I’m taking a break from talking about enzymes because I realize that there’s another important topic to broach before we can really get into a meaningful discussion about metabolism, and that’s one of ATP. Before now, I’ve made vague, hand-wavy allusions to this mystical magical molecule, assuming that my readers have a passing understanding of what I’m talking about, but now it’s time to get down to it. To get our hands dirty and ask the hard questions, such as, “Cell, how do you manage to keep around so much of this thermodynamically unfavored crud you call adenosine triphosphate?”
Whoops, spoilers, I guess. Oh well, at least I didn’t tell you that, in Hunter x Hunter, Gon—
If you’ve had a touch of organic chemistry, you’re probably familiar with the concept of an ester. (If not, you should control-click the little pink linkerino.) However, there’s a similar kind of bond of prevailing importance when we’re talking about ATP—the phosphoester bond.
Phosphoester bonds are formed when you condense an alcohol with phosphoric acid instead of a carboxylic acid. They’re very common in biochemistry, frequently as metabolites in various pathways. A relevant example of a phosphoester bond can be found in AMP, wherein the phosphate is bonded to adenosine by a phosphoester bond.
Another important kind of bond involving phosphate is the phosphoanhydride bonds. If you’re like me, this can be a little trickier to visualize, since I was always a bit fuzzy on the concept of an anhydride. When it comes to straight-up phosphoanhydride bonds, think of one phosphate bonded to another through an oxygen. This occurs in ADP and ATP, wherein the subsequent phosphates are bonded to each other through phosphoanhydride bonds. These phosphate groups are highly negative under physiological conditions, which will become important when we start talking about divalent magnesium.
On the other hand, mixed anhydrides are half phosphate, half carboxylic acid. We’ll be coming back to these, so in the back of your mind, remember that these are really freaking energetic. Is that enough emphasis for you? Shall I add some boldface type as well?
Okay, okay. This information is important because it helps us understand where ATP gets its energy content. If you’re familiar with the structure of ATP, you’re probably already counting types of bonds in your head. If you’re not, I’ll find you a public domain picture. Just a sec.
Here you go! In this picture, you can see the fully protonated form of ATP. The nitrogeny thing on the right is adenine (maybe you recognize it from my profile picture? 😉 ), and the ring in the middle with the OH groups is ribose (a sugar). You can see on the left that the first phosphate is attached to the sugar with a phosphoester bond. The remaining two phosphates are attached to each other and the first phosphate with two phosphoanhydride bonds. Those bonds are the primary source of energy in the molecule.
As a point of reference, it should be noted that, at pH 7 (physiological pH), those phosphates are either entirely deprotonated (four negative charges) or nearly deprotonated (three negative charges). Either way, ATP binds Mg2+ (that divalent magnesium I mentioned?) well.
Now that we’re familiar with the structure of ATP, let’s talk about what happens when you turn it into ADP. The general equation for that reaction as we’d expect it to happen goes like this:
ATP + H2O ⇌ ADP + Pi
where Pi is inorganic phosphate.
This is really slow in water, but we’re not too concerned about speed right now. Instead, let’s just figure out how much energy we lose or gain by running this reaction. If we look back on Gen Chem for a sec (#TT or whatever the kids are doing these days, eh?), we’ll remember that, for a reaction to be spontaneous, it will release energy. Turns out, this one does—at standard conditions, it releases a whopping 37 kJ per mole.
Why is that? Well, let’s take a step back. How much energy do we get if we just chop up any old phosphoanhydride bond? That is what’s happening in that hydrolysis reaction up there, after all.
Turns out, no matter which bond you cleave (ADP to AMP and Pi, ATP to ADP and Pi, ATP to AMP and PPi, or even PPi to 2 Pi), you’ll end up releasing 30-40 kJ for each mole of the molecule.
You can cleave phosphoester bonds, too, but you won’t get nearly as much energy. Taking AMP and making it into adenoside and Pi only gives you about 13 kJ/mol, and that’s well within the normal range for cleavage of a phosphoester bond (10-15 kJ). Knowing that, it makes sense that we don’t use AMP for our energy storage.
Here’s where the lightbulb goes off. Those phosphoanhydride bonds are where we’re getting the energy that’s stored in ATP, and that’s why the free energy change for the hydrolysis of ATP to ADP is so exothermic (releases so much energy). Knowing that, can we predict how much ATP, ADP and AMP there should be at an equilibrium?
