Spatial transcriptomics identified mucin-specific O-glycosylation as a key pathway in pancreatic cancer development and a promising therapeutic target

doi:10.21203/rs.3.rs-2095432/v1

Download PDF

Research Article

Spatial transcriptomics identified mucin-specific O-glycosylation as a key pathway in pancreatic cancer development and a promising therapeutic target

https://doi.org/10.21203/rs.3.rs-2095432/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

BACKGROUND.

Intraductal papillary mucinous neoplasm (IPMN) are the most prevalent pancreatic cystic neoplasms which may progress to pancreatic ductal adenocarcinoma (PDAC), the most lethal solid malignancy. Therefore, patients suffering from this condition represent the ideal population where to address the efforts of identifying prevention or interception strategies. Here, we used spatial transcriptomics on IPMNs of different grade to identify mechanisms that are associated to the progression of those lesions toward invasive carcinomas.

METHODS.

We analysed 43 IPMNs grouped according to their dysplasia grade by digital spatial whole transcriptome analysis (GeoMX Human Whole Transcriptome Atlas). The high-resolution of the technology gave us the opportunity to define the genes activated along progression of IPMN to cancer, ruling out the background given by the non-neoplastic cells. The gene signature identified was validated for expression in an external validation cohort of IPMN patients and in TCGA dataset and as therapeutic target in in vitro 3D models and in in vivo syngeneic orthotopic model of PDAC.

RESULTS.

By spatial transcriptome profiling of IPMNs with different dysplasia grades, we identified more than 3000 genes differentially expressed between LGD-, HGD- IPMNs and during transformation into invasive carcinoma. One of the top differentially regulated gene signature, mucins-specific O-Glycosylation, was both validated in a cohort of patients (n=9) by immunofluorescence (IF) analysis and in TCGA dataset. Preclinical models of pancreatic cancer, including in vitro 3D and in vivo experiments confirmed the role of GCNT3 and mucins in protecting tumor cells from T-cells recognition.

CONCLUSIONS.

We identified more than 3000 genes differentially expressed between LGD- and HGD- IPMNs and along the transformation from IPMN into invasive carcinoma. These results shed light on the role of mucin-specific O-glycosylation in the IPMN progression and in PDAC offering suitable markers for the early diagnosis. Moreover, we demonstrated in in vitro 3D models and in vivo experiment that GCNT3, the main regulator of mucins post-translational modification, is an actionable target in PDAC, paving the way for the development of novel strategies to target the protective mucin barrier to enhance PDAC chemotherapy efficacy.

Intraductal Mucinous Neoplasms (IPMNs)

Spatial transcriptomic

pancreatic ductal adenocarcinoma (PDAC)

Mucin O-Glycosylation

Intraductal papillary mucinous neoplasm (IPMN) are mucin-producing cystic neoplasms that represent putative precursors of pancreatic ductal adenocarcinoma (PDAC). The prevalence of these cysts is currently in constant increase reaching around 8% of the global population [1, 2]. The surgery is recommended for the IPMN with High Grade Dysplasia (HGD) or for the cysts located into the main duct of the pancreas. Of the other groups, Low Grade Dysplasia (LGD) IPMN patients who were assigned to clinical follow up, around 1–11% developed PDAC [3, 4].

PDAC is the third leading cause of cancer-related mortality among adults in the developed countries, with a median survival of a few months and a 5-year survival of less than 8%. It is projected to become the second cause of cancer death in Western societies within a decade [5].

Thus, the identification of earliest molecular events responsible for PDAC remains critical for early detection, prevention, and treatment interventions and also offers an opportunity to identify cancer vulnerability and new potential targets. PDAC arises from non-invasive precancerous lesions, including pancreatic intraepithelial neoplasia (PanIN) and IPMN, which are curable if detected and treated before they progress to invasive carcinoma [1, 6, 7].

The genetic and histopathological progression from PanINs to invasive carcinoma has been an intense area of investigation also thanks to the availability of several mouse models of the disease [8–10]. However, PanINs are microscopic lesions and therefore difficult to detect radiographically. On the other hand, IPMNs are larger lesions that can be detected radiographically, yet the molecular mechanisms driving the progression from these lesions to invasive carcinoma are poorly understood 11–14]. Given the prevalence of IPMNs in the general population and the possibility of their detection, the patients suffering from this condition represent the ideal population for which the identification of targetable mechanisms of progression would impact on the total PDAC burden [15].

Mucins are high molecular weight glycoproteins found in IPMN mucus. Secreted and transmembrane mucins form a physical barrier that play an important role in diverse biological functions, such as differentiation, adhesion, immune response, and cell signaling [16].

Changes in mucins synthesis and secretion were correlated with inflammation and carcinogenesis having potential value for diagnosis and follow-up of pancreatic neoplasms and for therapeutic interventions[17]. Several mucins are associated with PDAC and high-grade dyspalsia in PanIN and participate at several degrees to the malignancy [18].

The IPMN to pancreatic cancer progression is associated with local inflammation and failure of protective immunosurveillance [19, 20].

While antitumor immune components, including cytotoxic CD8⁺ T cells, infiltrates the early premalignant stages of IPMN lesions they are progressively lost during tumor progression, accompanied by the concomitant accumulation of immunosuppressive cells [21, 22].

