Single-Cell Analysis of Mycobacterial Persistence Using Microfabricated Tools

Bacterial cells behave as individuals despite being genetically identical and subject to the same environment. Although the underlying mechanisms of cellular individuality are not well understood, temporal variation of gene expression and protein levels in individual cells could act as sources for phenotypic heterogeneity of populations. Bacterial persistence in fluctuating environments relies on such diversity, allowing some individuals, by chance, to survive a challenge that would kill an unadapted and phenotypically homogeneous population. Bacterial persistence is a non-genetic, metastable phenomenon that is readily lost upon subculture of surviving cells in the absence of the environmental stress. A medically relevant example of persistence is the fractional killing of bacteria by antibiotics, whereby a subpopulation of cells survives indefinitely despite prolonged exposure to a bactericidal drug. It is hypothesized in the case of tuberculosis (TB) that drug therapy may require administration of multiple drugs for 6-9 months in order to eliminate a subpopulation of "persister" cells. The focus of this thesis is to understand how persisters tolerate antibiotics that kill their genetically identical siblings. To address this question, automated time-lapse fluorescence microcopy was applied in conjunction with microfluidic and microelectromechanical systems (MEMS) to study bacterial responses to antibiotics in real time at the single-cell level. A genetic approach was employed to identify mutants of Mycobacterium smegmatis with altered rates of persistence against the first-line anti-TB drug isoniazid (INH). A microfluidic platform was designed for high-throughput screening of transposon (Tn) mutant libraries. A pilot screen of 1,100 Tn mutants revealed 11 mutants that showed increased (persister-up, PU) or decreased (persister-down, PD) survival when exposed to isoniazid. These mutants were further characterized in comparison to wild-type cells using the single-cell microfluidics platform. This single-cell analysis revealed distinct modes of persistence within each category, which were otherwise masked at the population level. The results indicate that persistence behavior was the cumulative result of the dynamics of cell division and cell lysis. Mutants with decreased persistence exhibited behavior at single-cell level that challenges the accepted mechanism of isoniazid-mediated killing. Therefore, these results yield new insights into antibiotic persistence and isoniazid's mode of action. Persisters cannot be easily studied in non-fractionated populations of cells where they represent a small percentage of the total. As a complementary approach, a method was developed based on dielectrophoretic (DEP) purification of naturally occurred low frequency persister cells from an antibiotic-treated bacterial culture. Purification of persister cells would allow for downstream analysis using conventional bulk/population level transcriptomic and proteomic approaches. The crossover frequencies for live and dead cells were determined to enable selective trapping or release from the device. This approach enabled a proof-of-concept enrichment of live bacteria in the mixture of live and dead cells after INH treatment, based on their dielectric properties. Quantitative confirmation of the enriched/isolated cell populations was attempted using flow cytometry. The high-throughput microfluidic screening and single-cell approaches developed in this thesis could, in principal, be applied to the genetic analysis of a wide variety of dynamic cellular processes that can be imaged by time- lapse fluorescence microscopy, such as gene expression, cell division, chromosome replication and segregation, and protein localization. The biochemical analysis of live cells isolated by dielectrophoresis-based separation would provide useful information about the underlying persistence mechanism. Dielectrophoresis could also be a valuable analytical tool for mutant characterization based on their intrinsic dielectric properties, either at single-cell or population levels. In conclusion, this thesis presents several new microfabricated tools to identify and understand bacterial persistence. The thesis illustrates how integration of microfluidic tools with automated time-lapse microscopy allows hidden characteristics that are present but invisible in the macro- world of batch cultures to be examined in a quantitative and high-throughput manner in the micro-world of single-cell analysis.

