- Database
- Open Access
- Open Peer Review
Text-mining applied to autoimmune disease research: the Sjögren’s syndrome knowledge base
https://doi.org/10.1186/1471-2474-13-119
© Gorr et al.; licensee BioMed Central Ltd. 2012
- Received: 10 October 2011
- Accepted: 18 June 2012
- Published: 3 July 2012
Abstract
Background
Sjögren’s syndrome is a tissue-specific autoimmune disease that affects exocrine tissues, especially salivary glands and lacrimal glands. Despite a large body of evidence gathered over the past 60 years, significant gaps still exist in our understanding of Sjögren’s syndrome. The goal of this study was to develop a database that collects and organizes gene and protein expression data from the existing literature for comparative analysis with future gene expression and proteomic studies of Sjögren’s syndrome.
Description
To catalog the existing knowledge in the field, we used text mining to generate the Sjögren’s Syndrome Knowledge Base (SSKB) of published gene/protein data, which were extracted from PubMed using text mining of over 7,700 abstracts and listing approximately 500 potential genes/proteins. The raw data were manually evaluated to remove duplicates and false-positives and assign gene names. The data base was manually curated to 477 entries, including 377 potential functional genes, which were used for enrichment and pathway analysis using gene ontology and KEGG pathway analysis.
Conclusions
The Sjögren’s syndrome knowledge base (http://sskb.umn.edu) can form the foundation for an informed search of existing knowledge in the field as new potential therapeutic targets are identified by conventional or high throughput experimental techniques.
Keywords
- Salivary Gland
- Text Mining
- Lacrimal Gland
- Salivary Gland Hypofunction
- High Throughput Experimental Technique
Background
Sjögren’s syndrome is a tissue-specific autoimmune disease that affects exocrine tissues, especially salivary glands and lacrimal glands. It is one of the most common autoimmune disorders in the U.S., with an estimated prevalence of 2–4 million people. The autoimmune-mediated damage of the salivary and lacrimal glands in Sjögren’s syndrome leads to a decrease in the production of saliva and tears and to the development of dry mouth and dry eyes. Without the lubricating and protective functions of saliva and tears, the oral and ocular surfaces are subject to infections and discomfort, leading to a significantly reduced quality of life[1, 2].
Development of Sjögren’s syndrome requires a complex interplay between a number of genetic, hormonal and environmental factors, most of which have not been defined. Genetic linkages, especially involving major histocompatibility complex (MHC) genes, have been reported for Sjögren’s syndrome but it is not clear if, or how, the associated genes are involved in the development of the disease[3]. Additional non-MHC genes have also been linked with the development of Sjögren’s syndrome.
In addition to genetic predisposition, some studies suggest that infection of a genetically-susceptible individual by a virus or other pathogen might trigger the development of an autoimmune disease[4]. The proposed mechanisms include activation of the innate immune system, release of self antigens from damaged or apoptotic tissues, and molecular mimicry that results in activation of T cells and/or B cells that react with tissue antigens[4].
Both the innate and the adaptive immune systems are involved in the pathogenesis of Sjögren’s syndrome. The type I interferon (IFN) pathway, which plays an important role in the innate immune response to viruses, is also thought to play an important role in the development of Sjögren’s syndrome and other autoimmune disorders, including SLE[5, 6]. Moreover, type I IFNs can activate the adaptive immune system directly, by binding to IFN receptors on antigen presenting cells, T cells and B cells, or indirectly, by inducing the production and release of cytokines and chemokines that bind to these cells.
Autoantibodies to intracellular antigens, notably the nuclear proteins SSA/Ro and SSB/La, are found in the sera of many patients with Sjögren’s syndrome. These autoantibodies are thought to develop when intracellular antigens, some of which have undergone proteolytic cleavage that reveals new antigenic epitopes, become “visible” to the immune system in membrane blebs on the surface of apoptotic cells[7]. Alternatively, antigenic epitopes from bacteria and viruses, including Epstein-Barr virus (EBV) and coxsackie virus, may act as molecular mimics that trigger the development of antibodies that cross react with similar epitopes on target tissue autoantigens[2, 8, 9]. Although autoantibodies to intracellular antigens are useful in the diagnosis of Sjögren’s syndrome, it is not clear if they play a direct role in the development of salivary gland and lacrimal gland damage and hypofunction. In contrast, autoantibodies to the M3 muscarinic acetylcholine receptor (M3R) have been directly implicated in salivary gland hypofunction in the nonobese diabetic (NOD) mouse model of Sjögren’s syndrome[10]. Importantly, function-inhibiting anti-M3R autoantibodies are found in the sera of many patients with Sjögren’s syndrome[11].
Current therapy for Sjögren’s syndrome usually consists of palliative treatment that relieves the symptoms of dry eye and dry mouth, but fails to modify the underlying disease. Novel disease-modifying treatment strategies, based on recent immunological insights in Sjögren's syndrome and other autoimmune diseases, have met with mixed results[12]. For example, in recent clinical trials, treatment of Sjögren's syndrome patients with a B cell-depleting anti-CD20 monoclonal antibody (rituximab) led to significant improvement of the stimulated whole saliva flow rate and a reduction in parotid gland inflammation[13]. In contrast, TNFα inhibitors have been ineffective in the treatment of Sjögren's syndrome. Detailed studies on the immune response in Sjögren’s syndrome patients treated with one of the inhibitors (etanercept) revealed an increase in the circulating levels of TNFα[14]. These results suggest that TNFα may not play a pivotal role in the disease and that other therapeutic targets must be identified.
