- Email: [email protected]
Pathological features of primary sclerosing cholangitis identified by bile proteomic analysis
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
BBA - Molecular Basis of Disease journal homepage: www.elsevier.com/locate/bbadis
Pathological features of primary sclerosing cholangitis identified by bile proteomic analysis C. Ruppa, K.A. Bodeb, Y. Leopolda, P. Sauera, D.N. Gotthardta,⁎ a b
Department of Internal Medicine IV, University Hospital of Heidelberg, Im Neuenheimer Feld 410, 69120 Heidelberg, Germany Department of Infectious Diseases, Medical Microbiology and Hygiene, University of Heidelberg, Im Neuenheimer Feld 324, 69120 Heidelberg, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Sclerosing cholangitis Choledocholithiasis Cholangiocarcinoma Proteomics Cholestatic liver disease
Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease of unknown origin. Previous bile proteomic analyses in patients with PSC have revealed changes in disease activity specific to malignant transformation. In this study, we established a reference bile duct-derived bile proteome for PSC that can be used to evaluate biliary pathophysiology. Samples were collected from patients with PSC or with choledocholithiasis (control) (n = 6 each). Furthermore, patients with PSC-associated cholangiocarcinoma (CC) and with CC without concomitant PSC were analyzed. None of the patients showed signs of inflammation or infection based on clinical and laboratory examinations. Proteins overexpressed in patients with PSC relative to control patients were detected by two-dimensional difference gel electrophoresis and identified by liquid chromatographytandem mass spectrometry. Functional proteomic analysis was performed using STRING software. A total of 101 proteins were overexpressed in the bile fluid of patients with PSC but not in those of controls; the majority of these were predicted to be intracellular and related to the ribosomal and proteasomal pathways. On the other hand, 91 proteins were found only in the bile fluid of controls; most were derived from the extracellular space and were linked to cell adhesion, the complement system, and the coagulation cascade. In addition, proteins associated with inflammation and the innate immune response—e.g., cluster of differentiation 14, annexin-2, and components of the complement system—were upregulated in PSC. The most prominent pathways in PSC/ CC-patients were inflammation associated cytokine and chemokine pathways, whereas in CC-patients the Wnt signaling pathway was upregulated. In PSC/CC-patients DIGE-analysis revealed biliary CD14 and Annexin-4 expression, among others, as the most prominent protein that discriminates between both cohorts. Thus, the bile-duct bile proteome of patients with PSC shows disease-specific changes associated with inflammation and the innate immune response even in the absence of obvious clinical signs of cholangitis, malignancy, or inflammation. This article is part of a Special Issue entitled: Cholangiocytes in Health and Diseaseedited by Jesus Banales, Marco Marzioni and Peter Jansen.
1. Introduction Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease of unknown etiology that in most cases leads to cirrhosis [1,2]. The hallmark of the disease is multiple strictures of the bile duct system that cause intra- and/or extrahepatic cholestasis, leading to deterioration of liver function [3,4]. PSC is associated with inflammatory bowel disease—ulcerative colitis occurs in about 60%–70% of cases—and is more common in males [5]. Nonetheless, the pathophysiology of PSC at the level of hepatocyte and cholangiocyte biology is not well understood. Recent genome-wide association studies have revealed risk loci
that could shed light on disease etiology in the near future [6–8]. Several studies have used bile fluid to detect markers for disease activity and malignancy in PSC [9–12]; however, to date there is no established reference bile proteome for PSC that can be used to evaluate changes in biliary protein secretion in patients. To this end, we carried out a comprehensive analysis of bile fluid from patients with PSC without biliary infection or malignant disease in order to identify changes at the level of the biliary proteome.
Abbreviations: CC, cholangiocarcinoma; DIGE, difference gel electrophoresis; ERC, endoscopic retrograde cholangiography; PSC, primary sclerosing cholangitis; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis ⁎ Corresponding author at: Mediteo GmbH, Hauptstrasse 90, 69117 Heidelberg; Germany. E-mail address: [email protected] (D.N. Gotthardt). http://dx.doi.org/10.1016/j.bbadis.2017.09.012 Received 1 June 2017; Received in revised form 7 September 2017; Accepted 11 September 2017 0925-4439/ © 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Rupp, C., BBA - Molecular Basis of Disease (2017), http://dx.doi.org/10.1016/j.bbadis.2017.09.012
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
Table 1 Patient characteristics.
N F/M [male %] Age [years] AST [IU/l] ALT [IU/l] AP [IU/l] GGT [IU/l] Bilirubin [mg/dl] Albumin [g/dl] WBC [/nl] C-reactive protein [mg/l]
PSC
PSC/CCC
CCC
Control
Reference value
6 2/4 (66.7) 43.7 ± 11.7 57.0 ± 20.4 70.2 ± 41.7 266.3 ± 118.1 317.8 ± 239.6 1.6 ± 1.3 42.1 ± 2.8 5.8 ± 0.6 3.1 ± 1.8
6 2/4 (66.7) 51.2 ± 12.1 80.7 ± 33.1 90.9 ± 120.8 309.1 ± 193.1 330.1 ± 217.9 6.8 ± 3.9 38.9 ± 3.2 9.1 ± 2.7 26.2 ± 8.4
6 1/5 (83.3) 71.8 ± 9.2 90.0 ± 78.3 112.9 ± 58.2 241.6 ± 99.2 352.48 ± 291.6 2.3 ± 1.0 39.9 ± 2.7 8.8 ± 4.1 46.3 ± 29.0
6 4/2 (33.3) 58.1 ± 16.8 102.8 ± 88.2 219.5 ± 234.5 114.0 ± 50.0 291.2 ± 262.1 1.8 ± 1.4 40.0 ± 5.5 7.4 ± 3.1 4.9 ± 2.9
− − − − 19 IU/l − 23 IU/l 70–175 IU/l 6–28 IU/l − 1.1 mg/dl 35–55 g/dl − 10.0/nl − 5 mg/l
Values represent n (%) or mean ± SEM. ALT, alanine aminotransferase; AP, alkaline phosphatase; AST, aspartate transaminase; GGT, gamma-glutamyltransferase; F/M, female/male; PSC, primary sclerosing cholangitis; WBC, white blood cell. Fig. 1. Distribution of biliary proteins detected in patients with PSC and controls. The predominant pathways to which the majority of proteins belong are shown in boxes.
debris. Depending on the protein concentration, 25–100 μl of resincoupled antibodies against albumin and IgG (Albumin and IgG Removal kit; Amersham Biosciences, Little Chalfont, UK) were added after washing the resin with phosphate-buffered saline and incubating for 60 min. The resin was removed by centrifugation. To remove lipophilic substances, the following precipitation protocol was established based on a protocol for plant protein extraction [14]: 1 ml of 0.1 M ammonium acetate in 100% methanol was added to 200 μl of bile at − 20 °C, followed by vortexing and overnight incubation. The precipitated protein was pelleted at 13,000 × g, and after discarding the supernatant, the pellet was washed twice with 1.2 ml of 0.8 M ammonium acetate/ 80% methanol at − 20 °C, followed by two washes with 80% acetone. The resultant pellet was air dried on ice. Protein concentration was measured with the 2D-Quant kit (Amersham Biosciences).
2. Materials and methods 2.1. Patients and bile sampling during endoscopic retrograde cholangiography (ERC) Bile-duct samples (n = 24) were obtained from patients undergoing ERC after selective intubation of the papilla and prior to any therapeutic procedures or injection of radiocontrast agent. No patient had obvious haemobilia. The samples were mixed with protease inhibitors (Complete, EDTA-free; Roche, Mannheim, Germany), snap frozen in liquid nitrogen, and stored at −80 °C until use. Six of the patients had been diagnosed with choledocholithiasis and were considered as controls, six patients had PSC, six patients had CC and six patients had CC with concomitant PSC. PSC diagnosis was based on typical laboratory and ERC findings. All PSC patients were treated with ursodeoxycholic acid (UDCA). Clinical data are presented in Table 1. None of the patients had a history of variceal bleeding. The Mayo risk score was calculated for patients with PSC [13]. Samples from patients with apparent bacterial and/or fungal biliary infections were excluded. All procedures in this study complied with the Declaration of Helsinki and were approved by a local ethics committee. Patients provided written, informed consent for their participation in the study.
2.3. Two-dimensional difference gel electrophoresis (2D–DIGE) For 2D sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), the pellets were resuspended in lysis buffer containing 7 M urea, 2 M thiourea (Merck Millipore, Billerica, MA, USA), 2% (w/v) dimethyl[3-(propyl)azaniumyl]propane-1-sulfonate, 2% (v/v) Pharmalyte 3–10 (Amersham Biosciences), and protease inhibitor mixture. 50 mg protein of each sample were labeled with 200 pmol of amine reactive cyanine dyes, Cy3 or Cy5 and 50 mg protein of a mixture of both samples with Cy2 (GE Healthcare). Fluorescence labelling of the cell lysates was carried out according to the manufacturer's protocol for minimal CyDye labelling (GE Healthcare). The pooled
2.2. Sample preparation and mass spectrometry (MS) Excess albumin and IgG were removed from bile samples. A 200-μl volume of bile was centrifuged for 10 min at 13,000 × g to remove 2
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
Fig. 2. Predicted function and localization of biliary proteins in PSC and controls. A. Cellular localization. B. Biological function. C. Molecular function.
