Advanced search
1 file | 24.89 MB Add to list

A proteomic study on faecal samples of patients with cystic fibrosis and their unaffected siblings

Griet Debyser (UGent)
(2015)
Author
Promoter
(UGent) and (UGent)
Organization
Abstract
Cystic Fibrosis (CF) is the most common life threatening genetic disease in the Caucasian population. It affects multiple organs and has major consequences for the whole body. CF evolved from a strictly paediatric disease with poor prognosis to a chronic disease with a growing adult population for which the main focus of healthcare is to improve quality of life. Most of the CF research has focussed on the lung pathology. However, the disease also affects other parts of the body, including the gastrointestinal system. The human digestion system is associated with a complex microbiota and CF affects the gastrointestinal (GI) microbiota, both by biotic (altered intestinal environment) and abiotic (antibiotic treatment) factors. Furthermore, it is proposed that there is a link between GI dysbiosis, GI inflammation and pulmonary exacerbations. Insights in this GI dysbiosis and the host state (inflammation) is needed to adapt existing antimicrobial and anti-inflammatory therapies and/or develop alternative therapies for example through the administration of pre- and probiotics. Previous studies have shown that patients with CF have GI dysbiosis, multiple antibiotic resistant strains are detected in the faeces of the patients, inflammation occurs in the GI tract and probiotics are promising as potential treatments for GI dysbiosis and inflammation. The major goals of the present study were to (i) obtain deeper insights in the composition and functionality of the faecal microbiota of patients with CF and (ii) to investigate the faecal proteome and inflammatory state of the GI-tract of the patients. We received faecal samples from children with CF and their unaffected siblings from the University Hospital of Leuven, in collaboration with Prof. K. De Boeck. The first aim of this thesis was to develop a metaproteomic pipeline to obtain insights in the proteins of both the microbiota as well as the host. This ultimately led to a whole proteome extraction method, in combination with denaturing polyacrylamide gel electrophoresis and liquid chromatography-tandem mass spectrometry. The protocol was optimised and the reproducibility of the protein extraction was demonstrated using the whole proteome fraction from faecal samples (three replicates) from a patient with CF and her sibling. We showed that the metaproteomic analysis provided relevant information on the inflammation status of CF patients in combination with information on the bacterial dysbiosis. The results from this metaproteomic pipeline were used in the first study, a cross-sectional study to investigate the faecal dysbiosis by evaluating the proteins present in the faecal samples in combination with the host proteins. The faecal metaproteomes of fifteen patients with CF and their unaffected sibling were compared using label-free protein quantification methods in combined with computational and bioinformatics methods. The extracted proteins were digested with trypsin and the spectral counts of the peptides of this cross-sectional label-free proteomic study were statistically analysed. The retrieved peptides can be linked to the proteins and in total, 79 proteins were differentially detected between the patients with CF and their unaffected sibling. Of these proteins, 33 were detected more in the patients of which most were human and porcine proteins. The human proteins were associated with inflammation processes; the porcine proteins were administered to the patients via the pancreas enzyme replacement therapy, which is a standard care of treat for CF patients with pancreas insufficiency. In the siblings, 36 bacterial and 10 pancreas associated human proteins were more detected. Most of the bacterial proteins were involved in carbohydrate transport and metabolism. The retrieved spectral counts of the peptides from the LC-MS/MS runs were also analysed by Unipept in combination with statistical analysis. From every peptide, the lowest common ancestor was determined and 100 bacterial taxonomic units were retrieved, which contain at least four peptides. The statistical analysis revealed 27 taxa with taxon-specific peptides, which were significantly different in the patients or the siblings. Of these, seven were detected at species level. A significant reduction of members of the Clostridium cluster IV (Faecalibacterium prausnitzii, Ruminococcus bromii and Subdoligranulum variabile) was noticed in the CF group compared with the unaffected siblings. In addition, a significant reduction in the abundance of bacteria involved in butyrate and propionate metabolism, including Roseburia inulinivorans was detected in patients with CF. Also, an increased abundance of Blautia gnava, Clostridium clostridioforme and Clostridium nexile was found to be associated with dysbiosis in CF. These seven aforementioned species could potentially be used as biomarkers of dysbiosis in CF. The state of intestinal dysbiosis is found in all patients with CF, indicating that there is a disturbed microbiota due to the disease related intestinal dysfunction and/or the frequent use of antibiotics. We found that not only the bacterial