Shifts in the microbiome have been correlated with the physiology and pathophysiology of many organ systems both in humans and in mouse models. The gut microbiome has been typically studied through fecal specimen collections. The ease of obtaining fecal samples has resulted in many studies that have revealed information concerning the distal luminal gastrointestinal tract. However, few studies have addressed the importance of the microbiome in the proximal gut. Given that the duodenum is a major site for digestion and absorption, its microbiome is relevant to nutrition and liver disease and warrants further investigation. Here we detail a novel method for sampling the proximal luminal and mucosal gut microbiome in human subjects undergoing upper endoscopy by obtaining duodenal aspirate and biopsies. Specimen procurement is facile and unaffected by artifacts such as patient preparatory adherence, as might be the case in obtaining colonic samples during colonoscopy. The preliminary results show that the luminal and mucosal microbiomes differ significantly, which is likely related to environmental conditions and barrier functions. Therefore, a combination of duodenal aspirate and biopsies reveal a more comprehensive picture of the microbiome in the duodenum. Biopsies are obtained from the descending and horizontal segments of the duodenum, which are anatomically close to the liver and biliary tree. This is important in studying the role of bile acid biology and the gut-liver axis in liver disease. Biopsies and aspirate can be used for 16S ribosomal RNA sequencing, metabolomics, and other similar applications.
The intestinal microbiome has become an area of increased interest in recent years. It is now understood that the diverse bacterial population in the gut can differ based on a variety of factors, including genetics, diet, medication, and environmental influences1. Studies have also identified unique microbial profiles linked to varying gastrointestinal diseases, such as obesity, inflammatory bowel disease, and liver disease2,3. The majority of studies focus on profiling the microbiome of the large intestine through the analysis of fecal and distal mucosal samples4. Although the highest concentration of intestinal bacteria resides in the colon (1012 bacteria/gram), there nevertheless is a complex community of microbes residing in the duodenum (103/g), jejunum (104/g), and ileum (107/g) that plays a key role in digestive metabolism and absorption5.
The small intestine serves as the primary site of nutrient breakdown and absorption in the gastrointestinal tract. Commensal bacteria lining the small intestine play a fundamental role in aiding in the chemical breakdown of food substrates and in the release of bioactive compounds that aid in nutrient absorption6. These interactions contribute to a complex environment of microbe-microbe and host-microbe activity in the small intestine7. A study observing the small intestine microbiota in murine models found that germ-free mice fed a high fat diet had impaired lipid absorption but, when colonized with jejunal microbiota, had a direct increase in lipid absorption6. A human pilot study profiling the duodenal microbiota of obese and healthy individuals found that the duodenal microbiota of obese individuals had alterations in fatty acid and sucrose breakdown pathways, likely induced in a diet-dependent relationship8. Furthermore, dysbiosis in the small intestine microbiota has been identified in several diseases including small intestinal bacterial overgrowth, short-bowel syndrome, pouchitis, environmental enteric dysfunction, and irritable bowel syndrome7.
We are interested in the relationship between the microbiome and different stages of chronic liver disease. Specifically, the duodenum serves as the first site of chemical breakdown and nutritional absorption in the small intestine. Additionally, portal hepatic circulation brings nutrients and metabolites to the liver, where they are processed and regulated into the bloodstream. The anatomical proximity between the gut and the liver creates an environment susceptible to pro-inflammatory responses that can arise due to failure in the gut barrier or alterations in the gut microbiome9. Studies investigating the microbiome and liver disease progression have identified microbial dysbiosis in patients with non-alcoholic fatty liver disease (NAFLD), steatohepatitis (NASH), alcoholic liver disease, and cirrhosis10,11. While the majority of studies characterize the microbiome of the colon, we were interested in investigating the small intestine microbiome in relation to liver disease. By utilizing the novel method presented here, we have identified unique duodenal microbial profiles in patients with liver cirrhosis in relation to diet12.
