Ferroptosis: death by iron
The time has come at long last, as I’m sure everyone has been impatiently waiting for. Of course I am talking about oxylipids.
For some reason or another oxylipids have taken hold of my mind. My background is lipids in general, but oxylipids are another subdiscipline in that field that, in my opinion, is a little overlooked. Or probably better said, an emerging field of study just getting its footing.
Most lipids are composed of some kind of glycerol backbone with fatty acids attached. Those fatty acids are usually reduced hydrocarbons with the occasional double bond. This is what gives you your monounsaturated fatty acids or polyunsaturated fatty acids. But occasionally in nature you’ll find hydroxy fatty acids, a kind of oxylipid.
Fig. 1. Phosphatidylcholine.
The above is a phosphatidylcholine molecule as an example of a very common lipid that makes up cell membranes. I’ve labeled all the ‘parts’ that make up this particular lipid class. The glycerol backbone (magenta) has three oxygens (O) that can bind either fatty acids, in this case two fatty acids, or another type of chemical group, and in this case would be the phosphocholine (light blue) group.
The ‘oxy’ in ‘oxylipids’ refers to an oxygen. This oxygen would be bonded to one of the carbons of the fatty acids chains (i.e. one of the vertices shown in the structure above). It could form a double bond, in which case it would be a ketone, or form a single bond, in which case a hydrogen (H) would also be bonded to the ‘other side’ of the oxygen to form a hydroxyl functional group (see Fig. 2 below). One example of an oxylipid is ricinoleic acid from Ricinus communis, also known as Castor bean, seen in Fig. 2B.
Fig. 2. Different ways oxygen can bind to carbon in a fatty acid (A), and an example of an oxylipid, ricinoleic acid (B).
Oxylipids are rarer than non-oxylipids. (Rare might not be entirely correct when you think of bulk of biomass. Trees and plants that have a ton of cutin and suberin will have lots of oxylipids. But rare in this case refers more to the normal function of a cell, especially mammalian cells.) Non-oxylipids would make up everything from cellular membranes, such as the phospholipids and sphingolipids, to storage and energy lipids, such as the triglycerides, and there are a whole host of others, e.g. steroids, signaling lipids, extracellular protective/barrier lipids (i.e. waxes, suberin and cutin in plants, etc.; those last two are actually filled to the brim with oxylipids). Oxylipids are often found in the context of signaling lipids. For example lipid mediators are a form of oxygenated polyunsaturated fatty acids. During an inflammatory response, these mediators signal to the immune system that there’s something wrong here, start producing factors that will aid in this distress. They can also affect the vasodilation of the vascular system in the surrounding tissue (i.e. make the blood vessels wider or more narrow) allowing more or less nutrients and fluid to the affected area.
But oxylipids are not all good. Sometimes they’re a sign of oxidative damage. One thing I’ve found rather interesting studying science throughout my life is how reactive oxygen really is. Often in our grade school level of science, we learn that oxygen is one of those indispensable elements of life. And that’s true. But it’s also one of the more destructive elements too. Fire and combustion are not possible without oxygen. If you think about where you get antioxidants, you’ll notice they all come from plants, those oxygen producers. In plants those antioxidants help to ward off the destructive effects of oxygen while the plant continues to produce it.
Oxygen is highly reactive because of its lone pairs of electrons. (All chemical reactions are really just an accounting of electrons.) Sometimes reactive oxygen species can be produced. This is where you don’t just have oxygen, but forms of oxygen with even greater reaction capability. Reactive oxygen species include the awesomely named superoxide anion where a peroxide anion loses its hydrogen. Reactive oxygen species also include the run-of-the-mill peroxides as well (same stuff that comes in the brown bottle you put on cuts and then bubbles all fizzy-like and then stings a little).
Because these reactive oxygen species (like the first word in their name) are so, well, reactive, they tend to react with anything they can rather indiscriminately. This leads to oxidative damage. Oxidative damage can occur on DNA, proteins, and also lipids. Damage to DNA is a potent danger as it could affect the expression of the genes from DNA. If those genes are sufficiently damaged or impaired, and this has a subsequent deleterious affect on the cell, then the cell may undergo apoptosis–cell death. (And yes, it’s fun to imagine that the person who came up with the name ‘apoptosis’ thought that cells would ‘pop’ when dying, but the second ‘p’ is actually silent.) But sometimes apoptosis doesn’t happen. And in rare circumstances, such oxidative damage to DNA can lead to cancerous traits. In fact, cancer in a general sense is simply the mucking up of DNA so that it no longer cooperates with all its neighboring cells surrounding it, but rather decides to keep growing and growing. Luckily, through the genius of evolution, our immune system has ways of recognizing cancerous cells and destroys them when the regular process of apoptosis fails. True cancer comes about when even this secondary measure fails.
