Cheap DIY GFP (Green Fluorescent Protein) Illuminator

Cheap DIY GFP (Green Fluorescent Protein) Illuminator

It wasn’t some creepy-crawly fetish that got me reading The Worm Breeder’s Gazette. Rather, it was talking to Kathryn Hedges — a smart, passionate, and well-credentialed scientist and artist — about The Gazette’s tips on making a GFP illuminator on the cheap. GFP, or green fluorescent protein, is a fluorescent marker frequently used in molecular biology and neuroscience.

During her master’s program at Humbolt State University, Kathryn, like many other scientific researchers, was challenged with budget cuts and the difficulty of finding grant support. In order to continue her work on nematode neurons in a cash-strapped department when she had an 11th-hour need for new equipment, Kathryn unleashed the power of her “let’s see how can we do this better and cheaper” mentality to hack the DIY GFP illuminator into a better, research-worthy system, capable of visualizing neurons in tiny worms only 1 millimeter long.

Why was this important to Kathryn? In her words, “one of my strains kept selecting against GFP if I didn’t hand-select the brightest worms each generation. Without a way to see GFP on the “dissecting” microscope where I was picking up individual worms with a tiny wire spatula, I couldn’t guarantee that all their progeny would be fluorescent, which is a problem when I’m trying to see if a treatment damages them to make them lose fluorescence.”

The result: “The only microscopes we had with GFP capability were unsuitable for magnifying worms while moving them, and buying a suitable ‘scope would have cost thousands of dollars. I had $750 left in my budget, so I spent about $80 on LED parts and built an illuminator. It wasn’t bright enough to see long thin cells, so I spent $825 on top-quality filters, salvaged an old microscope eyepiece as a focusing lens, and used plumbing fittings to mount it on a micromanipulator we had in the lab. Then I could select my shiniest worms, raise reliably shiny worm larvae, and finish my experiments. If I had been tracking a larger feature in GFP, or I’d been using Red Fluorescent Protein, I probably could’ve used cheap filters and done the whole thing under $100. If I were to continue my experiments, I could use this illuminator when crossing GFP strains to see which progeny have both GFP markers.”

Kathryn’s thesis advisor encouraged this type of experimentation, as he and his own thesis advisor had developed their own DIY equipment to detect nerve conduction velocity in a small freshwater annelid. The prototype still hangs on a bulletin board over his desk. Not only does he teach animal physiology students how to do neurophysiology noninvasively, he and his research students have published papers using this method to collect data. He is a fan of the Backyard Brains Spiker Box and hopes to see it in action.

By hacking her own tool, without undergoing the too-high expense of buying an average illuminator for thousands of dollars (even used), Kathryn was able to finish her thesis. The illuminator is useful for any DIY scientists who need to view the expression of GFP or RFP (Red Fluorescent Protein) genes in bacteria or c. elegans. And Kathryn is proof you can do it on a dime.

Details on the basic GFP Illuminator from The Worm Breeder’s Gazette are below (after the jump). To learn more about Kathryn’s improvements, and to see her notes on worms, check out her post on Splendid Colors.

[Credit to Ian Chin-Sang and Weiwei Zhong]

Using LEDs as a low-cost source to detect GFP and DsRED

Ian Chin-Sang1 and Weiwei Zhong
Department of Biology, Queen’s University, Kingston, ON, Canada, 2Department of Biochemistry and Cell Biology, Rice University, Houston TX
Correspondence to: Weiwei Zhong (

Fluorescence detection components are too costly to be installed in every stereoscope. Often, our fluorescent markers are very bright and do not require the full capacity of such equipment. Here we present an LED setup that costs only about $100 and can detect bright fluorescent markers such as myo-2::GFP. The setup (Fig. 1) consists of:

  • X-acto helping hands or anything that can position the LED (
  • Lamp cord that has a switch (local hardware store)
  • 17-watt Xitanium LED driver (LED Supply)
  • Xitanium driver connector (LED Supply)
  • Heatsink (LED Supply)
  • Spot lens (LED Supply)
  • Luxeon V Star optic holder (LED Supply)
  • Royal-Blue Luxeon V Star LED for GFP detection (LED Supply), or Green Luxeon V Star LED for dsRED detection (LED Supply)
  • Optional excitation filter: Roscolux #4290 CalColor 90 Blue for GFP, or #389 Chroma Green for dsRED ( | Edmund Optics)
  • Emission filter: Roscolux #12 Straw for GFP, or #19 Fire for dsRED (availability same as excitation filters)

Details on how to actually put this device together can be found here and here.