As characterization of the small intestine microbiome continues to become an area of increased interest, it is necessary to develop uniform techniques for obtaining samples that accurately represent the small intestine microbiota. However, there are challenges associated with specimen procurement that have complicated the study of the small intestine microbiome environment. Current sampling methods require invasive procedures that are often subject to contamination, as outlined by Kastl et al7. Here we detail a novel method for obtaining duodenal aspirate and biopsies for microbial analysis from patients with liver disease undergoing esophagogastroduodenoscopy.
Duodenal samples were obtained at the Veteran Affairs Greater Los Angeles Healthcare System, Cedars-Sinai Medical Center, and the Ronald Reagan UCLA Medical Center after the clinical protocol for the Microbiome, Microbial Markers and Liver Disease (M3LD) study was accepted by the institutional review board of the local ethics review committee. Written informed consent was obtained from all participating patients.
1. Consent of participants
- Approach subjects with a diagnosis of liver cirrhosis of any etiology with endoscopies scheduled for variceal screening or portal hypertension for recruitment. Exclude patients with hepatocellular carcinoma (HCC).
- Introduce the study, carefully ensure participant understanding, and ask the participant if they would like to volunteer in the study. Complete an informed consent form if they agree.
- Ensure that a properly trained and credentialed medical professional performs the endoscopy.
2. Specimen collection
- Anesthetize the patient with 50 mg of fentanyl and 2 mg of midazolam. Increase the dosage as needed until the patient is moderately anesthetized and can tolerate the endoscope.
- Insert an endoscope (e.g., Pentax EG29-i10) into the patient’s esophagus and progress toward the pylorus.
NOTE: The endoscope size will vary based on the patient's physiology. Do not use the endoscope for any aspiration prior to sample collection to avoid contaminating the endoscope channel with gastric or oropharyngeal secretions.
- Obtain a microscope slide and blunt 18 Gauge syringe needle to prepare for specimen procurement. Prepare a sterile pad which will be used to transfer biopsies from the microscope slide to cryovials.
- When the scope arrives at the second portion of the duodenum distal to the ampulla of Vater, aspirate any fluid present through a sterile disposable aspiration catheter passed through the working channel of the upper endoscope and collect it into a 40 mL disposable specimen container. Flush up to 30 mL of additional sterile water by syringe into the duodenum and aspirate the fluid into a fluid specimen trap to collect a total of 15 mL.
- Immediately remove and cap the aspirate container attached to the suction component of the catheter. Place the container into a thermos over ice and into a specimen transport bag for transfer to the processing lab.
- Insert a 2.8 mm single-use biopsy forceps into the port on the side of the endoscope to remove tissue specimens for research. Perform 2 random passes of the forceps within the second portion of the duodenum, collecting two bites of tissue in each pass.
- Use a blunt 18 Gauge syringe needle to extract tissue samples from the forceps and onto the microscope slide before transferring the samples to four separate empty 2 mL cryovials. Tightly screw on the caps.
- Insert all 4 tightly capped cryovials into an empty container, such as a urine cup, and create a bath, consisting of 20 mL of ethanol and dry ice, to flash freeze the biopsies.
- Place the biopsy cryovials in an insulated Styrofoam container on a bed of dry ice and transfer the specimens to a -80 °C freezer for long term storage to preserve the specimens.
3. Aspirate processing
- Following biospecimen collection, immediately transfer the biospecimens to a BSL2 or higher laboratory for processing.
- Gently invert the container of aspirate fluid four times, and then aliquot approximately 1.5 mL of the fluid into 2 mL cryovials.
- Store specimens at -80 °C until batch analysis can be performed.
4. Questionnaire administration and collection of clinical data
- Ask participants to complete questionnaires to elicit information on diet, demographics, smoking /alcohol habits, medical conditions/medications, and quality of life13.
- Review medical charts to capture clinical data including clinical labs, imaging data, medications, and medical complications.
5. DNA extraction
- Perform DNA extraction immediately upon removing the duodenal biopsies and aspirate from storage at -80 °C. No preparatory steps are required to extract DNA from duodenal biopsy specimens.
