Understanding this phenomenon is the key to deciphering the antibiotic in vitro‑in vivo paradox

Classically, antibiotic drug therapy is delivered empirically and adjusted using data from the in vitro culture of pathogens isolated from patients. However, antibiotic therapy can fail even if the in vitro data on efficacy and pharmacology suggest a positive outcome. This phenomenon is known as an in vitro-in vivo paradox.

Antibiotic therapy failures are becoming widespread which in turn is leading to increasing multi-drug resistance globally. The antibiotic resistance catastrophe has led to severe medical complications such as sepsis, pneumonia and even death. A report last week from UK public health officials detailing the world’s first case of highly multi-drug resistant gonorrhea illustrates the frightening reality of this problem.

Despite the improvements in antimicrobial susceptibility tests, the issue of in vitro-in vivo paradox remains. This is due to the nature of antimicrobial tests being merely an in vitro representation of the real scenario, in much the same way that not all efficacious drug trials in small animals are efficacious in humans. Hence, all in vitro models are a precise depiction only of the behavior of that model under the conditions of the test, not of its potential clinical scenario.

This discrepancy is due to the complex interaction of the drug, the host and the microorganism. The complexity of this interaction is the failure of in vitro tests to precisely reflect the pharmacokinetics or pharmacodynamics of the drug and the immuno-modulatory effects of the antibiotic, as well as host changes in the drug’s activity and bacterial fitness.

The determination of the type of bacterial response seen when confronting a compound is dependent on the receptor. For instance, a compound can be a signal within the member of the same species but appear to be antimicrobial against a different species. What we know from several studies that have investigated the in vitro microbial response to antibiotic challenge is that antibiotics interact with signalling pathways that are already present in the organism and eventually lead to specific changes at the bacterial transcriptional level.

This phenomenon is described as autologous signalling with heterologous inhibition. It has been shown, for instance, that antibiotics promote the increase of mutation rates, and lead to recombination and horizontal gene transfer. Similar to the variation in the signalling mechanism, the same antibiotic can trigger different transcriptomic responses in different bacteria, and different antibiotics can have different responses to the same bacteria. Antibiotics have also shown to regulate the transcription of a large number of genes in S. typhimurium and E. coli with defective mutations at several pathways of the stress response.

To decipher this paradox, the most widespread suggestion has been to improve the in vitro techniques, such as a 3D in vitro models and ex vivo models. Another school of thought—trying to understand the bacteria-antibiotic regulation directly in animal models employing fluorescent imaging techniques or to directly investigate the bacterial transcriptome from the animal infection models.

Irrelevant of the approach researchers take, the light at the end of the tunnel will only come closer if there is a wholesome approach that combines the physiology of the pathogen—especially its level of fitness—the drug exposure levels, and host-response mechanisms, such as the innate immune system. Approaches that harmonize these three concepts will ensure the elucidation of this paradox and eventually aid us in combating the antibiotic resistance epidemic.

For further reading, please refer to:

  1. Thadepalli H, Bansal MB, Rao B, See R, Chuah SK, Marshall R, Dhawan VK. 1988. Ciprofloxacin: in vitro, experimental, and clinical evaluation. Rev. Infect. Dis. 10:505–15. 
  2. Bishai W. 2002. The in vivo-in vitro paradox in pneumococcal respiratory tract infections. J. Antimicrob. Chemother. 49:433–6. 
  1. Waddell SJ, Butcher PD, Stoker NG. 2007. RNA profiling in host-pathogen interactions. Curr. Opin. Microbiol. 10:297–302. 
  1. Salvador E, Wagenlehner F, Köhler C-D, Mellmann A, Hacker J, Svanborg C, Dobrindt U. 2012. Comparison of asymptomatic bacteriuria Escherichia coli isolates from healthy individuals versus those from hospital patients shows that long-term bladder colonization selects for attenuated virulence phenotypes. Infect. Immun. 80:668–78. 
  1. Wiles S, Hanage WP, Frankel G, Robertson B. 2006. Modelling infectious disease – time to think outside the box? Nat. Rev. Microbiol. 4:307–12.
  2.  Shaw KJ, Miller N, Liu X, Lerner D, Wan J, Bittner A, Morrow BJ. 2003. Comparison of the changes in global gene expression of Escherichia coli induced by four bactericidal agents. J. Mol. Microbiol. Biotechnol. 5:105–22.
  1. Goh E-B, Yim G, Tsui W, McClure J, Surette MG, Davies J. Transcriptional modulation of bacterial gene expression by subinhibitory concentrations of antibiotics. Proc. Natl. Acad. Sci. U. S. A. 99:17025–30. 
  1. Ng W-L, Kazmierczak KM, Robertson GT, Gilmour R, Winkler ME. Transcriptional Regulation and Signature Patterns Revealed by Microarray Analyses of Streptococcus pneumoniae R6 Challenged with Sublethal Concentrations of Translation Inhibitors. J. Bacteriol. 185:359–370.
  2. McElroy, MaryMaking Efficacy Models Count. Eureka Blog. Sept. 16, 2013.