PROTOCOL Building a morbidostat: an automated continuous- culture device for studying bacterial drug resistance under dynamically sustained drug inhibition Erdal Toprak1,2, Adrian Veres3, Sadik Yildiz2, Juan M Pedraza4, Remy Chait1, Johan Paulsson1,5 & Roy Kishony1,5 1Department of Systems Biology, Harvard Medical School, Boston, Massachusetts, USA. 2Faculty of Engineering and Natural Sciences, Sabanci University, Istanbul, Turkey. 3Health Sciences and Technology Program, Harvard Medical School, Boston, Massachusetts, USA. 4Department of Physics, Universidad de los Andes, Bogotá, Colombia. 5School of Engineering and Applied Sciences, Harvard University, Cambridge, Massachusetts, USA. Correspondence should be addressed to E.T. ([email protected]) or R.K. ([email protected]). Published online 21 February 2013; doi:10.1038/nprot.2013.021 We present a protocol for building and operating an automated fluidic system for continuous culture that we call the ‘morbidostat’. The morbidostat is used to follow the evolution of microbial drug resistance in real time. Instead of exposing bacteria to predetermined drug environments, the morbidostat constantly measures the growth rates of evolving microbial populations and dynamically adjusts drug concentrations inside culture vials in order to maintain a constant drug-induced inhibition. The growth rate measurements are done using an optical detection system that is based on measuring the intensity of back-scattered light from bacterial cells suspended in the liquid culture. The morbidostat can additionally be used as a chemostat or a turbidostat. The whole system can be built from readily available components within 2–3 weeks by biologists with some electronics experience or engineers familiar with basic microbiology. INTRODUCTION Antibiotic resistance is an important public health problem, ren- at a constant rate lower than the maximal growth rate of the popu- dering currently available drugs useless and threatening millions lation, the dilution rate of the morbidostat rdilution ≅ ∆V/(V·∆t) is of lives1–4. Evolution of resistance by spontaneous mutations can fixed. In steady state, the bacterial growth rate must reach a value be studied in the laboratory by exposing a bacterial culture to that matches this dilution rate. However, in contrast to a chemo- an environment in which growth is inhibited by antibiotics and stat, in which the bacterial growth is inherently limited by nutrient by characterizing the resistant mutants that emerge5,6. However, availability, in the morbidostat the cell density is kept low such that as bacteria are typically subjected to a fixed drug concentration, the population is not nutrient limited; instead, its growth rate is those studies are generally limited to only one or at most a few controlled by externally adjusting drug concentration. At the end Nature America, Inc. All rights reserved. Inc. Nature America, 3 mutational steps conferring resistance. Once such single mutations of each period ∆t, the growth rate (r, Fig. 1b, black lines) is calcu- emerge and sweep through the population, the inhibition by the lated on the basis of optical density (OD) measurements (Fig. 1b, gray dots). Next, depending on the calculated growth rate and the © 201 drug is relieved and there is no additional pressure to evolve higher levels of resistance. To be able to follow the evolution of resistance current OD of the culture, the morbidostat decides whether to add through sequential accumulation of multiple mutations, we must fresh medium or fresh medium plus drug (in either case, the same thus be able to keep increasing the drug concentration such that the fixed volume ∆V is added). Fresh medium with the drug is injected evolving bacterial population is constantly challenged. However, only if the OD exceeds a threshold (ODTHR) and if the growth rate carrying out experiments using environments in which drug con- is higher than the dilution rate (Fig. 1b, magenta filled circles); centrations are determined beforehand is almost impossible, as the in all other cases, fresh medium is injected. Drug concentrations phenotypic effects of emerging mutations on complex evolving inside the culture vials increase with drug injections and are gradu- populations cannot be predicted. Experimental systems that can ally reduced by dilution with successive fresh medium injections automatically adjust drug concentrations to maintain a fixed level (Fig. 1c, magenta line). The value of ODTHR is chosen to be small of growth inhibition on evolving bacterial populations are therefore enough such that the population is never nutrient limited, typically 7 useful for studying the evolution of drug resistance . Such experi- ODTHR = 0.15. Choosing the values of ∆t, V and ∆V such that the mental systems become particularly important for studies