A
B
C
(Cy2, Cy3 and Cy5) samples were cup loaded on a 24 cm IPG strips, pH 3–11. Isoelectric focusing was performed using the IPGphor II isoelectric focusing system for a total of 45 kVh at 20 ̊C. Thereafter the IPG gel was incubated in equilibration buffer (50 mM Tris HCl, pH 8.8, 6 M urea, 30% glycerol, 2% SDS, 0.002% bromophenol blue) supplemented with 0.5% DTT for 15 min, followed by 4.5% iodoacetamide in fresh equilibration buffer for an additional 15 min. The strip was immediately applied on a 8%–12% SDS Gel (25 × 20 cm; Ettan Dalt Six, GE Healthcare) and the gels were run at 20 °C with a constant current of 8 mA/gel for 1 h, followed by 16 mA/gel until the end of the run [15,16]. After electrophoresis, the gels were scanned using a Typhoon image scanner. ImageQuant software was used to generate the image presentation data including the single and overlay images. Comparative analysis of all spots was performed by DeCyder analysis software. After the spot picking design using DeCyder software, protein spots of interest were automatically picked from the 2D gel with the Ettan Spot Picker, followed by Protein Identification by Mass Spectrometry. False discovery rate for detection of individual proteins was determined by DeCyder analysis software for each individual experiment. A difference in the percent volume of spots > 2 fold was considered as differential
Table 2 Biliary proteins upregulated by at least 10 fold in patients with PSC relative to controls. Protein
Accession number
Size (kDa)
Fold difference
Aldehyde dehydrogenase, dimeric NADP-preferring Myosin-9 Annexin A2 Hemoglobin subunit gamma-2 Argininosuccinate synthase Isocitrate dehydrogenase [NADP] cytoplasmic ATP synthase subunit beta, mitochondrial 6-Phosphogluconate dehydrogenase, decarboxylating Glucose-6-phosphate isomerase Endoplasmin
ALDH3A1_HUMAN
50
75
MYH9_HUMAN ANXA2_HUMAN HBG2_HUMAN ASS1_HUMAN IDH1_HUMAN
227 39 16 47 47
25 15 15 13 11
ATPB_HUMAN
57
11
PGD_HUMAN
53
10
GPI_HUMAN HSP90B1_HUMAN
63 92
10 10
3
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
normalized to the sum of total spot volumes. Data on protein and gene names, predicted molecular weights, and isoelectric points were retrieved from UniProt (www.expasy.ch) [17]. Pathways associated with each protein were determined using UniProt (www.expasy.ch), Panther v.8.1 (www.pantherdb.org), and STRING 9.05 (http://string-db.org) databases. Statistical analyses were performed using the SPSS v.21 software (SPSS Inc., Chicago, IL, USA). Continuous data were analyzed with the non-parametric Mann-Whitney U test, and Fisher's exact test was used where applicable. Statistical significance was assumed at P values < 0.05.
Table 3 Top 20 proteins specific to primary sclerosing cholangitis. Protein name
Accession number
Protein disulfide-isomerase A3 Heterogeneous nuclear ribonucleoproteins A2/B1 Annexin A1 Gastric triacylglycerol lipase Band 3 anion transport protein Proteasome activator complex subunit 2 Rab GDP dissociation inhibitor beta 40S ribosomal protein S18 High mobility group protein B1 ATP synthase subunit alpha, mitochondrial Proteasome activator complex subunit 1 Keratin, type I cytoskeletal 14 Cathepsin G Keratin, type II cytoskeletal 6C Plasma cell-induced resident endoplasmic reticulum protein Elongation factor 1-gamma Prelamin-A/C Cathepsin B Thioredoxin domain-containing protein 5 Poly(rC)-binding protein 1
PDIA3_HUMAN ROA2_HUMAN ANXA1_HUMAN LIPG_HUMAN B3AT_HUMAN PSME2_HUMAN GDIB_HUMAN RS18_HUMAN HMGB1_HUMAN ATPA_HUMAN PSME1_HUMAN K1C14_HUMAN CATG_HUMAN K2C6C_HUMAN PERP1_HUMAN EF1G_HUMAN LMNA_HUMAN CATB_HUMAN TXND5_HUMAN PCBP1_HUMAN
2.5. Peptide mass fingerprinting (PMF) Identification of SDS-PAGE-separated proteins was performed on reduced (DTT or TCEP, 10 mM in 50 mM ammoniumbicarbonate (“ABC buffer”), pH 7.5, 50 °C, 30 min), alkylated (50 mM iodoacetamide in ABC buffer, 1 h, RT), and trypsin-digested (15 ng/ul, 37 °C, o/n) samples. Proteolytic digests were loaded using a nano-HPLC (Dionex UltiMate 3000RSLCnano HPLC) on reverse phase columns (trapping column: particle size 5 μm, C18, L = 20 mm; analytical column: particle size 3 μm, C18, L = 15 cm), and eluted in gradients of water (0.1% formic acid, buffer A) and acetonitrile (0.1% formic acid, buffer B). Typically, gradients were ramped from 5% to 55% B in 50 min at flowrates of 300 nl/min. Peptides eluting from the column were analyzed in a LTQ Orbitrap Elite mass spectrometer (Thermo Scientific). Peptide identification and quantification was achieved using the MaxQuant software package (1.5.3.8) with its built-in Andromeda search algorithm. Proteins were identified by matching the derived mass lists against the NCBI nr database (downloaded from http://www. ncbi.nlm.nih.gov/) on a local Mascot server (Matrix Science, UK). In general, a mass tolerance ± 0.05 Da for parent ion and fragment
expression. Spots with a P value ≤ 0.05 were considered statistically significant and were retained for protein identification by MS.
2.4. Image processing and data analysis Digitized images of 2D gels were analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA) for spot detection, gel matching, and background correction; these data were then
Fig. 3. Pathway Analysis of biliary proteins present in patients with PSC and controls. A. PSC. B. Controls.
4
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
Fig. 3. (continued) Fig. 4. Distribution of biliary proteins detected in patients with CCC and PSC. The predominant pathways to which the majority of proteins belong are shown in boxes.
5
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
A
Fig. 5. Representative 2D CyDye-maleimide DIGE. A PSC patients were labeled with Cy3-malemide (green). B Control patients were labeled with Cy5-maleimide (red). C The overlay of PSC patients and healthy controls. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
B PSC
C
Control
Overlay
Table 4 Proteins overexpressed in patients with PSC identified by difference gel electrophoresis. Protein
Accession number
Size (kDa)
Appearance
P value (analysis of variance)
Monocyte differentiation antigen CD14 Apolipoprotein A-I Small proline-rich protein 2E Peroxiredoxin-2 Quinone oxidoreductase Hemoglobin subunit beta Hemoglobin subunit alpha Annexin A2 Hornerin Ig lambda-2 chain C regions Complement C4-A Glyceraldehyde 3-phosphate dehydrogenase
CD14_HUMAN APOA1_HUMAN SPR2E_HUMAN PRDX2_HUMAN QOR_HUMAN HBB_HUMAN HBA_HUMAN ANXA2_HUMAN HORN_HUMAN LAC2_HUMAN CO4A_HUMAN G3P_HUMAN
40 31 8 22 35 16 15 39 282 11 193 36
3.90 4.39 10.68 2.58 9.8 3.24 2.18 3.08 4.94 2.76 5.61 10.61
0.0031 0.0030 0.0023 0.010 0.019 0.019 0.024 0.029 0.031 0.033 0.034 0.035
spectra, two missed cleavages, oxidation of Met and fixed modification of carbamidomethyl cysteine were selected as matching parameters in the search program (p = 0.05, 2 peptides per protein).
laboratory characteristics of the patients are shown in Table 1. There was no clinical or laboratory evidence of severe infection or inflammation in patients with PSC or controls. Bile cultures prepared from the samples revealed an absence of bacteria and fungi. The patients were re-evaluated yearly, but none developed any malignancies (including in the pancreaticobiliary system) during the 2 years after enrollment in the study. Patients with PSC-associated CC or CC without concomitant PSC had also no clinical evidence for acute cholangitis, but marginally elevated CRP levels.
3. Results 3.1. Baseline characteristics of study population We analyzed bile duct-derived bile collected during routine ERC in 12 patients with PSC or choledocholithiasis and 12 patients with PSCassociated CC or CC without concomitant PSC. The baseline clinical and 6
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
Fig. 6. Distribution of biliary proteins detected in patients with PSC-associated CC and CC without concomitant PSC. The predominant pathways to which the majority of proteins belong are shown in boxes.
Table 5 Proteins overexpressed in patients with PSC associated CC identified by difference gel electrophoresis. Protein
Accession number
Size (kDa)
Appearance
P value (analysis of variance)
Hemoglobin subunit beta Keratin, type II cytoskeletal 1 Monocyte differentiation antigen CD14 Carbonic anhydrase 1 Annexin A4 Malate dehydrogenase, mitochondrial Ig gamma-1 chain C region Cathepsin D Keratin, type I cytoskeletal 10 Serotransferrin
HBB_HUMAN K2C1_HUMAN CD14_HUMAN CAH1_HUMAN ANXA4_HUMAN MDHM_HUMAN IGHG1_HUMAN CATD_HUMAN K1C10_HUMAN TRFE_HUMAN
16 66 40 29 36 36 36 45 59 77
2.08 3.11 2.76 3.39 2.11 2.76 2.94 3.08 2.18 2.07
0.00005 0.0031 0.0042 0.0057 0.0061 0.0094 0.0096 0.017 0.029 0.034
Fig. 7. Pathway Analysis of biliary proteins present in patients with PSC-associated CC and CC without concomitant PSC. A. PSC-associated CC B. CC without concomitant PSC.
categories. The proteins were analyzed with respect to predicted molecular and biological function and cellular components. The majority of proteins were intracellular (65.1%) (Fig. 2A). The most prominent biological processes were metabolic (25.8%), cellular (13.2%), and immune system (10.8%) (Fig. 2B). The leading molecular functions were catalytic activity (33.0%) and binding (28.7%) (Fig. 2C).
3.2. Bile duct-derived bile proteome in PSC Protein extracts from bile samples were separated by SDS-PAGE and the gels were stained with Coomassie Blue. Proteins were subjected to in-gel trypsinization and extracted peptides were analyzed by liquid chromatography–tandem MS (LC-MS/MS). A total of 425 unique proteins were identified in bile duct-derived bile from patients with PSC (Supplementary Material Table S1), of which 61 had not been previously identified in bile and 422 could be assigned to Gene Ontology 7
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
were upregulated (Fig. 6). In PSC/CC-patients LC-MS/MS and DIGEanalysis revealed biliary CD14 and Annexin-4 expression, among others, as the most prominent protein that discriminates between both cohorts (Table 5) (Fig. 7).