Getting back to lipids, reactive oxygen species can also lead to oxidative damage to lipids. Sometimes oxygenation of lipids is needed and purposeful, such as in the case of lipid mediators synthesized by immune cells for signaling purposes. But sometimes something goes awry in the cell that then begins to produce a slew of reactive oxygen species that reek havoc to the cell. This is where I recently learned about a different type of cell death called ‘ferroptosis’. I’m assuming the word is pronounced like ‘Pharaoh-toes-is’. The first part of the word, ‘ferro-’, refers to iron. In cells, iron is an important metal for enzymes to work properly. Many people probably know that iron is an important metal for our heme proteins within red blood cells to carry oxygen. When our iron levels get low, our ability to sufficiently carry enough oxygen to power the rest of our body decreases and we become tired. Typically iron isn’t just floating about all willy-nilly like in the cytosolic juices of a cell, but rather it’s bound up with proteins. Besides heme, there are other enzymes that use iron as a cofactor for their enzymatic reaction. One of the major ones that I’m familiar with is, in fact, the desaturation of fatty acids. The iron within the enzyme helps to pull off a hydrogen from a saturated fatty acid with the help of oxygen to produce water and then a double bond to ultimately create an unsaturated fatty acid. (Interestingly, bacteria that are obligate anaerobes-―i.e. they must live without oxygen―-have a completely different mechanism for desaturating their fatty acids; I may write about this some time in another post.) But there are other proteins whose sole purpose is to bind up iron and to make it available to the cell when it’s in need, called ferritins. Iron can also form complex structures with sulfur too, which are called iron-sulfur clusters, that may act simply as another form of storage or be used in other enzymatic reactions.
But sometimes iron does get loose. And that’s a problem.
Free iron that is not bound with a protein can carry out non-enzymatic reactions throughout the cell. In particular, this free iron can react with oxygen to produce reactive oxygen species. This occurs through the Fenton cycle, shown below:
Fig. 3. The Fenton cycle of peroxide and hydroxide production (two reactive oxygen species)
This is able to make that superoxide anion (i.e. O2-•, produced by removing H+ from the peroxy radical in the lower left of Fig. 3.) and hydroxyl radicals (upper right of Fig. 3.). These ‘radicals’, as if they were some kind of molecular revolutionaries, have an unpaired electron surrounding them (the dot in the diagram above). Generally electrons are in pairs. When they’re not, they become quite reactive, capable of breaking bonds in other molecules, or simply binding itself to other molecules. This all leads to oxidative damage, to DNA, proteins, lipids, and every other metabolite within the cell.
With ferroptosis, this oxidative damage occurs on lipids, especially those at the cell membrane. The reactive oxygen species can end up reacting with polyunsaturated fatty acids. These types of fatty acids are particularly vulnerable to oxidative damage because of those extra double bonds that they carry. When enough of the lipids become oxidatively damaged, the cell undergoes cell death due to the loss of the integrity of its cell membrane. So, to recap, iron gets loose in the cell, it produces reactive oxygen species and also reacts with polyunsaturated fatty acids to produce oxidative damage, if left unmitigated, the damage becomes so great that the cell can no longer function properly and dies.
What’s particularly nefarious about the whole thing is that those reactive oxygen species can be self propagating. That means they can make more of themselves. So once there’s a little bit of oxidative damage to lipids, a lot more can be made due to a similar chemical reaction as that of the Fenton cycle, but occurring on the lipid itself. The oxidative damage to lipids at the membrane produce peroxy-phospholipids. The ‘peroxy’ means there are two oxygens with a single bond between themselves, and phospholipids are the main lipids that form cell membranes. Since those membrane lipids are bundled together to form that membrane, this type of self-propagating oxidative damage is like lighting a match to dry tinder. And it’s also clear how when these same lipids start to become damaged in this way, that it would ultimately lead to cell death. For what is a cell if not its membrane to define where it begins and ends in relation to the rest of the world?