Despite a large body of evidence gathered over the past 60 years, significant gaps still exist in our understanding of Sjögren’s syndrome. Recent gene expression and proteomic studies have identified many genes and pathways that may play a role in the pathogenesis of Sjögren’s syndrome[15–17]. However, validation of these data will require significant additional effort. As an initial step in this validation, we have compiled the published data on Sjögren’s syndrome that is not derived from gene expression or proteomic studies. No such unifying database currently exists. Through data curation, the existing data have been uniformly formatted to allow systematic retrieval and comparisons to newly generated gene expression data. As an example of its functionality, the Sjögren's Syndrome Knowledge Base (SSKB) was analyzed for biological functions and pathways that are likely to play a role in the disease.
Construction and content
Data mining
To catalog the existing knowledge in the field, we used text mining to generate the Sjögren’s Syndrome Knowledge Base (SSKB) of published gene/protein data (http://sskb.umn.edu/)[18]. The focus of this data-base is on individually identified genes and proteins. Thus, microarray experiments were not included. The raw data for SSKB was extracted from PubMed[19]) using the text mining program EBIMed (http://www.ebi.ac.uk/Rebholz-srv/ebimed/)[20] with the search term "Sjogren's Syndrome" restricted to "MeshHeadingsList". The foundational search identified over 7,700 abstracts and approximately 500 potential genes/proteins. The SSKB is continually updated by regular automated searches of PubMed followed by manual curation.
Curation of raw data
The identified abstracts were manually evaluated to remove duplicates and false-positives. In older publications, where gene names were not readily identifiable, names were assigned based on in depth evaluation of the protein name context and available gene data in public databases, including the National Center for Biotechnology Information’s Entrez search engine[21] and UniProt[22, 23]. The SSKB includes data from human studies and animal models. For the genes identified in animal models, the human homolog was identified by automated ortholog search, using WebGestalt 2.0[24, 25]. These steps reduced the database to 477 current entries. The online database contains the fully curated data and currently contains 413 entries, which can be accessed athttp://sskb.umn.edu. Updates and newly curated data are continually added.
The 477 entries were sorted to identify autoantigens and viral/bacterial antigens, resulting in 377 potential functional genes, which were used for enrichment and pathway analysis.
Enrichment analysis
The 377 human gene entries were used for subsequent enrichment analyses in Webgestalt[24, 25]. Gene enrichment in the SSKB gene set was compared to the human genome using the hypergeometric test with multiple test adjustment[26] and a significance level of P <0.01.
The Gene Ontology[27, 28] was accessed with Webgestalt and analysis was restricted to processes and functions represented by two or more genes. Pathway analysis was performed with Webgestalt in the Kyoto Encyclopedia of Genes and Genomes (KEGG)[29, 30]. The selection was restricted to pathways with 4 or more genes represented, resulting in identification of 72 KEGG pathways. The “salivary secretion” pathway (KO04970) was recently added to KEGG (11/9/10) and was not included in this analysis. This pathway contains 59 genes, seven of which are found in the SSKB gene set.
Utility and discussion
We constructed a database containing proteins and genes associated with Sjögren’s syndrome in human disease or animal models, as identified by text mining of published data. The public SSKB currently contains 413 genes/proteins and can be viewed online (http://sskb.umn.edu/). All genes have been assigned gene symbols and UniProt IDs, which allows rapid retrieval of gene-specific data from external databases. The SSKB data base can be used to determine whether a list of genes is enriched with known Sjögren’s syndrome genes and one can carry out a function enrichment analysis (hypergeometric distribution). Individual genes and the corresponding gene products, synonyms and alternate names can be searched by using a web browser search function. Autoantigens, viral antigens and bacterial antigens are separately identified under “Antigens”. The SSKB is continually maintained and updated and new genes are added as their analysis is completed.
Based on the abstracts used to retrieve the SSKB genes/proteins, 85 proteins were initially characterized as autoantigens and 15 proteins were characterized as viral (14) or bacterial (1) antigens. Not surprisingly, SSA/Ro and SSB/La were among the most frequently retrieved autoantigens. It has been proposed that viral or bacterial antigens act as autoimmune triggers by molecular mimicry of endogenous human proteins[2, 8, 9]. However, eight of the 14 putative viral antigens in SSKB were selected for BLAST analysis, which did not identify strong sequence similarity with human proteins (not shown).
The 377 proteins not identified as autoantigens or microbial antigens were considered candidates for functional genes that could play a role in the initiation and progression of Sjögren’s syndrome. Since the gene list contains data from humans and animals, the corresponding human genes were identified, with the assumption that genes identified in animal models of Sjögren’s syndrome may also be involved in the human disease.
