Characterization of multi-drug tolerant persister cells in Streptococcus suis
© Willenborg et al.; licensee BioMed Central Ltd. 2014
Received: 25 February 2014
Accepted: 6 May 2014
Published: 12 May 2014
Skip to main content
© Willenborg et al.; licensee BioMed Central Ltd. 2014
Received: 25 February 2014
Accepted: 6 May 2014
Published: 12 May 2014
Persister cells constitute a subpopulation of dormant cells within a microbial population which are genetically identical but phenotypically different to regular cells. Notably, persister cells show an elevated tolerance to antimicrobial agents. Thus, they are considered to represent a microbial ‘bet-hedging’ strategy and are of particular importance in pathogenic bacteria.
We studied the ability of the zoonotic pathogen Streptococcus (S.) suis to form multi-drug tolerant variants and identified persister cells dependent on the initial bacterial growth phase. We observed lower numbers of persisters in exponential phase cultures than in stationary growth phase populations. S. suis persister cells showed a high tolerance to a variety of antibiotics, and the phenotype was not inherited as tested with four passages of S. suis populations. Furthermore, we provide evidence that the persister phenotype is related to expression of genes involved in general metabolic pathways since we found higher numbers of persister cells in a mutant strain defective in the catabolic arginine deiminase system as compared to its parental wild type strain. Finally, we observed persister cell formation also in other S. suis strains and pathogenic streptococcal species.
Taken together, this is the first study that reports multi-drug tolerant persister cells in the zoonotic pathogen S. suis.
Formation of persister cells by bacteria is a phenomenon that, amongst others, contributes to tolerance of a bacterial subpopulation to antimicrobial agents. Notably, this antibiotic tolerance of persister cells is distinct from genetically inherited resistance. The persister cell subpopulation has been firstly described and named nearly 70 years ago  and research on persister cells has identified a number of typical characteristics as debated recently . Bacterial persister cells seem to represent a stage of dormancy that protects them from killing by antimicrobial substances, even in the presence of concentrations which vastly exceed the minimal inhibitory concentration (MIC). Persister cells are genetically identical to antibiotic sensitive bacteria within a population, but have a distinct phenotype in that they are tolerant to certain antibiotics . Since most antibiotics target bacterial components or pathways involved in replication, the dormancy stage in persister cells is thought to be the underlying mechanism of antibiotic tolerance . Nevertheless, persister celIs can switch from the dormant into a replicating stage. This ‘bet-hedging’ strategy is thought to be a survival strategy of microbial populations .
Two different types of persister cells have been postulated. Type I persister cells are formed in response to environmental stimuli, for instance during the initiation of the stationary growth phase, whereas type II persister cells arise stochastically within a dividing population [6, 7].
A recent report proposed a ‘persistence-if-stuff-happens’ hypothesis, i.e. persister cell formation is an inevitable process due to cellular errors that produce transient states of reduced replication and/or metabolic activity in a single bacterium . Nevertheless, in the last years many attempts have been made to identify molecular factors involved in the development of a persister cell subpopulation. There is increasing evidence that toxin-antitoxin modules, quorum-sensing molecules, global transcriptional regulators, and molecules of the stringent response like (p)ppGpp are involved in persister cell formation [4, 9–13].
Since the first report by Bigger in 1944 , bacterial persister cells have been described for a number of different species, including Escherichia coli, Staphylococcus aureus[14, 15], Pseudomonas aeruginosa, and Mycobacterium tuberculosis[17, 18]. For most of these bacterial species persister cells have also been found in biofilms, which contribute to recalcitrant and/or recurrent infections after antibiotic therapy [4, 19–25].
Little is known about persister cell formation in streptococci [9, 26]. Within pathogenic streptococci, the zoonosis Streptococcus suis (S. suis) is of particular interest since it can cause very severe diseases, such as sepsis, meningitis and streptococcal toxic shock like syndrome in humans who are in close contact to pigs or pig products [27–30]. Notably, S. suis has been shown to be one of the most frequent causes of adult bacterial meningitis in Asian countries including Vietnam and Thailand [31, 32]. S. suis infections are widely distributed in pigs, but can also occur in wildlife animals such as wild rabbits or wild boars [33, 34]. In pigs S. suis is a frequent early colonizer of the upper respiratory tract. In young pigs S. suis is also a major cause of meningitis, arthritis, and septicemia. Thus, S. suis infections are a major concern in the swine producing industry as they lead to high financial losses .