- Conduct DNA extraction and sequencing for all of the samples on the same day to decrease the risk of a batch effect in analysis. Set up a library specific to the collected specimens before the samples can be sequenced.
- Use a DNA microprep kit to extract DNA from duodenal biopsy specimens, per the manufacturer’s protocol14.
- Add the maximum amount of sample per protocol guidelines to a lysis tube containing 750 µL of lysis solution and a 0.7 mL dry volume of 2 mm glass beads.
- Secure the lysis tubes in a bead beater with a tube holder and run the beater at maximum speed for at least five minutes. The time will vary based on the bead beater and the sample being analyzed.
- Centrifuge the lysis tubes in a microcentrifuge at 10,000 x g for 1 min.
- Transfer up to 400 µL of supernatant through a filtered collection tube and centrifuge again at 8000 x g for 1 min.
- Add 1200 µL of binding buffer to the filtrate in a microcentrifuge tube and mix thoroughly until homogenous.
- Transfer 800 µL of the mixture to a spin column and tube assembly. Centrifuge at 10,000 x g for 1 min. Discard the flow-through.
- Place the spin column on a new centrifuge tube with 400 µL of DNA wash buffer and mix thoroughly until homogenous. Centrifuge at 10,000 x g for 1 min. Discard the flow-through.
- Repeat step 5.10 two times, with 700 µL and then 200 µL DNA wash buffer per wash.
- Add 20 µL of elution buffer to the spin column in a new centrifuge tube. Incubate for one minute, then centrifuge at 10,000 x g for 1 min to elute the DNA. Discard the column.
- Filter the DNA again. The DNA is now ready for PCR amplification.
6. DNA amplification
- Perform PCR amplification of the V4 region of the 16S ribosomal RNA gene by 250 × 2 paired-end sequencing on aMiSeq, HiSeq or NovaSeq sequencer15.
- Create a 96-well plate primer plate. Create a master mix of 165 μL of ILHS_515f conserved forward primer (100 mM stock) and 2970 μL of molecular biology grade water. Add 28.5 μL of this mix to each well.
- Add 1.5 μL of corresponding reverse primer from the 96 well plate of barcoded IL_806r unique primers (100 μM stock). These barcodes will identify unique patient samples.
- Create a master mix of water (23.1 μL x 3.3 x number of samples), PCR buffer (3 μL x 3.3 x number of samples), dNTPs (0.6 μL x 3.3 x number of samples), and JumpStart Taq DNA polymerase (0.3 μL x 3.3 x number of samples).
NOTE: 16S DNA amplification should be performed in triplicate.
- Add 81 μL of master mix to each well in the first 96 well plate. Add 6 μL of DNA and 3 mL of primer mix to each well. Use a P200 multichannel pipette to mix the wells, then pipette 30 μL into the corresponding well of each of the other two PCR plates. Seal the plate.
- Use the following thermal cycler settings: 94 °C x 3 min; 35 cycles: 94 °C x 45 s, 50 °C x 1 min, 72 °C x 1.5 min. After 35 cycles: 72 °C x 10 min, 94 °C x 5 min, 4 °C indefinitely.
7. Cleanup and library setup
- Employ a PCR or DNA cleanup kit to purify the PCR products. Elute the DNA with 30 μL of PCR grade (DEPC-treated) water.
- Use a UV-visible spectrum spectrophotometer that has the capability to quantify PCR samples.
- Open the spectrophotometer software on computer desktop. Click on Nucleic Acid.
- Wipe the port with distilled water. Add 1 μL of 1x PCR buffer/water to port and calibrate.
- Enter the sample ID and pipette 1 μL of sample onto the port.
- Close the port and hit Measure.
NOTE: Avoid air bubbles when adding samples onto the port.
- Wipe off the sample with lab wipes and distilled water, and then prepare the next sample. Record values with each sample. The computer should also auto-save.
- Combine 250 ng of each amplified PCR product in 1 tube. This is the DNA library.
- Deliver the library, along with reads 1 and 2 and the index sequencing primers to a sequencing lab for sequencing.
NOTE: If a local laboratory is not available, a commercial sequencing lab can be utilized for this step.