Gene ontology
Gene Ontology enrichment analysis
Rank | BIOLOGICAL PROCESS | GO ID | Reference Genes | Observed Genes | Ratio |
|---|---|---|---|---|---|
1 | regulation of lymphocyte proliferation | GO:0050670 | 81 | 32 | 39.51% |
2 | regulation of leukocyte proliferation | GO:0070663 | 82 | 32 | 39.02% |
3 | regulation of mononuclear cell proliferation | GO:0032944 | 82 | 32 | 39.02% |
4 | adaptive immune response based on somatic recombination of immune receptors built from immunoglobulin superfamily domains | GO:0002460 | 112 | 38 | 33.93% |
5 | adaptive immune response | GO:0002250 | 113 | 38 | 33.63% |
6 | lymphocyte proliferation | GO:0046651 | 112 | 37 | 33.04% |
7 | leukocyte proliferation | GO:0070661 | 114 | 37 | 32.46% |
8 | mononuclear cell proliferation | GO:0032943 | 114 | 37 | 32.46% |
9 | regulation of lymphocyte activation | GO:0051249 | 141 | 42 | 29.79% |
10 | regulation of cell activation | GO:0050865 | 168 | 46 | 27.38% |
11 | regulation of leukocyte activation | GO:0002694 | 159 | 43 | 27.04% |
12 | positive regulation of immune system process | GO:0002684 | 229 | 60 | 26.20% |
13 | regulation of immune response | GO:0050776 | 218 | 54 | 24.77% |
14 | immune effector process | GO:0002252 | 200 | 45 | 22.50% |
15 | regulation of immune system process | GO:0002682 | 362 | 79 | 21.82% |
16 | lymphocyte activation | GO:0046649 | 272 | 59 | 21.69% |
17 | leukocyte activation | GO:0045321 | 324 | 66 | 20.37% |
18 | inflammatory response | GO:0006954 | 359 | 71 | 19.78% |
19 | cell activation | GO:0001775 | 366 | 71 | 19.40% |
20 | immune response | GO:0006955 | 750 | 133 | 17.73% |
21 | regulation of response to stimulus | GO:0048583 | 441 | 75 | 17.01% |
22 | defense response | GO:0006952 | 657 | 100 | 15.22% |
23 | immune system process | GO:0002376 | 1066 | 162 | 15.20% |
24 | response to wounding | GO:0009611 | 560 | 85 | 15.18% |
25 | response to external stimulus | GO:0009605 | 904 | 110 | 12.17% |
26 | multi-organism process | GO:0051704 | 668 | 79 | 11.83% |
27 | regulation of programmed cell death | GO:0043067 | 812 | 92 | 11.33% |
28 | regulation of apoptosis | GO:0042981 | 805 | 91 | 11.30% |
29 | regulation of cell death | GO:0010941 | 815 | 92 | 11.29% |
30 | regulation of cell proliferation | GO:0042127 | 739 | 79 | 10.69% |
31 | apoptosis | GO:0006915 | 1063 | 102 | 9.60% |
32 | programmed cell death | GO:0012501 | 1071 | 102 | 9.52% |
33 | response to chemical stimulus | GO:0042221 | 1243 | 117 | 9.41% |
34 | cell proliferation | GO:0008283 | 1056 | 98 | 9.28% |
35 | death | GO:0016265 | 1171 | 107 | 9.14% |
36 | cell death | GO:0008219 | 1167 | 106 | 9.08% |
37 | response to stress | GO:0006950 | 1696 | 144 | 8.49% |
38 | positive regulation of biological process | GO:0048518 | 1865 | 153 | 8.20% |
39 | positive regulation of cellular process | GO:0048522 | 1699 | 130 | 7.65% |
40 | response to stimulus | GO:0050896 | 3471 | 221 | 6.37% |
Rank | CELLULAR COMPONENT | GO ID | Count | Observed | Ratio |
1 | calcineurin complex | GO:0005955 | 5 | 3 | 60.00% |
2 | external side of plasma membrane | GO:0009897 | 131 | 40 | 30.53% |
3 | platelet alpha granule lumen | GO:0031093 | 41 | 12 | 29.27% |
4 | MHC class II protein complex | GO:0042613 | 14 | 4 | 28.57% |
5 | nerve terminal | GO:0043679 | 14 | 4 | 28.57% |
6 | cytoplasmic membrane-bounded vesicle lumen | GO:0060205 | 44 | 12 | 27.27% |
7 | vesicle lumen | GO:0031983 | 46 | 12 | 26.09% |
8 | integrin complex | GO:0008305 | 29 | 7 | 24.14% |
9 | platelet alpha granule | GO:0031091 | 52 | 12 | 23.08% |
10 | high-density lipoprotein particle | GO:0034364 | 24 | 5 | 20.83% |
11 | MHC protein complex | GO:0042611 | 38 | 7 | 18.42% |
12 | plasma lipoprotein particle | GO:0034358 | 34 | 6 | 17.65% |
13 | protein-lipid complex | GO:0032994 | 34 | 6 | 17.65% |
14 | cell surface | GO:0009986 | 305 | 51 | 16.72% |
15 | axon part | GO:0033267 | 48 | 7 | 14.58% |
16 | extracellular space | GO:0005615 | 670 | 84 | 12.54% |
17 | receptor complex | GO:0043235 | 113 | 13 | 11.50% |
18 | secretory granule | GO:0030141 | 174 | 19 | 10.92% |
19 | membrane raft | GO:0045121 | 131 | 14 | 10.69% |
20 | extracellular region part | GO:0044421 | 939 | 94 | 10.01% |
21 | axon | GO:0030424 | 148 | 14 | 9.46% |
22 | cell soma | GO:0043025 | 155 | 13 | 8.39% |
23 | soluble fraction | GO:0005625 | 297 | 24 | 8.08% |
24 | cytoplasmic vesicle part | GO:0044433 | 177 | 13 | 7.34% |
