A team of scientists at Columbia University Mailman School of Public Health has developed a diagnostic platform that can detect all known human pathogenic bacterial species, plus antimicrobial resistance and virulence genes.
In a study yesterday in mBio, the scientists report that the bacterial capture sequencing (BacCapSeq) system outperformed conventional DNA sequencing methods in identifying pathogenic bacteria and resistance genes in blood samples. It also detected a pathogen that tests that are commonly used for diagnosing bacterial infections would not be able to identify.
The hope is that the platform, which still needs to be subjected to rigorous testing before it can be approved for use as a diagnostic assay in clinical settings, could enable earlier detection and treatment of antibiotic-resistant bacterial pathogens and enhance antibiotic stewardship. Because there is no current diagnostic test that can simultaneously identify bacteria and resistance genes, clinicians often have to treat bacterial infections with broad-spectrum antibiotics, which can hasten the development of resistance.
"BacCapSeq is a method for differential diagnosis of bacterial infections and defining antimicrobial sensitivity profiles that has the potential to reduce morbidity and mortality, health care costs, and the inappropriate use of antibiotics that contributes to the development of antimicrobial resistance," the authors of the study write.
Probes identify, bind to bacterial DNA
BacCapSeq identifies bacteria, resistance genes, and virulence using a probe set of 4.2 million oligonucleotides, which are short nucleic acid polymers designed to detect specific segments of DNA and then bind to them.
The probe set was based on a set of databases containing the genomes of all 307 known human-pathogenic bacterial species, 2,169 antimicrobial resistance gene sequences, and 30,178 virulence factor genes. After the probes bind to the corresponding DNA segment they're looking for, a magnetic process pulls out the genetic sequence, which can then be analyzed to identify the pathogen, resistance genes, and virulence genes.
To test the performance of the platform against standard sequencing methods, the researchers used extracts of blood spiked with the DNA of several species of pathogenic bacteria, whole blood spiked with bacterial cells, blood culture samples, and blood samples from two patients with sepsis of unknown cause.
In each set of tests, BacCapSeq detected a higher number of bacterial reads and obtained higher genome coverage for all bacterial targets than unbiased high-throughput sequencing (UHTS). The enhanced performance was particularly pronounced when the bacterial count in the samples was low. For example, in samples where the bacterial load was as low as 40 cells per milliliter, BacCapSeq detected multiple reads of Mycobacterium tuberculosis, Klebsiella pneumoniae, and Neisseria meningitides, while UHTS detected no sequences.
In the patients who had unexplained sepsis, the bacterial reads obtained with BacCapSeq were 1,000 times higher than those obtained with UHTS. In addition, BacCapSeq identified the bacterium Gardnerella vaginilis—which is only rarely associated with significant disease—as the cause of sepsis in one of the patients. That's a pathogen that multiplex polymerase chain reaction (PCR) assays, which are commonly used in hospitals but can identify only a handful of pathogenic bacterial species, would not have detected.
The platform was also able to identify antimicrobial resistance genes that matched standard antimicrobial sensitivity profiles and biomarkers that differentiate antibiotic-sensitive bacterial strains from antibiotic-resistant ones.
The current version of the BacCapSeq platform does not produce results fast enough to have a major clinical impact. The time from sample acquisition to results takes about 70 hours, while culture-based diagnosis of bacterial pathogens takes 48 to 72 hours. But the researchers believe BacCapSeq will eventually be able to provide faster results with the development of rapid sequencing systems.
That could be beneficial for treatment of sepsis, which can be deadly when it's not quickly recognized and treated with the right antibiotic.
Oct 23 mBio study