Figure out what you're doing! Ideally talk to someone that has done this before, get your basic questions out of the way. Don't worry if you don't understand everything right away. Understand why you're going about it this way.
I've already tested my tissue of interest (Aedes aegypti tarsi) in a bath/well-based system by Ibidi. This worked well, but I could only apply one stimulus per leg, and couldn't remove the stimulus from the bath. But I was able to get signal from a few different stimuli (at least once in a while - still not sure why some legs successfully respond and others don't). So I'm ready to invest some time/energy into a more sophisticated custom-perfusion set up for my mosquito legs.
The basic goals for my microfluidics chip:
Perfuse different solutions of stimuli over tissue in a controlled manner
Push leg against coverslip to improve image quality
Stabilize leg so that it doesn't move due to flow or muscle contraction1
Familiarizing myself with principles of making microfluidics chips
In brief, chips are made by putting 2D designs onto a "master mold" (this step has to be in a clean room), and then this is cast onto PDMS (can be done without a clean room). For the mold step I'm going to use MicroKosomos, a company affiliated with the Chronis lab at University of Michigan, and then I'll try to do the PDMS casting myself using instruments from a neighboring lab.
Read through articles that go about accomplishing something similar to what you're doing.
San-Miguel, Lu (2013) "Microfluidics as a tool for C. elegans research", WormBook → Overview and graphics of different types of microfluidics used in worm research
Chronis, Zimmer, Bargmann (2007), "Microfluidics for in vivo imaging of neuronal and behavioral activity in Caenorhabditis elegans" → Amazing example of laminar flow in chip for chemosensory imaging
Van Giesen... Sprecher (2016), "A microfluidics-based method for measuring neuronal activity in Drosophila chemosensory neurons" → Big inspiration for me trying this. Pretty sure this particular design wouldn't work for mosquito legs, but a great example nonetheless.
Mishra... Chronis, Collins (2014), "Using microfluidics chips for live imaging and study of injury responses in Drosophila larvae" → Clever use of a vacuum for mounting tissue that might not go through a biopsy punch or that you might not want to handle that roughly.
Lucchetta... Ismagilov (2005), "Characterization of the local temperature in space and time around a developing Drosophila embryo in a microfluidic device" → Basic example of laminar flow (to deliver stimuli of two different temperatures).
Other creative designs elements I probably won't use:
Ghannad-Rezaie... Collins, Chronis (2012), "Microfluidic chips for in vivo imaging of cellular responses to neural injury in Drosophila larvae" → Uses two layered PDMS chips for food and CO2 delivery to Drosophila larvae for longer term imaging (I don't think I will need to use two PDMS chips, but nevertheless was intrigued by this design).
Chaudhury... Collins, Chronis (2017), "On chip cryo-anesthesia of Drosophila larvae for high resolution in vivo imaging applications" → Creatively takes the last design a step further by using the top PDMS chip to deliver a thermal stimulus.
Albrecht, Bargmann (2011), "High-content behavioral analysis of Caenorhabditis elegans in precise spatiotemporal chemical environments" → Pegged chemotaxis behavior chamber with laminar flow for multiple C. elegans.
Keil, Kutscher, Shaham, Siggia (2017) "Long-Term High-Resolution Imaging of Developing C. elegans Larvae with Microfluidic" → 2-layer chip for long-term imaging set up that uses a vacuum to allow worms to move freely or keep them in place while imaging.
Big challenges for my chip design
Questions that I need to explore in my chip design brainstorm.
How to load my mosquito leg & seal chip
How will mosquito leg be stabilized but still exposed to flow
Whether or not to use laminar flow (single channel or multichannel)
How to get the mosquito leg out of the chip once imaging is complete
To understand the basic measurements and variability of the leg dimensions I used previously taken tile confocal images of the entire tarsi of adult mosquitoes and measured them using the graphic line tool in the Zen Blue software. (The Z dimension was harder - but generally my Z dimension was 30 - 35 microns).
(Made using HTML table generator)
It was suggested to me that I should have one mold with variations on my chip design that I can test, and then once I've settled on a final design, make a final mold with the same chip design iterated many times (so that I can have many chips). Pictured is the master mold design from the Van Giesen paper from the Sprecher lab. Their chip is something like 20mm x 7 mm and they're able to fit 28 design iterations on a mold. No idea what the final dimensions of my chip will be yet, but I'm doubtful that my chip will need to be so large, so I'm expecting I should be able to fit at least that many on my first mold.
Some of my super basic first pass on cartoons made in powerpoint - click to expand and view as slide show. Some of these are no doubt not great, but the major design features are there.
Loading Leg/Sealing Chip
The mosquito leg isn't very flexible because of it's exoskeleton, and also isn't naturally straight. I had been intrigued at some point to try building a leg-shaped chamber in the center of chip, loading the entire mosquito leg into the center of the chip, and then place a coverslip and seal with a vacuum.
For imaging just in the tarsi, there's a lot more that's possible, because the tarsi is flexible and pretty straight.
I still think the vacuum thing might be a good idea - although there might be a problem with this and bubbles in the chip (degassing the chip is a way of avoiding this and I wouldn't be able to do this with the tissue already loaded). Although perhaps this isn't a huge issue because this paper didn't have any degassing step?
→ Note to self: maybe still make a design for this? (Just to try even for future applications?)
But that leaves us with three main potential ways of sealing the chip.
Binding chip to coverslip → Would load via biopsy punch, but removing tissue afterwards seems difficult.
Binding to coverslip + agarose seal on periphery → Easy to take tissue in and out, would've
CURRENTLY FAVORED: Clamp system → Clamp for seal, load via biopsy punch, remove afterwards by unclamping.
→ Tests that I can do before mask
Get some legs
Biopsy punch (different sizes?) in PDMS without fluid channel → How easy is it to get leg in and out?
Microscope test on slides/slide mounted in wells?
Test KCl response in tarsi alone