25 | extracellular region | GO:0005576 | 1984 | 143 | 7.21% |
26 | basolateral plasma membrane | GO:0016323 | 190 | 13 | 6.84% |
27 | lysosome | GO:0005764 | 206 | 14 | 6.80% |
28 | integral to plasma membrane | GO:0005887 | 1183 | 72 | 6.09% |
29 | intrinsic to plasma membrane | GO:0031226 | 1206 | 73 | 6.05% |
30 | cytoplasmic membrane-bounded vesicle | GO:0016023 | 537 | 32 | 5.96% |
31 | membrane-bounded vesicle | GO:0031988 | 555 | 32 | 5.77% |
32 | extracellular matrix | GO:0031012 | 335 | 19 | 5.67% |
33 | neuron projection | GO:0043005 | 318 | 18 | 5.66% |
34 | plasma membrane part | GO:0044459 | 1918 | 104 | 5.42% |
35 | cell fraction | GO:0000267 | 1039 | 55 | 5.29% |
36 | cytoplasmic vesicle | GO:0031410 | 628 | 33 | 5.25% |
37 | vesicle | GO:0031982 | 655 | 33 | 5.04% |
38 | insoluble fraction | GO:0005626 | 803 | 34 | 4.23% |
39 | plasma membrane | GO:0005886 | 3650 | 139 | 3.81% |
40 | cytosol | GO:0005829 | 1251 | 47 | 3.76% |
Rank | MOLECULAR FUNCTION | GO ID | COUNT | Observed | RATIO |
1 | arginine binding | GO:0034618 | 3 | 3 | 100.00% |
2 | nitric-oxide synthase activity | GO:0004517 | 3 | 3 | 100.00% |
3 | tetrahydrobiopterin binding | GO:0034617 | 3 | 3 | 100.00% |
4 | C-X-C chemokine binding | GO:0019958 | 8 | 4 | 50.00% |
5 | beta-amyloid binding | GO:0001540 | 13 | 5 | 38.46% |
6 | tumor necrosis factor receptor binding | GO:0005164 | 21 | 8 | 38.10% |
7 | chemokine activity | GO:0008009 | 47 | 17 | 36.17% |
8 | chemokine receptor binding | GO:0042379 | 49 | 17 | 34.69% |
9 | coreceptor activity | GO:0015026 | 19 | 6 | 31.58% |
10 | tumor necrosis factor receptor superfamily binding | GO:0032813 | 31 | 9 | 29.03% |
11 | cytokine receptor binding | GO:0005126 | 178 | 46 | 25.84% |
12 | chemokine binding | GO:0019956 | 26 | 6 | 23.08% |
13 | cytokine activity | GO:0005125 | 196 | 45 | 22.96% |
14 | growth factor receptor binding | GO:0070851 | 67 | 14 | 20.90% |
15 | collagen binding | GO:0005518 | 35 | 7 | 20.00% |
16 | G-protein-coupled receptor binding | GO:0001664 | 107 | 20 | 18.69% |
17 | integrin binding | GO:0005178 | 58 | 9 | 15.52% |
18 | cysteine-type endopeptidase activity | GO:0004197 | 71 | 10 | 14.08% |
19 | growth factor activity | GO:0008083 | 161 | 19 | 11.80% |
20 | cytokine binding | GO:0019955 | 108 | 12 | 11.11% |
21 | protein heterodimerization activity | GO:0046982 | 189 | 21 | 11.11% |
22 | glycosaminoglycan binding | GO:0005539 | 139 | 14 | 10.07% |
23 | protein complex binding | GO:0032403 | 196 | 19 | 9.69% |
24 | receptor binding | GO:0005102 | 856 | 83 | 9.70% |
25 | receptor signaling protein activity | GO:0005057 | 159 | 15 | 9.43% |
26 | pattern binding | GO:0001871 | 153 | 14 | 9.15% |
27 | peptidase inhibitor activity | GO:0030414 | 154 | 14 | 9.09% |
28 | carbohydrate binding | GO:0030246 | 349 | 29 | 8.31% |
29 | endopeptidase activity | GO:0004175 | 370 | 28 | 7.57% |
30 | polysaccharide binding | GO:0030247 | 153 | 14 | 9.15% |
31 | protein dimerization activity | GO:0046983 | 514 | 36 | 7.00% |
32 | identical protein binding | GO:0042802 | 618 | 38 | 6.15% |
33 | enzyme binding | GO:0019899 | 505 | 29 | 5.74% |
34 | peptidase activity | GO:0008233 | 563 | 30 | 5.33% |
35 | peptidase activity, acting on L-amino acid peptides | GO:0070011 | 546 | 29 | 5.31% |
36 | molecular transducer activity | GO:0060089 | 2116 | 98 | 4.63% |
37 | signal transducer activity | GO:0004871 | 2116 | 98 | 4.63% |
38 | receptor activity | GO:0004872 | 1674 | 71 | 4.24% |
39 | protein binding | GO:0005515 | 8041 | 280 | 3.48% |
40 | binding | GO:0005488 | 12465 | 320 | 2.57% |
The most highly enriched biological processes (19 of 40; 18 of the top 20) were associated with immune function, including leukocyte proliferation, leukocyte activation, and regulation of the immune response. Other prominent biological processes were associated with apoptosis and cell death. Thus, the SSKB data set is consistent with recent microarray data[16] and reflects current models for the biological processes involved in the pathogenesis of Sjögren's syndrome[5, 31, 32].
The most highly enriched cellular component was the calcineurin complex, which plays a major role in the activation of T cells. Interestingly, in placebo-controlled clinical trials, treatment of Sjögren’s syndrome patients with eye drops that contain the calcineurin inhibitor cyclosporine, led to significant improvement in several of the signs and symptoms of dry eye[33].