3.3. Differential protein expression in PSC vs. control patients We carried out a comparative analysis of bile from control and patients with PSC by LC-MS/MS and found that 331 proteins were abundant in both groups (Fig. 1). There were 10 proteins that were upregulated at least 10 fold in patients with PSC as compared to controls (Table 2). A large fraction of these were related to secretion or the processing of secreted proteins (e.g., myosin-9, annexin A2, and endoplasmin) as well as to detoxification or response to oxidative stress (e.g., aldehyde and isocitrate dehydrogenase). There were 100 proteins unique to patients with PSC—i.e., they were not detected in control bile samples (Table 3). Besides lysosomal proteins such as cathepsin B and G, many ribosomal proteins (e.g., 40S ribosomal protein S18 and 60S ribosomal protein L12) were exclusively present in PSC patient samples. Furthermore, a large proportion of proteins were implicated in immunoproteasome assembly (e.g., proteasome activator complex subunits 1 and 2 and proteasome subunit alpha type 1). This was confirmed in the pathway analysis of proteins specific to bile fluid from patients with PSC using STRING software, which revealed distinct clusters of proteasomal and ribosomal proteins (Fig. 3). Conversely, 90 proteins were detected only in controls and were absent in PSC patient samples. Most of these proteins were associated with cell adhesion (intercellular adhesion molecule 1 and integrin α1 and β1) or the complement or coagulation cascades, although mucins were also highly represented (e.g., mucin 5B, 6, and 1). Cell adhesion and extracellular matrix–receptor interactions were prominent pathways active in bile fluid from control patients.
4. Discussion In this study, we carried out a comprehensive analysis of bile ductderived bile fluid from patients with PSC. Previous studies have focused on identifying early markers of malignancy [11]; this is the first report of a PSC-specific bile proteome analysis. We found that this proteome comprised disease-specific alterations in inflammation-associated proteins even in the absence of clinical or laboratory signs of cholangitis or malignancy in the patients. Notably, markers of fibrosis were not detected, emphasizing the role of inflammation in the pathogenesis of PSC. Bile fluid is secreted by hepatocytes and continuously modified by cholangiocytes. The major constituents include bile acids, phospholipids, and lipid components, while proteins account for only 3%–5% of bile fluid volume. This makes comprehensive analysis of biliary protein profiles by MS more challenging than analyses of other body fluids such as plasma or urine [18,19]. Furthermore, collection of bile might be hampered by contamination of radiocontrast or haemobilia. We aimed to avoid this bias by collecting bile right after intubation of the papilla before papillectomy or any further invasive procedure was performed. Many serum proteins secreted by the liver (e.g., albumin) are present in bile fluid, and patients with cholestatic liver disease show distinct peptide patterns [20,21]. Several biliary proteins associated with PSC severity and prognosis such as interleukin-8 and calprotectin (S100A8/ 9) have been identified by a proteomic approach [22]. S100A8 was previously identified by our group as a biliary protein activity marker in PSC [12,23,24]. The present study adds new proteins to the PSC-associated biliary proteome; proteins exclusively expressed in bile fluid of patients with PSC but not in controls were mainly involved in antigen processing and presentation [e.g., proteasome activator complex subunit (PSME)1, PSME2, cathepsin B, and protein disulfide isomerase A3]. Experiments with transgenic mice have shown that cholangitis mediated by CD8+ T cells activated in the gut-associated lymphoid tissue is a major cause of PSC pathogenesis [25]. The TLR pathway is critically involved in the response to bacteria or pathogen-associated molecular patterns [26,27]. The binding of lipopolysaccharide-binding protein by the CD14 receptor leads to the release of inflammatory modulators [28,29]. Interestingly, we found that CD14 was markedly upregulated in the bile of patients with PSC [30] along with annexin A2, which plays an important role in host defense against infections by negatively regulating TLR4-mediated signaling and thereby protecting the host from excessive inflammatory damage [31]. Furthermore, we were able to detect several other proteins upregulated in PSC patients. However, as some results only showed marginal differences these proteins need to be analyzed in further confirmatory studies. Additionally, also clinical baseline parameters were comparable between PSC and controls patients, differences in drug therapy (e.g. UDCA in PSC) might influence bile fluid composition. Several studies have sought to identify markers of disease activity or malignancy in PSC [10,32–34]; however, these focused on advanced or late-state disease, whereas little is known about changes in biliary protein secretion during PSC progression. For this reason, we analyzed patients with a well-established diagnosis and stable disease course. We detected 61 proteins not previously detected in bile fluid, most of which were ribosomal or proteasomal. Most of the 101 proteins present exclusively in the bile fluid of patients with PSC were intracellular, whereas the majority of the 90 proteins specific to the bile fluid of controls were extracellular (e.g., mucins). The absence of these secreted proteins in patients with PSC suggests a reduced capacity for secretion of hepatocytes and/or cholangiocytes. On the other hand, the many intracellular proteins present in the bile fluid of patients with PSC may
3.4. Differential protein expression in PSC vs. CC The comparative analysis of bile from patients with PSC and patients with CC by LC-MS/MS revealed 197 proteins that were abundant in both groups. There were 50 proteins unique to patients with CC. Several mucins (e.g. Mucin-1, Mucin-5 AC, Mucin-5B, Mucin-6) were exclusively present in CC patient samples. Furthermore, many proteins were implicated in CCKR signaling pathway and FGF signaling pathway, as confirmed in the pathway analysis using STRING software. Conversely, 228 proteins were detected only in PSC patients and were absent in CC patient samples. Many of these proteins were ribosomal proteins or associated with the immunoproteasome (Fig. 4). 3.5. DIGE analysis of differentially expressed proteins Superposition of gel images from DIGE experiments generated an overlay image containing 2462 protein spots (Fig. 5). Comparison of the color intensities using imaging software revealed 94 regulated protein spots; 78 (83.0%) of these were upregulated in PSC and 16 (17.0%) were upregulated in controls. Differentially expressed proteins included human cluster of differentiation (CD)14, annexin A2, small proline-rich protein IIE, complement component 4A, NADPH:quinone reductase, gene-3 protein, hemoglobin subunits beta and alpha, immunoglobulin lambda constant 2, hornerin, peroxiredoxin 2, and apolipoprotein A1 (Table 4). Pathway analysis of proteins identified by DIGE showed overrepresentation of inflammation mediated by chemokine and cytokine, Wnt, integrin, and Toll-like receptor (TLR) signaling pathways. 3.6. Differential protein expression in cholangiocarcinoma with and without concomitant PSC A total of 326 non-redundant biliary proteins were detected in all samples. The CC-Proteome comprises 246 proteins in contrast to 311 proteins in the PSC/CC-Proteome (Suppl. Table S2). 15 proteins were only found in CC-patients and 80 proteins only in PSC/CC-patients (Fig. 5). The most prominent pathways as analyzed by STRING software in CC-patients were the Wnt signaling pathway, whereas in PSC/CCpatients inflammation associated cytokine and chemokine pathways 8
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
sclerosing cholangitis, Hepatology 21 (3) (1995) 661–667. [4] D.N. Gotthardt, G. Rudolph, P. Kloters-Plachky, H. Kulaksiz, A. Stiehl, Endoscopic dilation of dominant stenoses in primary sclerosing cholangitis: outcome after longterm treatment, Gastrointest. Endosc. 71 (3) (2010) 527–534. [5] T.J. Weismuller, P.J. Trivedi, A. Bergquist, M. Imam, H. Lenzen, C.Y. Ponsioen, et al., Patient age, sex, and inflammatory bowel disease phenotype associate with course of primary sclerosing cholangitis, Gastroenterology (2017). [6] X. Liu, P. Invernizzi, Y. Lu, R. Kosoy, I. Bianchi, M. Podda, et al., Genome-wide meta-analyses identify three loci associated with primary biliary cirrhosis, Nat. Genet. 42 (8) (2010) 658–660. [7] T. Folseraas, E. Melum, P. Rausch, B.D. Juran, E. Ellinghaus, A. Shiryaev, et al., Extended analysis of a genome-wide association study in primary sclerosing cholangitis detects multiple novel risk loci, J. Hepatol. 57 (2) (2012) 366–375. [8] D. Ellinghaus, T. Folseraas, K. Holm, E. Ellinghaus, E. Melum, T. Balschun, et al., Genome-wide association analysis in primary sclerosing cholangitis and ulcerative colitis identifies risk loci at GPR35 and TCF4, Hepatology 58 (3) (2013) 1074–1083. [9] C.J. Scarlett, A.J. Saxby, A. Nielsen, C. Bell, J.S. Samra, T. Hugh, et al., Proteomic profiling of cholangiocarcinoma: diagnostic potential of SELDI-TOF MS in malignant bile duct stricture, Hepatology 44 (3) (2006) 658–666. [10] G.K. Bonney, R.A. Craven, R. Prasad, A.F. Melcher, P.J. Selby, R.E. Banks, Circulating markers of biliary malignancy: opportunities in proteomics? Lancet Oncol. 