codify for α1,3/4-fucts, FUT3, FUT4, FUT5 and FUT7 genes became downregulated at some time point and their expression was maintained or slightly downregulated with respect to the untreated cells, while FUT6, which is constitutively expressed at much higher levels than the other FUTs (FUT3, 5 and 7), increased its expression after IL-6 and TNFα treatments. Thus, in MDAPanc-3 cells the higher expression of ST3GAL3, ST3GAL4 and FUT6 may account for the higher SLe x levels at the cell membrane, in accordance with other authors. Treatment of lung cancer cells with TNFα enhanced SLe x membrane levels [44] as well as ST3GAL4 and FUT3 genes expression [45]. A moderate transcriptional up-regulation of ST3GAL3 and FUT3 mrna by TNFα was also demonstrated in cells derived from human tracheal glands [46] and of ST3GAL3 in human colon cells [31]. An increase in SLe x and 6-sulfo-SLe x epitopes in human bronchial mucosa tissues, specially located in mucins, was also observed after IL-6 or IL-8 treatments, what was mainly explained by the enhanced expression of ST3GAL6, FUT3 and FUT11 genes [22]. Although little information exists regarding the regulation of sialyltransferase and fucosyltransferase expression, the analyses of the promoter regions of ST3GAL3 and ST3GAL4 reveal the presence of consensus sites for nuclear factor kappa beta (NF-kB) and activator protein-1 (AP-1) transcription factors [47, 48]. Colomb et al. [47] identified in the upstream regulatory region of the ST3GAL4 BX transcript expressed in lung cells several consensus sites for NF-kB and AP-1, and described that TNFα controls the activity of this

17 putative promoter region. Regarding FUT6 promoter analysis, consensus sequences for NFkB or AP-1 were not detected, but they demonstrated that FUT6 gene expression is regulated by Hepatocyte Nuclear Factor-4α (HNF-4α) and Oct-1 transcription factors in HuH- 7 hepatocarcinoma cells, and that HNF-4α promoter activity can be affected by Chicken Ovalbumin Upstream Promoter Transcription Factors (COUP-TFs), whose promoter just contains NF-kB and AP-1 binding sites [47, 49]. In PDAC cells, NF-kB activation exerts an anti-apoptotic effect and is thought to play a key role in the acquisition of chemoresistance by cancer cells [50, 51]. For this reason, it has been suggested as a potential molecular target for pancreatic cancer therapy [52]. IL-6 binding to the IL-6 receptor can activate Janus kinases (JAKs)/signals transducers and activators of transcription (STATs, particularly STAT3), leading to the transcription of a key set of target genes involved in proliferation, survival, cell cycle progression, angiogenesis, and immunosupression [53]. IL-6 has been reported to play a role in cell invasion, tumor progression and chemoresistance, especially in PDAC [54, 55]. In MDAPanc-3 cells, pro-inflammatory cytokine stimuli also resulted in enhanced levels of Le y epitope at the cell membrane. FUT1 mrna was up-regulated after IL-6 and TNFα treatments along time; whereas increase in FUT2 was only detected at 24 h after TNFα treatment. A positive regulation for FUT1 and FUT2 genes was also described in MKN45 gastric cancer cells after IL-1β treatment, while IL-6 only activated FUT1 expression [25]. Interestingly, FUT1 and FUT2 genes are located at the same chromosome cluster (19q13.3) [56], what could explain a similar gene regulation. Recently, Taniuchi et al. [57] described that FUT1 promoter contains several putative binding sites for transcription factors such as Elk-1, c-rel, NF-κB, AREB6 and CREB, and demonstrated that FUT1 is transcriptionally regulated by Elk- 1 in colon cancer cells.