Other highly enriched cellular components include: 1) platelet alpha granules. Although platelet activation has been reported in the salivary glands of Sjögren's syndrome patients[34], a direct search of PubMed for “platelet alpha granules” with “sjogren’s” did not retrieve any published studies. Thus, while the proteins identified were retrieved from the literature, their potential association with platelet alpha granules in Sjögren’s syndrome has not previously been noted. 2) MHC protein complexes were identified and are presumably involved in the presentation of autoantigens[16]. 3) The finding that protein-lipid complexes and lipoprotein particles are associated with Sjögren's syndrome may be consistent with changes in serum lipid levels in Sjögren's syndrome patients[35] although the prevalence of anti-phospholipid antibodies is low in Sjögren's syndrome[36]. 4) Nerve terminals and axons were also prominent cellular components, consistent with the known neurological component of Sjögren's syndrome[37].
In molecular function, nitric oxide synthase (NOS) activity was the most highly enriched, although only three genes (NOS1-3) were identified. Nitric oxide (NO) signaling appears to be directly affected in salivary and lacrimal glands in Sjögren’s syndrome[38]. Other highly enriched molecular functions include chemokine and cytokine activity/receptor binding (8 of the top 15) and peptidase activities.
Pathway analysis
Biological pathways associated with SSKB genes
Rank | PATHWAY | SSKB Genes | ENRICHMENT | Raw P | Adjust P |
|---|---|---|---|---|---|
1 | Allograft rejection | 23 | 76.02 | 3.62E-39 | 6.82E-38 |
2 | Intestinal immune network for IgA production | 27 | 67.82 | 7.26E-44 | 2.05E-42 |
3 | Asthma | 14 | 58.61 | 4.14E-22 | 2.75E-21 |
4 | Type I diabetes mellitus | 20 | 57.09 | 9.13E-31 | 9.38E-30 |
5 | Graft-versus-host disease | 18 | 53.83 | 3.21E-27 | 2.79E-26 |
6 | Autoimmune thyroid disease | 22 | 52.13 | 1.29E-32 | 1.82E-31 |
7 | Primary immunodeficiency | 14 | 50.24 | 6.38E-21 | 3.79E-20 |
8 | Hematopoietic cell lineage | 33 | 47.1 | 1.39E-46 | 5.24E-45 |
9 | Toll-like receptor signaling pathway | 37 | 46.01 | 1.13E-51 | 6.38E-50 |
10 | Apoptosis | 25 | 35.68 | 5.55E-32 | 6.97E-31 |
11 | NOD-like receptor signaling pathway | 17 | 34.44 | 7.61E-22 | 4.78E-21 |
12 | Amyotrophic lateral sclerosis (ALS) | 14 | 33.18 | 5.81E-18 | 2.85E-17 |
13 | Other glycan degradation | 4 | 31.4 | 6.67E-06 | 1.24E-05 |
14 | Cytokine-cytokine receptor interaction | 66 | 31.05 | 5.91E-79 | 6.68E-77 |
15 | T cell receptor signaling pathway | 26 | 30.24 | 4.12E-31 | 4.66E-30 |
16 | RIG-I-like receptor signaling pathway | 17 | 30.07 | 9.98E-21 | 5.64E-20 |
17 | Cell adhesion molecules (CAMs) | 32 | 29.99 | 6.40E-38 | 1.03E-36 |
18 | Bladder cancer | 10 | 29.9 | 1.06E-12 | 3.24E-12 |
19 | Viral myocarditis | 17 | 29.25 | 1.68E-20 | 9.04E-20 |
20 | Cytosolic DNA-sensing pathway | 13 | 29.16 | 5.78E-16 | 2.42E-15 |
21 | Pancreatic cancer | 15 | 26.17 | 1.88E-17 | 8.50E-17 |
22 | Small cell lung cancer | 16 | 23.92 | 7.32E-18 | 3.45E-17 |
23 | Glycosaminoglycan degradation | 4 | 23.92 | 2.13E-05 | 3.65E-05 |
24 | Natural killer cell mediated cytotoxicity | 25 | 22.92 | 1.06E-26 | 8.56E-26 |
25 | ErbB signaling pathway | 13 | 22.16 | 2.51E-13 | 8.86E-13 |
26 | Epithelial cell signaling in Helicobacter pylori infection | 12 | 22.16 | 2.64E-13 | 9.04E-13 |
27 | Complement and coagulation cascades | 12 | 21.84 | 3.17E-13 | 1.05E-12 |
28 | B cell receptor signaling pathway | 13 | 21.77 | 3.38E-14 | 1.23E-13 |
29 | Prion diseases | 6 | 21.53 | 3.27E-07 | 6.84E-07 |
30 | Antigen processing and presentation | 15 | 21.17 | 5.49E-16 | 2.39E-15 |
31 | Colorectal cancer | 14 | 20.93 | 6.14E-15 | 2.48E-14 |
32 | Adipocytokine signaling pathway | 11 | 20.62 | 6.05E-12 | 1.80E-11 |
33 | Chemokine signaling pathway | 30 | 19.83 | 7.80E-30 | 7.35E-29 |
34 | Prostate cancer | 14 | 19.76 | 1.42E-14 | 5.53E-14 |
35 | Glioma | 10 | 19.32 | 1.10E-10 | 2.89E-10 |
36 | Jak-STAT signaling pathway | 23 | 18.64 | 1.67E-22 | 1.18E-21 |
37 | Non-small cell lung cancer | 8 | 18.61 | 1.13E-08 | 2.50E-08 |
38 | Melanoma | 10 | 17.69 | 2.71E-10 | 6.96E-10 |
39 | Pathways in cancer | 46 | 17.51 | 9.85E-43 | 2.23E-41 |
40 | Fc epsilon RI signaling pathway | 11 | 17.49 | 3.90E-11 | 1.05E-10 |
41 | Chronic myeloid leukemia | 10 | 16.75 | 4.74E-10 | 1.19E-09 |
42 | GnRH signaling pathway | 12 | 14.92 | 3.42E-11 | 9.43E-11 |
