Take a look at my gene gun! (Video at end of post)

Published: May 23rd, 2017   Last Modified: June 20th, 2018

At some point during your adventures you will find yourself having to deliver a payload of DNA/RNA into your organism of interest, usually through chemical means or by piggybacking on another organism. But what if, for whatever reason, you can’t use either method? Why not deliver your DNA coated onto gold/tungsten particles DIRECTLY INTO THE NUCLEUS at high speed through a gene gun!

Now, commercial gene guns are sold and manufactured by BioRad. However, the price tag of 10-30,000$ depending on the model is prohibitive for those who want to dip their toes into particle bombardment. How about doing it DIY style, safely, for less than 1000$ all in?

To give you a bit of background, my work recently has focused on constructing a plasmid encoding a 10 kb RNA plant virus under a 35S promoter and NOS terminator. I’ll spare you the gory details, but with the complete infectious clone in hand I was left scratching my head as to how to actually deliver plasmid into my plants to initiate infection. Rub innoculation (literally what it sounds like) has low efficiency, and the plasmid itself was toxic to bacteria so I couldn’t use agrobacterium to deliver it. Particle bombardment was the only option, however I wasn’t about to ask my C. elegans inclined brethren If I could spray their nicely sterilized bombardment chamber with dirt.

Looking for DIY gene gun alternatives yielded a few results, most notably How to build and use a gene gun. Full credit to the authors, who provide an excellent write up of the parts you need to build one as well as a protocol for coating tungsten beads with plasmid DNA. I used their article as a guide and freestyled off it. I’ll highlight a few key differences.

The first major design difference is the choice of materials for the main body. The authors used hardware store fittings through most of their build. Obviously their gene gun works and hasn’t exploded in their faces, so those fittings are probably adequate and demonstrate just how little you need to do particle bombardment. However, having tubing pressurized at 700 PSI a foot or two away from me tickled my spidey sense a little bit. To satisfy my paranoia of something going PINNNNNNNNGGGG and taking my eye out, it was worth it for me to spend a few hundred dollars more and buying proper Swagelok stainless steel fittings, capable of easily handling 5000 PSI all day, every day. Overkill? Yes. Shiny and pretty? Also yes! Do I cut corners later in the build? You bet!!! Let’s go a bit deeper and examine the gun bit by bit.

Like the original design, my gun uses 12 gram CO2 cartridges to provide the pressurized gas, the same ones used to power BB guns and carbonate your fizzy beverages. How to actually adapt these cartridges to the rest of your gun? You could use a bike tire inflator(~15-20$), however having tried this I can say that most bike tire inflators are pretty flimsy and don’t inspire much confidence, especially when you have to adapt them to standard NPT threads. Instead, for roughly 20$ more you can buy a CO2 cartridge adapter commonly used in paintball, which is made out of some solid aluminum and has a convenient 1/8in NPT thread drilled into it, no mods necessary. The adapter is made of two parts that screw together:

1 x https://www.paintballgear.ca/acm-quick-change-12-gram-co2-adapter
1 x https://www.paintballgear.ca/ninja-paintball-universal-fill-adapter

As you can see, I used a 1/8in NPT to 1/4in swagelok nipple union to adapt the CO2 cartridge to the rest of the system. Here are the exact parts of the adapter:

 

Once punctured, the cartridge releases the CO2, which is dosed out into the main firing chamber by a manual valve. These valves go for between 50-100$, a bit pricey but worth it.

The main valve releases CO2 into the main firing chamber (1/4in swagelok 4 way cross fitting) where it splits off in several directions.

On the right we have a 4000 PSI gauge ($50) adapted from a 1/4 inch NPT to 1/4 inch swagelok. Note the blue goo on the NPT threads to block the leakage path. Compression fittings like swagelok do not need sealant on the threads, and can even CAUSE leaks!

On the left is another valve which I use to release pressure after I’m done shooting, just so I don’t have to leave a pressurized vessel around If I don’t have to.

The bottom of the cross fitting connects to a Parker 1000 PSI 12V solenoid (Part #: 71216SN1FU00N0D100C1). This is essentially a switch that can be triggered by a 12v pulse, which releases the CO2 in bursts. The solenoid is controlled by a transistor operated by an arduino, nothing fancy. I’ll clean up the code and post it, but essentially the arduino waits for a serial command and then sends a signal to the gate of a transistor, which pulses the solenoid from a 12v power supply.

After the solenoid is activated, CO2 rushes into the barrel and makes contact with the macrocarrier, which is just a fancy word for the part which holds your gene bullet. The solenoid output is a 1/8 inch NPT thread, which is adapted to 1/4 inch swagelok and finally to a 1/2 inch swagelok male nipple. The gene bullet (a piece of parafilm between two washers) is held up against the nipple by a female 1/2 inch swagelok fitting that has been fused to 1/2 inch tubing with some epoxy putty. Welding would have been preferable here but I’m not set up for that kind of work just yet. Yep, looks pretty hokey and out of place, but I figure the gas only puts significant pressure on this part for a fraction of a second.