18 α2,6-sialic acid levels are enhanced in MDAPanc-28 cells after IL-1β treatments A significant increase in α2,6-sialic acid was observed in MDAPanc-28 cell membrane after IL-1β treatment, which can be explained by the enhanced expression of ST6GAL1 mrna levels at 24 h of stimuli. ST6GAL1 is expressed in a tissue-dependent manner, and in humans this is achieved by a different utilization of at least three physically distinct promoter regions, resulting in a family of mrnas differing only at their 5 -untranslated regions. The genomic segment 5 -adjacent to the transcription initiation point of ST6GAL1 is described to contain two potential NF-κB/Rel binding sites [58, 59], hence IL-1β may possibly modulate ST6GAL1 gene expression in these cells. In agreement with our results, TNFα and IL-1β treatments also increased ST6GAL1 mrna levels and sialyltransferase activity in human endothelial cells [60]. Several PDAC cells have been reported to express IL-1β, TNFα and IL-6 receptors through which the aforementioned pro-inflammatory cytokines can exert their effects. Taking into account all the above reported data, we could hypothesize that these effects may include the modulation of the cell glycosylation machinery. Inflammation and Lewis-type antigens expression in PDAC tissues SLe x and SLe a expression was found in the high and moderate inflamed tissues, as well as in some poor inflamed tissues. In contrast, Le y expression, which is not so generally expressed in PDAC, was found in both high and low inflamed tissues. The broad expression of SLe x and SLe a correlates with the reported high expression tendency of the α2,3-st genes ST3GAL3 and ST3GAL4 and of the α1,3/4-fut genes FUT3 and FUT6 in PDAC tissues compared to control tissues [39]. In this study, MDAPanc-3 and MDAPanc-28 PDAC cell lines treated with cytokines also up-regulated ST3GAL3, ST3GAL4 and FUT6 mrna expression; but FUT3, which was poorly expressed in these cell lines, did not show changes of expression after treatments.

19 The influence of inflammation in SLe x biosynthesis has been reported in several serum acute-phase proteins. Haptoglobin, fetuin, α1-antitrypsin and transferrin showed an increase in their SLe x content in chronic pancreatitis patients compared to control patients, while SLe x in α1-acid-glycoprotein (AGP) was increased in both chronic pancreatitis and advanced PDAC patients compared to control patients [61]. Several authors have associated the results of cytokine treatments to enhanced motility rates of tumor cells. For example, colon carcinoma cell lines showed higher adhesion to E-selectin via SLe x and SLe a epitopes after TNFα stimulation [62], as well as prostate cancer cells, which in turn had enhanced motility and invasiveness rates [20]. Furthermore, cytokines have also been described to modulate the adhesion of PDAC cells to ECM proteins [63]. In this sense, although the described changes in cell surface glycosylation and glycosyltransferase mrna levels in our cell lines upon cytokine treatments were not dramatic, it is plausible to hypothesize that the pro-inflammatory microambient may help to modulate the adhesive and invasive phenotype of PDAC cells in part through remodeling the cell glycome. In this regard, a more motile and metastatic profile for MDAPanc-28 cells transfected with ST3GAL3 or with ST3GAL4, which leads to an increase of cell surface SLe x was reported by our group [30, 39]. We also described that changes in the cell sialylation profile regulate the binding of PDAC cells to ECM components such as collagen, fibronectin and laminin [64], due to changes in α2β1 integrin glycosylation as well as the adhesiveness among cells through the modulation of E-cadherin function [65]. Conclusions The expression pattern of FUT and ST genes is tissue-, cell type- and stage-specific dependent and varies during malignant transformation [35, 36], contributing to the intrinsic heterogeneity of tumors and leading to the synthesis of tumor-associated carbohydrate antigens, such as SLe x. Pro-inflammatory cytokines act as a driving force in accelerating