8. Data formatting
- Sequencing data provides results in forward and reverse .fastq files. Convert this data into a data sheet and clean so that the information can be analyzed for downstream applications.
- Open the R programming software. Follow the instructions and code listed in the supplemental coding file to proceed through the DADA2 pipeline and clean the DNA sequencing data16. Download the SILVA reference database for the taxonomy assignment after the sequences have been passed through the pipeline17.
- Briefly, load the DADA2 package in R. Define a path in which all of the .fastq files are saved in the same directory and extract the files into R.
- Read the names of the fastq files and match the lists of forward and reverse fastq files.
- Inspect the quality of the forward reads and reverse reads.
NOTE: Nucleotide sequences can be truncated to maintain high quality sequences.
- Filter and trim the fastq files. Use the learnErrors method to learn the parametric error model that DADA2 uses. This will show the observed and expected error rates for each possible nucleotide substitution (A → C, A → T, A → G, etc.).
- Apply the sample inference algorithm to determine the number of unique sequences in each sample and further inspect the dada-class object to arrive at the number of true unique variants per sample.
- Merge the forward and reverse reads to obtain complete and denoised sequences.
- Construct an amplicon sequence variant (ASV) table.
- Remove chimeras from the ASV table. Chimeric sequences are a single sequence resulting from two more abundant parent sequences.
- Track the total number of reads that have passed through the pipeline. DADA2 will use SILVA to assign taxonomy to each unique sequence. The accuracy can be analyzed by including a community of known bacteria sequences in the DADA2 pipeline and comparing DADA2 sequence variants to the expected composition of the community.
NOTE: The DNA sequencing data is now cleaned and ready for analysis based on the specific scope and goals of the research project.
Population differences between mucosal and luminal microbiome of proximal gut
Previous studies have found differences in the microbial populations of luminal and mucosal colon specimens4,5,18. The preliminary results show that duodenal aspirate and biopsy specimens can measure for both luminal and mucosal microbiota in the proximal gut. Furthermore, we have found that these microbiome populations are distinct from one another (Figure 1). Additionally, bile acids in the aspirate correlate strongly with microbiome composition of the aspirate (Figure 2).
Figure 1: Microbial differences were identified based on sampling type and location in the proximal gut. (A) Principal coordinates (PC) analysis plot of microbial composition from aspirate samples representative of the luminal microbiome (red) compared to duodenal biopsy samples representative of the mucosal microbiome (blue). (B) Phylum summary for luminal versus mucosal populations. (C) Log2 fold changes (FC) show differential abundance by genus between duodenal aspirate (red) and biopsy (blue) specimens at q < 0.05. Dot size corresponds to genus abundance and color corresponds to phylum. (D) Genus level taxonomic summary by sampling type. Please click here to view a larger version of this figure.
Figure 2: Correlation Matrix of bile acids and microbiota from duodenal aspirate samples in which color represents correlation. The matrix was created with the corrplot and Hmisc packages in R which use Spearman correlation and adjust for false discovery rate. Please click here to view a larger version of this figure.
Supplemental File: DADA2 code Please click here to download this file.
Studies of the microbiome are incredibly important, as this complex ecosystem has a critical role in energy homeostasis, immunologic responses, and metabolism19. Regional microbiome differences exist that may reflect the distinct physiological functions of various regions of the gastrointestinal tract, which may affect different disease states20. The fecal microbiome is most commonly studied but more recently the small intestine microbiome has come under investigation6,21. There is a need to develop a uniform protocol for further study of the proximal gut microbiome and its effects on gastrointestinal health and disease states.