43 | Leukocyte transendothelial migration | 14 | 14.9 | 7.91E-13 | 2.48E-12 |
44 | VEGF signaling pathway | 9 | 14.87 | 1.04E-08 | 2.35E-08 |
45 | Hypertrophic cardiomyopathy (HCM) | 10 | 14.78 | 1.67E-09 | 4.10E-09 |
46 | p53 signaling pathway | 8 | 14.56 | 8.19E-08 | 1.75E-07 |
47 | Endometrial cancer | 6 | 14.49 | 3.65E-06 | 7.11E-06 |
48 | Systemic lupus erythematosus | 16 | 14.35 | 3.27E-14 | 1.23E-13 |
49 | MAPK signaling pathway | 30 | 14.01 | 3.15E-25 | 2.37E-24 |
50 | Focal adhesion | 22 | 13.75 | 1.21E-18 | 6.21E-18 |
51 | Dilated cardiomyopathy | 10 | 13.65 | 3.66E-09 | 8.44E-09 |
52 | Type II diabetes mellitus | 5 | 13.36 | 3.63E-05 | 6.12E-05 |
53 | Neurotrophin signaling pathway | 13 | 12.96 | 3.17E-11 | 8.96E-11 |
54 | ECM-receptor interaction | 8 | 11.96 | 3.85E-07 | 7.91E-07 |
55 | Alzheimer's disease | 16 | 11.89 | 6.32E-13 | 2.04E-12 |
56 | Lysosome | 11 | 11.81 | 2.86E-09 | 6.73E-09 |
57 | Arginine and proline metabolism | 5 | 11.63 | 7.15E-05 | 0.0001 |
58 | Renal cell carcinoma | 6 | 10.77 | 2.09E-05 | 3.63E-05 |
59 | Long-term depression | 6 | 10.77 | 2.09E-05 | 3.63E-05 |
60 | Long-term potentiation | 6 | 10.77 | 2.09E-05 | 3.63E-05 |
61 | Proteasome | 4 | 10.47 | 0.0006 | 0.0009 |
62 | Progesterone-mediated oocyte maturation | 7 | 10.22 | 6.00E-06 | 1.15E-05 |
63 | TGF-beta signaling pathway | 7 | 10.11 | 6.48E-06 | 1.22E-05 |
64 | Regulation of actin cytoskeleton | 16 | 9.3 | 2.69E-11 | 7.79E-11 |
65 | Calcium signaling pathway | 13 | 9.17 | 2.36E-09 | 5.67E-09 |
66 | Wnt signaling pathway | 11 | 9.15 | 4.17E-08 | 9.06E-08 |
67 | Gap junction | 6 | 8.37 | 8.67E-05 | 0.0001 |
68 | Cell cycle | 8 | 7.85 | 9.32E-06 | 1.70E-05 |
69 | Oocyte meiosis | 7 | 7.71 | 3.80E-05 | 6.31E-05 |
70 | Axon guidance | 7 | 6.82 | 8.33E-05 | 0.0001 |
71 | Endocytosis | 10 | 6.72 | 2.93E-06 | 5.81E-06 |
72 | Metabolic pathways | 26 | 2.96 | 1.12E-06 | 2.26E-06 |
The pathway analysis revealed dominant pathways associated with immune regulation. Indeed, the eight most highly enriched pathways were associated with antigen presenting cells and activation of T cells and B cells.
Several cancer associated pathways were identified. This is partly due to the overlap between cancer pathways. These pathways typically include cytokine or growth factor stimulation of cell cycle and cell death and were not further analyzed.
Pathways associated with apoptosis, cytokine signaling and inflammation were also highly enriched. To focus on the events associated with initiation of Sjögren's syndrome, we analyzed pathways with known triggers. Several of the highly enriched pathways are triggered by bacterial toxins, viral DNA, or viral RNA. These include signaling pathways for Toll-like receptor, NOD-like receptor, RIG-I-like receptor signaling pathways and the cytosolic DNA-sensing pathway.
Overlap with other autoimmune diseases
The KEGG pathways include several pathways for autoimmune diseases, including type I diabetes mellitus, autoimmune thyroid disease, and SLE. While about 50% of the genes associated with the first two pathways are also associated with Sjögren's syndrome, only 16 Sjögren's syndrome genes were identified in the 140-gene SLE pathway (KEGG ID: hsa05322). These findings suggest that significant differences exist in the pathogenesis of autoimmune diseases.
Conclusions
The results of this analysis can serve as a background and comparison for the increasing number of gene expression data sets available for Sjögren’s syndrome, e.g.[15–17]. Preliminary analysis of such data sets suggest that the biological pathways identified in the SSKB are very similar to those identified in human parotid tissue but quite different from those identified in human labial salivary glands[15]. Future analyses will further define these differences and focus on the comparison of biological pathways identified in human tissues and mouse models of Sjögren’s syndrome. It is envisioned that the SSKB data can also serve as the starting point for literature reviews and literature-based validation of identified genes; functional gene enrichment studies; protein-protein interaction networks and other bioinformatics analyses; it can be used to arrive at gene sets for SNP set enrichment analysis (pathway based GWAS studies); it can be used to define a gene set for gene set enrichment analysis (GSEA); as a starting point for bioinformatics analysis protein-protein interaction networks (based on yeast 2 hybrid) can be identified among the SSKB genes.
