Gut Microbiome analysis – Getting it right
Posted 8th March 2019 by Joshua Broomfield
The cascade of new discoveries relating health and disease to our gut microbiome has spurred the notion that we now find ourselves in the middle of a “microbiome revolution”. Just to mention some recent examples, mechanisms have been demonstrated for gut bacteria contributing to Parkinson’s disease, determining response to immune checkpoint inhibitor cancer therapy, and even autistic behavior when fecal material from autistic children was transplanted into mice.
DNA sequencing technology and its continuous development has played a significant part in growing our understanding of how our microbiome influences our health. The use of ever more advanced bioinformatic analysis coupled with incremental developments in sequencing hardware continuously changes the way in which microbiome analysis is performed and the level of detail it can provide.
This continuous development and advancement of DNA sequencing technology however leaves significant challenges in its wake. Primary among these is the lack of any accepted standards for microbiome analysis. Below are the core steps in microbiome analysis. How each of these steps is done affects the outcome, and each step can skew the results.
The most well-known example of how this variation in methodologies affects the outcome is the story about an individual who sent the exact same sample to two different DNA sequencing service providers, only to receive two entirely different sets of results.
At Bio-Me, we are very conscious of the biases different sampling, shipping, storage and DNA purification procedures can have on the result of the test. Our aim therefore is to determine which procedures are accurate and reproducible. As part of this work, we tested two different DNA purification methods in triplicate samples as part of an effort to document different fecal sampling methods.
All together 45 samples where split into two fractions. QIAamp PowerFecal DNA purification kit from Qiagen was used to purify DNA from one fraction, and ZymoBIOMICS Mini DNA purification kit was used to purify DNA from the other fraction. The two fractions where isolated on the same day by two experienced laboratory personnel. The summary of analysis of the samples from the two methods are given below.
As can be seen, the Zymo kit gave much better results with regards to average yield of DNA and number of samples with high enough quality DNA to be sequenced by a MiSeq Illumina instrument with a 16S amplicon sequencing protocol. The number of reads obtained was also higher with the Zymo kit.
However, the most important point is the observation that the ratio of Gram positive to Gram negative bacteria was much higher with the Zymo kit. A high Gram positive to Gram negative bacteria ratio is thought to be an indication of better or more complete lysis of all the bacteria in the sample, and therefore a more correct picture of the presence of the different bacteria. Further analysis of the samples confirmed this, although we primarily proceeded with analysis of the results from the Zymo kit, since all the samples where sequenced.
A couple of years ago, a great effort was made to try to unify the DNA purification used by different laboratories around the world. The different laboratories sent their samples to each other, and each laboratory used their established DNA purification method, and all samples were analyzed. After a list of many different parameters in the DNA purification protocol that influence the final results was made, they came up with one recommended protocol. Interestingly, the protocol they came up with was based on Qiagen’s QIAamp DNA stool kit. However, this kit has now been discontinued by Qiagen.
In spite of various efforts to unify the various steps of microbiome analysis leading to the final results, there is still a long way to go. As new methodologies are being developed, it may be more and more challenging to compare different studies. This is a universal challenge for all the different omics technologies used in microbiome analysis.
A possible solution forward could be to establish a reference method that is always used by everyone, for example a set of specific qPCR assays that can be used to normalize the results from various DNA analysis methods. Such an approach would make it possible to compare different studies and will benefit companies developing new live biotherapeutics, the CROs offering various microbiome services as well as the regulatory authorities.
Morten L. Isaksen is the founder and CEO of Bio-Me, Norway. He spoke at the 6th Microbiome R&D & Business Collaboration Forum.
See what’s coming up at the 7th Microbiome R&D Business Collaboration Forum: USA. Download the agenda.
 Sarkar R. & Banerjee S, Gut microbiota in neurodegenerative disorders. J Neuroimmunol. 2019 Jan 9; 328:98-104.
 Gopalakrishnan V. et al. The Influence of the Gut Microbiome on Cancer, Immunity, and Cancer Immunotherapy. Cancer Cell 33, April 9, 2018
 Sgritta et al. Mechanisms Underlying Microbial-Mediated Changes in Social Behavior in Mouse Models of Autism Spectrum Disorder. Neuron. 2019 Jan 16;101(2):246-259.e6.
 Costea et al. Towards standards for human fecal sample processing in metagenomic studies. Nature biotech Vol 35 No 11, 1069 – 1076 Nov 2017
Leave a Reply