This method allows the investigation of both the luminal and mucosal components of the microbiome of the proximal gut. This has advantages over a nasogastric tube method that cannot consistently procure post-pyloric samples and risks contamination by the oropharyngeal tract. Similarly, collecting biopsies via ileostomy risks exposure to the skin and external environment7. There are populations requiring routine endoscopies for screening and surveillance, such as cirrhotic patients or patients with gastric or esophageal cancer22. Obtaining samples from patients who are already undergoing screening removes the challenge of consenting patients to voluntarily tolerate an invasive sampling procurement, as described by Angelakis et al.8. However, this may create a potential limitation for studies aiming to investigate gastrointestinal physiology that does not require regular screening. Collecting samples by esophagogastroduodenoscopy allows the investigator to precisely analyze microbiome composition from the second and third portions of the duodenum, depending on the area of interest. Forceps are fed through the endoscope port and therefore avoid contamination by the oral cavity microbiome. Additionally, biopsies are associated with very minimal risk and aspiration is routinely used during endoscopy to clear and better visualize the mucosa. This method is valuable in that it is easy to simultaneously collect duodenal aspirates and biopsies for luminal and mucosal proximal gut microbiome analysis.
Modifications can be made to the specimen procurement protocol based on reagent availability and ease of use. In this protocol, a blunt 18 Gauge syringe needle was used to remove biopsies from the forceps but this could easily be substituted with a smaller or larger blunt syringe needle or dressing forceps. Using forceps may allow the sample procurer to transport specimens directly from endoscope forceps to cryovials and to allow the specimens to be frozen more quickly. Biopsies can be placed into cryovials with freezing medium and placed into a freezing container with isopropanol to freeze the tissue samples at a constant, gradual rate until they reach -80 °C. Liquid nitrogen tissue storage could also be valuable depending on the anticipated time from tissue collection to DNA extraction and analysis. Biopsies should not be fixed in formalin, which can disrupt duodenal microbiota and cause issues with DNA extraction and amplification. If the patient is hesitant to donate tissue biopsies, aspirate alone can provide useful information, however the mucosal microbiome cannot be evaluated in this case. Aspirates can also be analyzed by a metabolomics approach or by correlating luminal samples with bile acid concentrations. Further metabolomics analysis can be added by including a blood sample collection from a patient’s pre-procedure.
There are several bioinformatics pipelines that can be used in place of DADA2 that have tradeoffs in sensitivity, specificity, and degree of consensus between outputs23. Once samples have been sequenced and denoised, the possibilities for analysis are very broad. Briefly, sequencing data can be analyzed for alpha and beta diversity, coupled with a metabolomics approach to identify microbial byproducts. QIIME 2 is a useful microbiome analysis package that can generate a variety of different data manipulations from raw DNA sequencing inputs24. This allows flexibility in comparison of the microbial profiles in different disease states. The DNA extraction and sequencing protocols detailed in this manuscript can also be used to study oral, fecal and tissue-based microbiomes.
There are potential limitations to this method that should be considered when devising a protocol to study the duodenal microbiome. There is little correlation between duodenal and fecal microbiota, so results may not be generalizable to other regions of the gastrointestinal tract25. Studies comparing duodenal to oral and fecal microbiota in several disease states have found the communities to be unique as well26,27. Concomitant medications should be reviewed at the time of endoscopy. It is critical to ensure that participants have not used antibiotics in the past month, which are known to affect microbial diversity in the gut28,29. Cholestyramine or proton pump inhibitor usage could potentially affect microbial diversity and confound results30,31. Additionally, flexible endoscopes cannot be heat sterilized and it is possible for biofilms to form, potentially confounding microbial analysis of duodenal biopsies. Microbiological inspection of endoscopes should be considered when reviewing facility endoscope sterilization protocols.
This manuscript details a duodenal microbiome sampling protocol from initial biopsy and aspirate specimen procurement during endoscopy to DNA extraction, sequencing and data formatting to prepare for downstream applications. This method has direct applications specifically to research regarding liver disease. Absorbed macromolecules, microbiota, and their metabolites are carried from the gastrointestinal tract to the liver by portal circulation12. Cirrhosis impairs gastrointestinal epithelial barrier function, which can increase exposure of the liver to these potentially immunogenic features and has also been associated with microbial dysbiosis25. Dysbiosis of the gut microbiome has been linked to the pathogenesis of autoimmune diseases and other changes in immune function32,33. Following this reasoning, the proximal gut microbiota may also be important for the outcomes of liver and other gastrointestinal organ transplantation outcomes due to effects on the anti-graft immune response. Obtaining duodenal aspirate and mucosa samples may be useful for identifying and further investigating the specific roles of small bowel microbiota in gastrointestinal physiology and disease.
