By Monica Heger
A new method developed on the Illumina Genome Analyzer could make the platform useful for microbiome sequencing and bacterial diagnostics, according to researchers from Wayne State University.
The 16S rRNA gene sequence is a useful identifier of bacterial species, and population-based sequencing strategies that target just that region can characterize the entire spectrum of microbial diversity. Typically such experiments are done on Roche's 454 GS FLX because the platform's relatively long read lengths are needed to sequence the entire 300 bases of the region. However, the Wayne State researchers devised a method on the Illumina GA that effectively increased the read lengths so that they were able to sequence the entire 16S rRNA gene.
The method, published in the June issue of Systems Biology in Reproductive Medicine, could enable lower-cost epidemiological studies of bacteria's role in infection in the genital tract, gastrointestinal tract, and even skin, according to the authors, who believe that diagnostic tests could eventually be developed based on the approach.
"It is going to be very important, I think, in clinical applications simply because it will let us know far more about the human microbiome and the complexity of bacterial populations that live on our innards," said David Miller, a molecular andrologist in the division of reproduction and early development at the University Leeds Institute of Genetics, Health and Therapeutics, who also wrote an accompanying editorial to the study.
Jeffrey Ram, the lead author of the study, said that the key to the method was designing primers that corresponded to highly conserved regions of the 16S rRNA gene, which enabled the researchers to sequence the entire length of the region.
Instead of using Illumina's primers to initiate sequencing, the team designed its own primers. One set of primers was targeted to the middle of the 16S rRNA gene, while another set was targeted to either end of the gene.
Then, using a 75-base paired-end sequencing strategy, they were effectively able to produce Illumina reads of around 300 bases.
The method could have a range of clinical applications, in particular for reproductive medicine, said Ram.
"I do think it could be used as a diagnostic test," he said. However, "these microbiomes are just beginning to be analyzed, so a lot of epidemiological work needs to be done first," in order to differentiate pathogenic bacteria from normal bacteria.
The cost of the method would allow it to be used both for the initial research of characterizing human microbiomes and also for an eventual test. Ram estimated that the current price of the method on an Illumina GA would run about $350 per sample, which would include the cost of consumables and sequencing labor, but not the sample prep or analysis. Additionally, the cost assumes the multiplexing of 12 samples. Multiplexing up to 48 samples could bring the cost to under $100 per sample.
Miller added that the method could be useful in reproductive medicine by helping to pinpoint the role of bacteria in premature birth or miscarriage.
"There's always been an element of infection that's thought to be involved in [premature birth and miscarriage]," he said. "This technology will make a big difference in understanding the normal pattern of flora and how disturbances can lead to premature birth or even miscarriage."
In the proof-of-principle study, Ram and his team demonstrated the method on bacteria with known 16S rRNA gene sequences including E. coli, Streptococcus pyogenes, Bacteroides fragilis, and Desulfovibrio vulgaris, as well as several mixed samples from fecal matter whose bacterial contents were unknown.
The proof-of-principle confirmed that the self-designed primers worked, although it revealed that more work would need to be done on the bioinformatics pipeline.
Applying the method to epidemiological studies of vaginal samples could be useful in determining changes in the proportions of specific bacterial species in disease, and could help quantify species that are seen at low frequencies, the authors concluded.
Miller agreed, noting that he became interested in the study because he previously did research in preterm labor, for which bacteria has been thought to play a role. Large-scale sequencing studies of bacteria from the genital tract have not been done because of the cost, but this method greatly reduces that cost, making those studies possible, he said.
However, Miller said that while the method would be extremely useful for epidemiological studies and identifying a bacterial cause for diseases, it would have less utility as a diagnostic.
"Sequencing will be most powerful in understanding what's going wrong," he said. "It's unlikely that diagnostic tests will be based entirely on sequencing." Once the pathogen is identified, he said, it would be relatively cheap to screen for it on other technology, which could be culture-based, based on nucleic acid amplification, or a serology test, he said.
But because not all bacteria is culturable, and if certain pathogens are present at very small concentrations, a sequencing-based test would be advantageous, said Ram, since it could detect populations below 0.1 percent frequency, and would not require the bacteria to be cultured. Additionally, as sequencing prices continue to fall, cost may soon not be an issue.
Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at mheger [at] genomeweb [.] com.