composition is different; also, the host proteome shows that there is a pronounced difference between the healthy host proteome and the CF host proteome. In a second study, we developed a selected reaction monitoring (SRM) method for the validation of the potential biomarkers for intestinal dysbiosis in patients with CF. The faecal samples of four patients and their unaffected sibling were used to develop the method for the detection of two bacterial species that could serve as markers for the health state, i.e. Blautia gnava and Faecalibacterium prausnitzii. We targeted peptides from B. gnava, more frequently observed in the patients with CF and F. prausnitzii, which is typically observed at much lower abundance in the patients. Especially F. prausnitzii has the potential as a biomarker for the diagnosis of intestinal dysbiosis. This method could also be a promising alternative method to detect biomarkers of inflammation. By targeting these bacteria in combination with inflammation biomarkers, the infection could be monitored over time and efficiency of therapeutic interventions in the GI tract evaluated. The observed dysbiosis may represent a novel therapeutic target for the gastrointestinal complications in patients with CF. Following the administration of probiotics, the restoration of the intestinal microbiota could be monitored by using these biomarkers (taxa and inflammation biomarkers). In a final study, we aimed to reveal if the state of dysbiosis in patients with CF is temporally or more stable over time. In a longitudinal study, four faecal samples from five patients with CF and their unaffected sibling were analysed with our metaproteomic approach. Using the spectral count data, we could perform two descriptive statistical analyses, i.e. clustering analysis and a canonical correspondence analysis (CCA) for both the Unipept taxonomic data and protein cluster data. The clustering results of the taxonomic data show that the faecal microbiota of the siblings is stable over time, only perturbations due to the administration of antibiotics (AB) influenced the taxonomic units present in these samples. Moreover, 6 months after AB treatment, the faecal microbiota was restored to a similar state as the initial composition. For the patients, who received multiple doses of AB, not all samples from the same patients clustered together. Based on the protein data, the patients and siblings cluster separately, indicating that different proteins were achieved from the patients than from the siblings. The CCA analysis revealed similar results. For the taxonomic data, the siblings cluster together and the patients are more scattered. This is indicating that in the unaffected siblings a core microbiota is present and much temporal variation in presence of taxonomic units is seen in the patients. For the protein cluster data, the CCA revealed that the patients cluster together, indicating that the same proteins are present in the different patients. These proteins are mostly involved in inflammation pathways and proteins originating from the pancreatic enzyme replacement therapy. The siblings are more scattered over the CCA of the protein clusters data, indicating that different bacterial proteins in the patients were detected witch might be due to undersampling. In conclusion, the results from the cross-sectional and longitudinal data in this PhD provided the first metaproteomic data for the intestinal dysbiosis of patients with CF. The faecal dysbiosis that we noticed arised presumably from disease-associated factors such as abnormal mucus secretions and impaired digestion (all patients are pancreatic insufficient) and the destructive effects of the recurrent AB treatments. This all is accompanied with intestinal inflammation, which was found in the cross-sectional and longitudinal study. The intestinal inflammation also influences the bacterial dysbiosis and vice versa. The AB treatments of children with CF is necessary to prevent respiratory infectious outbreaks and a panel of biomarkers for dysbiosis could help in following the state of dysbiosis during CFTR corrector and potentiator therapy and/or treatments with AB. In addition, the panel of biomarkers could be used to follow the effects of probiotic therapies on the intestinal microbiota to improve the well-being of the patients with CF. In this respect, we observed a remarkable similarity with previous metaproteomic and metagenomic data of faecal samples from patients with inflammatory bowel disease, indicative for a common treatment strategy of the intestinal dysbiosis, using pre- or probiotics.
Keywords
Inflammation, Cystic Fybrosis (CF), Intestinal Dysbiosis, Proteomics

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 24.89 MB

Citation

Please use this url to cite or link to this publication:

MLA
Debyser, Griet. “A Proteomic Study on Faecal Samples of Patients with Cystic Fibrosis and Their Unaffected Siblings.” 2015 : n. pag. Print.
APA
Debyser, G. (2015). A proteomic study on faecal samples of patients with cystic fibrosis and their unaffected siblings. Ghent University. Faculty of Sciences, Ghent, Belgium.
Chicago author-date
Debyser, Griet. 2015. “A Proteomic Study on Faecal Samples of Patients with Cystic Fibrosis and Their Unaffected Siblings”. Ghent, Belgium: Ghent University. Faculty of Sciences.