Therefore, an immune permissive to immunosuppressive TME switch occurs during pancreatic carcinogenesis. However, the event that leads to changes of the TME during IPMN progression is just beginning to be understood. Both the cell- intrinsic and -extrinsic mechanisms were described to cooperate to escape from immunosurveillance in pancreatic cancer [26].

The immunosuppressive TME of pancreatic cancer can be largely attributed to genetic alterations that regulate the tumor microenvironment [27]. Strategies of immune evasion include also the production of post-translational modification mucin-type O-glycans on mucins as well as many other glycoproteins. This is a tumor cell-driven mechanism that pose the greatest barrier to NK and T cells and in turn to the success of immunotherapies [28, 29].

The first synthesis step of mucin O-Glycosylation is catalyzed by a family of polypeptide GalNAc-transferases attaching to specific serine and threonine residues in proteins the first carbohydrate residue [30]. GCNT3 is one of the main enzymes involved in mucin-specific O-Glycosylation modification.

In both mouse and human pancreatic cancer tumors, the glycosyltransferase Gcnt3 upregulation was correlated with increased expression of mucins [31].

Aberrant GCNT3 expression was also associated with increased mucin production, aggressive tumorigenesis, and reduced patient survival, and GCNT3 Knock-out in pancreatic cancer cells reduced proliferation and spheroid formation [31, 32]. Talniflumate, a small drug targeting GCNT3 and developed for cystic fibrosis treatment, can inhibit mucin-specific O-glycosylation and expression [31, 33].

Here, we analysed the complex tissue architecture of IPMN from 43 patients with different dysplasia grades by spatial transcriptome profiling (GeoMxwhole Human Whole Transcriptome Atlas) and we identified mucin-specific O-Glycosylation as one of the top rate gene signature differentially regulated between LGD- and HGD- IPMNs and along transformation of IPMN to invasive carcinoma. Theis shed more light on the role of mucin O-glycosylation in PDAC and in the IPMN progression, offering suitable markers for the early diagnosis. Moreover, we demonstrated with in vivo experiment that GCNT3 is an actionable target in PDAC, paving the way for the development of novel strategies to target the protective mucin barrier to enhance PDAC chemotherapy efficacy.

Patient Material

Two Tissue Macro Array (TMA) of clinically annotated IPMNs were kindly provided by Australian Pancreatic Cancer Genome Initiative (APCGI). Overall, the TMAs were composed of three normal pancreas samples, nine Low-Grade-Dysplasia (LGD), 17 Borderline, and 66 High-Grade-Dysplasia (HGD) IPMNs. TMA slides were stored prior to shipping at -80°C enclosed in a sealed bag with silica gel beads to avoid RNA oxidation and degradation. The collection and analysis of all samples were approved by the local and international ethical committee (IRB approval number n. 3536).

PDAC murine cell lines and 3D cultures

Murine PDAC cell lines DT4313, FC1242, and FC1245 were kindly provided by Dr D. Tuveson’s laboratory, at Cold Spring Harbor Laboratory (New York, USA) and Dr P. Cappello’s laboratory, at CeRMS laboratory (Turin, Italy), and passaged with original growth conditions. Cells were daily inspected and routinely tested to be mycoplasma free by PCR assay. We characterized these models in a previous paper [34]( 10.1136/jitc-2021-002876) for both transcriptome and immune profiling. From these cell lines we also established 3D cultures 13KC (DT4313), KPC06 (FC1242), KPC12 (FC1245). Briefly, after trypsinization, cell lines were resuspended in 50 μl Cultrex UltiMatrix Reduced Growth Factor Basement Membrane Extract (R&D, Minneapolis, USA) at a concentration of 3D structure (n=150) for each dome. 3D cultures were cultured in PancreaCult™ Organoid Growth Mouse Medium (STEMCELL Technologies, UK) and inspected daily.

Spatial Transcriptomics

Two IPMN TMAs obtained from APCG were analysed for Spatial Transcriptomics using GEOMX Human Whole Transcriptome Atlas (Nanostring, USA) following the provided protocol. TMAs were stained with GEOMX morphology kit to mark neoplastic cells (PanCK), and immune cells (CD45). We selected 60 ROI filtering out the areas with low cellularity (<100 nuclei) and that showed signs of tissue disruption and/or detachment. Segmentation was performed to isolate only the PanCK positive area to obtain only the IPMN specific transcriptome. GEOMX library was sequenced with NovaSeq 6000 at a coverage of 541 milion of total reads. Sequencing raw data was uploaded on Illumina BaseSpace hub and processed with Nanostring GeoMx Digital Spatial Profiling with DRAGEN to obtain .dcc files. The files were imported on R with the GeomxTools R package and quality control (QC) was performed using stringent (default) parameters. Only the 43 ROIs (6 LG, 11 Borderline, and 26 HG) that passed the QC were normalized by negative normalization. We then transformed the GeoMx dataset into a Seurat [35] to perform Differential Expression Analysis (DEA) and for data visualization. The genes resulting from DEA were used for a Gene Set Enrichment Analysis with the R package clusterProfiler interrogating the MSigDB database (www.gsea-msigdb.org). The Seurat function AddModuleScore() was used to calculate the average expression levels for the Mucin O-glycosylation Program (R-HSA-913709).