The authors have no competing financial conflicts of interest to report.
This research was funded by the National Institutes of Health/National Cancer Institute grant number RO1CA204145.
|2 mL cryovials||Corning||430659|
|96-well plates||Applied Biosystems||4306737|
|Dry ice||Provided by institution|
|EG29-i10 endoscope||Pentax||N/A||Endoscope size may vary depending on patient physiology|
|Epoch microplate spectrophotometer||Biotek||N/A|
|IL_806r reverse primer||IDT DNA technologies||custom||custom primers|
|ILHS_515f forward primer||IDT DNA technologies||custom||custom primers|
|JumpStart Taq DNA||Sigma||D4184|
|Mucus specimen trap||Busse Hospital Disposables||405||40 cc specimen trap with transport cap|
|Nanodrop Gen5 software||ThermoFisher Scientific|
|PCR cleanup kit||Zymo Research||D4204|
|Radial Jaw 4 Jumbo Forceps||Boston Scientific||M00513343||2.8mm Jaw OD|
|Vioscreen dietary questionnaire||VioCare||N/A|
|ZymoBIOMICS DNA Microprep Kit||Zymo Research||D4300||25 ug binding capacity|
- Kau, A., Ahern, P., Griffin, N., Goodman, A. L., Gordon, J. I. Human nutrition, the gut microbiome and the immune system. Nature. 474, 327-336 (2011).
- Young, V. B. The intestinal microbiota in health and disease. Current Opinion in Gastroenterology. 28, (1), 63-69 (2012).
- Kolodziejczyk, A. A., Zheng, D., Elinav, E. Diet-microbiota interactions and personalized nutrition. Nature Reviews Microbiology. 17, (12), 742-753 (2019).
- Tuddenham, S., Sears, C. L. The intestinal microbiome and health. Current Opinion in Infectious Diseases. 28, (5), 464-470 (2015).
- Dieterich, W., Schink, M., Zopf, Y. Microbiota in the Gastrointestinal Tract. Medical Sciences (Basel). 6, (4), 116 (2018).
- Martinez-Guryn, K., et al. Small Intestine Microbiota Regulate Host Digestive and Absorptive Adaptive Responses to Dietary Lipids. Cell Host Microbe. 23, (4), 458-469 (2018).
- Kastl, A. J., Terry, N. A., Wu, G. D., Albenberg, L. G. The Structure and Function of the Human Small Intestinal Microbiota: Current Understanding and Future Directions. Cellular and Molecular Gastroenterology and Hepatology. 9, (1), 33-45 (2020).
- Angelakis, E., et al. A Metagenomic Investigation of the Duodenal Microbiota Reveals Links with Obesity. PLoS One. 10, (9), 0137784 (2015).
- Yu, L. X., Schwabe, R. F. The gut microbiome and liver cancer: mechanisms and clinical translation. Nature Reviews. Gastroenterology and Hepatology. 14, (9), 527-539 (2017).
- Schnabl, B., Brenner, D. A. Interactions between the intestinal microbiome and liver diseases. Gastroenterology. 146, (6), 1513-1524 (2014).
- Shen, T. D., Pyrsopoulos, N., Rustgi, V. K. Microbiota and the liver. Liver Transplant. 24, (4), 539-550 (2018).
- Hussain, S. K., et al. Dietary Protein, Fiber and Coffee Are Associated with Small Intestine Microbiome Composition and Diversity in Patients with Liver Cirrhosis. Nutrients. 12, 1395 (2020).
- Kristal, A. R., et al. Evaluation of web-based, self-administered, graphical food frequency questionnaire. Journal of the Academy of Nutrition and Dietetics. 114, (4), 613-621 (2014).
