Say GFP! Today pre-teens got a lesson on proteins and GFP by making paper models with ME! GFP stands for Green Fluorescent Protein and it’s a protein that glows green, making it useful for all sorts of things – like tracking proteins inside of cells, seeing when certain genes get expressed, etc. And, to the long list of uses, add educational camp activity! Today I used print-and-fold models to teach middle-schoolers about proteins as part of the CSHL WiSE/ DNA Learning Center’s summer camp. And I learned a lot about GFP in the process. So today’s post is for those who want to know – what gives GFP its glow!
note: you can fold your own GFP models – the template & instructions (and lots of other awesome resources) are available through the PDB-101 website (which I learned about at the @ASBMB conference I attended and have been waiting to try out!) I found it a great tool for explaining the levels of protein structure as well as the biochemistry behind the green glow. http://bit.ly/2Hxyk29
Light is “just” little packets of energy called photons traveling in waves. You can think of them kinda like baseballs wiggling through the air that may or may not get “caught” depending on whether it runs into something with the “right mitt.” Different colors of light have packets with different amounts of energy (e.g. blue light has higher-energy photons than red light). And since all light has to travel at the same speed (the speed of light), the higher-energy photons “bounce up and down more” giving you closer together peaks (higher frequency, shorter wavelength). If you think of slalom racing, it’s like you have a bunch of skiers that all get to the finish line at the same time, but the skiers with higher energy make more S’s during that time.
Depending on their structure, different molecules can absorb different photons and hence “steal” some colors of the rainbow – or of the “infra” or “ultra” “rainbows” – there’s only a segment of the electromagnetic radiation (EMR) spectrum that we can actually see. We call this the visible spectrum – it spans from wavelengths of about 380-740 nanometers (nm)(a nanometer is 1 billionth of a meter so even the longest of these are pretty dang short).
Speaking of longest – that’d be the red light. It’s at the 740 end and then violet is at the 380 end with the OYGBI of ROYGBIV in between them in that order. Remember longer wavelength is lower energy & lower frequency. There’s a lot of other EMR that we can’t see – below red we have things like infrared & microwaves while above violet we have things like ultraviolet (UV) and x-rays
Just because we can’t see it doesn’t mean it’s not there – and just because we can’t see it doesn’t mean that other molecules can’t. And if we ask nicely, they might show it to us! The trick is to get it to “catch” the light we can’t see and give it back as light we can see.
Chromophores are molecules or specific parts of bigger molecules that absorb photons. If you think about different wavelengths of light as baseballs of different colors (photons of different energies) being thrown, chromophores are like catcher’s mitts that can only catch certain ones. They can only “catch a ball” if the photon’s energy is just right for exciting an electron in a molecule. An electron a type of subatomic particle that molecules use to interact & bond with one another. They can live in different “orbitals” and the highest energy electrons live in the furthest orbitals from the nucleus. More here: http://bit.ly/33RznDA
If they absorb light, they gain the energy of the photon and then can afford to live in the orbital “suburbs” – get excited to a higher energy level. But they can’t maintain this excited state. Some chromophores just steal the photons and dissipate the extra energy by just kinda shaking it off through vibrations, letting off heat, etc.
But a type of chromophore called a fluorophore steals a photon, which excites an electron into a higher orbital, but the excitement wears off and the electron falls back down and when it does so it releases the extra energy as a new photon, but some energy is lost as heat, etc. in the process so the new photon that’s released has lower energy and therefore a different color (left-shifted in ROYBGIV ordering) (i.e. GFP absorbs ultraviolet light and emits green light)
A lot of times chromophores involve metal cofactors and/or heterocyclic aromatic compounds (stuck-together rings in which the atoms share electrons “communally”). These are common chromatophores because their electrons are easily excited. So, many of the proteins that have color do because they’ve “bought a mitt” by binding to some cofactor that’s not part of the “protein” as written – for example, the heme of hemoglobin. One of the cool things about GFP it is that it “makes it own mitt” – its chromophore is “built into” the protein – no cofactors required – although some post-translational assembly *is* required.