Gene expression and survival analyses

RNA counts from murine models were normalized with the DESeq2 R package. Mucin O-glycosilation gene set was obtained from MSigDB database and the normalized counts for these genes were used to perform an explorative multivariate data analysis with the R libraries FactoMineR and FactoExtra. Gene expression profiles of GCNT3, MUC1, and MUC5AC were checked on the TCGA/GTEX PAAD database with the R library UCSCXenaShiny. Prognostic significance for these genes was evaluated with Kaplan-Meier plots generated with KM-plotter [36].

Histology, immunohistochemistry (IHC), and Immunofluorescence (IF) analyses

Whole 3 µm tissue sections were dewaxed and rehydrated. Hematoxylin and Eosin was performed according to standard protocols, Mallory Trichrome stain was purchased from DIAPATH, Italy. For IHC Antigen retrieval was performed using Bond Epitope Retrieval Solution 2 (AR9640, Leica Microsystems). IF was performed using retrieval agent accordin to antibodies manufacturers’ indication. The following antibodies were used for IF analyses: MUC1 (ab109185), MUC5AC(ab3649), GCNT3(PA5-24455). Images were acquired by EVOS FLAUTO2 (Thermo Scientific, US).

For 3D cultures FFPE preparation we used the following protocol. Cells were collected with Cultrex Organoid Harvesting Solution (R&D, Minneapolis, USA) and incubated in 2% PFA for 15 mins. After a brief centrifugation, PFA was discarded, and 3D cultures were included in a mold with Histogel (Thermo Fisher Scientific, USA) and left on ice for 3 min. The included 3D cultures were then moved in a histology cassette and incubated in 2% PFA overnight at 4°C. The next day 3D cultures were included in Paraffin.

In Vitro proliferation and inhibition assays

To identify Talniflumate IC50 dose in our model 3D culture models, we plated them in 24 well plates in growth medium containing increasing concentration of drug: 50μM, 100μM, 200μM, 500μM. After 72h MTT assay (???) was used to obtain relative estimates of viable cell number according to manufacturer instruction.

To estimate Talniflumate inhibitory effect on GCNT3, MUC1, and MUC5AC expression, murine 3D cultures were plated at a concentration of 150 structures/dome and cultured in a growth medium with 100μM Talniflumate. After 48h 3D cultures were collected and included in paraffin as described above and used for IF as described above.

3D Culture-T cells interaction platform.

- T cell priming

T cells were isolated from peripheral blood samples from syngeneic C57BL/6 mice by whole blood EasySep™ MouseT Cell Isolation Kit (#17951, StemCell Technologies) following manufacturer protocol and cultured in 12 wells plate at concentration of 1*106 cells/well in 1mL of T-Cell Culture Medium: RPMI, 10% FBS, 1% L-Glutamine, 1% Hepes, 1% Penicillin/Streptomycin, 10ng/ml IL-2, 10 ng/ml IL-7. After 4h of resting in incubator T Cells were activated with Dynabeads Human T-Activator CD3/CD28 (Gibco, USA) for 24h. 3D cultures (n=150) were plated in the xenofree matrix Vitrogel ORGANOID-3 (The Well Bioscience,UK) and cultured in a medium containing 500U/mL IFNγ to stimulate epitope production. After 48h hours 3D cultures were disrupted by pipetting and added to activated T-cells for priming.

- Interaction Platform

Murine PDAC 3D cultures [1000/well] were plated in Xeno-free Vitrogel ORGANOID-3 (The Well Bioscience,UK) in a 24 wells plate. After 24h 3D cultures were stained with caspase-3/7 activity tracer CellEvent™ Caspase-3/7 Green Detection Reagent (GFP) (ThermoFisher Scientific, USA) to detect apoptosis. Educated T cells were recovered from the lysated- 3D cultures/lymphocytes admixture prepared for priming, stained with vital staining CellTracker™ Red CMPTX Dye (Texas Red) Invitrogen, USA) and added to 3D cultures. Platform was checked daily and fluorescence images were acquired with EVOS FL Auto 2 Cell Imaging System (ThermoFisher Scientific) for 72h. The GFP channel raw images were acquired to detect caspase-3/7 activity and quantified with Fiji ImagJ software and Integrated Density (IntDen) and corrected total cell fluorescence (CTCF) were calculated.

In vivo experiments

To generate syngeneic orthotopic mouse models an incision was made in the left abdominal side at the level of the spleen of C57BL/6 mice. KPC06 3D culture were suspended in Cultrex and injected into the tail region of the pancreas using insulin syringes (BD micro-fine 28 Gauge). The injection was considered successful if there were no signs of leakage and the Cultrex sphere containing 3D cultures was visible. The wound was sutured with short-term absorbable suture (Vetsuture). Following weekly manual palpation starting 10 days following transplantation, tumor-bearing mice were subjected to high-contrast ultrasound imaging using the Vevo 2100 System with a MS250, 13–24 MHz scan-head (VisualSonics, Amsterdam, The Netherlands) and randomly assigned to the experimental group in order to correct the dimension bias. Mice were treated after 14 days from tumor injection according to each group: CTR, (saline solution), Gemcitabine/Abraxane (GEM, 100mg/kg i.p. once a week for 2 weeks; ABX 10mg/Kg i.p. once a week for 2 weeks), Talniflumate (dedicated diet containing 400ppm of Talniflumate) and combination of Talniflumate with Gem/ABX. When tumor CTR reached 200mm3, all mice were sacrificed, and excised tumor growth was determined by measuring the long and short axis of the tumor by caliper. All mouse models were housed and treated in accordance with the power analysis and guidelines of the Institutional OPBA and Italian Ministry of Health Ethic Committee. The methods for animal study followed the ARRIVE Guidelines 2.0 ⁵¹.