A Microfluidic Platform Enables Large-Scale Single-Cell Screening to Identify Genes Involved in Bacterial Persistence

Amanda Verpoorte

Persistent cells, representing a small fraction of a bacterial population, are refractory to antibiotic therapy. It is a phenotypic transient state. The underlying mechanisms of persistence remain largely unknown, but even divergent hypotheses agree on a fundamental component: persistence arises from phenotypic heterogeneity. Novel approaches are needed to elucidate which factors allow some bacteria to persist but not others. High-content screening (HCS), which uses microscopy as a screening tool, enables the visualization of single-cell phenotypes and is thus particularly suited for studying bacterial persistence. Traditionally performed with eukaryotic cells in micro-well plate format, executing HCS with prokaryotic cells required the development of a screening novel strategy due to their much smaller size. Microfluidics has proved to be a remarkable toolbox for bacterial studies. In combination with time-lapse microscopy, it permits the study of dynamic processes with single-cell resolution in a tightly controlled microenvironment. Though numerous microfluidic platforms for prokaryotes imaging have been developped, the current state of the art does not allow the imaging of thousands of genotypes in parallel, as is required for high-content screening. The focus of this thesis is to develop a high-throughput microfluidic screening platform enabling high-content imaging of prokaryotic cells. This platform allows 1152 bacterial strains to be imaged in parallel on both the single-cell and sub-cellular level with precise control over the growth environment. First, we demonstrate the ability of the platform to effectively screen for both altered cell division phenotypes through subcellular markers imaging in a static environment and for phenotypes emerging in complex medium profiles, such as bacterial persistence. Second, we employed our platform in a genetic screen, exposing a mutant library of Mycobacterium smegmatis to the antibiotic Isoniazid (INH), and identified a total of 31 mutants with altered persistence rates during antibiotic exposure as compared to the wild-type. Eight of these mutants had decreased persistence rates while 23 mutants displayed increased persistence. Third, using the same platform we performed in depth analysis of mutants with altered rates of persistence in order to identify the cell division and cell death dynamics underlying persistence. This single-cell analysis revealed unexpected dynamics where a mutant with decreased persistence displayed a lower killing rate compared to the wild-type. This high-throughput microfluidic screening platform is highly versatile and can be used in a wide range of bacterial studies, as it is capable of imaging not only single cells, but also subcellular markers, such as those based on the cell cycle or bacterial septa. This novel technology thus opens new doors for microbiologists.

EPFL2017

The Phosphate Specific Transport (Pst) System in Fast- and Slow-Growing Mycobacteria

Cyntia De Piano

Although mycobacteria are known to accumulate polyphosphate (PolyP), a linear polymer of orthophosphate, when subjected to stresses, the exact roles of this molecule are not yet clearly understood. We found that deletion of the pstA gene encoding for a transmembrane domain of the high-affinity phosphate-specific ABC transporter system (Pst system) resulted in accumulation of high amounts of PolyP during growth in phosphate (Pi)-replete conditions in Mycobacterium tuberculosis and Mycobacterium smegmatis. These PolyP stores were gradually used during Pi starvation by the ApstA mutants in order to sustain growth and this allowed the mutant strains to reach a higher cell density than the wild-type (WT) parental strain. Neither the mRNA nor the protein levels of polyphosphate kinase 1 (PPK1) could explain the PolyP accumulation observed in the mutant, suggesting that another regulatory mechanism is involved. The ApstA mutant of M. smegmatis was hypersensitive to several in vitro stresses, including detergent, oxidative stress, and cell wall targeting antibiotics. Deletion of the main enzyme involved in PolyP synthesis, PPK1, in the ApstA mutant background of M. smegmatis completely reversed the PolyP accumulation and partially reversed the stress sensitivity as well as the derepression of PHO regulon genes observed in the ApstA mutant. The mechanism of PolyP accumulation in M. smegmatis was investigated using a genetic approach. We found that this mechanism differs from the (p)ppGpp-mediated PolyP accumulation observed in Escherichia coli, since deletion of the RelA enzyme in a ApstA mutant background did not prevent PolyP accumulation. The mechanism of PolyP degradation during Pi starvation in the M. smegmatis ApstA mutant was also analyzed. Deletion of both polyphosphatase enzymes (PPX1 and PPX2) as well as the polyphosphate kinase 2 (PPK2) enzyme did not have any impact on PolyP degradation. PPK1 is a bifunctional enzyme that can either synthesize or degrade PolyP. Our current hypothesis is that PPK1 might perform the reverse reaction (PolyP degradation) during Pi starvation. The aim of the second part of my project was to make a comparative study of the expression and regulation of the phosphate specific transporter (pst) operon in fast-growing [M. smegmatis] and slow-growing [M. tuberculosis] mycobacteria. To achieve this goal, pst reporter strains of M. smegmatis and M. tuberculosis were constructed by precise insertion of a modified dest_gfp gene (provided by Giulia Manina) encoding a destabilized green fluorescent protein (dest_GFP) into the corresponding pst locus as transcriptional ) fusions. The transcriptional reporter construct [p) was introduced in both the WT and the ApstA mutant backgrounds. The kinetics of induction of the pst-gfp reporters in WT and ApstA mutant strains in response to Pi starvation was analyzed. Population-averaged measurements were made using qRT-PCR and Western blotting. Single-cell measurements using flow cytometry and timelapse fluorescence microscopy were also performed. For the microscopy experiments, the bacteria were cultured in custom-made microfluidic devices that permit rapid switching of the culture medium, e.g., switching between Pi-replete and Pi-depleted media. Because these studies require precise environmental control and resolution of individual and temporal phenotypes, my principal experimental tool was the integrated microfluidics and timelapse fluorescence microscopy systems that have recently been developed in the laboratory. Although there have been many studies published on the population-averaged behavior of bacteria responding to external stresses such as Pi limitation, we are not aware of any studies addressing the individuality of such responses using a real-time single-cell approach. We observed that the kinetics of induction of the Pst operon upon Pi starvation scaled with the growth rate of the bacteria and peak induction occurred within about two generations for both smegmatis and tuberculosis. Interestingly, we found that the basal level of Pst expression and the steady state level reached upon induction by Pi starvation were similar in tuberculosis and M. smegmatis, although the kinetics of induction were much faster in M. smegmatis. These studies contributed to improve the understanding of the regulation and the kinetics of response of the PHO regulon in fast- and slow-growing mycobacteria. The finding that PolyP plays a role in this process could also apply to other organisms where the PHO regulon is regulated in a similar way.