Since antibiotics are widely used to control S. suis infections (in humans and in animals), we examined the ability of S. suis to produce antibiotic tolerant persister cells. We analyzed the effects of the initial bacterial growth phase on persister cell formation, the tolerance of these cells to different types of antibiotics, as well as persister cell levels of different S. suis strains and other human pathogenic streptococci. Our results show for the first time that S. suis forms high levels of persister cells that confer tolerance to a variety of antimicrobial compounds. We also present evidence that persister cell formation is not only found in S. suis but also in other streptococcal species.
Next we studied the persister cell levels of stationary grown S. suis since for several other bacterial species a drastic increase in persister levels has been reported at the onset of stationary growth phase . Antibiotic treatment of stationary cultures of S. suis with 100-fold MIC resulted in a substantial drug tolerance, i.e. a distinct biphasic killing pattern such as seen with exponential cultures was not observed (Figure 1A vs. B). Only a slight decrease in numbers of CFU was detected over time after treatment with β-lactams, ciprofloxacin and gentamicin. In the case of gentamicin a relative difference of approximately three logarithmic orders in CFU was recorded after the first hour of antibiotic treatment, when comparing populations of exponential and stationary grown S. suis. Notably, growth to the stationary growth phase did not enhance the tolerance of S. suis to the cyclic lipopeptide daptomycin which completely killed the S. suis population after only one hour of treatment. Taken together, the killing kinetics revealed that under the conditions tested S. suis develops a growth phase dependent subpopulation showing antibiotic tolerance to a variety of antimicrobial compounds except daptomycin.
In order to dissect whether type I or type II persisters are responsible for gentamicin tolerance, we performed a persister cell elimination assay. In this assay the formation of type I persisters is suppressed by sequential re-inoculation of an early exponential culture. This procedure leads to dilution of type I persisters whilst stochastically build type II persister cell levels should remain constant. As depicted in Figure 2B the percentage of antibiotic tolerant persisters decreased sequentially after 100-fold MIC gentamicin challenge when the bacterial culture was kept in the early growth phase for three cycles. This data indicate that gentamicin tolerant persisters are not or only rarely produced in the early exponential growth phase and that most of the tolerant bacteria represented type I persisters. These were probably ‘left overs’ from the overnight culture and became diluted within repeated cycles of exponential growth.
Generation of bacterial persister cells is important not only with respect to the understanding of population dynamics but also concerning antibiotic tolerance in respective therapy of infections . Accordingly, there is growing evidence that bacterial persisters are involved in relapses of refractory bacterial infections and in the establishment of resistance mechanisms in bacteria . Owing to this it seems not surprising that persister cells have been described for numerous pathogenic bacteria. In this study we have shown for the first time that S. suis forms multi-drug tolerant persister cells. Even though 100-fold MIC is unlikely to be achieved in therapy of natural infections, we assumed that this treatment would be a method of choice to identify highly drug-tolerant persister cells of S. suis in accordance to results reported for S. aureus. By this we identified persister cell formation in three different S. suis strains, suggesting that this phenomenon may be a general trait among this species. Though this has to be further confirmed by testing more S. suis strains and antibiotics that are of higher clinical relevance to treat S. suis infections in pigs and humans, persister cells should be considered in the future in cases of ineffective antibiotic treatments or when studying antibiotic tolerance of S. suis.
In line with several previous studies [3, 14, 22, 46] the number of persisters observed was higher during stationary growth of S. suis when compared to exponential grown bacteria. Type I persisters were found to be the main source of antibiotic tolerance in our experiments. Among other stress signals, nutrient limitation in stationary growth is thought to be a trigger inducing down-regulation of the metabolic activity and bacterial dormancy in energy-deprived cells which can protect the bacteria from antibiotic killing. We found some hints for involvement of the catabolic enzyme system ADS, since approximately two log-fold higher levels of persister cells were found in the exponential growth phase of an arginine deiminase knock-out strain (10ΔAD) as compared to its wild type strain. In S. suis the arginine deiminase system metabolizes arginine as a substrate to produce energy in form of ATP . The diminished ATP levels may lead to reduced general metabolic activity of strain 10ΔAD that might explain the slower growth rate (see Additional file 2: Figure S1) and enhanced number of antibiotic tolerant persister cells. Furthermore, the ccpA deficient strain exhibited lower numbers of persister cells in the stationary growth phase when compared to the wild type. This is in agreement with studies in S. gordonii showing that a ccpA knock-out resulted in an increased sensitivity of the bacteria to penicillin treatment . Since CcpA is a pleiotropic regulator that is important for a balanced metabolic flux in the central carbon metabolism, the alteration of central metabolic processes may influence persister cell formation of S. suis. Accordingly, an interplay between carbohydrate consumption and formation of persisters has recently been demonstrated for E. coli. Further studies are needed to clarify the mechanisms involved in CcpA and/or arginine deiminase dependent changes in antibiotic tolerance of S. suis.