Spatial Transcriptomics

We analyzed a total of 43 IPMNs (6 LG, 11 Borderline, and 26 HG) for whole spatial transcriptomics (Fig. 1A). To identify pathways deregulated in high-grade lesions, we performed DEA followed by GSEA (Fig. 1B and C). Mucin-specific O-glycosylation was one of the main pathway enriched in HGD IPMN. In keeping with that, both Mucins (MUC1, MUC3A, MUC5AC, MUC5B, MUC6, MUC13) and genes responsible for their O-glycosilation (GCNT3, A4GNT, B3GNT3, GALNT2, GALNT3, GALNT4) were consistently upregulated in HG IPMN (Fig. 1C). Among those genes, GCNT3, MUC1, and MUC5AC were the most upregulated in HG-IPMN (Fig. 1D). Finally, we calculated the gene set activity for the Mucin O-glycosilation pathway (R-HSA-913709) using the AddModuleScore() function and we found a consistent activation of this pathway in HG IPMN in respect to both Borderline and LG IPMNs (Fig. 1E).

GCNT3, MUC1, and MUC5AC expression in IPMN and PDAC patients

We initially confirmed that GCNT3 was expressed at higher levels in PDAC tissue with respect to normal adjacent tissue in the TCGA dataset (Fig. 2A). Furthermore, we confirmed that GCNT3 expression correlates with MUC1 and MUC5AC expression (Pearsons’ coefficient (Cor) 0,48 and 0,51 respectively) (Fig. 2B) and that all these mucins and GCNT3 were biomarkers of worse prognosis in pancreatic cancer patients (GCNT3, HR= 1.64),(MUC1, HR= 2.56), (MUC5AC, HR=2.04) (Fig. 2C). We then validated GCNT3, MUC1 and MUC5AC gene expression data on our cohort of IPMN tissues. IF analysis demonstrated that GCNT3 and the related mucins were more expressed in PDAC tissues and in the HGD, both associated and not, to pancreatic cancer tissue with respect to LGD- IPMNs (Fig. 2d).

Mucin O-glycosylation in murine models

To test the hypothesis that GCNT3 could be a promising therapeutic target, we assessed the effect of GCNT3 specific inhibitors, the talniflumate, in pancreatic cancer models with different immunogenic potential based on the ability of these models to evoke an immune response when injected in a syngeneic recipient mouse model and recently established by our group [34]. Briefly, cancer derived grafts (CDG) PDAC murine cell lines DT4313, FC1242, FC1245 were characterized by the molecular profile and TME displaying distinct immune features. The DT4313 cell line (classified as immunogenic and classical according to Moffitt and Bailey classification, respectively) has high immunogenic potential cancer cells. The FC1242 has low immunogenic potential (classified as progenitor and classical according to Moffitt and Bailey classification, respectively) and the FC1245 without any immunogenic potential (classified as ADEX and Basal according to Moffitt and Bailey classification, respectively) with typical features of a cold tumor with absent immune infiltration.

We assessed the expression of the gene involved in mucins synthesis in murine models. PCA analysis (Fig. 3A) showed that there is a considerable variance in the expression of genes involved in Mucin-specific O-glycosylation in the models. We found that GCNT3, MUC1, and MUC5AC, were consistently upregulated in the low (FC1242) or no immunogenic (FC1245) in respect to high-immunogenic (DT4313) cell line (Fig.3B). To study the role of mucins in a more realistic model where the mucins could form a physical barrier able to protect tumors and exclude or suppress immune cells, we established 3D cultures from our 2D CDG model cell lines. KC13 (DT4313), KPC06 (FC1242), and KPC12 (FC1245).

Talniflumate reduce GCNT3 and mucins expression in In Vitro experiments

To demonstrate that the specific GCNT3 inhibition affects mucins expression, the 3D models were treated with talniflumate. GCNT3 and mucins expression were evaluated by IF analysis following 96 hours of treatment. Talniflumate reduced both GCNT3 and Mucins (MUC1 and MUC5AC) expression in all our models (Fig. 4A). In vitro, activated T cells were able to recognize and kill 3D cancer cells only if this latter were pre-treated with talniflumate (Fig. 4 B ).

These data indicate that GCNT3 participates in the formation of mucin's physical barrier surrounding cancer cells and that its inhibition, disrupting this barrier, facilitates the recognition of the tumor from T cells.