The gene bullet is held within the yellow blob of epoxy, which holds the female swagelok fitting, here’s a view inside:

***UPDATE*** Instead of the ugly epoxy I used silver brazing to secure the pieces together, producing a much cleaner and stronger joint. Considering silver brazing was used for the Helios gene gun, I feel safe using this updated barrel. As well, I decided to copy Bio-rad’s barrel tip design, it should be superior to my solid piece. Was done on a 3D printer.

***UPDATE – May 21/2018*** I’ve attached two sets of files. The first design is meant to be press fit onto 1/4 inch stainless tubing. Holds well, won’t fly off. For the second design you first need to put about 13 mm of M7 thread onto 1/4 inch SS tubing, and the barrel should thread onto that. Holds much more securely. Also added a bit more instruction on how to assemble this part of the gun. The hardest part to make here is the 1/4 inch swagelok male to 1/2 inch swagelok nut. Essentially what I did was silver braze a 3/8 in to 1/4in reducer to a 1/2 inch nut, fits together quite nice.

V1 – Press fit:

V2 – Threaded:

Having a bit of fun playing with barrel designs, trying to force the expanding gas out and away from the target, here are the files. Also added grips to help with threading the barrel onto the tubing.

The gene bullet goes into that fitting, bead side up, and is screwed onto the barrel.

To make a bullet you sandwich some parafilm between two washers that fit into the female fitting. Green marker signifies deposited beads coated in plasmid. The procedure of coating the particles is covered thoroughly in the article I mentioned in the beginning. One caveat here is that the particles must be deposited directly in the middle of the parafilm. If any ethanol touches the washers capillery action will wick away the ethanol and beads, ruining your bullet.

Here’s what the bullet looks after firing. As you can see the parafilm is nicely mushroomed out.

So, pretty cool stuff, but was all that time building the gun worth it? Does it work? Before I started bombarding plants with infectious plasmid, I shot a plasmid encoding for GFP at union pieces as suggested in the original article. My particles managed to penetrate several mm into the onion slice, making them quite difficult to find, but it was worth it!

Here are two nice impacts I got pictures of, very clear GFP expression, I was thrilled! Now for the real experiment, could I induce virus infection by shooting my newly constructed viral plasmid?

Above is a preliminary (read:ugly) western blot of plant leaves shot with the plasmid using an antibody against my virus. Looks like particle bombardment produces a similar result as plants inoculated with intact viral particles, yay!

***UPDATE*** Atleast for me, the GUS assay seems to be an easier way to visualize whether a gene gun works. Here are some samples of onion epidermal peel that I bombarded using my DIY gene gun with 0.5 mg of tungsten beads + 0.5 ug plasmid per shot at 250 psi at a distance of 2cm away from the tip of my barrel. FYI, to do an onion peel assay you cut up an onion into manageable, 1 inch square pieces WITHOUT removing that thin, wispy layer of skin on the inside. Bombard the skin and incubate in a petri plate with a bit of water to keep things moist. After incubation, THEN you remove the wispy epidermal layer and stain with X-GlucA or look for GFP expression etc. Onion peels are nice for this application as the peel is essentially a one layer thick sheet of cells that have a very distinct shape.

Without getting into the nitty gritty, the presence of blue coloration indicates that GUS enzyme has been produced, and it has converted X-GlucA into a blue substance. This is catalyzed by oxygen, so be careful incubating for too long otherwise things will start turning blue on their own.

*** UPDATE 19/06/18 ***

I’ve received a fair amount of attention of people building their own gun, which has spurred me on to improve the current design which is lacking primarily in the area of usability.

It will be a few more days (a week, tops) before I’m satisfied/have more proof of the bombardment efficiency, but to tide everyone over here’s a little “teaser” of the gun, and some more visual proof of the bombardment efficiency.

This is currently the best shot of pSIM24-GUS that I’ve done so far. It’s starting to look more and more like the Helios (picture source). Won’t go into detail, but the efficiency depends mostly on how you coat the beads and the bullet.

Overhauled control panel. No more stupid laptop/serial control.

Three solenoids, powered by a helium tank, no more dinky little CO2 cartridges that run out after a few shots.

Mmmmmm, wire sleeving :3

One thing to keep in mind is that unless you’re following any design to the letter, and then using that gun to fire at the same type of specimen, expect to do a fair bit of optimizing. This isn’t quite a “plug and play” solution.

I recorded a brief video giving an overview of the gun and a test firing, hopefully it gives you some ideas. Also I go over some little gotchas I had to figure out on my own when I was building the gun.

Other gene-gun-ish thoughts:

• You could totally give yourself a GFP tattoo with this thing, if it can go a few mm in onion then your epidermis shouldn’t be too difficult (I’m pretty sure human genetic engineering is illegal though, don’t actually do this)
• The swagelok fittings were “new old” stock, purchased off ebay, took a while to get all these
• Be careful shopping for solenoids and check the voltage they require to actuate. Switching 200v DC isn’t as trivial as 12V.