Chicago author-date (all authors)
Debyser, Griet. 2015. “A Proteomic Study on Faecal Samples of Patients with Cystic Fibrosis and Their Unaffected Siblings”. Ghent, Belgium: Ghent University. Faculty of Sciences.
Vancouver
1.
Debyser G. A proteomic study on faecal samples of patients with cystic fibrosis and their unaffected siblings. [Ghent, Belgium]: Ghent University. Faculty of Sciences; 2015.
IEEE
[1]
G. Debyser, “A proteomic study on faecal samples of patients with cystic fibrosis and their unaffected siblings,” Ghent University. Faculty of Sciences, Ghent, Belgium, 2015.
@phdthesis{5839956,
  abstract     = {{Cystic Fibrosis (CF) is the most common life threatening genetic disease in the Caucasian population. It affects multiple organs and has major consequences for the whole body. CF evolved from a strictly paediatric disease with poor prognosis to a chronic disease with a growing adult population for which the main focus of healthcare is to improve quality of life. Most of the CF research has focussed on the lung pathology. However, the disease also affects other parts of the body, including the gastrointestinal system. The human digestion system is associated with a complex microbiota and CF affects the gastrointestinal (GI) microbiota, both by biotic (altered intestinal environment) and abiotic (antibiotic treatment) factors. Furthermore, it is proposed that there is a link between GI dysbiosis, GI inflammation and pulmonary exacerbations. Insights in this GI dysbiosis and the host state (inflammation) is needed to adapt existing antimicrobial and anti-inflammatory therapies and/or develop alternative therapies for example through the administration of pre- and probiotics. 
Previous studies have shown that patients with CF have GI dysbiosis, multiple antibiotic resistant strains are detected in the faeces of the patients, inflammation occurs in the GI tract and probiotics are promising as potential treatments for GI dysbiosis and inflammation. The major goals of the present study were to (i) obtain deeper insights in the composition and functionality of the faecal microbiota of patients with CF and (ii) to investigate the faecal proteome and inflammatory state of the GI-tract of the patients. We received faecal samples from children with CF and their unaffected siblings from the University Hospital of Leuven, in collaboration with Prof. K. De Boeck. 
The first aim of this thesis was to develop a metaproteomic pipeline to obtain insights in the proteins of both the microbiota as well as the host. This ultimately led to a whole proteome extraction method, in combination with denaturing polyacrylamide gel electrophoresis and liquid chromatography-tandem mass spectrometry. The protocol was optimised and the reproducibility of the protein extraction was demonstrated using the whole proteome fraction from faecal samples (three replicates) from a patient with CF and her sibling. We showed that the metaproteomic analysis provided relevant information on the inflammation status of CF patients in combination with information on the bacterial dysbiosis.
The results from this metaproteomic pipeline were used in the first study, a cross-sectional study to investigate the faecal dysbiosis by evaluating the proteins present in the faecal samples in combination with the host proteins. The faecal metaproteomes of fifteen patients with CF and their unaffected sibling were compared using label-free protein quantification methods in combined with computational and bioinformatics methods. The extracted proteins were digested with trypsin and the spectral counts of the peptides of this cross-sectional label-free proteomic study were statistically analysed. The retrieved peptides can be linked to the proteins and in total, 79 proteins were differentially detected between the patients with CF and their unaffected sibling. Of these proteins, 33 were detected more in the patients of which most were human and porcine proteins. The human proteins were associated with inflammation processes; the porcine proteins were administered to the patients via the pancreas enzyme replacement therapy, which is a standard care of treat for CF patients with pancreas insufficiency. In the siblings, 36 bacterial and 10 pancreas associated human proteins were more detected. Most of the bacterial proteins were involved in carbohydrate transport and metabolism.
The retrieved spectral counts of the peptides from the LC-MS/MS runs were also analysed by Unipept in combination with statistical analysis. From every peptide, the lowest common ancestor was determined and 100 bacterial taxonomic units were retrieved, which contain at least four peptides. The statistical analysis revealed 27 taxa with taxon-specific peptides, which were significantly different in the patients or the siblings. Of these, seven were detected at species level. A significant reduction of members of the Clostridium cluster IV (Faecalibacterium prausnitzii, Ruminococcus bromii and Subdoligranulum variabile) was noticed in the CF group compared with the unaffected siblings. In addition, a significant reduction in the abundance of bacteria involved in butyrate and propionate metabolism, including Roseburia inulinivorans was detected in patients with CF. Also, an increased abundance of Blautia gnava, Clostridium clostridioforme and Clostridium nexile was found to be associated with dysbiosis in CF. These seven aforementioned species could potentially be used as biomarkers of dysbiosis in CF. The state of intestinal dysbiosis is found in all patients with CF, indicating that there is a disturbed microbiota due to the disease related intestinal dysfunction and/or the frequent use of antibiotics. We found that not only the bacterial composition is different; also, the host proteome shows that there is a pronounced difference between the healthy host proteome and the CF host proteome. 