When using antibiotics with varying modes of action, resulting killing profiles were quite different, ranging from pronounced biphasic killing patterns to nearly plane curves, at least for exponential grown S. suis. These findings seem to be highly dependent on the type of antibiotic used, which is also emphasized by the fact that treatment with the β-lactam antibiotics amoxicillin and penicillin resulted in similar killing curves. Thus we speculated a common mechanism of tolerance for certain antibiotics and tested the gentamicin tolerance in other streptococcal species. S. suis strain 10 highly tolerated 100-fold MIC of gentamicin, whereas the other streptococcal strains were completely killed after one hour. These data suggest that a specific mechanism for gentamicin tolerance of S. suis persisters may have evolved and that this is, most likely, not due to a shared genetic background within the genus Streptococcus. Interestingly, after gentamicin treatment of S. suis we also observed a small-colony-variant (SCV) like phenotype (data not shown) that has also been reported for S. aureus upon aminoglycoside treatment [15, 48]. Although it reverted to the typical large-colony phenotype after subcultivation, it remains to be elucidated if this phenotype will change to a stable phenotype after longer exposure times and altered antibiotic tolerance to aminoglycosides. However, at the stationary growth phase the investigated S. suis strain 10 highly tolerated several antimicrobials targeting different bacterial components over time. Given the high rate of multi-drug tolerant cells produced by S. suis strain 10 during stationary growth, it was remarkable that the cyclic lipopeptide daptomycin efficiently eradicated this subpopulation. This is in contrast to observations that in S. aureus 100-fold MIC of daptomycin failed to eradicate stationary phase cultures . Even though the MIC for daptomycin is rather high when compared to that of other streptococcal species  this treatment eradicated S. suis persister cells in vitro.
In the last years bacterial persistence and enhanced antibiotic tolerance was intensively discussed in the context of recurrent infections caused by bacterial pathogens. Interestingly, a human case of recurrent septic shock due to a S. suis serotype 2 infection has previously been reported . Together with our present study this suggests a clinical relevance of S. suis persisters. Although experimental evidence for S. suis persister cell and biofilm formation in vivo is yet missing, S. suis is able to produce biofilms in vitro that tolerate antibiotic challenge [51, 52]. Given the fact that the S. suis colonization rate of pigs is nearly 100% [35, 53, 54] and that antibiotic treatment with penicillin, ampicillin, or ceftiofur failed to eliminate the tonsillar carrier state of S. suis in swine , it is plausible to speculate that persister cells, possibly also as part of biofilm structures, may contribute to the observed problems in antibiotic treatments. Indeed, P. aeruginosa persister cells have been described as the dominant population responsible for drug tolerance in biofilms .
Our study showed that the zoonotic pathogen S. suis is able to form a multi-drug tolerant persister cell subpopulation. S. suis persister cells tolerated a variety of antimicrobial compounds that were applied at 100-fold of MIC and could be detected in different S. suis strains. Thus, our study provides a basis for future studies on the role of S. suis persister cells in bacterial colonization of host tissues, general antibiotic tolerance, and recurrent infections.
Bacterial strains used in this study
Virulent serotype 2 strain, porcine isolate
Strain 10 ccpA mutant; ccpA::EmR
Strain 10 arginine deiminase operon mutant; arcA::SpcR
Virulent serotype 2 strain, isolate from human outbreak in China
Virulent serotype 9 strain, porcine isolate
A clinical isolate belonging to serotype III
A clinical isolate belonging to M type 12
Daptomycin (commercial Cubicin®) analytic grade powder was purchased from Novartis Pharma. Penicillin G, ciprofloxacin, amoxicillin, and rifampicin were purchased from Sigma, and gentamicin from Roth. The antimicrobial solutions were prepared freshly prior to each application according to the manufacturers’ recommendations.