9 (2) (2008) 149–158. [11] T.O. Lankisch, J. Metzger, A.A. Negm, K. Vosskuhl, E. Schiffer, J. Siwy, et al., Bile proteomic profiles differentiate cholangiocarcinoma from primary sclerosing cholangitis and choledocholithiasis, Hepatology 53 (3) (2011) 875–884. [12] L. Reinhard, C. Rupp, H.D. Riedel, T. Ruppert, T. Giese, C. Flechtenmacher, et al., S100A9 is a biliary protein marker of disease activity in primary sclerosing cholangitis, PLoS One 7 (1) (2012) e29821. [13] W.R. Kim, T.M. Therneau, R.H. Wiesner, J.J. Poterucha, J.T. Benson, M. Malinchoc, et al., A revised natural history model for primary sclerosing cholangitis, Mayo Clin. Proc. 75 (7) (2000) 688–694. [14] W.J. Hurkman, C.K. Tanaka, Solubilization of plant membrane proteins for analysis by two-dimensional gel electrophoresis, Plant Physiol. 81 (3) (1986) 802–806. [15] A. Gorg, C. Obermaier, G. Boguth, A. Harder, B. Scheibe, R. Wildgruber, et al., The current state of two-dimensional electrophoresis with immobilized pH gradients, Electrophoresis 21 (6) (2000) 1037–1053. [16] D. Gotthardt, V. Blancheteau, A. Bosserhoff, T. Ruppert, M. Delorenzi, T. Soldati, Proteomics fingerprinting of phagosome maturation and evidence for the role of a Galpha during uptake, Mol. Cell. Proteomics 5 (12) (2006) 2228–2243. [17] The Universal Protein Resource (UniProt) in 2010, Nucleic Acids Res. 38 (2010) D142–8 (Database issue). [18] T.Z. Kristiansen, J. Bunkenborg, M. Gronborg, H. Molina, P.J. Thuluvath, P. Argani, et al., A proteomic analysis of human bile, Mol. Cell. Proteomics 3 (7) (2004) 715–728. [19] A. Farina, J.M. Dumonceau, P. Lescuyer, Proteomic analysis of human bile and potential applications for cancer diagnosis, Expert Rev Proteomics 6 (3) (2009) 285–301. [20] V. Parviainen, S. Joenvaara, E. Tukiainen, M. Ilmakunnas, H. Isoniemi, R. Renkonen, Relative quantification of several plasma proteins during liver transplantation surgery, J Biomed Biotechnol 2011 (2011) 248613. [21] H.C. Masuoka, R. Vuppalanchi, R. Deppe, P. Bybee, M. Comerford, S. Liangpunsakul, et al., Individuals with Primary Sclerosing Cholangitis Have Elevated Levels of Biomarkers for Apoptosis but Not Necrosis, Dig. Dis. Sci. 60 (12) (2015) 3642–3646. [22] S.J. Zweers, A. Shiryaev, M. Komuta, M. Vesterhus, J.R. Hov, M.J. Perugorria, et al., Elevated interleukin-8 in bile of patients with primary sclerosing cholangitis, Liver Int. 36 (9) (2016) 1370–1377. [23] T. Voigtlander, J. Wlecke, A.A. Negm, H. Lenzen, M.P. Manns, T.O. Lankisch, Calprotectin in bile: a disease severity marker in patients with primary sclerosing cholangitis, J. Clin. Gastroenterol. 48 (10) (2014) 866–869. [24] A. Gauss, P. Sauer, A. Stiehl, C. Rupp, J. Krisam, Y. Leopold, et al., Evaluation of Biliary Calprotectin as a Biomarker in Primary Sclerosing Cholangitis, Medicine (Baltimore) 95 (17) (2016) e3510. [25] D. Seidel, I. Eickmeier, A.A. Kuhl, A. Hamann, C. Loddenkemper, E. Schott, CD8 T cells primed in the gut-associated lymphoid tissue induce immune-mediated cholangitis in mice, Hepatology 59 (2) (2014) 601–611. [26] K. Inamori, S. Ariki, S. Kawabata, A Toll-like receptor in horseshoe crabs, Immunol. Rev. 198 (2004) 106–115. [27] Z. Zhou, M. Ding, L. Huang, G. Gilkeson, R. Lang, W. Jiang, Toll-like receptormediated immune responses in intestinal macrophages; implications for mucosal immunity and autoimmune diseases, Clin. Immunol. 173 (2016) 81–86. [28] A. Haziot, N. Hijiya, S.C. Gangloff, J. Silver, S.M. Goyert, Induction of a novel mechanism of accelerated bacterial clearance by lipopolysaccharide in CD14-deficient and Toll-like receptor 4-deficient mice, J. Immunol. 166 (2) (2001) 1075–1078. [29] I. Zanoni, R. Ostuni, L.R. Marek, S. Barresi, R. Barbalat, G.M. Barton, et al., CD14 controls the LPS-induced endocytosis of Toll-like receptor 4, Cell 147 (4) (2011) 868–880. [30] K. Friedrich, M. Smit, M. Brune, T. Giese, C. Rupp, A. Wannhoff, et al., CD14 is associated with biliary stricture formation, Hepatology 64 (3) (2016) 843–852. [31] J.F. Swisher, N. Burton, S.M. Bacot, S.N. Vogel, G.M. Feldman, Annexin A2 tetramer activates human and murine macrophages through TLR4, Blood 115 (3) (2010) 549–558. [32] M. Vesterhus, A. Holm, J.R. Hov, S. Nygard, E. Schrumpf, E. Melum, et al., Novel serum and bile protein markers predict primary sclerosing cholangitis disease severity and prognosis, J. Hepatol. 66 (6) (2017) 1214–1222, http://dx.doi.org/10.
result from an increased breakdown of cholangiocytes or infiltrated immune cells [35]. It is worth noting that we analyzed patients with PSC who showed no evidence of inflammatory activity, biliary infection, or malignancy. Therefore, the observed changes in the bile fluid of patients with PSC were not attributable to infection-associated alterations or pre-malignant lesions. In addition, we found specific biliary protein pattern in patients with PSC-associated CC, compared to patients without concomitant PSC. Proteins belonging to or associated with the Wnt-pathway were highly expressed in sporadic CCs. Several studies already highlighted that presumably human CC, is a Wnt-driven tumor and probably a promising therapeutic target [36,37]. On the contrary pathogenesis of PSC-associated CC seems to be more complex and is only poorly understood. Pro-inflammatory cytokines seem to play an important role in the carcinogenic process. In line with this we were able to demonstrate inflammation associated cytokine and chemokine pathways to be more prevalent in PSC-associated CC [38]. Among others, biliary proteome analysis revealed biliary CD14 and Annexin-4 expression, as the most prominent protein that discriminates between both cohorts. Invasion of CD14 + macrophages is associated with a more invasive phenotype and poor prognosis in cholangiocarcinoma [39]. Furthermore Annexin-4 up-regulation promotes tumor progression and confers chemoresistance in several human cancers [40]. Cholangiocarcinoma is a heterogeneous group of malignancies with different features of biliary tract differentiation. Beside different histological grading systems, two main histological subtypes can be differentiated. The bile ductular type (mixed), arising from the small intrahepatic bile ducts, and the bile duct type (mucinous), arising from the larger bile ducts. In PSC, the mucinous type is more frequently found. It shares several phenotypic features with pancreatic cancers and can be preceded by pre-malignant lesions such as biliary intraepithelial neoplasm or intraductal papillary neoplasm [41]. In conclusion, we identified disease-specific alterations in the bile duct-derived bile proteome in PSC and PSC-associated CC. Our findings suggest that perturbation in cholangiocyte secretion may underlie the pathogenesis of this disorder, and provide potential therapeutic targets for PSC treatment. Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bbadis.2017.09.012. Transparency document The Transparency document associated with this article can be found, in online version. Acknowlegement We thank Prof. R. Ruppert at the ZMBH, Dept. of Proteomics, University of Heidelberg for protein sequencing. Grant support C.R. and D.G. were supported by a grant from the Deutsche Forschungsgemeinschaft (DFG) (RU 1936/1-2). Disclosures The authors have no disclosures and declare no conflicts of interest. References [1] R.W. Chapman, B.A. Arborgh, J.M. Rhodes, J.A. Summerfield, R. Dick, P.J. Scheuer, et al., Primary sclerosing cholangitis: a review of its clinical features, cholangiography, and hepatic histology, Gut 21 (10) (1980) 870–877. [2] K.N. Lazaridis, N.F. LaRusso, Primary Sclerosing Cholangitis, N. Engl. J. Med. 375 (12) (2016) 1161–1170. [3] J.G. Lee, S.M. Schutz, R.E. England, J.W. Leung, P.B. Cotton, Endoscopic therapy of
9
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al. 1016/j.jhep.2017.01.019. [33] J. Metzger, A.A. Negm, R.R. Plentz, T.J. Weismuller, J. Wedemeyer, T.H. Karlsen, et al., Urine proteomic analysis differentiates cholangiocarcinoma from primary sclerosing cholangitis and other benign biliary disorders, Gut 62 (1) (2013) 122–130. [34] U. Navaneethan, V. Lourdusamy, P. Gk Venkatesh, B. Willard, M.R. Sanaka, M.A. Parsi, Bile proteomics for differentiation of malignant from benign biliary strictures: a pilot study, Gastroenterol Rep. (Oxf). 3 (2) (2015) 136–143. [35] E. Schrumpf, C. Tan, T.H. Karlsen, J. Sponheim, N.K. Bjorkstrom, O. Sundnes, et al., The biliary epithelium presents antigens to and activates natural killer T cells, Hepatology 62 (4) (2015) 1249–1259. [36] P. Carotenuto, M. Fassan, R. Pandolfo, A. Lampis, C. Vicentini, L. Cascione, et al., Wnt signalling modulates transcribed-ultraconserved regions in hepatobiliary cancers, Gut 66 (7) (2017) 1268–1277, http://dx.doi.org/10.1136/gutjnl-2016312278. [37] L. Boulter, R.V. Guest, T.J. Kendall, D.H. Wilson, D. Wojtacha, A.J. Robson, et al.,
[38]
[39]
[40] [41]
10
WNT signaling drives cholangiocarcinoma growth and can be pharmacologically inhibited, J. Clin. Invest. 125 (3) (2015) 1269–1285. M.R. Timmer, U. Beuers, P. Fockens, C.Y. Ponsioen, E.A. Rauws, K.K. Wang, et al., Genetic and epigenetic abnormalities in primary sclerosing cholangitis-associated cholangiocarcinoma, Inflamm. Bowel Dis. 19 (8) (2013) 1789–1797. C. Subimerb, S. Pinlaor, V. Lulitanond, N. Khuntikeo, S. Okada, M.S. McGrath, et al., Circulating CD14(+) CD16(+) monocyte levels predict tissue invasive character of cholangiocarcinoma, Clin. Exp. Immunol. 161 (3) (2010) 471–479. B. Wei, C. Guo, S. Liu, M.Z. Sun, Annexin A4 and cancer, Clin. Chim. Acta 447 (2015) 72–78. J.M. Banales, V. Cardinale, G. Carpino, M. Marzioni, J.B. Andersen, P. Invernizzi, et al., Expert consensus document: Cholangiocarcinoma: current knowledge and future perspectives consensus statement from the European Network for the Study of Cholangiocarcinoma (ENS-CCA), Nat. Rev. Gastroenterol. Hepatol. 13 (5) (2016) 261–280.
