The above reaction has a half-life somewhere between 7 to 200 years. As we discussed in earlier sections, the stability of the amide bond against this reaction is due to the fact that the carbonyl attacked in hydrolysis is not particularly electrophilic because the carbonyl is electron rich due to amide resonance. Remember, the amide bond has a zwitterionic resonance structure too. This makes the nitrogen sp2 hybridized as well. In fact, the nitrogen donates electrons to the carbonyl so well that the zwitterionic resonance structure actually affects the overall structure more than the neutralized form. Electrophillicity is also hindered by steric strain produced in the transition state cause by pushing the R group and the N from 120 degrees apart to 109.5 degrees apart. We can see the resonance structures below:
The resonance stability of the amide bond makes it difficult enough to break, but keep in mind also that the intermediate (rate-determining step) requires both a significant geometrical change of the previously sp2 hybridized carbonyl carbon to sp3 and charge separation. As humans made largely up of complex proteins held together by amide bonds and approximately 70% water, we should be particularly grateful for this stability. Otherwise, our bodies would be completely hydrolyzed.
Above is an energy reaction coordinate, the best representation of these energy differences for this purpose. We see that the very high activation energy necessary for this endothermic reaction is caused by the high-energy transition state. Using Hammond’s Postulate, we can speculate that the TS is similar to the tetrahedral intermediate. The transition state is closer in energy to the intermediate, so it must resemble the intermediate more closely. We know this intermediate is unstable (remember, this is because of charge separation and the loss of resonance stabilization) and because transition states are inherently high in energy due to properties such as pentavalent carbons. To successfully catalyze a reaction, we simply have to lower the activation energy.In order to be a functional and effective enzyme, the enzyme has to be able to hold down organic substrates through a combination of ionic bonding, hydrogen bonding, and so-called "hydrophobic" bonds. When the substrate joins the enzyme, the energy decreases. You can see from the very apparent difference between ∆∆G‡ and the ∆G-cat‡ in the energy reaction coordinate above that the protease is working by stabilizing the transition state. This idea is consistent with the Pauling Principle which claims that since ∆G-cat‡ < ∆G-uncat‡ for catalysis, enzymes must stabilize the TS‡ better than any other state along the reaction (the rate-determining step). So inorder for proteases to be effective, nature had to evolve a way to stabilize the below transition state.
Common Enzyme Mechanisms
Below are the names of 6 enzymes and the 1 step mechanisms that they facilitate. Try to recognize patterns in the mechanistic pathways of enzymes.
There are quite a few functioning proteolysis enzymes, but the subjects of this section are the serine proteases. Below is a diagram of the active site of thrombin, an example of a well-known and frequently studied serine protease, immediately after its strike on the carbonyl carbon of the substrate.
On the top left is thrombin’s specificity pocket. The specificity pocket contains an amino acid that binds the substrate's R-group with salt bridges and hydrogen bonding, a player in the “strapping down” process before an enzyme really gets its job done. Each protease targets different amino acids. In the case of thrombin, the targeted amino acids are lysine and arginine, as can be seen by the negatively charged carboxyl group perfectly set up to target the positively charged imine as well as the hydrogen bonding between the amide and the other oxygen of the carboxyl group. This is just one example of an intricate and ideal folding of the enzyme that facilitates optimal spacing for the given function. Notice the transition state shown here is the same as the general one shown earlier. The oxyanion generate in the transition state is stabilized by hydrogen bonding with the hydrogens from the oxyanion hole. This makes the carbonyl much more electrophilic. As a result, the carbonyl carbon is much easier to attack. This, as well as other intricacies that will be explained later, show the amazing enzymatic microenvironment that allows for speedy, facile amide bond hydrolysis without the need of an oxyanion. Too bad scientists can't be as awesome as enzymes...
Let's start with the captured substrate.
Above, we witness the type of chain gang partnership of the catalytic triad. Aspartic acid hydrogen bonds with the proton on the nitrogen of the histidine side-chain. This causes the histidine nitrogen to "feel" more negative facilitating the donation of its lone-pair. Subsequently the lone-pair is donated and histidine rearanges itself to deprotonate the serine which in turn increases the nucleophilicity of that oxygen “driving” it to attack the carbonyl carbon forming a tetrahedral intermediate. Amazingly, this all happens in one concerted step due to the proximity allowed by the environment of the enzyme - in an enzyme, everything is effectively intramolecular and very, very fast. This is yet another example of the microenvironment that makes the enzyme so amazingly efficient in catalyzing these reactions.
Above, we can see that the amide bond of the substrate is broken and its electrons pick up histidine's proton, putting histidine back to its original state. The complex is now an acyl-enzyme; the substrate is covalently bonded to the enzyme. However, this cannot be too stable of a bond since the enzyme is intended to catalyze hydrolysis and the job is only half done. We can't let the products of the reaction be too comfortable or else they would never leave the enzyme. If we can't even let the products stay, we certainly can't let this intermediate stay. This concept has a lot to do with inhibition as well, which will be discussed in a bit.
The oxyanion, once again stabilized by hydrogen bonding, slams back down to form a double bond and throw serine back where it came from.