Talniflumate reduce GCNT3 and mucins expression in in vivo experiments

To test in vivo the effects of talniflumate, we orthotopically transplanted into immunocompetent and syngeneic mice (n = 20) the CGD cells displaying the lowest immunogenicity (KPC06). Tumor bearing mice were randomly assigned to 4 groups (n=5) to receive saline as control (CTR), Talniflumate diet regimen (TALN), or a human standard therapy Gemcitabine/Abraxane (GEM/ABX) alone or in combination with the talniflumate diet regimen (COMB). As expected, tumor-bearing mice were relatively insensitive to the standard chemotherapy regimen (GEM/ABX) (Fig. 5 A). Talniflumate as single agent did not significantly affect tumor growth, yet tumors were on average smaller in this arm (Fig. 5A). The combination of Taniflumate with with standard therapy significantly reduced tumor growth (p<0.01) (Fig. 5 A and B). IF analysis of the excised tumors demonstrated that in vivo talniflumate reduced GCNT3 and mucins expression also as a single agent. However, only the combination treatment of talniflumate with standard therapy was able to also increase the number CD8⁺ T cells infiltrating the tumors which might explain the tumor volume shrinkage. Overall, these data indicate GCNT3 as a promising therapeutic target and the administration of talniflumate in combination with standard treatment a viable therapeutic option.

IPMN are little cysts with malignant potential into the pancreas of patients, deriving from epithelium of the main or collateral branch ducts. The prevalence of these cysts is currently in constant increase reaching around 8% of the global population [1]. The surgery is recommended for the IPMN with HGD or for the cysts into the main duct of the pancreas. Of the other group of IPMN patients who were assigned to clinical follow up, around 1–11% developed pancreatic cancer [4]. Unfortunately, pancreatic cancer has currently a 5-yrs overall survival of less than 8%, being metastatic relapse after surgery is very common, and it is projected to become the second leading cause of death in less than ten years [5]. Thus, the identification of earliest molecular events responsible for PDAC remains critical for early detection, prevention, and treatment interventions and offers also an opportunity to identify vulnerability and new potential targets.

Several papers are aimed on the identification of IPMN cancer progression proposing mutational aberration [37], extravesical (EV) proteins [38], and proteins[39], are the main factors of IPMN transformation. The main part of these studies derived from retrospective patients’ cohorts and focused on specific alteration indicative of pancreatic cancer transformation with no attempt to validate mechanisms in valid preclinical models (da mitigare).

In this paper using the GeoMx Human Whole Transcriptome Atlas (WTA) panel, which measures ~ 18,000 genes, we identified more than 3000 genes differentially expressed by LGD and HGD IPMN and activated during transformation of IPMN to invasive carcinoma. This is the first paper that links tissue IPMN architecture and dysplasia grade to genes activation. Recently, Iyer M.K., and colleagues filed in bioarchive a paper concerning the spatial transcriptome profiling for the establishing of a molecular framework for risk stratification of IPMN patients [40]. As acknowledged by the authors, the work has several important limitations. First, the CTA kit panel (8,584 independent probes targeting 1,829 unique gene identifiers) used for the analysis is very limited for the discovery purpose of the study and several of the IPMN mucin marker genes, such as MUC2, MUC5AC, and others, were not measured by this probe panel. Then, also the the number of specimen analyzed is very limited (12 FFPE) especially if we then take into consideration the number of region of interest (ROIs) selected (n = 98) the number of histologies (Pancreatobiliary, PB n = 24; Gastric Foveolar, GF n = 21; Intestinal, INT n = 38) and dysplasia of different degrees contained in them (HGD, n = 47; LGD, n = 36). Every specimen in this study contained the different areas of LGD and HGD, and it is conceivable that LGD from specimens completely free from HGD and malignant tissues may differ in gene expression. Moreover no validation of identified genes was attempted in any external/different validation cohort.

Among top rate activated gene signatures identified, we focused our attention on the “Mucin-specific O-Glycosylation” gene signature. Glycosyltransferase enzyme GCNT3, a master regulator of mucin-specific O-Glycosylation, aberrant expression was associated with increased mucins production, reduced patient survival and chemoresistance [41]. Specific inhibition by talniflumate a orally available, small-molecule inhibitor, and a potential muco-regulator that inhibits mucin synthesis, blocks mucous overproduction in patients with chronic mucous respiratory problems, and reduces cystic fibrosis [42], moreover it has been recently reported its activity in blocking pancreatic cancer cell proliferation and tumor formation [43].

Several papers reported the role of mucins in forming the physical barrier that protects normal epithelia from injury [44]. It has been reported an aberrant expression of mucins in many types of disease including solid tumors[45] and that mucins could modulate directly immune response [46–50]. It has also been demonstrated that Mucin-specific O-Glycosylation influences a broad range of signaling pathways to promote disease onset and progression sustaining proliferative and pro-tumorigenic signaling, contributing to pancreatic cancer phenotypes [51, 52]. and facilitating tumor immune escape [53]. In particular, aberrant mucins glycosylation on cancer cells lead to expression of atypical epitope resulting in a specific recognition and binding of cancer cell membrane glycosylation patterns, that trigger to apoptosis of cancer specific effector T-cells [54]. Altered glycosylation patterns affect the binding of galectin-1 to mucins [55], which hampers cytotoxic T-lymphocyte (CTL) activation and results in immunosuppression[54]. In addition these new epitopes could inhibit the cytotoxic activity of natural killer (NK) cells, and the effect is further enhanced in the presence of ammonium ions, which are known inhibitors of NK cell function [56]. To test a realistic model in which the mucins can form the protective barrier, we transformed our 2D models into 3D structures and initially validated mucins expression in these models. Then, to address the role of GCNT3 in immunosuppressive effects, we set up an in vitro recognition platform between cancer and T cells highlighting that talniflumate abrogating mucins secretion and modifications could facilitate the immune cells task. We confirmed these data also in an in vivo orthotopic syngeneic model of pancreatic cancer with low immunogenic potential.