Translation is the process of using RNA instructions to make proteins by linking together amino acids (protein letters) which have a generic backbone & unique side chains (we call the order of these letters the “primary structure”). GFP has 238 amino acids (giving it a mass of 26.9 kDa). More on translation: http://bit.ly/31IwofL
Normally, in proteins (with the exception of disulfide bonds) the only strong covalent bonds are the peptide bonds linking the generic backbones of amino acids into chains. When amino acids link together, they lose the equivalent of a water, leaving a “residual” part consisting of most of the generic backbone with the side chain sticking out – so in the context of a chain, we frequently call amino acids “residues”
The generic backbones also interact with one another, leading to common “secondary structure” motifs like beta strands (which can interact with each other to give you β-sheets
parallel β-sheet – at least 2 β-strands & they’re oriented in same direction, like
antiparallel β-sheet – β-strands oriented in opposite directions like
and a-helices. GFP has 11 β-strands and 3 short α-helices
But, unlike the primary structure, the interactions leading to secondary structure are just weaker bonds called hydrogen bonds, in which electrons aren’t really shared (no electron housing changes just partial charge attractions).
Since the side chains are sticking out, they can interact with one another, leading to “tertiary structure.” GFP has a lot of residues residues which like water (are hydrophilic) and a lot of others which want to avoid it (are hydrophobic). So it folds up into a barrel structure so that the hydrophobic parts huddle together inside an outer shield of the hydrophilic ones. Different chains can interact to give you quaternary structure (leading to things like dimers, trimers, etc.).
Above the primary level (with the exception of disulfide bonds between cysteine residues) all the structure comes from non-covalent (weak) interactions.
But at the core of GFP (hidden in the center, where water molecules can’t “steal its energy”) are a few amino acids that make special bonds with their side chains. A special linkage of a threonine, a glycine, & a tyrosine (Thr65-Tyr66-Gly67) where their side chains link up to form a cyclic structure.
The folding to optimize secondary and tertiary interactions positions these key residues in position to form a chromophore – but actually forming the chromophore requires them to link together in weird ways to make them more “dye-like” (bring in some resonance. The Ser carbonyl & Gly nitrogen link up, cyclizing & kicking out water (dehydration). This makes a ring, but the ring isn’t resonance-stabilized yet (electrons aren’t being spread out around the ring). In order to get that resonance stabilization (pi-delocalization) the Tyr has to get oxidized (desaturated). Now it has a delocalized electron network that’s easier for electrons to get excited from.
So when it gets hit by high-energy light it can absorb a photon & use the energy to excite an electron – but then the electron “crashes” and falls back down, releasing a photon of lower energy (longer wavelength) – in this case light our brains tell us is green.
GFP is a naturally-occurring protein. The “classic” comes from a jellyfish called Aequorea victoria & was first isolated in the 1960s. In the 1990s it took off as a major part of biochemists’ & biologists’ tool boxes. Some uses of GFP & it’s variants:
- fuse it to a protein of interest to see where in the cell the protein ends up
- stick a miRNA target site after it to see if the corresponding miRNA is expressed & the miRNA machinery’s working well
- stick it under control of a promoter to see when that promoter’s turned on
The classic GFP from Aequorea victoria (avGFP) has a major excitation peak at 395nm (blue to UV range) minor excitation peak at 475 emission peak at 509 (lower green portion of visible spectrum). But scientists have made modifications to it to get it to give off different colors, like the yellow glow of YFP.
These special bonds also make the chromophore really hardy as I found out a while back when I was having a purification problem where I wasn’t getting any of my protein sticking to the affinity column even though it had the matching tag. more here: http://bit.ly/2W8zo2b
In addition to sticking a tag on the end of my protein, when expressing my proteins in insect cells I I coexpress YFP as a reporter. It’s on the same bacmid as my protein and under a promoter that gets activated at the same time so it helps me ensure that the backed got taken in & used to produce baculovirus & protein. (but unlike the tag it’s not actually part of my protein). more here: http://bit.ly/2YQtPFJ
Thanks to the YFP, the lysate (soluble stuff you get when you break open (lyse) the cells & centrifuge them (spin them really fast) to separate out the membrane parts) looks really yellow. But the YFP doesn’t have the affinity tag so it doesn’t stick to the column. But a few months ago, the column was turning yellow & staying yellow until I added the competitor. Normally when I add competitor, it’s only my protein on the column and so it’s only my protein that gets pushed off.
Turned out there was a problem with my buffer (pH-stabilizing salt water) – instead of the pH of 8 that I wanted the “buffer” was at a pH of ~4. And this was denaturing my protein (unfolding all that beautiful structure) so that the protein no longer bound. But the YFP was so hardy, it survived, and still looked yellow! And, because the YFP was the only survivor and without competition it could weakly bind the column without being outcompeted. When I added desthiobiotin (competitor), the YFP got pushed out giving me a yellow-tinted eluant that didn’t have my protein in it.
We also played with 3D models from the MSOE molecular lending library – thanks!
More on protein structure: http://bit.ly/2KBFRiS
Slides on protein structure and structural biology: http://bit.ly/2Zzog2K