Ubiquitin tags proteins for the bin! UBIQUITIN (Ub) is a small protein that acts like a pre-addressed shipping label that gets attached to proteins your cell wants to get rid of. The UPS (Ubiquitin Proteasome System) delivers these damaged or unwanted packages to a “trash incinerator” called the PROTEASOME.
the UPS has multiple steps (w/substeps) & “job positions”
- tagging (UBIQUITINYLATION): stick on ubiquitin “shipping label”
- delivery: take the labeled protein to its proteasomal destination (our incinerator)
- degradation: proteasome “scans label” & (if all checks out) chops it up
Before we discuss how it works, Why’s it important? UPS is important in several ways including: needing stuff, not needing stuff, & needing to get rid of stuff
- needing PARTS: Your cells have limited resources – you can’t make something from nothing, so you need to salvage parts from things you no longer need & use them to make things you do need. Kinda like how you sometimes have to break up your awesome LEGO structure in order to build a new one
- needing SPACE: Your cells have limited space – they don’t wind your DNA tightly around histone proteins to stuff into the nucleus just so that you can fill the cell with useless proteins!
not needing stuff: once a protein’s done its job there’s no need to keep it around. Like one of those awkward meetings where you think it’s over – your part’s done at least, but then the big whigs start chatting with each other and you’re not sure what do do so you’re just waiting to be dismissed… ubiquitin gives a protein “permission” to leave (& get sent to the trash incinerator…)
needing to get rid of stuff: other than wanting their parts, the reason the proteins get sent for destruction is because their existence is harmful to the cell
🔹 sometimes their “normal” function is harmful (wrong time, wrong place) for example, regulatory proteins that are involved in controlling cell cycles need to be able to cycle, so you want to be able to get rid of them quickly when it’s not their turn
🔹 sometimes they’ve gained some “abnormal” function
🔹🔹 PARTIAL PROTEINS can be partially functional – say a protein gets truncated (it’s shorter than it should be) -> this can happen for multiple reasons, including no-go decay, where the ribosome stalls when making it so it has to be extracted before the protein-making’s done http://bit.ly/2GACRzp or proteolytic cleavage, where the protein gets cut after it’s made. No matter how they form, partial proteins can be complete nightmares!These proteins might only have part of what they need to work, so they can wreak havoc, like if they can bind to something, but can’t do what they’re supposed to to it, so they just hog it so that even if there are still “good” proteins they can’t bind cuz the bad one’s there
🔹🔹 MISFOLDED PROTEINS can aggregate – proteins fold in a way that maximizes favorable interactions between their amino acid building blocks. One major folding determinant is sticking hydrophilic (water-loving) parts on the outside facing water and shielding hydrophobic (water-avoiding) parts from the water by grouping them together and in the interior of the protein. (this is assuming you’re talking proteins in the liquidy parts of your body – the opposite holds for proteins embedded in the lipidy parts (fatty membranes, which are hydrophobic))
- if proteins don’t fold properly, hydrophobic parts can be left facing water, and they freak out -> panic and clump together -> form insoluble aggregates & can sequester other stuff (trap important molecules with them)
So, how does UPS get rid of these bad guys? The tagging part itself (UBIQUITINYLATION) has several steps and involves molecules called UBIQUITIN LIGASES – there are 3 types that work in a concerted process: E1, E2, & E3
Unlike other forms of post-translational modification (changes to a protein in its protein form (not at the DNA or RNA level)) like phosphorylation (addition of a phosphate group) or glycosylation (addition of a sugar chain), ubiquitinylation adds a whole protein onto the protein – hard to miss that! Especially when you add whole chains of it on (polyubiquitination) and there are different ubiquitin-binding domains (UBDs) on various proteins that recognize different types of them.
These chains of Ub can be linear or branched because Ub has multiple attachment spots. When amino acids link up to form a protein, they do so using their generic backbone which has an amiNo (-NH2/3) group on one end and a Carboxyl (C=O-OH/O-) group on the other end. So you’re left with an amino-group-free “N-terminal” at the start of the chain and a carboxyl-group-free C-terminus at the end of the chain. The initial attachment’s through ubiquitin’s C-terminus and it attaches to lysine residues and/or the N-terminal of another protein. Ub has 7 lysines, so it can form linkages in different places, but lysine 48 (K48) seems to be most important for degradation.
jargon note: lysine is one of 20 common amino acids (protein letters) & “residue” is just the name we give to any amino acid when it’s part of a larger chain (it’s the residual leftovers after parts of its backbone are used for the linking).
Why lysine? Lysine has an amino group (one of those nitrogen/hydrogen things) at the end of its long side chain which allows for a special type of linkage between lysine & the C-terminal of Ub. Normally, it’s only the generic backbone part of amino acid letters that form strong, covalent, bonds between one another – amino group of one + carboxyl group of one next to it -> peptide bond
But unlike those end-to-end linkages, the linkages to ubiquitin are ISOPEPTIDE BONDS – similarly to how isopropanol & propanol or isoleucine & leucine have the same atoms, but in a different arrangement, isopeptide bonds are peptide bonds that are formed “differently” from the ones we’re used to, with the amine group from a lysine “mimicking” an N-terminus. Only 1 end required and the lysine can be anywhere.