Contents lists available at ScienceDirect
BBA - Molecular Basis of Disease journal homepage: www.elsevier.com/locate/bbadis
Pathological features of primary sclerosing cholangitis identified by bile proteomic analysis C. Ruppa, K.A. Bodeb, Y. Leopolda, P. Sauera, D.N. Gotthardta,⁎ a b
Department of Internal Medicine IV, University Hospital of Heidelberg, Im Neuenheimer Feld 410, 69120 Heidelberg, Germany Department of Infectious Diseases, Medical Microbiology and Hygiene, University of Heidelberg, Im Neuenheimer Feld 324, 69120 Heidelberg, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Sclerosing cholangitis Choledocholithiasis Cholangiocarcinoma Proteomics Cholestatic liver disease
Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease of unknown origin. Previous bile proteomic analyses in patients with PSC have revealed changes in disease activity specific to malignant transformation. In this study, we established a reference bile duct-derived bile proteome for PSC that can be used to evaluate biliary pathophysiology. Samples were collected from patients with PSC or with choledocholithiasis (control) (n = 6 each). Furthermore, patients with PSC-associated cholangiocarcinoma (CC) and with CC without concomitant PSC were analyzed. None of the patients showed signs of inflammation or infection based on clinical and laboratory examinations. Proteins overexpressed in patients with PSC relative to control patients were detected by two-dimensional difference gel electrophoresis and identified by liquid chromatographytandem mass spectrometry. Functional proteomic analysis was performed using STRING software. A total of 101 proteins were overexpressed in the bile fluid of patients with PSC but not in those of controls; the majority of these were predicted to be intracellular and related to the ribosomal and proteasomal pathways. On the other hand, 91 proteins were found only in the bile fluid of controls; most were derived from the extracellular space and were linked to cell adhesion, the complement system, and the coagulation cascade. In addition, proteins associated with inflammation and the innate immune response—e.g., cluster of differentiation 14, annexin-2, and components of the complement system—were upregulated in PSC. The most prominent pathways in PSC/ CC-patients were inflammation associated cytokine and chemokine pathways, whereas in CC-patients the Wnt signaling pathway was upregulated. In PSC/CC-patients DIGE-analysis revealed biliary CD14 and Annexin-4 expression, among others, as the most prominent protein that discriminates between both cohorts. Thus, the bile-duct bile proteome of patients with PSC shows disease-specific changes associated with inflammation and the innate immune response even in the absence of obvious clinical signs of cholangitis, malignancy, or inflammation. This article is part of a Special Issue entitled: Cholangiocytes in Health and Diseaseedited by Jesus Banales, Marco Marzioni and Peter Jansen.
1. Introduction Primary sclerosing cholangitis (PSC) is a chronic cholestatic liver disease of unknown etiology that in most cases leads to cirrhosis [1,2]. The hallmark of the disease is multiple strictures of the bile duct system that cause intra- and/or extrahepatic cholestasis, leading to deterioration of liver function [3,4]. PSC is associated with inflammatory bowel disease—ulcerative colitis occurs in about 60%–70% of cases—and is more common in males [5]. Nonetheless, the pathophysiology of PSC at the level of hepatocyte and cholangiocyte biology is not well understood. Recent genome-wide association studies have revealed risk loci
that could shed light on disease etiology in the near future [6–8]. Several studies have used bile fluid to detect markers for disease activity and malignancy in PSC [9–12]; however, to date there is no established reference bile proteome for PSC that can be used to evaluate changes in biliary protein secretion in patients. To this end, we carried out a comprehensive analysis of bile fluid from patients with PSC without biliary infection or malignant disease in order to identify changes at the level of the biliary proteome.
Abbreviations: CC, cholangiocarcinoma; DIGE, difference gel electrophoresis; ERC, endoscopic retrograde cholangiography; PSC, primary sclerosing cholangitis; SDS-PAGE, sodium dodecyl sulfate polyacrylamide gel electrophoresis ⁎ Corresponding author at: Mediteo GmbH, Hauptstrasse 90, 69117 Heidelberg; Germany. E-mail address: [email protected] (D.N. Gotthardt). http://dx.doi.org/10.1016/j.bbadis.2017.09.012 Received 1 June 2017; Received in revised form 7 September 2017; Accepted 11 September 2017 0925-4439/ © 2017 Elsevier B.V. All rights reserved.
Please cite this article as: Rupp, C., BBA - Molecular Basis of Disease (2017), http://dx.doi.org/10.1016/j.bbadis.2017.09.012
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
Table 1 Patient characteristics.
N F/M [male %] Age [years] AST [IU/l] ALT [IU/l] AP [IU/l] GGT [IU/l] Bilirubin [mg/dl] Albumin [g/dl] WBC [/nl] C-reactive protein [mg/l]
PSC
PSC/CCC
CCC
Control
Reference value
6 2/4 (66.7) 43.7 ± 11.7 57.0 ± 20.4 70.2 ± 41.7 266.3 ± 118.1 317.8 ± 239.6 1.6 ± 1.3 42.1 ± 2.8 5.8 ± 0.6 3.1 ± 1.8
6 2/4 (66.7) 51.2 ± 12.1 80.7 ± 33.1 90.9 ± 120.8 309.1 ± 193.1 330.1 ± 217.9 6.8 ± 3.9 38.9 ± 3.2 9.1 ± 2.7 26.2 ± 8.4
6 1/5 (83.3) 71.8 ± 9.2 90.0 ± 78.3 112.9 ± 58.2 241.6 ± 99.2 352.48 ± 291.6 2.3 ± 1.0 39.9 ± 2.7 8.8 ± 4.1 46.3 ± 29.0
6 4/2 (33.3) 58.1 ± 16.8 102.8 ± 88.2 219.5 ± 234.5 114.0 ± 50.0 291.2 ± 262.1 1.8 ± 1.4 40.0 ± 5.5 7.4 ± 3.1 4.9 ± 2.9
− − − − 19 IU/l − 23 IU/l 70–175 IU/l 6–28 IU/l − 1.1 mg/dl 35–55 g/dl − 10.0/nl − 5 mg/l
Values represent n (%) or mean ± SEM. ALT, alanine aminotransferase; AP, alkaline phosphatase; AST, aspartate transaminase; GGT, gamma-glutamyltransferase; F/M, female/male; PSC, primary sclerosing cholangitis; WBC, white blood cell. Fig. 1. Distribution of biliary proteins detected in patients with PSC and controls. The predominant pathways to which the majority of proteins belong are shown in boxes.
debris. Depending on the protein concentration, 25–100 μl of resincoupled antibodies against albumin and IgG (Albumin and IgG Removal kit; Amersham Biosciences, Little Chalfont, UK) were added after washing the resin with phosphate-buffered saline and incubating for 60 min. The resin was removed by centrifugation. To remove lipophilic substances, the following precipitation protocol was established based on a protocol for plant protein extraction [14]: 1 ml of 0.1 M ammonium acetate in 100% methanol was added to 200 μl of bile at − 20 °C, followed by vortexing and overnight incubation. The precipitated protein was pelleted at 13,000 × g, and after discarding the supernatant, the pellet was washed twice with 1.2 ml of 0.8 M ammonium acetate/ 80% methanol at − 20 °C, followed by two washes with 80% acetone. The resultant pellet was air dried on ice. Protein concentration was measured with the 2D-Quant kit (Amersham Biosciences).
2. Materials and methods 2.1. Patients and bile sampling during endoscopic retrograde cholangiography (ERC) Bile-duct samples (n = 24) were obtained from patients undergoing ERC after selective intubation of the papilla and prior to any therapeutic procedures or injection of radiocontrast agent. No patient had obvious haemobilia. The samples were mixed with protease inhibitors (Complete, EDTA-free; Roche, Mannheim, Germany), snap frozen in liquid nitrogen, and stored at −80 °C until use. Six of the patients had been diagnosed with choledocholithiasis and were considered as controls, six patients had PSC, six patients had CC and six patients had CC with concomitant PSC. PSC diagnosis was based on typical laboratory and ERC findings. All PSC patients were treated with ursodeoxycholic acid (UDCA). Clinical data are presented in Table 1. None of the patients had a history of variceal bleeding. The Mayo risk score was calculated for patients with PSC [13]. Samples from patients with apparent bacterial and/or fungal biliary infections were excluded. All procedures in this study complied with the Declaration of Helsinki and were approved by a local ethics committee. Patients provided written, informed consent for their participation in the study.
2.3. Two-dimensional difference gel electrophoresis (2D–DIGE) For 2D sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), the pellets were resuspended in lysis buffer containing 7 M urea, 2 M thiourea (Merck Millipore, Billerica, MA, USA), 2% (w/v) dimethyl[3-(propyl)azaniumyl]propane-1-sulfonate, 2% (v/v) Pharmalyte 3–10 (Amersham Biosciences), and protease inhibitor mixture. 50 mg protein of each sample were labeled with 200 pmol of amine reactive cyanine dyes, Cy3 or Cy5 and 50 mg protein of a mixture of both samples with Cy2 (GE Healthcare). Fluorescence labelling of the cell lysates was carried out according to the manufacturer's protocol for minimal CyDye labelling (GE Healthcare). The pooled
2.2. Sample preparation and mass spectrometry (MS) Excess albumin and IgG were removed from bile samples. A 200-μl volume of bile was centrifuged for 10 min at 13,000 × g to remove 2
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
Fig. 2. Predicted function and localization of biliary proteins in PSC and controls. A. Cellular localization. B. Biological function. C. Molecular function.