Free iron occurs when it is on its way to becoming bound to proteins or other chemical groups (like heme). Or it can become ‘loose’ when metabolic pathways are not working correctly due to some other pathological disease or disorder. Another, though I would assume rarer, situation would be an influx of iron coming into the system where the cells are unable to store or get rid of the excess iron before it starts doing damage. This is likely a form of iron toxicity. In any case, if the pool of free iron becomes too great, reactive oxygen species can start to form and then lead to cell death via ferroptosis.
But free iron will always be around, and yet cells aren’t ‘popping’ themselves to death constantly, are they? Well, yes, cells are constantly being replaced over time, but no they aren’t all dying at the same time so that the body begins to deteriorate unless there’s a serious illness at hand. So cells must have a way for dealing with any reactive oxygen species produced by this free iron.
And boy do they. There are tons and tons of mechanisms and enzymes and pathways to deal with reactive oxygen species that muck about the cell when they go haywire. Too many to cover here. But there’s one in particular that is helpful. The main chemical group that leads to more oxidative damage is that peroxy group on oxidatively damaged phospholipids. Two oxygens bound by a single bond can easily break apart, and since oxygens are rather ‘greedy’ when it comes to electrons, they’ll split that single bond evenly, each side getting a single electron to form a hydroxyl radical and a hydroxyl-phospholipid radical. Both can go off and start the whole oxidative damage all over again. But the enzyme glutathione peroxidase-4 (GPX4) will convert that peroxyphospholipid into a hydroxyphospholipid, a far less reactive lipid molecule. And if need be, the cell has mechanisms to remove that hydroxy fatty acid on that phospholipid and replace it with a regular nonhydroxy fatty acid (almost like it were replacing a flat tire to a car). GPX4 also seems to be a major player in preventing ferroptosis since any knockout of the gene producing it is lethal. So ferroptosis may be a rather common phenomenon, or would be if it weren’t prevented by GPX4 converting oxidatively damaged lipids back to their undamaged selves, and other enzymes and molecules to capture or mediate reactive oxygen species, like superoxide anion and hydroxyl radicals, and, of course, all the iron-binding proteins that prevent too much of it becoming loose.
So, ferroptosis sounds bad, right? Well, not exactly. It’s a little more complicated than that (isn’t that the case for all science?). For example, if there is an increasing amount of free iron accumulating in the cell, then something’s gone haywire. And if the amount only becomes greater and greater, producing more and more oxidative damage, to the point where the cell can’t fix it all, it’s better for the cell to undergo ferroptosis and die rather than risk such oxidative damage occurring to its DNA, which could lead to cancerous traits should just the right combination of genes become damaged/mutated.
Which can, in fact, happen. Some studies have shown that cancerous cells have gained the ability to suppress ferroptosis so that they do not undergo cell death^(see references in review below)^. This allows them to continue their unchecked growth. They can do this in a few different ways. One is by diverting fatty acid synthesis away from polyunsaturated fatty acids, those that are most susceptible to oxidative damage and becoming hydroperoxyphospholipids, and towards monounsaturated fatty acids (i.e. fatty acids with a single double bond). Monounsaturated fatty acids are far less reactive to reactive oxygen species than the polyunsaturated fatty acids. Another way cancerous cells can avoid ferroptosis is by promoting the degradation of fatty acids for energy rather than making any fatty acids. This would be like cleaning out the kindling that could catch fire. No fuel to burn, no way for the whole house to burn down. Some other ways would be to increase the amount of enzymes, such as GPX4, that could resolve oxidatively damaged lipids or others that that trap or disarm reactive oxygen species.
So for a cell that is not cancerous, but is dysfunctional in some way, ferroptosis is a process of cell death through the oxidative damage of its membrane lipids due to ‘loose’ iron. Cell death in this case is a beneficial process, as it would prevent such oxidative damage that could lead to cancerous cells. And in such cancerous cells, the suppression of ferroptosis is one mechanism (one of many) that allows its survival and continued growth. For this reason, treatments and drugs that might target the process of ferroptosis, in order to activate it, in cancerous cells may be one way of treatment.
I was intrigued by this topic from the recent review article by Liang et al., 2022 and would recommend reading it if you can