They may be weird, but they’re still strong, covalent bonds, not like normal protein-protein interactions. The non-peptide bonds between amino acids, like the ones that lead to proper folding, are (individually) weak bonds based on opposite charge attractions (these charges are often partial and/or transient). This extends to binding between proteins -> proteins typically bind to one another through weak attractive bonds, often involving interactions between the unique parts (side chains) of amino acids. It’s important that these interactions aren’t permanent because you need to allow for dynamics -> proteins coming & going in response to intercellular signals, etc.
But when it comes for applying shipping labels, you want that to stick! so ubiquitin is added through strong covalent bonds. It’s still reversible, but you need special “erasers” called deubiquitinating enzymes (DUBs) to cut them off.
But these bonds are also harder to form and you want to make sure you put them on correctly. So you’re gonna need to invest some energy money and use multiple proteins to ensure specificity.
It starts with E1 (ubiquitin-activating enzyme) There are just 2 of these in mammals (as far as we know). E1 activates the Ub – it’s kinda like “licking the stamp” & it’s really similar to how you activate amino acids in translation. Once again, it involves a trip to the ATP ATM for some energy money.
ATP is often used to “bribe” reactions to occur. You see, molecules like to be free. So getting molecules to come together is often a challenge. But you can pair reactions that don’t like to happen (require energy) to reactions that do like to happen (release energy) to help them happen.
One reaction that likes to happen is ATP hydrolysis (water-mediated splitting). ATP has 3 phosphate groups which are like 3 balls of negative charge being held together against their will (opposite charges repel), so if you let them split up it’s like releasing a compressed spring -> like a bimolecular slinky, you can use the released energy to do things (in this case forming new bonds rather than moving down stairs)
1st E1 facilitates Ub activation: Ub + ATP -> AMP sticks on (for now), giving you adenylated Ub (Ub~AMP) & PPi is lost as a “transfer fee.” now you have AMP stuck to Ub, but you want it stuck (temporarily) to E1. So you swap: E1 + Ub~AMP -> Ub~E1 + AMP
Now the energy money’s all been spent but that’s ok because the next steps are more favorable – before you were forcing 2 molecules together, but now you’re just transferring one to another. So you don’t have to bribe them, you just need to get the molecules close together and in the right orientation so that they realize the other partner’s better.
now E2 comes in. E2s are ubiquitin-conjugating enzymes. There are ~ 40 of these. They all have Ub-transferring domains but they also have differences that allow for some specificity.
E1 passes off the Ub to E2 (giving you E2~ubiquitin): Ub~E1 + E2 -> Ub~E2 + E1
That’s nice and all, but where you *really* want that Ub to be is on the protein you want to label (the SUBSTRATE). For this you need an E3.
E3s are ubiquitin protein ligases. There are lots of these (~600) because, unlike the others (E1 & E2) they have to bind the protein to be tagged, so they need to have some specificity so you don’t tag the wrong thing, and there are lots of things that need tagging. The E3s can be split into 3 major classes: HECT, RING and RING-between-RING (RBR)
RING ligases rely on E2 to do the final transfer (E2~ub + substrate -> E2 + ub~substrate)
HECT & RBR ligases do the final transfer themselves. But the Ub is on E2, so you have to go through an additional intermediate transfer of Ub from E2 to E3 and then to the substrate.
You could stop with just 1 (monoubiquitinylation) but the proteasome goes by the Lay’s slogan -> it can’t eat just one! You need at least 4 ubiquitin in order for the chain to be long enough that the ubiquitin can bind the ubiquitin-binding part of the proteasome and the attachment part can reach the scissors part that cuts the Ub off & starts letting the protein in & chopping it up (more on this to come).
And because DUBs (Ub erasers) are around, if you want to be safe, you’ll want to add more than 4. So, many E3s are proccessive -> can keep adding without falling off the protein substrate. But they need more to add. So the E2s come and go, with freshly Ub-loaded ones replacing the Ub-less ones. The stronger the interaction between the E3 & the protein, the longer it can stick around before falling off, and so the more Ub that can be added.
This is one way you can get some regulatory control. For example, maybe you phosphorylate part of a protein to get the E3 to bind strongly when you’re ready to degrade it. If it binds at the “wrong time” it’ll be a weaker interaction so it will fall off before it adds 4 or more and the DUBs can remove them.
There are other functions of ubiquitinylation as well, and scientists are starting to figure out the “ubiquitin code” – how different lengths and branching types, etc. can send different messages, such as recruiting other proteins to help with things like gene regulation
good video primer: http://bit.ly/2L4RFvD
link to review article model’s from: https://www.nature.com/articles/nsmb.2780#f2
more about all sorts of stuff: #365DaysOfScience All (with topics listed) 👉 http://bit.ly/2OllAB0