Fig 1. The LED setup

Fig 2. Using the LED setup to detect myo-2::GFP in an mIn1 animal.

Fig 3. Using the LED setup to detect myo-2::dsRED.

When assembling the LED to the driver, make sure that the +/- ends on the LED and the neutral/line ends on the lamp cord match those on the driver. Use a non-conductive glue such as silicone adhesive (Devcon, part No.12045, local hardware store) to put the lens holder onto the LED. The excitation filter can be glued to the lens by applying a small amount of glue on the edge. The emission filter can be simply taped under the microscope objective.

This setup has a long operating life, requires no warm-up or cool-down time, and has no radiation. However, it can only detect strong signals. We have used the setup to detect the following markers: myo-2::GFP (Fig. 2), sur-5::GFP, ajm-1::GFP, and myo-2::dsRED (Fig. 3).

Using LEDs as a low-cost source to detect GFP and DsRED

Bio: Eri Gentry is a biotech entrepreneur, citizen science community organizer, and the co-founder of BioCurious, the first hackerspace for biotech, in the San Francisco Bay Area.

4 thoughts on “Cheap DIY GFP (Green Fluorescent Protein) Illuminator

  1. Cathal Garvey says:

    Love it, even if it worries me that it takes so much to detect fluorescence.. I suppose worms can’t make enough GFP to be clearly visible without so much input?

    Makes me want to find some lab strains of worms and grow them for love. I’d probably end up neglecting them though. Poor worms.

    1. Kathryn Hedges says:

      Hi Cathal,

      Thanks for being interested!

      My main problems were (a) low signal-to-noise ratio and (b) looking for features barely visible at those magnifications.

      The main technical problem in viewing GFP is that the blue light used for excitation is close on the spectrum to the green emission, and since you get some spread around the maxima, there’s overlap–that cyan background you see in Ian’s photos. (I haven’t used dsRED but the spectra are so far apart [green excitation, red emission] I’m sure the cheap filters would be fine; I’ve tried it with similar dyes.) Plus, the worms are transparent but reflective, so you see glare on their surface.

      Viewing neurons (or the neuron-like excretory cell) in a moving animal 1 mm long is challenging. I was using around 240-500X magnification on a Wild M5 dissecting microscope, a classic with high-quality optics and great depth of field.

      But GFP-tagged worms are undeniably gorgeous. I wish I had taken time to film some video of them through the scope, particularly the strain with tagged GABA neurons (CZ1200). They looked like strings of Christmas lights, or living necklaces, gliding around in sine waves. The dopamine neurons look like sets of headlights and taillights, and the excretory cell is a racing stripe. Watching them through a stereomicroscope as they undulate through the agar is amazing. The GFP is easier to see when they’re in the agar than on top of it, because you don’t see reflections from the body wall.

      Regarding neglect, C. elegans is an amazingly tough animal. In the wild, their typical ecological cycle is boom-bust-spread-wait. Eat and reproduce, devour any food (bacteria) in their environment, go into starvation-resistant forms at three separate points of the lifecycle, and disperse while waiting for more food to grow. You can find live dauer larvae on old dried-up plates that will mature almost normally after you put them on fresh food. You can chunk new plates every 3-4 weeks and maintain a culture, though there will be epigenetic changes from starvation that may affect some experiments.

      I hope I answered your questions without going overboard!

  2. dontlistentothepundits says:

    Would this be effective in seeing DsRed in mosquitoes and other small insects ?

  3. Old microscope + £100 (+ 3D printer) = GFP fluorescent microscope | David Molnar says:

    […] comparison to another DIY fluorescence microscope design which can only detect strong fluorescence, the use of interference filters seems an improvement. […]

Comments are closed.

Discuss this article with the rest of the community on our Discord server!


Maker Faire Bay Area 2023 - Mare Island, CA

Escape to an island of imagination + innovation as Maker Faire Bay Area returns for its 15th iteration!

Buy Tickets today! SAVE 15% and lock-in your preferred date(s).