Of course we can, but the results aren’t pretty. Unsurprisingly, because it contains the least energy, AMP is the most stable of the three. ADP is next, and ATP is last. What that means is that equilibrium is going to favor AMP, because converting ATP to ADP and then AMP is going to give off a lot of energy (73 kJ/mol worth!). The opposite is also true; making ATP from AMP would take 73 kJ for every mole of AMP. Yikes, that’s a lot of energy.
Because of that, at equilibrium, it turns out that there would be one molecule of ATP for every million molecules of ADP. The same is true of ADP in comparison to AMP. Are you cringing yet? You get why this is a problem; ATP is the energy currency of the cell, and it looks like thermodynamics very much want our cells to go broke.
Obviously, our cells aren’t going to be able to successfully make ATP from ADP and Pi using the equilibrium I showed you above. So, what do they do? Do they sit around and cry, the way I did when I realized I had another biochem test coming up?
Of course not, because our cells are awesome. (Like, seriously. You should thank yours sometime. Not sure how that would work, but these things are freaking cool.) They just use a workaround! (Much like I’m doing right now to study for my biochem test!)
To understand their workaround, first, let’s talk about why this reaction is so problematic in the first place.
When a molecule of ATP sets out to become a molecule of ADP (yes, I know I’m giving molecules sentience, whatever, I blame Hetalia), it has to overcome an energy barrier called the activation energy. The activation energy is the energy difference between ATP and a high-energy in-between structure, the transition state. Once ATP surmounts that energy difference, though, it becomes ADP and Pi. On the other hand, if ADP and Pi want to become ATP, they also have to work their way up to the energy level of that same transition state.
So, the question becomes, who has to work harder, the ATP or the ADP and Pi? As you’ve already guessed, it’s a lot harder for ADP and Pi, which are both relatively low energy, to climb up that hill than it is for ATP, which is already pretty high energy. The difference between the amount ATP has to climb and the amount ADP and Pi have to climb is what we think of as the free energy change of the reaction. This makes sense, since we know this reaction gives off energy—the products must be less energetic than the reactants, to compensate for that.
That means that, while the ATP will eventually convert into ADP and Pi, that reaction is essentially irreversible. Under standard conditions, ADP and Pi just don’t have enough “umph” to clear that hurdle and become ATP. So, what’s a cell to do? Lower that energy hump, perhaps?
Well, what causes it in the first place? The transition state, right? Well, what’s so terrible about the transition state?
In the case of ATP, it’s a pentacoordinate transition state.
Yes, if you’re feeling cringy, that’s probably appropriate.
You see, what happens is that water attacks the outermost phosphorus on ATP, opening up one of the P=O double bonds and putting (another) negative charge on (another) oxygen. Water isn’t very nucleophilic, so it doesn’t do this very well. After it does, phosphorus has five bonds, which isn’t as stable as phosphorus with four bonds. Finally, another water has to come in and pull a hydrogen off of the water that attacked the phosphorus (and water isn’t that basic at pH 7, mind you) while the electrons on that oxygen move back in and eject the phosphorus from the ATP. You end up with ADP and Pi, but at what cost?
This reaction doesn’t even really work well in the forward direction, the only direction it can occur in—it occurs at a rate of four ATP each 105 minutes. That’s four every seventy days.
So, basically, this reaction is a no-go. So, I ask again, what’s a cell to do?
Hmm… well, what have we studied that lowers activation energy? That would certainly do the trick, wouldn’t it…
(Forgive me, I’ve spent the whole day being sarcastic in the privacy of my own room. It’s starting to leak out…)
You got it! Enter Enzyme-san, stage left!
If you’ve read my posts about enzymes, or if you’ve studied a lot about enzymes already, you know that this is exactly what enzymes live for. They earn their keep by lowering the activation energies of reactions, increasing the speed at which both the forward and the reverse reactions happen. However, because the difference in the energy hump (smaller though it may be) between both sides is still the same, they don’t fiddle with the energy change or the position of the equilibrium.
How do they pull it off? By choosing a different, lower energy transition state! Enzymes are really clever in the way they do this, and the enzyme that does this with ATP is no different.
The reaction still proceeds through the same pentacoordinate transition state, but an enzyme changes things up a little bit. First off, you remember that unhappy little water that had to act as a nucleophile? Psh, forget that! This enzyme has a basic group on it that plucks a proton off of a water molecule, plain and simple, right there where it needs to react. Hydroxide is significantly better at being a nucleophile, and it happily reacts with the phosphorus the way water did before.
