On‐chip spectroscopic assessment of microbial susceptibility to antibiotics within 3.5 hours

Abstract

In times of rising antibiotic resistances, there is a high need for fast, sensitive and specific methods to determine antibiotic susceptibilities of bacterial pathogens. Here, we present an integrated microfluidic device in which bacteria from diluted suspensions are captured in well-defined regions using on-chip dielectrophoresis and further analyzed in a label-free and non-destructive manner using Raman spectroscopy. Minimal sample preparation and automated sample processing ensure safe handling of infectious material with minimal hands-on time for the operator. Clinical applicability of the presented device is demonstrated by antibiotic susceptibility testing of Escherichia coli towards the commonly prescribed second generation fluoroquinolone ciprofloxacin. Ciprofloxacin resistant E. coli were differentiated from sensitive E. coli with high accuracy within roughly three hours total analysis time paving the way for future point-of-care devices. Spectral changes leading to the discrimination between sensitive and resistant bacteria are in excellent agreement with expected metabolic changes in the bacteria due to the mode of action of the drug. The robustness of the method was confirmed with experiments involving different chip devices with different designs, both electrode as well as microfluidics design, and material. Furthermore, general applicability was demonstrated with different operators over an extended time period of half a year.

Raman spectral signature reflects transcriptomic features of antibiotic resistance in Escherichia coli

“…In the past decade, the application of vibrational spectroscopy to identify antibiotic resistance has become a promising avenue of research leading to the development of diagnostic methods 14, 16–18 . Previous studies have focused on monitoring phenotypic responses by bacteria in response to direct drug exposure to evaluate the impact of antibiotic treatment or to investigate the antibiotic mode of action 4–18 .…”

“…The accuracy of the predictions demonstrates the ability of Raman spectroscopy to produce fine-grained, reliable, phenotypic signatures for the characterization of antibiotic resistance even though strains are closely related. Further investigations could address the antibiotic resistance dynamics in unknown genetic background or a broader range of species by performing analytical models trained on spectral database established on a large variety of species, as suggested in previous studies 16–18 .…”

“…Approximately 50 min were required to scan and entire 96-well plate. Cells were kept alive in a microchamber at 37°C during analysis to avoid metabolic stress 14, 16 , rather than measuring the Raman spectra of fixed or dried bacteria cells 5, 7, 9–11 . Measurements were performed using WinSpec (Princeton, NJ, USA) synchronized to a home-made automated program developed in IGOR (IGOR Pro v6, WaveMetrics, Inc., Portland, USA).…”

“…Raman spectroscopy has been successfully used to monitor the phenotypic responses of bacterial species exposed to various drugs, at various concentrations, and for various durations of exposure 5–17 . Cost-efficient and fast identification measurements of various types and subtypes of pathogenic and commensal bacteria and mycobacteria exposed to drugs were reported 14, 16, 17 , demonstrating the potential of this technique by comparison to more conventional approaches. In addition, some studies also hinted at possible metabolic or structural relations between cells and spectral signatures 15, 17, 18 .…”

Imaging the invisible—Bioorthogonal Raman probes for imaging of cells and tissues

“…Therefore, antibiotic resistance represents a potential danger for humans. Hence, it is regarded as a major public health concern [51,52].…”