A
B
C
(Cy2, Cy3 and Cy5) samples were cup loaded on a 24 cm IPG strips, pH 3–11. Isoelectric focusing was performed using the IPGphor II isoelectric focusing system for a total of 45 kVh at 20 ̊C. Thereafter the IPG gel was incubated in equilibration buffer (50 mM Tris HCl, pH 8.8, 6 M urea, 30% glycerol, 2% SDS, 0.002% bromophenol blue) supplemented with 0.5% DTT for 15 min, followed by 4.5% iodoacetamide in fresh equilibration buffer for an additional 15 min. The strip was immediately applied on a 8%–12% SDS Gel (25 × 20 cm; Ettan Dalt Six, GE Healthcare) and the gels were run at 20 °C with a constant current of 8 mA/gel for 1 h, followed by 16 mA/gel until the end of the run [15,16]. After electrophoresis, the gels were scanned using a Typhoon image scanner. ImageQuant software was used to generate the image presentation data including the single and overlay images. Comparative analysis of all spots was performed by DeCyder analysis software. After the spot picking design using DeCyder software, protein spots of interest were automatically picked from the 2D gel with the Ettan Spot Picker, followed by Protein Identification by Mass Spectrometry. False discovery rate for detection of individual proteins was determined by DeCyder analysis software for each individual experiment. A difference in the percent volume of spots > 2 fold was considered as differential
Table 2 Biliary proteins upregulated by at least 10 fold in patients with PSC relative to controls. Protein
Accession number
Size (kDa)
Fold difference
Aldehyde dehydrogenase, dimeric NADP-preferring Myosin-9 Annexin A2 Hemoglobin subunit gamma-2 Argininosuccinate synthase Isocitrate dehydrogenase [NADP] cytoplasmic ATP synthase subunit beta, mitochondrial 6-Phosphogluconate dehydrogenase, decarboxylating Glucose-6-phosphate isomerase Endoplasmin
ALDH3A1_HUMAN
50
75
MYH9_HUMAN ANXA2_HUMAN HBG2_HUMAN ASS1_HUMAN IDH1_HUMAN
227 39 16 47 47
25 15 15 13 11
ATPB_HUMAN
57
11
PGD_HUMAN
53
10
GPI_HUMAN HSP90B1_HUMAN
63 92
10 10
3
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
normalized to the sum of total spot volumes. Data on protein and gene names, predicted molecular weights, and isoelectric points were retrieved from UniProt (www.expasy.ch) [17]. Pathways associated with each protein were determined using UniProt (www.expasy.ch), Panther v.8.1 (www.pantherdb.org), and STRING 9.05 (http://string-db.org) databases. Statistical analyses were performed using the SPSS v.21 software (SPSS Inc., Chicago, IL, USA). Continuous data were analyzed with the non-parametric Mann-Whitney U test, and Fisher's exact test was used where applicable. Statistical significance was assumed at P values < 0.05.
Table 3 Top 20 proteins specific to primary sclerosing cholangitis. Protein name
Accession number
Protein disulfide-isomerase A3 Heterogeneous nuclear ribonucleoproteins A2/B1 Annexin A1 Gastric triacylglycerol lipase Band 3 anion transport protein Proteasome activator complex subunit 2 Rab GDP dissociation inhibitor beta 40S ribosomal protein S18 High mobility group protein B1 ATP synthase subunit alpha, mitochondrial Proteasome activator complex subunit 1 Keratin, type I cytoskeletal 14 Cathepsin G Keratin, type II cytoskeletal 6C Plasma cell-induced resident endoplasmic reticulum protein Elongation factor 1-gamma Prelamin-A/C Cathepsin B Thioredoxin domain-containing protein 5 Poly(rC)-binding protein 1
PDIA3_HUMAN ROA2_HUMAN ANXA1_HUMAN LIPG_HUMAN B3AT_HUMAN PSME2_HUMAN GDIB_HUMAN RS18_HUMAN HMGB1_HUMAN ATPA_HUMAN PSME1_HUMAN K1C14_HUMAN CATG_HUMAN K2C6C_HUMAN PERP1_HUMAN EF1G_HUMAN LMNA_HUMAN CATB_HUMAN TXND5_HUMAN PCBP1_HUMAN
2.5. Peptide mass fingerprinting (PMF) Identification of SDS-PAGE-separated proteins was performed on reduced (DTT or TCEP, 10 mM in 50 mM ammoniumbicarbonate (“ABC buffer”), pH 7.5, 50 °C, 30 min), alkylated (50 mM iodoacetamide in ABC buffer, 1 h, RT), and trypsin-digested (15 ng/ul, 37 °C, o/n) samples. Proteolytic digests were loaded using a nano-HPLC (Dionex UltiMate 3000RSLCnano HPLC) on reverse phase columns (trapping column: particle size 5 μm, C18, L = 20 mm; analytical column: particle size 3 μm, C18, L = 15 cm), and eluted in gradients of water (0.1% formic acid, buffer A) and acetonitrile (0.1% formic acid, buffer B). Typically, gradients were ramped from 5% to 55% B in 50 min at flowrates of 300 nl/min. Peptides eluting from the column were analyzed in a LTQ Orbitrap Elite mass spectrometer (Thermo Scientific). Peptide identification and quantification was achieved using the MaxQuant software package (1.5.3.8) with its built-in Andromeda search algorithm. Proteins were identified by matching the derived mass lists against the NCBI nr database (downloaded from http://www. ncbi.nlm.nih.gov/) on a local Mascot server (Matrix Science, UK). In general, a mass tolerance ± 0.05 Da for parent ion and fragment
expression. Spots with a P value ≤ 0.05 were considered statistically significant and were retained for protein identification by MS.
2.4. Image processing and data analysis Digitized images of 2D gels were analyzed using ImageJ software (National Institutes of Health, Bethesda, MD, USA) for spot detection, gel matching, and background correction; these data were then
Fig. 3. Pathway Analysis of biliary proteins present in patients with PSC and controls. A. PSC. B. Controls.
4
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
Fig. 3. (continued) Fig. 4. Distribution of biliary proteins detected in patients with CCC and PSC. The predominant pathways to which the majority of proteins belong are shown in boxes.
5
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
A
Fig. 5. Representative 2D CyDye-maleimide DIGE. A PSC patients were labeled with Cy3-malemide (green). B Control patients were labeled with Cy5-maleimide (red). C The overlay of PSC patients and healthy controls. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
B PSC
C
Control
Overlay
Table 4 Proteins overexpressed in patients with PSC identified by difference gel electrophoresis. Protein
Accession number
Size (kDa)
Appearance
P value (analysis of variance)
Monocyte differentiation antigen CD14 Apolipoprotein A-I Small proline-rich protein 2E Peroxiredoxin-2 Quinone oxidoreductase Hemoglobin subunit beta Hemoglobin subunit alpha Annexin A2 Hornerin Ig lambda-2 chain C regions Complement C4-A Glyceraldehyde 3-phosphate dehydrogenase
CD14_HUMAN APOA1_HUMAN SPR2E_HUMAN PRDX2_HUMAN QOR_HUMAN HBB_HUMAN HBA_HUMAN ANXA2_HUMAN HORN_HUMAN LAC2_HUMAN CO4A_HUMAN G3P_HUMAN
40 31 8 22 35 16 15 39 282 11 193 36
3.90 4.39 10.68 2.58 9.8 3.24 2.18 3.08 4.94 2.76 5.61 10.61
0.0031 0.0030 0.0023 0.010 0.019 0.019 0.024 0.029 0.031 0.033 0.034 0.035
spectra, two missed cleavages, oxidation of Met and fixed modification of carbamidomethyl cysteine were selected as matching parameters in the search program (p = 0.05, 2 peptides per protein).
laboratory characteristics of the patients are shown in Table 1. There was no clinical or laboratory evidence of severe infection or inflammation in patients with PSC or controls. Bile cultures prepared from the samples revealed an absence of bacteria and fungi. The patients were re-evaluated yearly, but none developed any malignancies (including in the pancreaticobiliary system) during the 2 years after enrollment in the study. Patients with PSC-associated CC or CC without concomitant PSC had also no clinical evidence for acute cholangitis, but marginally elevated CRP levels.
3. Results 3.1. Baseline characteristics of study population We analyzed bile duct-derived bile collected during routine ERC in 12 patients with PSC or choledocholithiasis and 12 patients with PSCassociated CC or CC without concomitant PSC. The baseline clinical and 6
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
Fig. 6. Distribution of biliary proteins detected in patients with PSC-associated CC and CC without concomitant PSC. The predominant pathways to which the majority of proteins belong are shown in boxes.
Table 5 Proteins overexpressed in patients with PSC associated CC identified by difference gel electrophoresis. Protein
Accession number
Size (kDa)
Appearance
P value (analysis of variance)
Hemoglobin subunit beta Keratin, type II cytoskeletal 1 Monocyte differentiation antigen CD14 Carbonic anhydrase 1 Annexin A4 Malate dehydrogenase, mitochondrial Ig gamma-1 chain C region Cathepsin D Keratin, type I cytoskeletal 10 Serotransferrin
HBB_HUMAN K2C1_HUMAN CD14_HUMAN CAH1_HUMAN ANXA4_HUMAN MDHM_HUMAN IGHG1_HUMAN CATD_HUMAN K1C10_HUMAN TRFE_HUMAN
16 66 40 29 36 36 36 45 59 77
2.08 3.11 2.76 3.39 2.11 2.76 2.94 3.08 2.18 2.07
0.00005 0.0031 0.0042 0.0057 0.0061 0.0094 0.0096 0.017 0.029 0.034
Fig. 7. Pathway Analysis of biliary proteins present in patients with PSC-associated CC and CC without concomitant PSC. A. PSC-associated CC B. CC without concomitant PSC.
categories. The proteins were analyzed with respect to predicted molecular and biological function and cellular components. The majority of proteins were intracellular (65.1%) (Fig. 2A). The most prominent biological processes were metabolic (25.8%), cellular (13.2%), and immune system (10.8%) (Fig. 2B). The leading molecular functions were catalytic activity (33.0%) and binding (28.7%) (Fig. 2C).
3.2. Bile duct-derived bile proteome in PSC Protein extracts from bile samples were separated by SDS-PAGE and the gels were stained with Coomassie Blue. Proteins were subjected to in-gel trypsinization and extracted peptides were analyzed by liquid chromatography–tandem MS (LC-MS/MS). A total of 425 unique proteins were identified in bile duct-derived bile from patients with PSC (Supplementary Material Table S1), of which 61 had not been previously identified in bile and 422 could be assigned to Gene Ontology 7
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
were upregulated (Fig. 6). In PSC/CC-patients LC-MS/MS and DIGEanalysis revealed biliary CD14 and Annexin-4 expression, among others, as the most prominent protein that discriminates between both cohorts (Table 5) (Fig. 7).