EPFL2013

Single-Cell Tracking Reveals Antibiotic-Induced Changes in Mycobacterial Energy Metabolism

Zeljka Maglica, John McKinney

ATP is a key molecule of cell physiology, but despite its importance, there are currently no methods for monitoring single-cell ATP fluctuations in live bacteria. This is a major obstacle in studies of bacterial energy metabolism, because there is a growing awareness that bacteria respond to stressors such as antibiotics in a highly individualistic manner. Here, we present a method for long-term single-cell tracking of ATP levels in Mycobacterium smegmatis based on a combination of microfluidics, time-lapse microscopy, and Forster resonance energy transfer (FRET)-based ATP biosensors. Upon treating cells with antibiotics, we observed that individual cells undergo an abrupt and irreversible switch from high to low intracellular ATP levels. The kinetics and extent of ATP switching clearly discriminate between an inhibitor of ATP synthesis and other classes of antibiotics. Cells that resume growth after 24 h of antibiotic treatment maintain high ATP levels throughout the exposure period. In contrast, antibiotic-treated cells that switch from ATP-high to ATP-low states never resume growth after antibiotic washout. Surprisingly, only a subset of these nongrowing ATP-low cells stains with propidium iodide (PI), a widely used live/dead cell marker. These experiments also reveal a cryptic subset of cells that do not resume growth after antibiotic washout despite remaining ATP high and PI negative. We conclude that ATP tracking is a more dynamic, sensitive, reliable, and discriminating marker of cell viability than staining with PI. This method could be used in studies to evaluate antimicrobial effectiveness and mechanism of action, as well as for high-throughput screening. IMPORTANCE New antimicrobials are urgently needed to stem the rising tide of antibiotic-resistant bacteria. All antibiotics are expected to affect bacterial energy metabolism, directly or indirectly, yet tools to assess the impact of antibiotics on the ATP content of individual bacterial cells are lacking. The method described here for single-cell tracking of intracellular ATP in live bacteria has many advantages compared to conventional ensemble-averaged assays. It provides a continuous real-time readout of bacterial ATP content, cell vitality, and antimicrobial mechanism of action with high temporal resolution at the single-cell level. In combination with high-throughput microfluidic devices and automated microscopy, this method also has the potential to serve as a novel screening tool in antimicrobial drug discovery.

American Society for Microbiology2015