Characterizing Microbe-Environment Interactions Through Experimental Evolution: The Autonomous Adaptive Directed Evolution Chamber

Abstract:

We are developing a laboratory system for studying micro- to meso-scale interactions between microorganisms and their physicochemical environments. The Autonomous Adaptive Directed Evolution Chamber (AADEC) cultures microorganisms in controlled,small-scale geochemical environments. It observes corresponding microbial interactions to these environments and has the ability to adjust thermal, chemical, and other parameters in real time in response to these interactions. In addition to the sensed data, the system allows the generation of time-resolved ecological, genomic, etc. samples on the order of microbial generations.

The AADEC currently houses cultures in liquid media and controls UVC radiation, heat exposure, and nutrient supply. In a proof-of-concept experimental evolution application, it can increase UVC radiation resistance of Escherichia coli cultures by iteratively exposing them to UVC and allowing the surviving cells to regrow. A baseline characterization generated a million fold resistance increase. This demonstration uses a single-well growth chamber prototype, but it was limited by scalability. We have expanded upon this system by implementing a microwell plate compatible fluidics system and sensor housing. This microwell plate system increases the diversity of microbial interactions seen in response to the geochemical environments generated by the system, allowing greater control over individual cultures' environments and detection of rarer events.

The custom microfluidic card matches the footprint of a standard microwell plate. This card enables controllable fluid flow between wells and introduces multiple separate exposure and sensor chambers, increasing the variety of sensors compatible with the system. This gives the device control over scale and the interconnectedness of environments within the system. The increased controllability of the multiwell system provides a platform for implementing machine learning algorithms that will autonomously adjust geochemical environmental parameters.