In a second study, we developed a selected reaction monitoring (SRM) method for the validation of the potential biomarkers for intestinal dysbiosis in patients with CF. The faecal samples of four patients and their unaffected sibling were used to develop the method for the detection of two bacterial species that could serve as markers for the health state, i.e. Blautia gnava and Faecalibacterium prausnitzii. We targeted peptides from B. gnava, more frequently observed in the patients with CF and F. prausnitzii, which is typically observed at much lower abundance in the patients. Especially F. prausnitzii has the potential as a biomarker for the diagnosis of intestinal dysbiosis. This method could also be a promising alternative method to detect biomarkers of inflammation. By targeting these bacteria in combination with inflammation biomarkers, the infection could be monitored over time and efficiency of therapeutic interventions in the GI tract evaluated. The observed dysbiosis may represent a novel therapeutic target for the gastrointestinal complications in patients with CF. Following the administration of probiotics, the restoration of the intestinal microbiota could be monitored by using these biomarkers (taxa and inflammation biomarkers). 
In a final study, we aimed to reveal if the state of dysbiosis in patients with CF is temporally or more stable over time. In a longitudinal study, four faecal samples from five patients with CF and their unaffected sibling were analysed with our metaproteomic approach. Using the spectral count data, we could perform two descriptive statistical analyses, i.e. clustering analysis and a canonical correspondence analysis (CCA) for both the Unipept taxonomic data and protein cluster data. The clustering results of the taxonomic data show that the faecal microbiota of the siblings is stable over time, only perturbations due to the administration of antibiotics (AB) influenced the taxonomic units present in these samples. Moreover, 6 months after AB treatment, the faecal microbiota was restored to a similar state as the initial composition. For the patients, who received multiple doses of AB, not all samples from the same patients clustered together. Based on the protein data, the patients and siblings cluster separately, indicating that different proteins were achieved from the patients than from the siblings. The CCA analysis revealed similar results. For the taxonomic data, the siblings cluster together and the patients are more scattered. This is indicating that in the unaffected siblings a core microbiota is present and much temporal variation in presence of taxonomic units is seen in the patients. For the protein cluster data, the CCA revealed that the patients cluster together, indicating that the same proteins are present in the different patients. These proteins are mostly involved in inflammation pathways and proteins originating from the pancreatic enzyme replacement therapy. The siblings are more scattered over the CCA of the protein clusters data, indicating that different bacterial proteins in the patients were detected witch might be due to undersampling.
In conclusion, the results from the cross-sectional and longitudinal data in this PhD provided the first metaproteomic data for the intestinal dysbiosis of patients with CF. The faecal dysbiosis that we noticed arised presumably from disease-associated factors such as abnormal mucus secretions and impaired digestion (all patients are pancreatic insufficient) and the destructive effects of the recurrent AB treatments. This all is accompanied with intestinal inflammation, which was found in the cross-sectional and longitudinal study. The intestinal inflammation also influences the bacterial dysbiosis and vice versa. The AB treatments of children with CF is necessary to prevent respiratory infectious outbreaks and a panel of biomarkers for dysbiosis could help in following the state of dysbiosis during CFTR corrector and potentiator therapy and/or treatments with AB. In addition, the panel of biomarkers could be used to follow the effects of probiotic therapies on the intestinal microbiota to improve the well-being of the patients with CF. In this respect, we observed a remarkable similarity with previous metaproteomic and metagenomic data of faecal samples from patients with inflammatory bowel disease, indicative for a common treatment strategy of the intestinal dysbiosis, using pre- or probiotics.}},
  author       = {{Debyser, Griet}},
  isbn         = {{9789461972637}},
  keywords     = {{Inflammation,Cystic Fybrosis (CF),Intestinal Dysbiosis,Proteomics}},
  language     = {{eng}},
  pages        = {{XXI, 244}},
  publisher    = {{Ghent University. Faculty of Sciences}},
  school       = {{Ghent University}},
  title        = {{A proteomic study on faecal samples of patients with cystic fibrosis and their unaffected siblings}},
  year         = {{2015}},
}