All models are wrong, but some are useful: Averting the 'microbial apocalypse'

This article refers to:

Comparative virulence of urinary and bloodstream isolates of extra-intestinal pathogenic Escherichia coli in a Galleria mellonella model

Alexander Fleming's discovery of penicillin, an antibiotic produced by the fungus Penicillium rubens, saw the dawn of a golden age for humankind. The routine use of antibiotics has since prevented a great deal of suffering and saved countless lives.^1,2 Worryingly, that era is now coming to an end. A 2014 report by the World Health Organization describes how antibiotic-resistant superbugs are present in every region of the world, and that many countries lack even the basic systems to track and monitor these dangerous microbes.³ The report calls for drastic action to avert a return to the pre-antibiotic era, where cancer chemotherapy and routine surgery becomes life-threateningly risky and common infections will no longer be treatable. Economist Jim O’Neill was recently commissioned by the UK Prime Minister, the Rt Hon David Cameron MP, to chair the Review on Antimicrobial Resistance.⁴ The first report published by the Review committee makes sober reading; failing to tackle drug-resistant infections will see attributable deaths rise from the current 700,000 per year to 10 million per year by 2050.⁴ The cost to the world's economy will be considerable, at US$100 trillion a year. The committee warns that even these frightening figures do not fully capture what a world without antibiotics would look like; they have not accounted for the impact of being unable to carry out routine procedures, such as joint and hip operations, which allow people to live active lives for longer and enable them to remain in the workforce.⁴

To stay ahead of this 'microbial apocalypse' the world desperately needs new antibiotics and vaccines, and novel ways to tackle infectious microbes. Key to this is a better understanding of the organisms themselves. Despite some limitations, animals play an important role as surrogate hosts for investigating the mechanisms microbes employ to cause disease.⁵ Importantly, the use of animals in scientific research is guided by the promotion of an ethical framework called the 3Rs. First described by William Russell and Rex Birch over 50 y ago,⁶ the 3Rs stipulate that animals must be replaced where possible, or where this is not possible, studies should use methods which reduce the numbers of animals required and procedures should be refined to minimise pain, suffering and distress. Popular replacement hosts include invertebrates such as the nematode Caenorrhabditis elegans and the fruitfly Drosophila melanogaster.⁷ One major drawback of these model organisms is that they cannot be housed at the temperature of the human body, 37°C, an important consideration when investigating the virulence of human and other mammalian pathogens.⁸ These model invertebrates are also very small (1–3 mm) which hinders handling and sample collection, and specialized equipment is needed to house and infect D. melanogaster.

The larvae of the Greater wax moth Galleria mellonella provide a convenient and affordable alternative surrogate host for infectious diseases research. They can be obtained from breeders of insects as food for exotic pets, and can be conveniently housed in disposable laboratory plasticware. Most importantly, they are easily infected via injection and survive incubation at 37°C. Similar to mammals, G. mellonella and other insects possess a circulatory system and a comparable innate immune system,⁹ which enable them to rapidly and effectively respond to the threat of microbial pathogens and parasites. Their cellular immune response consists of haemocytes which engulf and encapsulate invading cells, while their humoral immune response is mediated by a range of antimicrobial peptides, the formation of melanin and the coagulation of haemolymph to seal ruptures in the cuticle and stop the entry of pathogens.⁹ Although the genome of G. mellonella has not yet been deciphered, a recently published comprehensive transcriptome map of infected G. mellonella caterpillars¹⁰ provides a valuable platform for detailed analyses of the organism's immune responses.

It is clear from the published literature that G. mellonella is increasing in popularity as a surrogate host for studying infectious diseases; a Pubmed search of ‘Galleria mellonella infection’ reveals that approximately 70 articles were published in 2014 alone. Many researchers are now using G. mellonella to study the impact of particular genes on microbial virulence, to compare the virulence of different clinical isolates, and to assess the efficacy of potential treatments.^11-30 In this issue of Virulence, Ciesielczuk and colleagues have used G. mellonella to study the pathogenicity of 40 well-characterized clinical isolates of extra-intestinal pathogenic E. coli (ExPEC).²⁸ This is a task for which G. mellonella is well suited as it can become prohibitive to carry out such experiments using more traditional surrogate hosts, such as mice. ExPEC are a significant cause of urinary tract and blood stream infections worldwide.^31,32 Falling into distinct phylogenetic groups, ExPEC isolates also vary in their virulence gene and antibiotic resistance profiles. We have previously demonstrated a correlation between the number of virulence genes and caterpillar mortality using a smaller number of clinical ExPEC isolates.²² In their study, Ciesielczuk and colleagues have further identified sequence type (ST) 131, a genitourinary-source of bacteraemia and the presence of genes for pyelonephritis associated pili (pap) and outer membrane protease OmpT as important variables in the pathogenicity of ExPEC, at least where G. mellonella is concerned.²⁸ Interestingly, a similar recent characterization of 71 clinical uropathogenic E. coli (UPEC) isolates concluded that ST131 isolates were the least virulent,²⁷ and suggested that ST127 strains should be more widely monitored. There may be a number of reasons for these discrepancies,^21,33,34 including that Alghoribi and colleagues studied 10–20 urinary tract infection isolates of each sequence type,²⁷ compared to the 3–4 blood or urine isolates of each sequence type examined Ciesielczuk and colleagues.²⁸ These disparities suggest the need for caution when drawing conclusions based on small sample sizes, but may also reflect the heterogeneous nature of clinical isolates around the world.

As statistician George EP Box (1919–2013) is reported to have said: “All models are wrong, but some are useful.”³⁵ While not without its limitations,^21,33,34 it is clear that G. mellonella has its uses in infectious diseases research. If we are to see G. mellonella earn its place as a true model host organism, it is time for us to standardise our assays, and press for the genomes of the organism known to many researchers around the world as the ‘wax worm’ to be sequenced.