In this scenario, magnesium is also bound to the negative charges in the phosphates. That stabilizes the large amount of negative charge. Once the electrons fall back into that P=O double bond and eject that phosphate, the basic group on the enzyme transfers the proton it picked up to water, regenerating the original enzyme.
So, that’s how we mediate the hydrolysis of ATP to ADP and Pi, but I still haven’t told you how we keep up our levels of ATP. Well, first, we should take note of a fact that is perhaps a bit surprising: there are a lot of compounds in our body that are higher energy than ATP. Phosphocreatine, found in our muscles, releases 43 kJ/mol upon hydrolysis of its phosphamide bond. Phosphoenolpyruvate is hugely energetic (take note of that—we’ll use it in glycolysis!) releasing a ridiculous 62 kJ per mole phosphoester bond that you break. (Yes, that is anomalous for a phosphoester bond—we’ll get to that!)
Now, let’s back up a little bit and take note of another important fact: usually, when phosphate is transferred from a molecule like ATP as part of a metabolic reaction, it isn’t done with water, like I showed above. Rather, it’s mediated by a kind of transferase called a kinase, which attaches it directly to the other molecule in a basically irreversible reaction. These reactions are often irreversible because they form bonds less energetic than the phosphoanhydride bond in ATP.
For example, when ATP is used to turn glucose into glucose-6-phosphate, an enzyme called hexokinase mediates the reaction, and it’s irreversible. A phosphate from the ATP is transferred directly to the glucose without the use of water, and the reaction releases 32 kJ/mol. (How do you get that number? Add up the energy it takes to make glucose-6-phosphate by condensation with phosphate and the energy it takes to hydrolyze a phosphoanhydride bond in ATP. Even though that’s not what happens, the enzyme doesn’t alter free energy change, remember?) At equilibrium, there are 10,000 more product molecules than reactants.
Okay, that’s great, but, again, obviously we can’t make ATP this way. Why is this relevant?
Because ATP can be made the same way!
Yep, that’s right. Knowing what you know of the driving force in these reactions, it shouldn’t come as too much of a surprise that ATP can be made this way, provided you use reactant molecules that are higher in energy than ATP. Remember those really high energy species I mentioned? Phosphocreatine? Phosphoenolpyruvate? Yup, this is where they come in.
For example, in the last step of glycolysis (which aims to make ATP without using oxygen), phosphoenolpyruvate, a phosphoester of the higher energy enol form of enolpyruvate, is used to make ATP and pyruvate. Pyruvate kinase (a transferase for phosphate, ‘member) takes the phosphate group from phosphoenolpyruvate and sticks it on a molecule of ADP to make ATP. This reaction actually releases energy in addition to the formation of the ATP molecule—a whole 25 kJ/mol’s worth, to be precise.
Why? Well, if you were to just hydrolyze the phospoester bond in phosphoenolpyruvate, you would release 62 kJ/mol. You get so much energy for a measly phosphoester bond because the phosphate is forcing the molecule to adopt its enol form, which is high energy. Once that phosphoate is lopped off, the enolate tautomerizes to its keto form (pyruvate), netting you a lot of energy. Meanwhile, making that phosphoanhydride bond in ATP takes 37 kJ/mol of energy. Add those numbers up, and what do you get? The energy change for the reaction!
(How do we get such a high energy compound as phosphoenolpyruvate, though? Well, I’ll talk about that when we get to glycolysis. 😉 )
Something slightly different happens with phosphocreatine, which hangs out mostly in your muscles. The free energy change for the transfer of the phosphate on phosphocreatine to ADP is about -4 kJ/mol, meaning that ATP and creatine are favored, but only slightly. When your muscles are running low on energy, they pop phosphates off of molecules of phosphocreatine to get some more ATP. If, instead, they’ve got plenty of ATP hanging around, they’ll put the phosphates back on creatine to replenish their emergency energy reserve.
Ta-da! Don’t you feel like your life is so much better now that you know this…?
Well, either way, this is really important to making the energy that our cells need, so it’s something handy to know. It will help us understand glycolysis better, as well. However, before we get into that, we need to look at the driving force behind the whole setup, a lovely little thing called “metabolic flux.”
Questions? Comments? Suggestions? Corrections? I’m listening!