20 tumor progression, up-regulating several tumor promoting genes, and here we have shown that these may include some FUT and ST genes that lead to SLe x and related carbohydrate antigens biosynthesis. This study reinforces the theory that pancreatic tumors take advantage of their cancer-associated inflammation microenvironment and that interruption of this inflammatory status could be an important strategy in attempting to improve the currently poor outcome of PDAC. Abbreviations FucT, fucosyltransferase; FUT, fucosyltransferase gene, Le x, Lewis x; Le y, Lewis y; MAA, Maackia amurensis; SLe x, sialyl-lewis x; SLe a, sialyl-lewis a; SNA, Sambuccus nigra agglutinin; ST, sialyltransferase. Competing interest The authors declare that they have no competing interests. Author s contributions Conceived and designed the experiments: SB, RP, CdB. Performed the experiments: SB, HA, LC, MRO, ELl, RP. Analyzed the data: SB, HA, LC, MRO, ELl, CdB, RP. Wrote the paper: SB, ELl, RP. Acknowledgements This work was supported by the Spanish Ministerio de Ciencia e Innovación (grant BIO , awarded to RP) and by the Fundació La Marató de TV3 (grant , awarded to RP). SB acknowledges the Government of Catalonia for a pre-doctoral fellowship.

21 We thank Dr. Marsha L. Frazier for kindly donating the MDAPanc-3 and MDAPanc-28 cell lines. References 1. Siegel R, Ma J, Zou Z, Jemal A: Cancer statistics, CA Cancer J Clin 2014, 64: Duffy JP, Eibl G, Reber HA, Hines OJ: Influence of hypoxia and neoangiogenesis on the growth of pancreatic cancer. Mol Cancer 2003, 2: Hernandez-Munoz I, Skoudy A, Real FX, Navarro P: Pancreatic ductal adenocarcinoma: cellular origin, signaling pathways and stroma contribution. Pancreatology 2008, 8: Inman KS, Francis AA, Murray NR: Complex role for the immune system in initiation and progression of pancreatic cancer. World J Gastroenterol 2014, 20: Esposito I, Menicagli M, Funel N, Bergmann F, Boggi U, Mosca F, Bevilacqua G, Campani D: Inflammatory cells contribute to the generation of an angiogenic phenotype in pancreatic ductal adenocarcinoma. J Clin Pathol 2004, 57: Mahadevan D, Von Hoff DD: Tumor-stroma interactions in pancreatic ductal adenocarcinoma. Mol Cancer Ther 2007, 6: Weber A, Wasiliew P, Kracht M: Interleukin-1 (IL-1) pathway. Sci Signal 2010, 3:cm1. 8. Karin M, Greten FR: NF-kappaB: linking inflammation and immunity to cancer development and progression. Nat Rev Immunol 2005, 5: Norman JG, Fink GW, Denham W, Yang J, Carter G, Sexton C, Falkner J, Gower WR, Franz MG: Tissue-specific cytokine production during experimental acute pancreatitis. A probable mechanism for distant organ dysfunction. Dig Dis Sci 1997, 42: Viedma JA, Perez-Mateo M, Dominguez JE, Carballo F: Role of interleukin-6 in acute pancreatitis. Comparison with C-reactive protein and phospholipase A. Gut 1992, 33:

22 11. Matull WR, Pereira SP, O'Donohue JW: Biochemical markers of acute pancreatitis. J Clin Pathol 2006, 59: Motoo Y, Xie MJ, Mouri H, Sawabu N: Expression of interleukin-8 in human obstructive pancreatitis. JOP 2004, 5: Satomura Y, Sawabu N, Takemori Y, Ohta H, Watanabe H, Okai T, Watanabe K, Matsuno H, Konishi F: Expression of various sialylated carbohydrate antigens in malignant and nonmalignant pancreatic tissues. Pancreas 1991, 6: Peracaula R, Tabares G, Lopez-Ferrer A, Brossmer R, de Bolos C, de Llorens R: Role of sialyltransferases involved in the biosynthesis of Lewis antigens in human pancreatic tumour cells. Glycoconj J 2005, 22: Nakamori S, Nishihara S, Ikehara Y, Nagano H, Dono K, Sakon M, Narimatsu H, Monden M: Molecular mechanism involved in increased expression of sialyl Lewis antigens in ductal carcinoma of the pancreas. J Exp Clin Cancer Res 1999, 18: Rambaruth ND, Dwek MV: Cell surface glycan-lectin interactions in tumor metastasis. Acta Histochem 2011, 113: Harduin-Lepers A, Krzewinski-Recchi MA, Colomb F, Foulquier F, Groux-Degroote S, Delannoy P: Sialyltransferases functions in cancers. Front Biosci (Elite Ed) 2012, 4: Dall'Olio F, Malagolini N, Trinchera M, Chiricolo M: Mechanisms of cancer-associated glycosylation changes. Front Biosci (Landmark Ed) 2012, 17: Higai K, Miyazaki N, Azuma Y, Matsumoto K: Interleukin-1beta induces sialyl Lewis X on hepatocellular carcinoma HuH-7 cells via enhanced expression of ST3Gal IV and FUT VI gene. FEBS Lett 2006, 580: Radhakrishnan P, Chachadi V, Lin MF, Singh R, Kannagi R, Cheng PW: TNFalpha enhances the motility and invasiveness of prostatic cancer cells by stimulating the expression of selective glycosyl- and sulfotransferase genes involved in the synthesis of selectin ligands. Biochem Biophys Res Commun 2011, 409:

23 21. Colomb F, Vidal O, Bobowski M, Krzewinski-Recchi MA, Harduin-Lepers A, Mensier E, Jaillard S, Lafitte JJ, Delannoy P, Groux-Degroote S: TNF induces the expression of the sialyltransferase ST3Gal IV in human bronchial mucosa via MSK1/2 protein kinases and increases FliD/sialyl-Lewisx mediated adhesion of Pseudomonas aeruginosa. Biochem J Groux-Degroote S, Krzewinski-Recchi MA, Cazet A, Vincent A, Lehoux S, Lafitte JJ, Van Seuningen I, Delannoy P: IL-6 and IL-8 increase the expression of glycosyltransferases and sulfotransferases involved in the biosynthesis of sialylated and/or sulfated Lewisx epitopes in the human bronchial mucosa. Biochem J 2008, 410: Frazier ML, Pathak S, Wang ZW, Cleary K, Singletary SE, Olive M, Mackay B, Steck PA, Levin B: Establishment of a new human pancreatic adenocarcinoma cell line, MDAPanc-3. Pancreas 1990, 5: Frazier ML, Fernandez E, de Llorens R, Brown NM, Pathak S, Cleary KR, Abbruzzese JL, Berry K, Olive M, Le Maistre A, Evans DB: Pancreatic adenocarcinoma cell line, MDAPanc-28, with features of both acinar and ductal cells. Int J Pancreatol 1996, 19: Padro M, Mejias-Luque R, Cobler L, Garrido M, Perez-Garay M, Puig S, Peracaula R, de Bolos C: Regulation of glycosyltransferases and Lewis antigens expression by IL- 1beta and IL-6 in human gastric cancer cells. Glycoconj J 2011, 28: Petretti T, Kemmner W, Schulze B, Schlag PM: Altered mrna expression of glycosyltransferases in human colorectal carcinomas and liver metastases. Gut 2000, 46: Julien S, Krzewinski-Recchi MA, Harduin-Lepers A, Gouyer V, Huet G, Le Bourhis X, Delannoy P: Expression of sialyl-tn antigen in breast cancer cells transfected with the human CMP-Neu5Ac: GalNAc alpha2,6-sialyltransferase (ST6GalNac I) cdna. Glycoconj J 2001, 18: Schneider F, Kemmner W, Haensch W, Franke G, Gretschel S, Karsten U, Schlag PM: Overexpression of sialyltransferase CMP-sialic acid:galbeta1,3galnac-r alpha6-