In conclusion, this work contributes to the identification of markers of IPMN transformation into invasive carcinoma and to a new putative vulnerability target of pancreatic cancer. It focuses on mucin-specific O-glycosylation as one of the main gene signature activated along IPMN- cancer progression. Since mucins and their post-translational glycosylation play an important role in diverse biological functions, such as differentiation, adhesion, immune response, and cell signaling, they could be considered not only as a potential tool in disease prognosis but also as a therapeutic target. Overall, these results pave the way of next translational evaluation to validate the use of mucin-specific O-Glycosylation genes as a risk biomarker and for a new approach to treat PDAC that may be translated into a clinical setting.

IPMN

Intraductal mucinous Neoplasm

PDAC

Pancreatic ductal adenocarcinoma

HGD

High-Grade Dysplasia

LGD

Low-Grade Dysplasia

ROI

Region of Interest

PCA

Principal component Analysis

FFPE

Formalin Fixed Paraffin Embedded

TCGA

The Cancer Genome Atlas

Immunofluorescence

IHC

Immunohistochemistry

Cor

Pearsons’ coefficient

Hazard Ratio

CTR

Control

TALN

Talniflumate

GEM/ABX

Gemcitabine/Abraxane

COMB

Combination

Ethics approval and concept to partecipate

The experimental protocol was approved by the local ethics committee (Fondazione Policlinico Gemelli IRCCs, Ethical Committee approval Prot. Gen. 3536) and followed EU regulations

Consent for publication

Not applicable.

Avalaibility of Data and Materials

All data generated or analyzed during this study are included in this published article.

Competing interests

The authors declare no competing interests.

Funding

This work was supported by My First AIRC Grant "Luigi Bonatti e Anna Maria Bonatti Rocca,” grant number 23681 to C.C.; AIRC IG grant number 26330 to G.T., AIRC IG grant number 20583 to E.B. and Convenzione Gemelli-FIMP Progetto CUP J38D19000690001 to G.T.; Ministry of Health (CO 2019‐12369662) to G.T.,

Author Contributions

AA, GP, AE, AC, LR performed the experiments. GQ, SA, AL, IG and VC evaluated the data. FI performed and analyzed IHC; CC, AA, GT, EB contributed to the design of the experiments and to the writing of the manuscript. CC and AA designed the experiments and wrote the manuscript. All authors read and approved the final manuscript.

Acknowledgements

We thank the Australian Pancreatic Cancer Initiative (APGI) for sharing IPMN derived patient tissues; Fondazione Nadia Valsecchi; Italian pancreatic cancer Community (IPCC).

