How the Environmental Sample Processor Works
Sensors: Underwater Research of the Future (SURF Center)
The Second-Generation ESP
The current second-generation (2G) ESP uses small chambers called “pucks” to house the filter material. Water is pulled through the puck until one liter has been filtered or the filter is so loaded with biomass that no more water will pass, whichever comes first. Once filtering is complete, two things can happen, depending on the wishes of the researcher.
First, the sample can be “archived”. This means preservative is added to the filter, and the puck is put away for processing once the instrument is recovered. We use RNALater™, which locks all gene expression at the moment it’s applied, without need for freezing. We have had remarkable results using this preservative, obtaining high quality DNA and RNA from the sample even with deployments over eight months long!
Second, the sample can be processed. Here we add a lytic agent and heat to the puck, which breaks open cells, releasing proteins and nucleic acids from every microbe captured on the filter. This new liquid, called a homogenate, is a slurry of molecules that will be very useful in downstream analysis.
The homogenate can be used in a number of ways, depending on how the ESP is configured. Normally, we perform a sandwich-hybridization assay (SHA), introducing some homogenate to a puck containing a pre-printed set of nucleic acid probes. If any of the nucleic acids within the homogenate are complementary to the printed array they will bind and, after addition of a few more chemicals, the printed spot will glow. SHA is particularly useful when biomass is overly abundant; no purification is required and there is no inhibition when arrays are presented with concentrated homogenate. The glowing spots on the array image are then captured with an on-board camera, saved as a .TIFF file, and sent via radio or cell phone back to shore for analysis.
If the ESP has a module attached called the micro-fluidic block (MFB), we can perform quantitative PCR unattended, under the surface of the ocean. The MFB was designed for the manipulation of not milliliters but microliters of fluid, a requirement for qPCR analysis. A small qPCR module was developed in collaboration with Lawrence Livermore National Laboratory, and we have performed in situ qPCR from surface waters (depths of 10 meters) to hydrothermal vents at underwater volcanos (1,800 meters). qPCR is quite sensitive at finding rare targets, but is easily inhibited by large amounts of biomass. Thus SHA and qPCR represent two ends of a continuum depending on the targets of interest and the biomass one expects to encounter.
The Third-Generation ESP
The third generation (3G) ESP improves on the 2G in a number of ways and is currently under development (2017). The initial goal of the 3G ESP has always been to mount it on an autonomous underwater vehicle (AUV); giving this ecogenomic sensor mobility will be a transformative event in oceanography, as sampling events are no longer locked in one location, sampling water that happens to drift past.
To recast everything the 2G ESP does such that it will fit within a 12-inch-by-24-inch cylinder introduces some enormous engineering challenges, in particular how to shrink the entire instrument into something the size of two basketballs! Obviously we would not have room for bags of communal reagents, and large waste containers; we had to think of a way to shrink the liquid requirements to perform the biology.
While we stayed with the idea of filtering water within a puck, we developed the idea of a “cartridge” where all reagents were carried on each cartridge creating a self-contained, use-once entity arranged around a toroid ring. This resulted in a 60-cartridge “turbine” with each cartridge individually accessed and processed at a shared processing station. Cartridges are of two types: archival and lyse-n-go and follow the functionality of the pucks in the 2G ESP.
Archival cartridges filter water, apply preservative, and await recovery once the vehicle’s mission is complete.
Lyse-n-go cartridges are more complex, requiring heater circuitry and slightly more convoluted fluid pathways through the cartridge. The goal of these cartridges is to create homogenate and then pass that homogenate off the cartridge to a downstream analytical processing module. Currently we are working with collaborators to develop a Surface Plasmon Resonance (SPR) module, a digital droplet PCR (ddPCR) module, and a Total Internal Reflection Fluorescence (TIRF) module. The future is very exciting given the various modalities we will be able to use to explore the world oceans.
Bowers, H.A., Marin, R.III, Birch, J.A., Scholin, C.A., and Doucette, G.J. (2016). Recovery and identification of Pseudo-nitzschia frustlules from natural samples acquired using the Environmental Sample Processor (ESP). Journal of Phycology, 52:135–140. http://doi.org/10.1111/jpy.12369
Herfort, L., Seaton, C., Wilkin, M., Roman, B., Preston, C., Marin, R., Seitz, K., Smith, M., Haynes, V., Scholin, C., Baptista, A., Simon, H. (2016). Use of continuous, real-time observations and model simulations to achieve autonomous, adaptive sampling of microbial processes with a robotic sampler. Limnology and Oceanography: Methods, 14:50-67. http://doi.org/10.1002/lom3.10069