3.3. Differential protein expression in PSC vs. control patients We carried out a comparative analysis of bile from control and patients with PSC by LC-MS/MS and found that 331 proteins were abundant in both groups (Fig. 1). There were 10 proteins that were upregulated at least 10 fold in patients with PSC as compared to controls (Table 2). A large fraction of these were related to secretion or the processing of secreted proteins (e.g., myosin-9, annexin A2, and endoplasmin) as well as to detoxification or response to oxidative stress (e.g., aldehyde and isocitrate dehydrogenase). There were 100 proteins unique to patients with PSC—i.e., they were not detected in control bile samples (Table 3). Besides lysosomal proteins such as cathepsin B and G, many ribosomal proteins (e.g., 40S ribosomal protein S18 and 60S ribosomal protein L12) were exclusively present in PSC patient samples. Furthermore, a large proportion of proteins were implicated in immunoproteasome assembly (e.g., proteasome activator complex subunits 1 and 2 and proteasome subunit alpha type 1). This was confirmed in the pathway analysis of proteins specific to bile fluid from patients with PSC using STRING software, which revealed distinct clusters of proteasomal and ribosomal proteins (Fig. 3). Conversely, 90 proteins were detected only in controls and were absent in PSC patient samples. Most of these proteins were associated with cell adhesion (intercellular adhesion molecule 1 and integrin α1 and β1) or the complement or coagulation cascades, although mucins were also highly represented (e.g., mucin 5B, 6, and 1). Cell adhesion and extracellular matrix–receptor interactions were prominent pathways active in bile fluid from control patients.
4. Discussion In this study, we carried out a comprehensive analysis of bile ductderived bile fluid from patients with PSC. Previous studies have focused on identifying early markers of malignancy [11]; this is the first report of a PSC-specific bile proteome analysis. We found that this proteome comprised disease-specific alterations in inflammation-associated proteins even in the absence of clinical or laboratory signs of cholangitis or malignancy in the patients. Notably, markers of fibrosis were not detected, emphasizing the role of inflammation in the pathogenesis of PSC. Bile fluid is secreted by hepatocytes and continuously modified by cholangiocytes. The major constituents include bile acids, phospholipids, and lipid components, while proteins account for only 3%–5% of bile fluid volume. This makes comprehensive analysis of biliary protein profiles by MS more challenging than analyses of other body fluids such as plasma or urine [18,19]. Furthermore, collection of bile might be hampered by contamination of radiocontrast or haemobilia. We aimed to avoid this bias by collecting bile right after intubation of the papilla before papillectomy or any further invasive procedure was performed. Many serum proteins secreted by the liver (e.g., albumin) are present in bile fluid, and patients with cholestatic liver disease show distinct peptide patterns [20,21]. Several biliary proteins associated with PSC severity and prognosis such as interleukin-8 and calprotectin (S100A8/ 9) have been identified by a proteomic approach [22]. S100A8 was previously identified by our group as a biliary protein activity marker in PSC [12,23,24]. The present study adds new proteins to the PSC-associated biliary proteome; proteins exclusively expressed in bile fluid of patients with PSC but not in controls were mainly involved in antigen processing and presentation [e.g., proteasome activator complex subunit (PSME)1, PSME2, cathepsin B, and protein disulfide isomerase A3]. Experiments with transgenic mice have shown that cholangitis mediated by CD8+ T cells activated in the gut-associated lymphoid tissue is a major cause of PSC pathogenesis [25]. The TLR pathway is critically involved in the response to bacteria or pathogen-associated molecular patterns [26,27]. The binding of lipopolysaccharide-binding protein by the CD14 receptor leads to the release of inflammatory modulators [28,29]. Interestingly, we found that CD14 was markedly upregulated in the bile of patients with PSC [30] along with annexin A2, which plays an important role in host defense against infections by negatively regulating TLR4-mediated signaling and thereby protecting the host from excessive inflammatory damage [31]. Furthermore, we were able to detect several other proteins upregulated in PSC patients. However, as some results only showed marginal differences these proteins need to be analyzed in further confirmatory studies. Additionally, also clinical baseline parameters were comparable between PSC and controls patients, differences in drug therapy (e.g. UDCA in PSC) might influence bile fluid composition. Several studies have sought to identify markers of disease activity or malignancy in PSC [10,32–34]; however, these focused on advanced or late-state disease, whereas little is known about changes in biliary protein secretion during PSC progression. For this reason, we analyzed patients with a well-established diagnosis and stable disease course. We detected 61 proteins not previously detected in bile fluid, most of which were ribosomal or proteasomal. Most of the 101 proteins present exclusively in the bile fluid of patients with PSC were intracellular, whereas the majority of the 90 proteins specific to the bile fluid of controls were extracellular (e.g., mucins). The absence of these secreted proteins in patients with PSC suggests a reduced capacity for secretion of hepatocytes and/or cholangiocytes. On the other hand, the many intracellular proteins present in the bile fluid of patients with PSC may
3.4. Differential protein expression in PSC vs. CC The comparative analysis of bile from patients with PSC and patients with CC by LC-MS/MS revealed 197 proteins that were abundant in both groups. There were 50 proteins unique to patients with CC. Several mucins (e.g. Mucin-1, Mucin-5 AC, Mucin-5B, Mucin-6) were exclusively present in CC patient samples. Furthermore, many proteins were implicated in CCKR signaling pathway and FGF signaling pathway, as confirmed in the pathway analysis using STRING software. Conversely, 228 proteins were detected only in PSC patients and were absent in CC patient samples. Many of these proteins were ribosomal proteins or associated with the immunoproteasome (Fig. 4). 3.5. DIGE analysis of differentially expressed proteins Superposition of gel images from DIGE experiments generated an overlay image containing 2462 protein spots (Fig. 5). Comparison of the color intensities using imaging software revealed 94 regulated protein spots; 78 (83.0%) of these were upregulated in PSC and 16 (17.0%) were upregulated in controls. Differentially expressed proteins included human cluster of differentiation (CD)14, annexin A2, small proline-rich protein IIE, complement component 4A, NADPH:quinone reductase, gene-3 protein, hemoglobin subunits beta and alpha, immunoglobulin lambda constant 2, hornerin, peroxiredoxin 2, and apolipoprotein A1 (Table 4). Pathway analysis of proteins identified by DIGE showed overrepresentation of inflammation mediated by chemokine and cytokine, Wnt, integrin, and Toll-like receptor (TLR) signaling pathways. 3.6. Differential protein expression in cholangiocarcinoma with and without concomitant PSC A total of 326 non-redundant biliary proteins were detected in all samples. The CC-Proteome comprises 246 proteins in contrast to 311 proteins in the PSC/CC-Proteome (Suppl. Table S2). 15 proteins were only found in CC-patients and 80 proteins only in PSC/CC-patients (Fig. 5). The most prominent pathways as analyzed by STRING software in CC-patients were the Wnt signaling pathway, whereas in PSC/CCpatients inflammation associated cytokine and chemokine pathways 8
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al.
sclerosing cholangitis, Hepatology 21 (3) (1995) 661–667. [4] D.N. Gotthardt, G. Rudolph, P. Kloters-Plachky, H. Kulaksiz, A. Stiehl, Endoscopic dilation of dominant stenoses in primary sclerosing cholangitis: outcome after longterm treatment, Gastrointest. Endosc. 71 (3) (2010) 527–534. [5] T.J. Weismuller, P.J. Trivedi, A. Bergquist, M. Imam, H. Lenzen, C.Y. Ponsioen, et al., Patient age, sex, and inflammatory bowel disease phenotype associate with course of primary sclerosing cholangitis, Gastroenterology (2017). [6] X. Liu, P. Invernizzi, Y. Lu, R. Kosoy, I. Bianchi, M. Podda, et al., Genome-wide meta-analyses identify three loci associated with primary biliary cirrhosis, Nat. Genet. 42 (8) (2010) 658–660. [7] T. Folseraas, E. Melum, P. Rausch, B.D. Juran, E. Ellinghaus, A. Shiryaev, et al., Extended analysis of a genome-wide association study in primary sclerosing cholangitis detects multiple novel risk loci, J. Hepatol. 57 (2) (2012) 366–375. [8] D. Ellinghaus, T. Folseraas, K. Holm, E. Ellinghaus, E. Melum, T. Balschun, et al., Genome-wide association analysis in primary sclerosing cholangitis and ulcerative colitis identifies risk loci at GPR35 and TCF4, Hepatology 58 (3) (2013) 1074–1083. [9] C.J. Scarlett, A.J. Saxby, A. Nielsen, C. Bell, J.S. Samra, T. Hugh, et al., Proteomic profiling of cholangiocarcinoma: diagnostic potential of SELDI-TOF MS in malignant bile duct stricture, Hepatology 44 (3) (2006) 658–666. [10] G.K. Bonney, R.A. Craven, R. Prasad, A.F. Melcher, P.J. Selby, R.E. Banks, Circulating markers of biliary malignancy: opportunities in proteomics? Lancet Oncol. 