The MIC of each antibiotic was determined in duplicate by the microdilution technique in 96-well plates. Serial two-fold dilutions of different antibiotics prepared in RPMI 1640 medium were inoculated each with 5 × 105 colony forming units (CFU) of exponential grown cryo-conserved bacteria per well. MICs were also determined for freshly prepared bacterial cultures and did not significantly differ from MICs of cryo-conserved bacteria (data not shown). Plates were covered with a Breathe-Easy® sealing membrane to avoid evaporation and incubated for 24 hours at 37°C. The lowest antibiotic concentration that inhibited visible bacterial growth was defined the MIC. The determined MIC values are listed in Additional file 1: Table S1.
Chemically defined RPMI 1640 medium was inoculated with 1 × 107 CFU of either exponential or stationary grown cryo-conserved bacteria. Freshly prepared antimicrobial substances were added at a final concentration of 100-fold MIC, if not stated otherwise. Suspensions were incubated with end-over-end rotation at 37°C. Samples were taken after 1, 2, 4, 6, and 8 hours for determination of CFU by serial dilution and plating. For this 100 μl of bacterial suspensions were immediately harvested by centrifugation, once washed in sterile 0.85% NaCl solution and spotted as 10 μl aliquots on sheep blood Columbia agar plates in serial dilutions. Plating of the aliquots was performed in triplicates and all antibiotic killing experiments were performed at least with two biological replicates. Bacterial colonies were counted 24 and 48 hours after incubation at 37°C to ensure detection of slow growing bacteria. The results were analyzed with the GraphPad Prism 5 software and expressed in CFU/ml on a logarithmic scale. The limit of detection was defined as 100 CFU/ml and lower bacterial numbers were considered not detectable (n. d.). If indicated statistical significance was determined by one-sided Student t test.
An overnight culture was diluted to an OD600 of 0.02 in fresh THB medium and further incubated until the early exponential growth phase was reached. Then bacteria were harvested by centrifugation, once washed with PBS, and inoculated in fresh RPMI medium containing 100-fold MIC of the respective antibiotic to a final bacterial concentration of 1 × 107 CFU/ml. The suspensions were incubated at 37°C with moderate end-over-end rotation. Samples were taken hourly as indicated and the CFUs were determined after removal of remaining antibiotics by washings as described above. After 3 hours of antibiotic treatment (surviving) bacteria were collected by centrifugation, once washed in PBS, inoculated in fresh THB medium and grown overnight. This culture was then used to start a new cycle of antibiotic treatment with exponential grown bacteria. This procedure was repeated with three consecutive cycles and the experiment performed at least with two biological replicates. Colonies were counted and CFUs calculated as described above.
To dissect whether the antibiotic tolerant persister population of S. suis strain 10 comprises type I or type II persister cells, we performed a persister cell elimination test as described by Keren et al., with some modifications. Briefly, an overnight culture of S. suis strain 10 was adjusted to OD600 = 0.02 in fresh THB medium and further incubated until bacteria reached OD600 = 0.2. Then, aliquots of this culture were used to inoculate fresh THB medium to OD600 = 0.02 for a further cycle and to determine persister cell levels after a 100-fold MIC gentamicin challenge. A gentamicin challenge was done as described for the test of heritability of persistence with the exception that the antibiotic treatment lasted one hour. Dilution-growth cycles with subsequent antibiotic challenge were repeated thrice. For each cycle the initial inoculum before and the surviving bacteria after antibiotic challenge were determined by CFU counting. Data were expressed as percentage of surviving bacteria in relation to the initial inoculum before antibiotic treatment.
This work was supported by the Deutsche Forschungsgemeinschaft (DFG, Germany) as part of the Priority Programme SPP1316 (grants GO983-3/1 and BE4038/2-2). We gratefully acknowledge the following researchers for providing bacterial strains or antibiotics: Hilde Smith (Central Veterinary Institute, Wageningen University, Lelystad; S. suis strain 10), Susanne Talay (Helmholtz Centre for Infection Research, Braunschweig; S. pyogenes strain A40), Christoph Baums (University of Veterinary Medicine, Institute of Microbiology, Hannover; S. suis strain A3286/94), Jiaqi Tang (Research Institute for Medicine of Nanjing Command, Nanjing; S. suis strain 05ZYH33), and Mathias Hornef (Hannover Medical School, Hannover; Daptomycin/Cubicin®).
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.