Availability and requirements
The Sjögren’s syndrome knowledge base is freely available at sskb.umn.edu.
Declarations
Acknowledgements
The authors thank Dr. Ammon Peck, University of Florida Dr. Michael Zhou, UCLA for helpful discussions. The Minnesota Supercomputing Institute provides web hosting for the SSKB database. This work was supported by U.S. PHS grants R01DE019255 (SUG, SM, DTW) and R01DE014385 (SM) from NIDCR and a research grant from the Sjögren’s Syndrome Foundation (SM).
Authors’ Affiliations
References
- Meijer JM, Meiners PM, Huddleston Slater JJ, Spijkervet FK, Kallenberg CG, Vissink A, Bootsma H: Health-related quality of life, employment and disability in patients with Sjogren's syndrome. Rheumatology. 2009, 48 (9): 1077-1082. 10.1093/rheumatology/kep141.View ArticlePubMedGoogle Scholar
- Rhodus NL: Sjogren's syndrome. Quintessence Int. 1999, 30 (10): 689-699.PubMedGoogle Scholar
- Williams PH, Cobb BL, Namjou B, Scofield RH, Sawalha AH, Harley JB: Horizons in Sjogren's syndrome genetics. Clin Rev Allergy Immunol. 2007, 32 (3): 201-209. 10.1007/s12016-007-8002-9.View ArticlePubMedPubMed CentralGoogle Scholar
- Ercolini AM, Miller SD: The role of infections in autoimmune disease. Clin Exp Immunol. 2009, 155 (1): 1-15. 10.1111/j.1365-2249.2008.03834.x.View ArticlePubMedPubMed CentralGoogle Scholar
- Mavragani CP, Crow MK: Activation of the type I interferon pathway in primary Sjogren's syndrome. J Autoimmun. 2010, 35 (3): 225-231. 10.1016/j.jaut.2010.06.012.View ArticlePubMedGoogle Scholar
- Peck AB, Nguyen CQ, Sharma A, McIndoe RA, She JX: The Interferon-Signature of Sjögren's Syndrome: What does It say about the etiopathology of autoimmunity?. J Clin Rheumatol Musculoskel Med. 2011, 3 (1):Google Scholar
- Rosen A, Casciola-Rosen L: Altered autoantigen structure in Sjogren's syndrome: implications for the pathogenesis of autoimmune tissue damage. Crit Rev Oral Biol Med. 2004, 15 (3): 156-164. 10.1177/154411130401500304.View ArticlePubMedGoogle Scholar
- Stathopoulou EA, Routsias JG, Stea EA, Moutsopoulos HM, Tzioufas AG: Cross-reaction between antibodies to the major epitope of Ro60 kD autoantigen and a homologous peptide of Coxsackie virus 2B protein. Clin Exp Immunol. 2005, 141 (1): 148-154. 10.1111/j.1365-2249.2005.02812.x.View ArticlePubMedPubMed CentralGoogle Scholar
- Zhao Z-S, Granucci F, Yeh L, Schaffer PA, Cantor H: Molecular mimicry by herpes simplex virus-type 1: autoimmune disease after viral infection. Science. 1998, 279 (5355): 1344-1347. 10.1126/science.279.5355.1344.View ArticlePubMedGoogle Scholar
- Nguyen KH, Brayer J, Cha S, Diggs S, Yasunari U, Hilal G, Peck AB, Humphreys-Beher MG: Evidence for antimuscarinic acetylcholine receptor antibody-mediated secretory dysfunction in NOD mice. Arthritis Rheum. 2000, 43 (10): 2297-2306. 10.1002/1529-0131(200010)43:10<2297::AID-ANR18>3.0.CO;2-X.View ArticlePubMedGoogle Scholar
- Li J, Ha Y, Ku N, Choi S, Lee S, Oh S, Kim J, Lee J, Lee E, Song Y, et al.: Inhibitory effects of autoantibodies on the muscarinic receptors in Sjogren's syndrome. Lab Invest. 2004, 84 (11): 1430-1438. 10.1038/labinvest.3700173.View ArticlePubMedGoogle Scholar
- von Bultzingslöwen I, Sollecito TP, Fox PC, Daniels T, Jonsson R, Lockhart PB, Wray D, Brennan MT, Carrozzo M, Gandera B, et al.: Salivary dysfunction associated with systemic diseases: systematic review and clinical management recommendations. Oral Surg Oral Med Oral Pathol Oral Radiol Endod. 2007, 103 (Suppl): S75.e1-e15.-Google Scholar
- Meijer JM, Meiners PM, Vissink A, Spijkervet FK, Abdulahad W, Kamminga N, Brouwer E, Kallenberg CG, Bootsma H: Effectiveness of rituximab treatment in primary Sjogren's syndrome: a randomized, double-blind, placebo-controlled trial. Arthritis Rheum. 2010, 62 (4): 960-968. 10.1002/art.27314.View ArticlePubMedGoogle Scholar
- Moutsopoulos NM, Katsifis GE, Angelov N, Leakan RA, Sankar V, Pillemer S, Wahl SM: Lack of efficacy of etanercept in Sjogren's syndrome correlates with failed suppression of TNF{alpha} and systemic immune activation. Ann Rheum Dis. 2008, 67 (10): 1437-1443. 10.1136/ard.2007.077891.View ArticlePubMedGoogle Scholar