that aim dilution rate rdilution is substantially lower than the maximal growth to quantify the evolutionary dynamics of bacterial populations in rate of the bacteria forces the system to adjust the drug concentra- different drugs or drug combinations8,9. tion to reduce the growth rate accordingly. Typically, for bacteria − 1 We recently introduced an automated continuous culture device, growing at a maximal rate of r0 = 0.8 h , the morbidostat is set to the morbidostat, for studying evolution of drug resistance in a con- V = 12 ml, ∆V = 1 ml and ∆t = 12 min, such that the dilution rate trolled environment containing an antibiotic at a concentration of the system is half of the growth rate of bacteria in the absence of − 1 that is dynamically adjusted such that the bacterial population is drugs (rdilution ≅ ∆V/(V·∆t) = 0.4 h ; Fig. 1b, green filled circles). constantly challenged (Fig. 1)5. The bacterial population is growing Throughout the entire experiment, the volumes of the cultures are in a fixed volume (V) with continuous stirring, and at fixed time kept constant using a suction pump (Fig. 1a). intervals (∆t) the culture is diluted by injection of a fixed amount As the bacterial population growth is limited by drug inhibi- (∆V) of fresh medium or fresh medium containing dissolved drugs. tion and not by nutrient depletion, the evolutionary changes Similar to a chemostat, in which cell cultures are periodically diluted occurring in the populations are likely to be associated with drug NATURE PROTOCOLS | VOL.8 NO.3 | 2013 | 555 PROTOCOL Figure 1 | The morbidostat is an automated a b 0.17 continuous-culture device that maintains a Media pump ODTHR constant level of growth inhibition on evolving Drug pump 0.15 bacterial populations. (a) The working algorithm Control algorithm Waste OD of the morbidostat. (b) Representative bacterial pump ∆OD Add �V r �V/V growth in the morbidostat. OD values are shown No Grow media V OD > ODTHR t with gray dots. Growth rates (r) of bacterial Detector & � LED for �t ∆OD > 0 Add �V populations are periodically calculated by fitting Yes 0.10 drug 52 53 54 exponential growth functions (black lines). Markers Time (h) with magenta and green colors represent dilutions with drug solution and fresh medium, respectively. c d 4 e ) (TMP) r (c) Final OD values at the end of each growth cycle. –1 0 3 h (d) Growth rates (blue) and drug concentrations 10–1 ( µ (magenta) between two consecutive drug solution 2 g ml OD injections. (e) The drug concentration that –1 50% growth 1 ) inhibits growth by 50% (IC50) is calculated by inhibition Growth rate (1 analyzing growth rates and corresponding drug IC50 10–2 0 0 concentrations. TMP, trimethoprim. 47 51 55 59 63 67 47 51 55 59 63 67 10–2 100 Time (h) Time (h) (TMP) (µg ml–1) resistance. Once the population evolves and becomes more resistant, it starts to grow faster in the presence of the drug, and in response the drug conventional selection techniques such as the disc diffusion assay concentration is adjusted such that the population is again inhib- or selection on agar plates or liquid with fixed drug concentrations, ited using a higher drug dose and the growth rate converges to which selects just once for bacteria capable of surviving particular the fixed dilution rate of the system. The continuous drug inhibi- drug levels, the morbidostat does not relax selection as the bacteria tion challenges the bacterial population to keep evolving by accu- become increasingly resistant7,9,10. Instead, it follows the changes mulating multiple resistance mutations. We typically continue in drug sensitivity and adjusts the drug concentration accordingly morbidostat experiments until a diminishing rate of increase to maintain bacterial growth inhibition (Fig. 1d). This feature in resistance is observed. For all three drugs used in our recent makes it possible to follow the evolution of multistep increases in studies, all of the cultures attained high, steady resistance levels drug resistance in real time. In contrast, microbial selection experi- within 3 weeks5. ments that are carried out using antibiotic-containing agar plates Construction of the morbidostat involves assembly of the mor- or growth medium have the main advantage of relative simplicity bidostat culture vials, assembly of the optical detection system and compared with the morbidostat. However, bacterial mutants that assembly of a computer-controlled array of peristaltic pumps used are selected in constant drug environments tend to develop only for liquid transfer. Our design currently allows growing 15 inde- one mutation that will release the selection pressure, unlike clinical pendently controlled cultures in parallel
Details
-
File Typepdf
-
Upload Time-
-
Content LanguagesEnglish
-
Upload UserAnonymous/Not logged-in
-
File Pages13 Page
-
File Size-