Patra KC, Bardeesy N, Mizukami Y. Diversity of Precursor Lesions For Pancreatic Cancer: The Genetics and Biology of Intraductal Papillary Mucinous Neoplasm. Clin Transl Gastroenterol. 2017;8:e86.
Keane MG, Afghani E. A Review of the Diagnosis and Management of Premalignant Pancreatic Cystic Lesions. J Clin Med. 2021;10:1284.
Levink I, Bruno M, Cahen D. Management of Intraductal Papillary Mucinous Neoplasms: Controversies in Guidelines and Future Perspectives. Curr Treat Options Gastroenterol. 2018;16:316–32.
del Chiaro M, Ateeb Z, Hansson MR, Rangelova E, Segersvärd R, Kartalis N, et al. Survival Analysis and Risk for Progression of Intraductal Papillary Mucinous Neoplasia of the Pancreas (IPMN) Under Surveillance: A Single-Institution Experience. Ann Surg Oncol. 2017;24:1120–6.
Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA Cancer J Clin. 2022;72:7–33.
Jacob HKC, Banerjee S. Intraductal Papillary Mucinous Neoplasms: Attack of the Clones. Gastroenterology. 2019;157:929–32.
Liffers S-T, Godfrey L, Frohn L, Haeberle L, Yavas A, Vesce R, et al. Molecular heterogeneity and commonalities in pancreatic cancer precursors with gastric and intestinal phenotype. Gut. 2022;gutjnl-2021-326550.
Hruban RH, Maitra A, Goggins M. Update on pancreatic intraepithelial neoplasia. Int J Clin Exp Pathol. 2008;1:306–16.
Dennaoui R, Shrestha H, Wagner K-U. Models of pancreatic ductal adenocarcinoma. Cancer Metastasis Rev. 2021;40:803–18.
Miyabayashi K, Baker LA, Deschênes A, Traub B, Caligiuri G, Plenker D, et al. Intraductal Transplantation Models of Human Pancreatic Ductal Adenocarcinoma Reveal Progressive Transition of Molecular Subtypes. Cancer Discov. 2020;10:1566–89.
Matthaei H, Wu J, Dal Molin M, Shi C, Perner S, Kristiansen G, et al. GNAS sequencing identifies IPMN-specific mutations in a subgroup of diminutive pancreatic cysts referred to as “incipient IPMNs”. Am J Surg Pathol. 2014;38:360–3.
Maker A, Katabi N, Gonen M, DeMatteo RP, D’Angelica MI, Fong Y, et al. Pancreatic cyst fluid and serum mucin levels predict dysplasia in intraductal papillary mucinous neoplasms of the pancreas. Ann Surg Oncol. 2011;18:199–206.
Rosenberger LH, Stein LH, Witkiewicz AK, Kennedy EP, Yeo CJ. Intraductal papillary mucinous neoplasm (IPMN) with extra-pancreatic mucin: a case series and review of the literature. J Gastrointest Surg. 2012;16:762–70.
Stiles ZE, Khan S, Patton KT, Jaggi M, Behrman SW, Chauhan SC. Transmembrane mucin MUC13 distinguishes intraductal papillary mucinous neoplasms from non-mucinous cysts and is associated with high-risk lesions. HPB. 2019;21:87–95.
Brosens LAA, Hackeng WM, Offerhaus GJ, Hruban RH, Wood LD. Pancreatic adenocarcinoma pathology: changing “landscape”. J Gastrointest Oncol. 2015;6:358–74.
Terris B, Dubois S, Buisine M-P, Sauvanet A, Ruszniewski P, Aubert J-P, et al. Mucin gene expression in intraductal papillary-mucinous pancreatic tumours and related lesions. J Pathol. 2002;197:632–7.
Machado NO, al Qadhi H, al Wahibi K. Intraductal Papillary Mucinous Neoplasm of Pancreas. N Am J Med Sci. 2015;7:160–75.
Niv Y. Mucin expression and the pancreas: A systematic review and meta-analysis. World J Metaanal. 2017;5:63.
Beatty PL, van der Geest R, Hashash JG, Kimura T, Gutkin D, Brand RE, et al. Immunobiology and immunosurveillance in patients with intraductal papillary mucinous neoplasms (IPMNs), premalignant precursors of pancreatic adenocarcinomas. Cancer Immunol Immunotherapy. 2016;65:771–8.
Roth S, Zamzow K, Gaida MM, Heikenwälder M, Tjaden C, Hinz U, et al. Evolution of the immune landscape during progression of pancreatic intraductal papillary mucinous neoplasms to invasive cancer. EBioMedicine. 2020;54:102714.
Bernard V, Semaan A, Huang J, San Lucas FA, Mulu FC, Stephens BM, et al. Single-Cell Transcriptomics of Pancreatic Cancer Precursors Demonstrates Epithelial and Microenvironmental Heterogeneity as an Early Event in Neoplastic Progression. Clin Cancer Res. 2019;25:2194–205.
Hiraoka N, Yamazaki–Itoh R, Ino Y, Mizuguchi Y, Yamada T, Hirohashi S, et al. CXCL17 and ICAM2 Are Associated With a Potential Anti-Tumor Immune Response in Early Intraepithelial Stages of Human Pancreatic Carcinogenesis. Gastroenterology. 2011;140:310–21.e4.
Fukunaga A, Miyamoto M, Cho Y, Murakami S, Kawarada Y, Oshikiri T, et al. CD8 + Tumor-Infiltrating Lymphocytes Together with CD4 + Tumor-Infiltrating Lymphocytes and Dendritic Cells Improve the Prognosis of Patients with Pancreatic Adenocarcinoma. Pancreas. 2004;28:e26–31.
Roth S, Zamzow K, Gaida MM, Heikenwälder M, Tjaden C, Hinz U, et al. Evolution of the immune landscape during progression of pancreatic intraductal papillary mucinous neoplasms to invasive cancer. EBioMedicine. 2020;54:102714.
Yang S, Liu Q, Liao Q. Tumor-Associated Macrophages in Pancreatic Ductal Adenocarcinoma: Origin, Polarization, Function, and Reprogramming. Front Cell Dev Biol. 2021;8.
Vonderheide RH, Bayne LJ. Inflammatory networks and immune surveillance of pancreatic carcinoma. Curr Opin Immunol. 2013;25:200–5.
Wang S, Zheng Y, Yang F, Zhu L, Zhu X-Q, Wang Z-F, et al. The molecular biology of pancreatic adenocarcinoma: translational challenges and clinical perspectives. Signal Transduct Target Ther. 2021;6:249.