- Benhammou, J. N., et al. Novel Lipid Long Intervening Noncoding RNA, Oligodendrocyte Maturation-Associated Long Intergenic Noncoding RNA, Regulates the Liver Steatosis Gene Stearoyl-Coenzyme A Desaturase As an Enhancer RNA. Hepatology Communications. 3, (10), 1356-1372 (2019).
- Earth Microbiome Project. 16S Illumina Amplicon Protocol. Available from: https://earthmicrobiome.org/protocols-and-standards/16s/ (2020).
- Callahan, B. J., McMurdie, P. J., Rosen, M. J., Han, A. W., Johnson, A. J. A., Holmes, S. P. DADA2: High-resolution sample inference from Illumina amplicon data. Nature Methods. 13, 581-583 (2016).
- Quast, C., et al. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Research. 41, 590-596 (2013).
- Eckburg, P. B., et al. Diversity of the human intestinal microbial flora. Science. 308, (5728), 1635-1638 (2005).
- Nieuwdorp, M., Gilijamse, P. W., Pai, N., Kaplan, L. M. Role of the microbiome in energy regulation and metabolism. Gastroenterology. 146, (6), 1525-1533 (2014).
- Jacob, N., et al. Inflammation-independent TL1A-mediated intestinal fibrosis is dependent on the gut microbiome. Mucosal Immunology. 11, (5), 1466-1476 (2018).
- Fujimori, S. What are the effects of proton pump inhibitors on the small intestine. World Journal of Gastroenterology. 21, (22), 6817-6819 (2015).
- Garcia-Tsao, G., Abraldes, J. G., Berzigotti, A., Bosch, J. Portal hypertensive bleeding in cirrhosis: Risk stratification, diagnosis, and management: 2016 practice guidance by the American Association for the study of liver diseases. Hepatology. 65, (1), 310-335 (2016).
- Prodan, A., Tremaroli, V., Brolin, H., Zwinderman, A. H., Nieuwdorp, M., Levin, E. Comparing bioinformatic pipelines for microbial 16S rRNA amplicon sequencing. PLoS One. 15, (1), (2019).
- Bolyen, E., et al. interactive, scalable and extensible microbiome data science using QIIME 2. Nature Biotechnology. 37, 852-857 (2019).
- Li, G., et al. Diversity of Duodenal and Rectal Microbiota in Biopsy Tissues and Luminal Contents in Healthy Volunteers. Journal of Microbiology and Biotechnology. 25, (7), 1136-1145 (2015).
- de Groot, P. F., et al. Distinct fecal and oral microbiota composition in human type 1 diabetes, an observational study. PLoS One. 12, (12), (2017).
- Kong, X., et al. New and Preliminary Evidence on Altered Oral and Gut Microbiota in Individuals with Autism Spectrum Disorder (ASD): Implications for ASD Diagnosis and Subtyping Based on Microbial Biomarkers. Nutrients. 11, (9), 2128 (2019).
- Dethlefsen, L., Huse, S., Sogin, M. L., Relman, D. A. The Pervasive Effects of an Antibiotic on the Human Gut Microbiota, as Revealed by Deep 16S rRNA Sequencing. PLOS Biology. 6, (11), 280 (2008).
- Dethlefsen, L., Relman, D. A. Incomplete recovery and individualized responses of the human distal gut microbiota to repeated antibiotic perturbation. PNAS. 108, (1), 4554-4561 (2011).
- Williams, R. C., Showalter, R., Kern, F. In vivo effect of bile salts and cholestyramine on intestinal anaerobic bacteria. Gastroenterology. 69, (2), 483-491 (1975).
- Seto, C. T., Jeralod, P., Orenstein, R., Chia, N., DiBiase, J. K. Prolonged use of a proton pump inhibitor reduces microbial diversity: implications for Clostridium difficile susceptibility. Microbiome. 2, (42), (2014).
- De Luca, F., Shoenfeld, Y. The microbiome in autoimmune diseases. Clinical and Experimental Immunology. 195, (1), 74-85 (2019).
- Belkaid, Y., Hand, T. W. Role of the microbiota in immunity and inflammation. Cell. 157, (1), 121-141 (2014).