24 Sialyltransferase is related to poor patient survival in human colorectal carcinomas. Cancer Res 2001, 61: Mas E, Pasqualini E, Caillol N, El Battari A, Crotte C, Lombardo D, Sadoulet MO: Fucosyltransferase activities in human pancreatic tissue: comparative study between cancer tissues and established tumoral cell lines. Glycobiology 1998, 8: Perez-Garay M, Arteta B, Pages L, de Llorens R, de Bolos C, Vidal-Vanaclocha F, Peracaula R: alpha2,3-sialyltransferase ST3Gal III modulates pancreatic cancer cell motility and adhesion in vitro and enhances its metastatic potential in vivo. PLoS One 2010, Higai K, Ishihara S, Matsumoto K: NFkappaB-p65 dependent transcriptional regulation of glycosyltransferases in human colon adenocarcinoma HT-29 by stimulation with tumor necrosis factor alpha. Biol Pharm Bull 2006, 29: De Bolos C, Garrido M, Real FX: MUC6 apomucin shows a distinct normal tissue distribution that correlates with Lewis antigen expression in the human stomach. Gastroenterology 1995, 109: Lopez-Ferrer A, de Bolos C, Barranco C, Garrido M, Isern J, Carlstedt I, Reis CA, Torrado J, Real FX: Role of fucosyltransferases in the association between apomucin and Lewis antigen expression in normal and malignant gastric epithelium. Gut 2000, 47: Weiman S, Dahesh S, Carlin AF, Varki A, Nizet V, Lewis AL: Genetic and biochemical modulation of sialic acid O-acetylation on group B Streptococcus: phenotypic and functional impact. Glycobiology 2009, 19: Harduin-Lepers A, Mollicone R, Delannoy P, Oriol R: The animal sialyltransferases and sialyltransferase-related genes: a phylogenetic approach. Glycobiology 2005, 15: Ma B, Simala-Grant JL, Taylor DE: Fucosylation in prokaryotes and eukaryotes. Glycobiology 2006, 16:158R-184R.

25 37. Takashima S, Tsuji S, Tsujimoto M: Comparison of the enzymatic properties of mouse beta-galactoside alpha2,6-sialyltransferases, ST6Gal I and II. J Biochem 2003, 134: Varki A, Schauer R: Sialic Acids. In Essentials of Glycobiology. 2nd edition. Edited by Varki A, Cummings RD, Esko JD, Freeze HH, Stanley P, Bertozzi CR, Hart GW, Etzler ME. New York: Cold Spring Harbor Laboratory Press; 2009: Perez-Garay M, Arteta B, Llop E, Cobler L, Pages L, Ortiz R, Ferri MJ, de Bolos C, Figueras J, de Llorens R, Vidal-Vanaclocha F, Peracaula R: alpha2,3-sialyltransferase ST3Gal IV promotes migration and metastasis in pancreatic adenocarcinoma cells and tends to be highly expressed in pancreatic adenocarcinoma tissues. Int J Biochem Cell Biol 2013, 45: Coussens LM, Werb Z: Inflammation and cancer. Nature 2002, 420: Bracci PM, Wang F, Hassan MM, Gupta S, Li D, Holly EA: Pancreatitis and pancreatic cancer in two large pooled case-control studies. Cancer Causes Control 2009, 20: Farrow B, Evers BM: Inflammation and the development of pancreatic cancer. Surg Oncol 2002, 10: Azuma Y, Murata M, Matsumoto K: Alteration of sugar chains on alpha(1)-acid glycoprotein secreted following cytokine stimulation of HuH-7 cells in vitro. Clin Chim Acta 2000, 294: Kuninaka S, Yano T, Yokoyama H, Fukuyama Y, Terazaki Y, Uehara T, Kanematsu T, Asoh H, Ichinose Y: Direct influences of pro-inflammatory cytokines (IL-1beta, TNFalpha, IL-6) on the proliferation and cell-surface antigen expression of cancer cells. Cytokine 2000, 12: Ishibashi Y, Inouye Y, Okano T, Taniguchi A: Regulation of sialyl-lewis x epitope expression by TNF-alpha and EGF in an airway carcinoma cell line. Glycoconj J 2005, 22:53-62.

