Physiologically Informed Imagining and Fluidic System for Vertical Swimming of Marine Organisms

Introduction: We demonstrated an integrated system that combines DIY microscopy and vertically oriented microfluidics to investigate the hydrodynamics of marine larvae, enabling vertical swimming behaviors (nicknamed “Aquavert System”). Using the system, the vertical swimming behaviors with and without external flow were examined across three marine larvae, including  Dendraster Excentricus (sand dollar), Strongylocentrotus purpuratus (sea urchin), and Pycnopodia helianthoides (starfish)  larvae with and without external flow were examined. The life cycle of most marine invertebrates includes a planktonic stage. Swimming behaviors are critical for dispersal and survival of individual larvae and adult populations. Marine larvae navigate through the ocean by propelling themselves within water column layers. Ciliary bands across the organisms' bodies accomplish much of these swimming activities. Therefore, marine larvae could be used as a model organism to study ciliary biomechanics. Yet, investigating the swimming behaviors in the field has remained difficult. Therefore, tractable lab-based approaches that mimic physiological conditions complimented with microscopy (to visualize ciliary behaviors) are needed to examine the swimming behaviors under well-controlled environmental conditions. Microfluidics could provide a suitable approach for mimicking the environmental factors. However, standard microfluidics systems are typically placed horizontally (chip parallel to the XY-plane on a conventional microscope), thus limiting critical vertical swimming behaviors. The Aquavert system addressed these gaps by combining vertical microfluidics and customizable imaging strategies. The system could be used to investigate the functional consequences of phenotypic plasticity (an organism's ability to change in response to environmental stimuli) and the impact of climate change on marine organisms.

Materials and

Methods: The Aquavert system included two main components: imaging and microfluidics (Fig 1). The microscope was built by integrating off-the-shelf components (objective, 50 mm tube lens, connectors, precision stages, camera, and optical breadboard) and 3D printed parts. Single channels (for free swimming) and multi-inlet channels (for stratified flow) were developed. Figure 2 demonstrates microfluidic for the stratified boundary layer experiments. Microfluidics system was made by silicone casting from laser-cut acrylic molds. PDMS parts were plasma bonded to glass slides (25 or 50 mm wide). Tube connectors (for 1/16" ID) were used at the inlet and outlets to prevent potential leakage. Throughout the experiment, silicone tubing was used. 3D printed holders were built to secure a microfluidics system vertically in either "portrait" or "landscape" configuration. The chip holder was secured on an XYZ stage for precision focusing of the sample. Syringe pumps (up to three) were used to induce stratification. Volumetric flow rates were between 0.5–2 mL/hr. To verify stratified flow experiments, food color dye was used to visualize the flow boundaries. 
Reynolds numbers were calculated to confirm laminar flow (Re# < 1) for all stratified flow experiments. Laminar flow is needed for stable flow boundaries. Larvae were grown according to the standard protocols with 4 or 20 L culture vessels at 16 C ± 1 with motor-driven paddles set to 6 rpm. Larvae were fed Rhodomonas sp. (unicellular red algae). Water change occurred before each feeding. Larvae were then released to the Microfluidics system, and their swimming behaviors were analyzed with Kinovea.

Results, Conclusions, and Discussions: Key environmental factors (food concentrations or flows) surrounding the larvae are vertically structured in nature. Therefore, navigating these environments via vertical swimming is critical for the organisms' survival. The Aquavert system enabled vertical swimming under well-controlled environmental conditions. First, using a single channel (cross-section 1.6 mm × 4 mm2, length 65 mm, positioned in the portrait profile) with no-flow baseline vertical swimming of marine organisms was examined (Fig 2). Organisms could traverse the length of the channel. Swimming parameters were quantified, including vertical displacements, velocities, and relative position (measured via angle relative to the chamber base). Second, the vertical swimming of sand dollar larvae in stratified flow was investigated using a two-inlet device oriented in landscape mode. Volumetric flow rates in the top and bottom layers were 0.5 mL/hr and 1.0 mL/hr, respectively. In this experiment, three larvae were introduced from the lower inlet. Two larvae navigated across the boundary (from the bottom layer to the top). Vertical displacement, velocity, and position inside the channel were quantified for all larvae. While the flow in this experiment was directed parallel to the Y-direction (horizontal stratification), the microfluidic channel could also be oriented in portrait mode to enable flow parallel to the Z-direction. In the third experiment, algae were introduced inside the channel to confirm the possibility of investigating swimming/feeding behaviors. Similar to the second experiment, a two-inlet device created a layered flow. Instead of food dye, algae were used to visualize the layers. Sea urchin larvae were introduced from the top inlet (layer without algae). Larvae successfully crossed the lower boundary layer, and feeding behaviors were observed. In this experiment, various objectives (2X, 4X, and 10X) were applied. Additionally, as part of this experiment, the current optical resolution of the system was examined using a standard USAF resolution target for all objectives. Beyond marine larvae, the Aquavert system could be used to investigate the behavior of aquatic organisms (Hydra) in vertically- oriented environments that have been absent from standard microfluidics.

Acknowledgements (Optional): Authors express gratitude to members of Ahrar-lab and Pace-lab.
Authors are thankful to CSULB Marine Lab.

This work was supported in part by a CSUPERB new investigator grant to S. Ahrar.