- Gottenberg J-E, Cagnard N, Lucchesi C, Letourneur F, Mistou S, Lazure T, Jacques S, Ba N, Ittah M, Lepajolec C, et al.: Activation of IFN pathways and plasmacytoid dendritic cell recruitment in target organs of primary Sjogren's syndrome. Proc Natl Acad Sci U S A. 2006, 103 (8): 2770-2775. 10.1073/pnas.0510837103.View ArticlePubMedPubMed CentralGoogle Scholar
- Hu S, Zhou M, Jiang J, Wang J, Elashoff D, Gorr S, Michie SA, Spijkervet FK, Bootsma H, Kallenberg CG, et al.: Systems biology analysis of Sjogren's syndrome and mucosa-associated lymphoid tissue lymphoma in parotid glands. Arthritis Rheum. 2009, 60 (1): 81-92. 10.1002/art.24150.View ArticlePubMedPubMed CentralGoogle Scholar
- Nguyen CQ, Sharma A, Lee BH, She JX, McIndoe RA, Peck AB: Differential gene expression in the salivary gland during development and onset of xerostomia in Sjogren's syndrome-like disease of the C57BL/6.NOD-Aec1Aec2 mouse. Arthritis Res Ther. 2009, 11 (2): R56-10.1186/ar2676.View ArticlePubMedPubMed CentralGoogle Scholar
- Sjogren's Syndrome Knowledge Base.http://sskb.umn.edu/,
- PubMed.http://www.ncbi.nlm.nih.gov/pubmed/,
- Rebholz-Schuhmann D, Kirsch H, Arregui M, Gaudan S, Riethoven M, Stoehr P: EBIMed—text crunching to gather facts for proteins from Medline. Bioinformatics. 2007, 23 (2): e237-e244. 10.1093/bioinformatics/btl302.View ArticlePubMedGoogle Scholar
- Entrez.http://www.ncbi.nlm.nih.gov/Entrez/,
- UniProt.http://www.uniprot.org/,
- Consortium TU: Ongoing and future developments at the Universal Protein Resource. Nucleic Acids Res. 2011, 39 (suppl 1): D214-D219.View ArticleGoogle Scholar
- WebGestalt.http://bioinfo.vanderbilt.edu/webgestalt/,
- Zhang B, Kirov S, Snoddy J: WebGestalt: an integrated system for exploring gene sets in various biological contexts. Nucleic acids research. 2005, 33 (Web Server issue): W741-W748.View ArticlePubMedPubMed CentralGoogle Scholar
- Benjamini Y, Hochberg Y: Controlling the false discovery rate - a practical and powerful approach to multiple testing. J Roy Stat Soc B Met. 1995, 57 (1): 289-300.Google Scholar
- The Gene Ontology.http://www.geneontology.org/,
- Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, Cherry JM, Davis AP, Dolinski K, Dwight SS, Eppig JT, et al.: Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000, 25 (1): 25-29. 10.1038/75556.View ArticlePubMedPubMed CentralGoogle Scholar
- Kyoto Encyclopedia of Genes and Genomes.http://www.genome.jp/kegg/,
- Kanehisa M, Goto S: KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Res. 2000, 28 (1): 27-30. 10.1093/nar/28.1.27.View ArticlePubMedPubMed CentralGoogle Scholar
- Nguyen CQ, Peck AB: Unraveling the pathophysiology of Sjogren syndrome-associated dry eye disease. Ocul Surf. 2009, 7 (1): 11-27. 10.1016/S1542-0124(12)70289-6.View ArticlePubMedPubMed CentralGoogle Scholar
- Mariette X, Gottenberg JE: Pathogenesis of Sjogren's syndrome and therapeutic consequences. Curr Opin Rheumatol. 2010, 22 (5): 471-477. 10.1097/BOR.0b013e32833c36c5.View ArticlePubMedGoogle Scholar
- Sall K, Stevenson OD, Mundorf TK, Reis BL: Two multicenter, randomized studies of the efficacy and safety of cyclosporine ophthalmic emulsion in moderate to severe dry eye disease. CsA Phase 3 Study Group. Ophthalmology. 2000, 107 (4): 631-639. 10.1016/S0161-6420(99)00176-1.View ArticlePubMedGoogle Scholar
- Oxholm P, Winther K, Manthorpe R: Platelets in blood and salivary glands of patients with primary Sjogren's syndrome. Scand J Rheumatol Suppl. 1986, 61: 170-172.PubMedGoogle Scholar
- Lodde BM, Sankar V, Kok MR, Leakan RA, Tak PP, Pillemer SR: Serum lipid levels in Sjögren's syndrome. Rheumatology. 2006, 45 (4): 481-484. 10.1093/rheumatology/kei190.View ArticlePubMedGoogle Scholar
- Ostrowski RA, Robinson JA: Antiphospholipid antibody syndrome and autoimmune diseases. Hematol Oncol Clin North Am. 2008, 22 (1): 53-65. 10.1016/j.hoc.2007.10.003.View ArticlePubMedGoogle Scholar
- Chai J, Logigian EL: Neurological manifestations of primary Sjogren's syndrome. Curr Opin Neurol. 2010, 23 (5): 509-513. 10.1097/WCO.0b013e32833de6ab.View ArticlePubMedGoogle Scholar
- Caulfield VL, Balmer C, Dawson LJ, Smith PM: A role for nitric oxide-mediated glandular hypofunction in a non-apoptotic model for Sjogren's syndrome. Rheumatology. 2009, 48 (7): 727-733. 10.1093/rheumatology/kep100.View ArticlePubMedGoogle Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2474/13/119/prepub
Pre-publication history
Copyright
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