26 46. Delmotte P, Degroote S, Merten MD, Van Seuningen I, Bernigaud A, Figarella C, Roussel P, Perini JM: Influence of TNFalpha on the sialylation of mucins produced by a transformed cell line MM-39 derived from human tracheal gland cells. Glycoconj J 2001, 18: Colomb F, Krzewinski-Recchi MA, El Machhour F, Mensier E, Jaillard S, Steenackers A, Harduin-Lepers A, Lafitte JJ, Delannoy P, Groux-Degroote S: TNF regulates sialyl-lewisx and 6-sulfo-sialyl-Lewisx expression in human lung through up-regulation of ST3GAL4 transcript isoform BX. Biochimie 2012, 94: Taniguchi A, Saito K, Kubota T, Matsumoto K: Characterization of the promoter region of the human Galbeta1,3(4)GlcNAc alpha2,3-sialyltransferase III (hst3gal III) gene. Biochim Biophys Acta 2003, 1626: Higai K, Miyazaki N, Azuma Y, Matsumoto K: Transcriptional regulation of the fucosyltransferase VI gene in hepatocellular carcinoma cells. Glycoconj J 2008, 25: Neesse A, Gress TM, Michl P: Therapeutic targeting of apoptotic pathways: novel aspects in pancreatic cancer. Curr Pharm Biotechnol 2012, 13: Fujioka S, Son K, Onda S, Schmidt C, Scrabas GM, Okamoto T, Fujita T, Chiao PJ, Yanaga K: Desensitization of NFkappaB for overcoming chemoresistance of pancreatic cancer cells to TNF-alpha or paclitaxel. Anticancer Res 2012, 32: Holcomb B, Yip-Schneider M, Schmidt CM: The role of nuclear factor kappab in pancreatic cancer and the clinical applications of targeted therapy. Pancreas 2008, 36: Scheller J, Ohnesorge N, Rose-John S: Interleukin-6 trans-signalling in chronic inflammation and cancer. Scand J Immunol 2006, 63: Huang C, Yang G, Jiang T, Huang K, Cao J, Qiu Z: Effects of IL-6 and AG490 on regulation of Stat3 signaling pathway and invasion of human pancreatic cancer cells in vitro. J Exp Clin Cancer Res 2010, 29:51.

27 55. Toyonaga T, Nakano K, Nagano M, Zhao G, Yamaguchi K, Kuroki S, Eguchi T, Chijiiwa K, Tsuneyoshi M, Tanaka M: Blockade of constitutively activated Janus kinase/signal transducer and activator of transcription-3 pathway inhibits growth of human pancreatic cancer. Cancer Lett 2003, 201: Rouquier S, Lowe JB, Kelly RJ, Fertitta AL, Lennon GG, Giorgi D: Molecular cloning of a human genomic region containing the H blood group alpha(1,2)fucosyltransferase gene and two H locus-related DNA restriction fragments. Isolation of a candidate for the human Secretor blood group locus. J Biol Chem 1995, 270: Taniuchi F, Higai K, Tanaka T, Azuma Y, Matsumoto K: Transcriptional regulation of fucosyltransferase 1 gene expression in colon cancer cells. ScientificWorldJournal 2013, 2013: Wang X, Vertino A, Eddy RL, Byers MG, Jani-Sait SN, Shows TB, Lau JT: Chromosome mapping and organization of the human beta-galactoside alpha 2,6-sialyltransferase gene. Differential and cell-type specific usage of upstream exon sequences in B- lymphoblastoid cells. J Biol Chem 1993, 268: Lo NW, Lau JT: Novel heterogeneity exists in the 5'-untranslated region of the betagalactoside alpha 2,6-sialytransferase mrnas in the human B-lymphoblastoid cell line, louckes. Biochem Biophys Res Commun 1996, 228: Hanasaki K, Varki A, Stamenkovic I, Bevilacqua MP: Cytokine-induced betagalactoside alpha-2,6-sialyltransferase in human endothelial cells mediates alpha 2,6- sialylation of adhesion molecules and CD22 ligands. J Biol Chem 1994, 269: Sarrats A, Saldova R, Pla E, Fort E, Harvey DJ, Struwe WB, de Llorens R, Rudd PM, Peracaula R: Glycosylation of liver acute-phase proteins in pancreatic cancer and chronic pancreatitis. Proteomics Clin Appl 2010, 4: Majuri ML, Niemela R, Tiisala S, Renkonen O, Renkonen R: Expression and function of alpha 2,3-sialyl- and alpha 1,3/1,4-fucosyltransferases in colon adenocarcinoma cell lines: role in synthesis of E-selectin counter-receptors. Int J Cancer 1995, 63:

28 63. Stefani AL, Basso D, Panozzo MP, Greco E, Mazza S, Zancanaro F, De Franchis G, Plebani M: Cytokines modulate MIA PaCa 2 and CAPAN-1 adhesion to extracellular matrix proteins. Pancreas 1999, 19: Bassaganas S, Perez-Garay M, Peracaula R: Cell surface sialic acid modulates extracellular matrix adhesion and migration in pancreatic adenocarcinoma cells. Pancreas 2014, 43: Bassaganas S, Carvalho S, Dias AM, Perez-Garay M, Ortiz MR, Figueras J, Reis CA, Pinho SS, Peracaula R: Pancreatic cancer cell glycosylation regulates cell adhesion and invasion through the modulation of alpha2beta1 integrin and E-cadherin function. PLoS One 2014, 9:e98595.

29 Figure legends Figure 1. SLe x, Le y and α2,6-sialic acid surface levels in MDAPanc-28 and MDAPanc-3 cells after pro-inflammatory cytokines stimuli. MDAPanc-28 and MDAPanc.3 cells at subconfluence were stimulated with IL-6, IL-1β or TNFα for 40 h, and afterwards their cell membrane content on Lewis-type and sialylated glycans was analyzed by flow cytometry. A. SLe x and α2,6-sialic acid levels in MDAPanc-28 cells treated with IL-1β. B. SLe x and Le y levels in MDAPanc-3 cells treated with IL-6 or TNFα. Black line for the untreated cells; grey line for the treatments; spaced line for the control. Representative histograms are shown. Figure 2. Effect of IL-1β treatments on the relative expression of ST and FUT genes in MDAPanc-28 cells. MDAPanc-28 cells at subconfluence were stimulated with IL-1β for 4, 12 and 24 h. The mrna expression of several ST and FUT genes was determined by PCR, and the ST3GAL3, ST3GAL4, ST6GAL1, FUT5, FUT6 and FUT7 expression levels are shown. Results were normalized with actin and HPRT as internal control genes for STs and FUTs, respectively. Bars on graphs represent SD. * Significantly different when comparing to the untreated cells (p<0.01); # significantly different compared to the 0 h control (p<0.001). Figure 3. Effect of IL-6 treatments on the relative expression of STs and FucTs genes in MDAPanc-3 cells. MDAPanc-3 cells at subconfluence were stimulated with IL-6 for 4, 12 and 24 h The mrna expression of several ST and FucT genes was determined by PCR, and the ST3GAL3, ST3GAL4 and FUT1 to FUT7 mrna expression levels are represented. Results were normalized with TBP as internal control gene, and bars represent SD. * Significantly different

30 when comparing to the untreated cells (p<0.01); # significantly different compared to the 0 h control (p<0.01). Figure 4. Effect of TNFα treatments on the relative expression of STs and FucTs genes in MDAPanc-3 cells. MDAPanc-3 cells at subconfluence were stimulated with TNFα for 4, 12 and 24 h. The mrna expression of several ST and FUT genes was determined by PCR, and the ST3GAL3, ST3GAL4 and FUT1 to FUT7 mrna expression levels are represented. Results were normalized with TBP as internal control gene, and bars represent SD. * Significantly different when comparing to the untreated cells (p<0.01); # significantly different compared to the 0 h control (p<0.001). Figure 5. Representative examples of SLe a, SLe x and Le y immunostaining in PDAC tissues and their corresponding inflammation status. A. SLe a staining in a grade 1 inflamed PDAC tissue. B. Le y staining in a grade 1 inflamed PDAC tissue C. SLe a staining in a grade 2 PDAC inflamed tissue. D. SLe x staining in a grade 2 PDAC inflamed tissue. Arrows indicate lymphocytes, arrows with a ball at the base indicate polymorphonuclear (PMN) cells, and arrowheads indicate plasma cells. Original magnification (100X).

31 Figure 1

32 Figure 2

33 Figure 3

34 Figure 4

35 Figure 5