Reily C, Stewart TJ, Renfrow MB, Novak J. Glycosylation in health and disease. Nat Rev Nephrol. 2019;15:346–66.
Agrawal B, Gupta N, Konowalchuk JD. MUC1 Mucin: A Putative Regulatory (Checkpoint) Molecule of T Cells. Front Immunol. 2018;9.
Tarp MA, Clausen H. Mucin-type O-glycosylation and its potential use in drug and vaccine development. Biochimica et Biophysica Acta (BBA) -. Gen Subj. 2008;1780:546–63.
Rao C, Janakiram NB, Madka V, Kumar G, Scott EJ, Pathuri G, et al. Small-Molecule Inhibition of GCNT3 Disrupts Mucin Biosynthesis and Malignant Cellular Behaviors in Pancreatic Cancer. Cancer Res. 2016;76:1965–74.
Boj SF, Hwang C-I, Baker LA, Chio IIC, Engle DD, Corbo V, et al. Organoid Models of Human and Mouse Ductal Pancreatic Cancer. Cell. 2015;160:324–38.
Walker NM, Simpson JE, Levitt RC, Boyle KT, Clarke LL. Talniflumate Increases Survival in a Cystic Fibrosis Mouse Model of Distal Intestinal Obstructive Syndrome. J Pharmacol Exp Ther. 2006;317:275–83.
Carbone C, Piro G, Agostini A, Delfino P, de Sanctis F, Nasca V, et al. Intratumoral injection of TLR9 agonist promotes an immunopermissive microenvironment transition and causes cooperative antitumor activity in combination with anti-PD1 in pancreatic cancer. J Immunother Cancer. 2021;9:e002876.
Hao Y, Hao S, Andersen-Nissen E, Mauck WM, Zheng S, Butler A, et al. Integrated analysis of multimodal single-cell data. Cell. 2021;184:3573–87.e29.
Lánczky A, Győrffy B. Web-Based Survival Analysis Tool Tailored for Medical Research (KMplot): Development and Implementation. J Med Internet Res. 2021;23:e27633.
Lee J-H, Kim Y, Choi J-W, Kim Y-SKRAS, GNAS. and RNF43 mutations in intraductal papillary mucinous neoplasm of the pancreas: a meta-analysis. Springerplus. 2016;5:1172.
Yang KS, Ciprani D, O’Shea A, Liss AS, Yang R, Fletcher-Mercaldo S, et al. Extracellular Vesicle Analysis Allows for Identification of Invasive IPMN. Gastroenterology. 2021;160:1345–58.e11.
Allen PJ, Qin L-X, Tang L, Klimstra D, Brennan MF, Lokshin A. Pancreatic Cyst Fluid Protein Expression Profiling for Discriminating Between Serous Cystadenoma and Intraductal Papillary Mucinous Neoplasm. Ann Surg. 2009;250:754–60.
Matthew K, Iyer C, Shi AM, Eckhoff A, Fletcher DP, Nussbaum PJ, Allen. Spatial Transcriptome Profiling of Intraductal Papillary Mucinous Neoplasms Establishes a Molecular Framework for Risk Stratification.
Fernández LP, Sánchez-Martínez R, Vargas T, Herranz J, Martín-Hernández R, Mendiola M, et al. The role of glycosyltransferase enzyme GCNT3 in colon and ovarian cancer prognosis and chemoresistance. Sci Rep. 2018;8:8485.
Walker NM, Simpson JE, Levitt RC, Boyle KT, Clarke LL. Talniflumate increases survival in a cystic fibrosis mouse model of distal intestinal obstructive syndrome. J Pharmacol Exp Ther. 2006;317:275–83.
Rao C, Janakiram NB, Madka V, Kumar G, Scott EJ, Pathuri G, et al. Small-Molecule Inhibition of GCNT3 Disrupts Mucin Biosynthesis and Malignant Cellular Behaviors in Pancreatic Cancer. Cancer Res. 2016;76:1965–74.
Melhem H, Regan-Komito D, Niess JH. Mucins Dynamics in Physiological and Pathological Conditions. Int J Mol Sci. 2021;22:13642.
Marimuthu S, Rauth S, Ganguly K, Zhang C, Lakshmanan I, Batra SK, et al. Mucins reprogram stemness, metabolism and promote chemoresistance during cancer progression. Cancer Metastasis Rev. 2021;40:575–88.
Apostolopoulos V, McKenzie IFC. Cellular Mucins: Targets for Immunotherapy. Crit Rev Immunol. 2017;37:421–37.
Hanson R, Hollingsworth M. Functional Consequences of Differential O-glycosylation of MUC1, MUC4, and MUC16 (Downstream Effects on Signaling). Biomolecules. 2016;6:34.
Sharma P, Allison JP. The future of immune checkpoint therapy. Science (1979). 2015;348:56–61.
Rao C, Janakiram NB, Mohammed A. Molecular Pathways: Mucins and Drug Delivery in Cancer. Clin Cancer Res. 2017;23:1373–8.
Gendler SJ. MUC1, the renaissance molecule. J Mammary Gland Biol Neoplasia. 2001;6:339–53.
Lumibao JC, Tremblay JR, Hsu J, Engle DD. Altered glycosylation in pancreatic cancer and beyond. Journal of Experimental Medicine. 2022;219.
Kaur S, Kumar S, Momi N, Sasson AR, Batra SK. Mucins in pancreatic cancer and its microenvironment. Nat Rev Gastroenterol Hepatol. 2013;10:607–20.
Varki A, Cummings RD, Esko JD, Stanley P, Hart GW, Aebi M, et al. Essentials of Glycobiology. 3rd edition. Cold Spring Harbor (NY): Cold Spring Harbor Laboratory Press; 2015.
Häuselmann I, Borsig L. Altered Tumor-Cell Glycosylation Promotes Metastasis. Front Oncol. 2014;4.
Cousin J, Cloninger M. The Role of Galectin-1 in Cancer Progression, and Synthetic Multivalent Systems for the Study of Galectin-1. Int J Mol Sci. 2016;17:1566.
Ogata S, Maimonis PJ, Itzkowitz SH. Mucins bearing the cancer-associated sialosyl-Tn antigen mediate inhibition of natural killer cell cytotoxicity. Cancer Res. 1992;52:4741–6.

Download PDF

Version 1

posted

You are reading this latest preprint version

Spatial transcriptomics identified mucin-specific O-glycosylation as a key pathway in pancreatic cancer development and a promising therapeutic target

Status:

Version 1

Abstract

Figures

Background

Methods

Results

Discussion

Conclusions

Abbreviations

Declarations

References

Status:

Version 1