Below we provide the list of parts from Thorlabs, Cairns Research and other suppliers that we used to build our fluorescent mini microscopes. These microscopes will be used to perform calcium imaging in undergraduate practicals and small lab projects for level 3 students. Slightly improved, they can be used for research projects. For this we recommend using more advanced camera (e.g. Retiga R3). Due to their size, they can be used in a CO2 incubators for chronic imaging of cultured cells. See our recent paper how chronic calcium imaging be performed using conventional microscopes: https://www.future-science.com/doi/pdf/10.2144/btn-2018-0024

The microscopes were designed and implemented by Ash Cadby: https://www.sheffield.ac.uk/physics/people/academic/ashley-cadby @imagine2022 @Cadby_lab and Anton Nikolaev: https://www.sheffield.ac.uk/hearing/sensory-integration @CalcNeuro

Cheap LEDs were designed by Federico Ceriani (Marcotti’s lab).

Overall price of the microscope is £1600-1700 but can be reduced to ~1200 if you design the LEDs yourself and use filter cubes from old microscopes (we found plenty) and old objectives.

Objective part

Option 1.

CCM1-G01 30mm Cage mounted aluminium Mirror

S1TM06 SM1 to 6 mm Lens Cell adapter

C392TME-A lens (f = 2.75 mm. NA = 0.64

ER2-P4 extension rods to connect the objective to the filter cube

Option 2.

CCM1-G01 30mm Cage mounted aluminium Mirror

Objective canibalised from your old microscope (NON INFINITY CORRECTED!)

Appropriate SM1 adapter (e.g. RMS to SM1) – will depend on the objective lens

ER2-P4 extension rods to connect the objective to the filter cube

Option 1 will require a bit of fiddling with the position of the camera lens. Option 1 will also require appropriate focusing system. With option 2 we simply screw the objective lens up and down (NB! This affects the size of the sample a little).

Filter cube part

Option 1 (expensive)

Thorlabs C6W 4-way mounting cube

Thorlabs B3C/M – Rotatable cover plate

Thorlabs MF469-35 GFP excitation filter

Thorlabs MF525-39 GFP emission filter

Thorlabs MD498 Di filter R band = 452-490

Thorlabs FFM1 cage dichroic filter mount

Two Thorlabs SM1L05 short tubes to connect the excitation and emission filters to the C6W cube

Option2 (cheap)

Thorlabs C6W 4-way mounting cube

Thorlabs B3C/M – Rotatable cover plate

Filter cube cannibalised from an old microscope and glued to B3C/M using glue gun

The rest of the microscope

Thorlabs SM1L05 lens tube

Thorlabs AC254-100-A camera lens

Thorlabs SML35 3.50’’ lens tube

Thorlabs SM1A9 adapter with external C-mount thread and external SM1 thread

Thorlabs SM1T2 coupler

Camera

DCV/BFS-U3-16S2M-CS Blackfly camera  NB! This camera heats a lot and requires a heatsink or to be switched of between frame acquisitions (ideally both). Otherwise its really good — lightwave, cheap (350-370 pounds) and pretty good quality.

LED

Option 1 (Thorlabs)

Thorlabs mount LED (e.g. M490L4)

Thorlabs T-cube LED driver (LEDD1B)

Thorlabs SM1L35 tube

Thorlabs SM1T2 coupler

Thorlabs ACL2520U lens (F=20mm)

You will need to spend some time to properly position the lens

Option 2 (Cairns)

MonoLED light source

Both these options do great job but are quite expensive (around £500). Here is option 3 for electronics lovers:

Option 3

Arduino uno/mega http://uk.rs-online.com/web/p/processor-microcontroller-development-kits/7154081/   (2 for monitor+controller configuration)

AD5171 digital potentiometer http://uk.rs-online.com/web/p/digital-potentiometers/8099301/

Surface Mount (SMT) Board SOT Epoxy Glass Double-Sided 23.5 x 13.5 x 1.5mm FR4 https://uk.rs-online.com/web/p/surface-mount-smt-to-through-hole-adapter-boards/7288838/?searchTerm=re906&relevancy-data=636F3D3126696E3D4931384E53656172636847656E65726963266C753D656E266D6D3D6D61746368616C6C7061727469616C26706D3D5E2E2A2426706F3D31333326736E3D592673743D43415443485F414C4C5F44454641554C542673633D592677633D4E4F4E45267573743D7265393036267374613D726539303626&dym=re90

RCD24 LED driver http://uk.rs-online.com/web/p/led-drivers/6689882/

2X 4.7 KOhm resistors

Power supply

LED (1-3 W) blue LED. E.g – https://futureeden.co.uk/products/3w-royal-blue-led-epiled-440-450nm-with-star-pcb-heatsink

If you don’t need to regulate the brightness of the LED you do not need the potentiometer. We are still working on the best way to mount these LEDs but the idea is to glue it to a Thorlab cap (SM1CP2M) and screw this onto CP02. You will also need a lens to focus the LED.

Past couple of months we have been busy building equipment for our project. We need the following pieces of equipment:

• Fluorescent microscope allowing to image activity in newly defined neurons and, hopefully other things. This was done using Thorlabs optical parts and mechanics. It’s very simple, lightweight and can be (in theory) used directly in a CO2 incubator if we figure out how to save the camera from wet environment. Here is an example of calcium imaging of HeLa cells. As you can see the spatial resolution can be improved but this is not critical for our goals as we don’t need to define calcium dynamics on a sub-cellular level.

• Automatic syringe pumps (at least 4 or 8). Commercially available pumps are ridiculously expensive and we have decided to build our own. Mechanical parts were printed and modified from available internet analogue, https://www.instructables.com/id/3D-Printed-Syringe-Pump-Rack/ although ours is more elegant and stylish 🙂 We used Nema 17 step motors, which may be a bit of an overkill as the pumps quite substantially vibrate when moving the plunger. The motors are controlled using Arduino and CNC expansion shield. Here is an example of a mock experiment:

• Our next step is building a programmable solution application robot to be used with the above two. It will automatically apply solution in a 96 well plate and will allow subsequent imaging of calcium dynamics. Sounds like lots of fun.

Who are we and what do we want?

We are a bunch of scientists from Sheffield who are interested in understanding how neuronal circuits work. One of the ways to answer this fundamental question is to build neural circuits from scratch. We will therefore i) turn stem (NTERA-2) into neurons, make these neurons ii) grow and iii) form synapses with other neurons in a controlled way.

Has this been done before?

Yes, to some extent. There are many wonderful papers published on this topic and some of them will be discussed in this blog.

How do we differ?

Our approach is to marry classical biological techniques with machine learning techniques. By doing so, we will optimise the conditions for NTERA2 cells differentiation, regulation of axon growth and triggering long-term potentiation and depression. All this will be discussed in this blog.

What is this blog about?

Every individual experiment, it’s analysis and main results will be immediately published here. We will also discuss scientific ideas, research papers and share acquisition and analysis software (mostly written on Python).