[psst: want an exciting job? read on!]
During a decade of progress, bacterial genomics has already primed the development of new vaccines, novel anti-bacterial agents and innovations in identifying, culturing and tracking the spread and evolution of bacterial pathogens. Yet, despite its success as a research discovery platform, genome sequencing has, until now, proven too cumbersome and expensive to be used as a routine diagnostic or epidemiological tool in clinical bacteriology. But now, with the arrival of “next-generation” high-throughput sequencing (HTS) technologies it may be possible to deliver sequence data quickly and easily enough (i.e. within days or weeks) for it to have a real-time impact in the clinical arena.
Several pioneering efforts have established the potential of high-throughput sequencing in pathogen discovery, albeit primarily in virology. Unbiased 454 sequencing of cDNA from three Australian transplant recipients with fatal infections yielded >100,000 sequences, of which 14 resembled sequences from lymphocytic choriomeningitis virus (Palacios et al., 2008, N Engl J Med, 358, 991-8). Specific PCR and serological assays were then used to confirm the infection. Similar approaches made it possible to identify Lujo virus, a novel arenavirus responsible for a nosocomial outbreak of haemorrhagic fever in South Africa, within 72 hours of sample receipt (Briese et al., 2009, PLoS Pathog, 5, e1000455) and to obtain 70% of the genome of a novel species of Ebola virus (Bundibugyo ebolavirus) from Uganda in less than 10 days (Towner et al., 2008, PLoS Pathog, 4, e1000212).
A metagenomics survey of faeces has shown the feasibility of detecting bacterial pathogens such as Campylobacter through unbiased high-throughput sequencing of microbial DNA (Nakamura et al., 2008, Emerg Infect Dis, 14, 1784-6), while several publications have shown that high-throughput sequencing can be applied to the analysis of amplified molecular bar-codes, such as 16S rDNA sequences (a topic to be covered in a later posting). However, at present, the most exciting application of high-throughout sequencing in clinical bacteriology is in unravelling the fine-grained epidemiology of pathogen spread. In this context, genome sequencing represents the ultimate epidemiological typing method—a universally applicable, digital, “library” typing method, portable internationally and across time; capable of distinguishing strains that differ by as little as a millionth of a genome. In detecting differences between strains that appear indistinguishable by current methods, high-throughput sequencing can certainly augment existing typing approaches and might give us the chance to unrvael chains of transmission of pathogens. A more provocative thought is that high-throughput whole-genome sequencing might soon consign all existing typing methods to the dustbin of history!
In a recent study published in Science, Harris et al (Harris et al., 2010, Science, 327, 469-74) used Illumina sequencing of 63 isolates, twinned with alignment against a reference, to obtain a high-resolution view of the genomic epidemiology of an epidemic strain of methicillin-resistant Staphylococcus aureus (MRSA). The resulting SNP-based phylogeny revealed a geographic structuring within the lineage spanning five continents and person-to-person transmission within the hospital environment. Beres et al, in a PNAS paper, used a similar approach to unravel the genomic epidemiology of invasive Streptococcus pyogenes infections in Ontario, Canada. They sequenced the genomes of 95 strains from three epidemics, which coupled with analysis of 280 SNPs in 344 strains, revealed a complex population structure and dynamic mixture of clonal complexes. They were also able to identify genes with increased rates of nonsynonymous SNPS, which were probably the result of adaptive changes.
Recently, we have also turned our attention to the potential of high-throughput genome sequencing in the epidemiology of bacterial pathogens. Acinetobacter baumannii is an important cause of healthcare-associated infection, particularly ventilator-associated pneumonia and bloodstream infections in critically ill patients. Two features of the organism stand out as particularly alarming: firstly, its propensity to cause outbreaks in the hospital setting, particularly in intensive care units; and secondly, its ability to acquire resistance to multiple antibiotics. Indeed, this bacterium is frontrunner in the progression towards what has been called the “post-antibiotic apocalypse”, where pathogens become resistant to all available antibiotics. A. baumannii infections and outbreaks have proven a problem in hospitals across the developed world, with three “international clones” (European Clones I, II and III) spreading widely. Acinetobacter infection has also emerged as a threat to British and American casualties of the Iraq and Afghanistan wars, prompting high-level coverage in the news media and even on a patient advocacy site (www.acinetobacter.org). Our own experience in Birmingham, where UK military casualties are treated, confirms that the problem extends beyond the theatre of war, with strains introduced by returning casualties cross-infecting other military or civilian patients.
In a modestly priced survey (~£5000 in sequencing costs, kindly funded by the Hospital Infection Society), we applied 454 whole-genome sequencing to six isolates of Acientobacter baumannii that were indistinguishable using standard typing techniques. The identification of SNPs in three loci facilitated discrimination between alternative epidemiological hypotheses and provided evidence of transmission between a military and civilian patient.
This small scale study, which was published recently in the Journal of Hospital Infection, primed a successful proposal to the UK Medical Research Council to use high-throughput bacterial genome sequencing to advance our understanding of the biology, epidemiology and evolution of multi-drug resistant Acinetobacter baumannii and to explore the potential of genome sequencing as a tool in clinical microbiology and infection control. To do this, we will be working with partners from the National Health Service and the Health Protection Agency.
We have ambitious plans—we aim to genome-sequence at least 180 Acinetobacter isolates over the next three years and perform high-throughput phenotypic assays, pioneering an approach that could be applied to a wide range of human pathogens…
And so we are looking to recruit an ambitious, energetic and successful Research Fellow in Bacterial Pathogenomics to bring biological, epidemiological and/or clinical expertise to our high-throughput genotypic and phenotypic analyses of Acinetobacter isolates. If you think you fit the bill, take a look at the Job Description and follow this link for instructions on how to apply. Closing Date: 9 Jul 2010. Interview Date: 22 July 2010. For informal enquiries please e-mail me on email@example.com. Feel free to forward this blog post to anyone who might be interested.