And there you have it! So now we have a means by which our bodies can act more than 86,400,000 times faster than laboratory processes can (a thousand times a second versus once a day). Too bad each enzyme is specific to one or two amino acids, or we'd probably just go ahead and harvest them in large quantities… wait, we do (remember trypsin?). However, a couple of adjustments have to be made when we use these proteases for functions such as Edmund digestions. We can't exactly afford to have thrombin going around cleaving every arginine or lysine it sees. Since this and several other proteases are used within our own system, things would get messy and think of the poor flustered graduate students. For example, we've got this really lazy little protein called fibrinogen that roams around our cells. It does absolutely nothing...unless you get a cut. Once you get this cut, thrombin (remember this being mentioned before?) is utilized to break fibrinogen into fibrin which can aggregate and form blood clots that are excellent for protecting wounds from their immediate environment and, well, unfortunately, also the culprit of heart attacks and strokes. Strokes happen when unwanted clots dislodge and get stuck in the brain, blocking many neurological functions. We've got to block the clotting so that we can regulate WHEN the thrombin makes cuts...otherwise we'd have a problem. But really now, the body has proven time and time again that it can do these types of things...so are we really surprised that the body also has a way, which of course labs have mimicked, of also stopping the activity of thrombin and other serine proteases? [Side note: recall that thrombin’s specificity pocket was arg. oriented---fibrinogen contains lots of this amino acid. See the connection]
As was mentioned earlier, the covalent bond that forms between the serine residue and the substrate must later be cleaved so that the thrombin can go on working. Check out what happens when this doesn’t happen. If serine were to attack, say... one of these:
Interesting things would happen. Organisms synthesize this natural inhibitor for the sole purpose of interrupting various serine proteases from doing their jobs. The mechanism by which it does so is, to say the least, complex. There are two possible areas that the side chain can attack. It attacks the indicated carbonyl. Why doesn't it attack the amide? It is likely that the enzyme binds the inhibitor in such a way that the serine side chain cannot reach the amide. However, if the serine were given the choice it would likely still attack the indicated carbonyl. Why? Amide resonance. The indicated carbonyl is uncommonly electrophillic. The C=O bond is polarized to begin with; however this is true in every carbonyl. Look immediately to the left and the right of the carbonyl carbon, do you see any electron donating groups (ie are there any atoms with lone pairs)? No, there is an sp3 hybridized carbon to the left and an sp2 carbonyl to the right. Neither one of these groups has a lone-pair. The sp3 hybridized carbon can however donate a small amount of electron density through induction. This is one of the reasons ketones are less electrophillic than aldehydes. But this small inductive stabilization cannot overcome the electron withdraing effects of the carbonyl carbon of the ajacent amide. Carbonyl carbons are understood to be partially positive and therefore are by nature electron withdrawing groups. Amides are typically thought of as having electron rich carbons due to strong resonance from the nitrogen. This is true; however, this added electron density is in the pi system only, the sigma system is still subject to inductive withdrawal from the adacent oxygen and nitrogen. In effect, although the amide carbon has an electron rich pi system, it still has a quite electro-positive sigma system. Consequently, in an attempt at stability, it withdraws electron density from the indicated carbon. So not only does the carbonyl carbon not have any pi donators, but it also has a sigma electron withdrawing neighbor. As a result, this carbonyl is extremely electrophillic and thus an excellent electrophile/inhibitor for a serine protease. Attack by the serine protease creates a tetrahedral intermediate (which looks like the transition state but isn't - remember inhibitors are designed to mimic the transition state of the enzymes normal substrate). Since there is no available leaving group, the enzmye just sits there until something comes along and puts the humiliated protease out of its misery.
This cyclotheonamide, as well as other serine protease inhibitors, are irreversible inhibitors, meaning they form a covalent bond with the enzyme.
Chemists are always attempting to create inhibitors that mimic the action of those found in nature. So effective inhibitors are aldehydes, boronic acids, and alpha-chloro-ketones.
Like the cyclotheonamide, the aldehydes and alpha-chloro-ketones act as serine protease inhibitors by making it so that there is no good leaving group. The only thing that would leave when the oxyanion recollapses is the serine, which would simply attack again. Boron doesn't have a full octet, in fact it has an empty p orbital that can accept electrons from the serine with only the penalty of a formal negative charge. This makes it an excellent serine protease inhibitor.
- Enzymes stabilize the transition state better than it does any other ground state of the reaction it catalyzes, according to the Pauling Principle
- Three key parts to a serine protease: specificity pocket, oxyanion hole, and catalytic triad (aspartic acid, histidine, serine).
- When you think about it, Serine proteolysis is really just two simple addition-elimination reactions.
- The enzyme provides optimal alignment of the substrate through creating a microenvironment so that this can be done quickly in four steps due as an intramolecular interaction.
- To inhibit an enzyme, bind it to something it can't let go of.
- Serine Proteases are truly amazing. Compare the speed of this reaction to the one at the top of the page:
D9-1. Trifluoromethylketones are known to act as inhibitors of serine proteases by the mechanism shown to the right. Clearly and concisely explain why forming the adduct formed between the enzyme and inhibitor is favorable.
D9-3. Serine proteases are inhibited by TPCK. Draw (in abbreviated form) the structure of the product formed when serine proteases are treated with TPCK, and describe whether inhibition is reversible or irreversible.
D9-5. As far as I know, no one has ever discovered a tyrosine protease. But, if one ever is discovered, I’ll bet it will have only a catalytic dyad (His, Tyr), akin to the thiol proteases, which do exist (see problem set). Is this hypothesis reasonable? Why?
D9-2. When the histidine in the catalytic triad of a serine protease is mutated to a tryptophan, the protease loses all of its catalytic activity. Why?
D9-4. During early studies on the serine protease chymotripsin, diisopropylphosphofluoridate (DIPF) was used to probe the active site. It was allowed to react with the enzyme to form an acid-stable covalent adduct. The enzyme was degraded into its component amino acids by acidic hydrolysis (thus hydrolyzing all of the peptide bonds), and the following molecule was among those isolated. Show the mechanism of the reaction that produced the serine derivative shown (do not show the mechanism of amide bond hydrolysis).