9 (2) (2008) 149–158. [11] T.O. Lankisch, J. Metzger, A.A. Negm, K. Vosskuhl, E. Schiffer, J. Siwy, et al., Bile proteomic profiles differentiate cholangiocarcinoma from primary sclerosing cholangitis and choledocholithiasis, Hepatology 53 (3) (2011) 875–884. [12] L. Reinhard, C. Rupp, H.D. Riedel, T. Ruppert, T. Giese, C. Flechtenmacher, et al., S100A9 is a biliary protein marker of disease activity in primary sclerosing cholangitis, PLoS One 7 (1) (2012) e29821. [13] W.R. Kim, T.M. Therneau, R.H. Wiesner, J.J. Poterucha, J.T. Benson, M. Malinchoc, et al., A revised natural history model for primary sclerosing cholangitis, Mayo Clin. Proc. 75 (7) (2000) 688–694. [14] W.J. Hurkman, C.K. Tanaka, Solubilization of plant membrane proteins for analysis by two-dimensional gel electrophoresis, Plant Physiol. 81 (3) (1986) 802–806. [15] A. Gorg, C. Obermaier, G. Boguth, A. Harder, B. Scheibe, R. Wildgruber, et al., The current state of two-dimensional electrophoresis with immobilized pH gradients, Electrophoresis 21 (6) (2000) 1037–1053. [16] D. Gotthardt, V. Blancheteau, A. Bosserhoff, T. Ruppert, M. Delorenzi, T. Soldati, Proteomics fingerprinting of phagosome maturation and evidence for the role of a Galpha during uptake, Mol. Cell. Proteomics 5 (12) (2006) 2228–2243. [17] The Universal Protein Resource (UniProt) in 2010, Nucleic Acids Res. 38 (2010) D142–8 (Database issue). [18] T.Z. Kristiansen, J. Bunkenborg, M. Gronborg, H. Molina, P.J. Thuluvath, P. Argani, et al., A proteomic analysis of human bile, Mol. Cell. Proteomics 3 (7) (2004) 715–728. [19] A. Farina, J.M. Dumonceau, P. Lescuyer, Proteomic analysis of human bile and potential applications for cancer diagnosis, Expert Rev Proteomics 6 (3) (2009) 285–301. [20] V. Parviainen, S. Joenvaara, E. Tukiainen, M. Ilmakunnas, H. Isoniemi, R. Renkonen, Relative quantification of several plasma proteins during liver transplantation surgery, J Biomed Biotechnol 2011 (2011) 248613. [21] H.C. Masuoka, R. Vuppalanchi, R. Deppe, P. Bybee, M. Comerford, S. Liangpunsakul, et al., Individuals with Primary Sclerosing Cholangitis Have Elevated Levels of Biomarkers for Apoptosis but Not Necrosis, Dig. Dis. Sci. 60 (12) (2015) 3642–3646. [22] S.J. Zweers, A. Shiryaev, M. Komuta, M. Vesterhus, J.R. Hov, M.J. Perugorria, et al., Elevated interleukin-8 in bile of patients with primary sclerosing cholangitis, Liver Int. 36 (9) (2016) 1370–1377. [23] T. Voigtlander, J. Wlecke, A.A. Negm, H. Lenzen, M.P. Manns, T.O. Lankisch, Calprotectin in bile: a disease severity marker in patients with primary sclerosing cholangitis, J. Clin. Gastroenterol. 48 (10) (2014) 866–869. [24] A. Gauss, P. Sauer, A. Stiehl, C. Rupp, J. Krisam, Y. Leopold, et al., Evaluation of Biliary Calprotectin as a Biomarker in Primary Sclerosing Cholangitis, Medicine (Baltimore) 95 (17) (2016) e3510. [25] D. Seidel, I. Eickmeier, A.A. Kuhl, A. Hamann, C. Loddenkemper, E. Schott, CD8 T cells primed in the gut-associated lymphoid tissue induce immune-mediated cholangitis in mice, Hepatology 59 (2) (2014) 601–611. [26] K. Inamori, S. Ariki, S. Kawabata, A Toll-like receptor in horseshoe crabs, Immunol. Rev. 198 (2004) 106–115. [27] Z. Zhou, M. Ding, L. Huang, G. Gilkeson, R. Lang, W. Jiang, Toll-like receptormediated immune responses in intestinal macrophages; implications for mucosal immunity and autoimmune diseases, Clin. Immunol. 173 (2016) 81–86. [28] A. Haziot, N. Hijiya, S.C. Gangloff, J. Silver, S.M. Goyert, Induction of a novel mechanism of accelerated bacterial clearance by lipopolysaccharide in CD14-deficient and Toll-like receptor 4-deficient mice, J. Immunol. 166 (2) (2001) 1075–1078. [29] I. Zanoni, R. Ostuni, L.R. Marek, S. Barresi, R. Barbalat, G.M. Barton, et al., CD14 controls the LPS-induced endocytosis of Toll-like receptor 4, Cell 147 (4) (2011) 868–880. [30] K. Friedrich, M. Smit, M. Brune, T. Giese, C. Rupp, A. Wannhoff, et al., CD14 is associated with biliary stricture formation, Hepatology 64 (3) (2016) 843–852. [31] J.F. Swisher, N. Burton, S.M. Bacot, S.N. Vogel, G.M. Feldman, Annexin A2 tetramer activates human and murine macrophages through TLR4, Blood 115 (3) (2010) 549–558. [32] M. Vesterhus, A. Holm, J.R. Hov, S. Nygard, E. Schrumpf, E. Melum, et al., Novel serum and bile protein markers predict primary sclerosing cholangitis disease severity and prognosis, J. Hepatol. 66 (6) (2017) 1214–1222, http://dx.doi.org/10.
result from an increased breakdown of cholangiocytes or infiltrated immune cells [35]. It is worth noting that we analyzed patients with PSC who showed no evidence of inflammatory activity, biliary infection, or malignancy. Therefore, the observed changes in the bile fluid of patients with PSC were not attributable to infection-associated alterations or pre-malignant lesions. In addition, we found specific biliary protein pattern in patients with PSC-associated CC, compared to patients without concomitant PSC. Proteins belonging to or associated with the Wnt-pathway were highly expressed in sporadic CCs. Several studies already highlighted that presumably human CC, is a Wnt-driven tumor and probably a promising therapeutic target [36,37]. On the contrary pathogenesis of PSC-associated CC seems to be more complex and is only poorly understood. Pro-inflammatory cytokines seem to play an important role in the carcinogenic process. In line with this we were able to demonstrate inflammation associated cytokine and chemokine pathways to be more prevalent in PSC-associated CC [38]. Among others, biliary proteome analysis revealed biliary CD14 and Annexin-4 expression, as the most prominent protein that discriminates between both cohorts. Invasion of CD14 + macrophages is associated with a more invasive phenotype and poor prognosis in cholangiocarcinoma [39]. Furthermore Annexin-4 up-regulation promotes tumor progression and confers chemoresistance in several human cancers [40]. Cholangiocarcinoma is a heterogeneous group of malignancies with different features of biliary tract differentiation. Beside different histological grading systems, two main histological subtypes can be differentiated. The bile ductular type (mixed), arising from the small intrahepatic bile ducts, and the bile duct type (mucinous), arising from the larger bile ducts. In PSC, the mucinous type is more frequently found. It shares several phenotypic features with pancreatic cancers and can be preceded by pre-malignant lesions such as biliary intraepithelial neoplasm or intraductal papillary neoplasm [41]. In conclusion, we identified disease-specific alterations in the bile duct-derived bile proteome in PSC and PSC-associated CC. Our findings suggest that perturbation in cholangiocyte secretion may underlie the pathogenesis of this disorder, and provide potential therapeutic targets for PSC treatment. Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bbadis.2017.09.012. Transparency document The Transparency document associated with this article can be found, in online version. Acknowlegement We thank Prof. R. Ruppert at the ZMBH, Dept. of Proteomics, University of Heidelberg for protein sequencing. Grant support C.R. and D.G. were supported by a grant from the Deutsche Forschungsgemeinschaft (DFG) (RU 1936/1-2). Disclosures The authors have no disclosures and declare no conflicts of interest. References [1] R.W. Chapman, B.A. Arborgh, J.M. Rhodes, J.A. Summerfield, R. Dick, P.J. Scheuer, et al., Primary sclerosing cholangitis: a review of its clinical features, cholangiography, and hepatic histology, Gut 21 (10) (1980) 870–877. [2] K.N. Lazaridis, N.F. LaRusso, Primary Sclerosing Cholangitis, N. Engl. J. Med. 375 (12) (2016) 1161–1170. [3] J.G. Lee, S.M. Schutz, R.E. England, J.W. Leung, P.B. Cotton, Endoscopic therapy of
9
BBA - Molecular Basis of Disease xxx (xxxx) xxx–xxx
C. Rupp et al. 1016/j.jhep.2017.01.019. [33] J. Metzger, A.A. Negm, R.R. Plentz, T.J. Weismuller, J. Wedemeyer, T.H. Karlsen, et al., Urine proteomic analysis differentiates cholangiocarcinoma from primary sclerosing cholangitis and other benign biliary disorders, Gut 62 (1) (2013) 122–130. [34] U. Navaneethan, V. Lourdusamy, P. Gk Venkatesh, B. Willard, M.R. Sanaka, M.A. Parsi, Bile proteomics for differentiation of malignant from benign biliary strictures: a pilot study, Gastroenterol Rep. (Oxf). 3 (2) (2015) 136–143. [35] E. Schrumpf, C. Tan, T.H. Karlsen, J. Sponheim, N.K. Bjorkstrom, O. Sundnes, et al., The biliary epithelium presents antigens to and activates natural killer T cells, Hepatology 62 (4) (2015) 1249–1259. [36] P. Carotenuto, M. Fassan, R. Pandolfo, A. Lampis, C. Vicentini, L. Cascione, et al., Wnt signalling modulates transcribed-ultraconserved regions in hepatobiliary cancers, Gut 66 (7) (2017) 1268–1277, http://dx.doi.org/10.1136/gutjnl-2016312278. [37] L. Boulter, R.V. Guest, T.J. Kendall, D.H. Wilson, D. Wojtacha, A.J. Robson, et al.,
[38]
[39]
[40] [41]
10
WNT signaling drives cholangiocarcinoma growth and can be pharmacologically inhibited, J. Clin. Invest. 125 (3) (2015) 1269–1285. M.R. Timmer, U. Beuers, P. Fockens, C.Y. Ponsioen, E.A. Rauws, K.K. Wang, et al., Genetic and epigenetic abnormalities in primary sclerosing cholangitis-associated cholangiocarcinoma, Inflamm. Bowel Dis. 19 (8) (2013) 1789–1797. C. Subimerb, S. Pinlaor, V. Lulitanond, N. Khuntikeo, S. Okada, M.S. McGrath, et al., Circulating CD14(+) CD16(+) monocyte levels predict tissue invasive character of cholangiocarcinoma, Clin. Exp. Immunol. 161 (3) (2010) 471–479. B. Wei, C. Guo, S. Liu, M.Z. Sun, Annexin A4 and cancer, Clin. Chim. Acta 447 (2015) 72–78. J.M. Banales, V. Cardinale, G. Carpino, M. Marzioni, J.B. Andersen, P. Invernizzi, et al., Expert consensus document: Cholangiocarcinoma: current knowledge and future perspectives consensus statement from the European Network for the Study of Cholangiocarcinoma (ENS-CCA), Nat. Rev. Gastroenterol. Hepatol. 13 (5) (2016) 261–280.