Complement system

We come, now, to the complement system - the other prong of the innate immune system (although it has a relationship with the adaptive immune system, too). The complement system complements the activities of immunocompetent cells and of antibodies. The complement system is made up of proteins produced predominantly by the liver which are generally present in the blood plasma, but in inactive states. These proteins take part in a biochemical cascade initiated by means of three possible pathways - the classical complement pathway, the alternative complement pathway, and the lectin pathway - each of which we will look at. As we have already seen to an extent, this cascade serves a number of important functions:

• Some of the proteins serve as chemotactic factors for immunocompetent cells involved in the inflammatory response


•  Some proteins opsonise pathogens, "marking" them as targets for phagocytosis


•  As we saw above, the membrane attack complex destroys pathogens through the process called lysis


• Finally, "clearing up" antigen-antibody complexes (more on these shortly) is the last role that the complement system plays

Since the complement system can cause severe damage to the body it's defending, it is of critical importance that it be well regulated. The primary way of doing this is through complement control proteins, which we will talk about very soon, as they're very relevant and have an important role to play in the effective functioning of the system itself, as well.

Before we talk about the complement system, however, we have to discuss antibodies in a lot more detail. We've already made reference to them, but it is time to consider them properly. We will have more to say about them when we look at adaptive immunity; but, for now, we need only understand what they are and what they do.

Antibodies (which have the abbreviation Ab and are also referred to as immunoglobulin (or Ig)) are proteins which play a critical role in the immune system by instigating immunocompetent cells to attack a pathogen, or by directly interfering with the pathogen's ability to invade the body. Importantly, antibodies are capable of engendering particular responses to particular pathogens, which is obviously important for the effective and efficient removal of a threat.

Antibodies function by binding to antigens (antibody generators). An antigen is any substance which binds with an antibody. This is getting to be quite circular and we need to be more specific. An antigen is any substance which is able to be bound by the antigen-binding site of an antibody. Antigen-binding sites are referred to as paratopes and they are highly variable; each antibody will have a slightly different paratope. Each paratope will bind to a particular epitope,¹ which is the part of an antigen which actually binds to an antibody. The analogy which is so often used is that of a lock and key. Estimates for the number of different antibodies produced by the human body reach 10 billion, which gives a vast range of paratopes. This, of course, means that there are a vast number of possible epitopes. Antibodies work because of their huge diversity. Billions are produced meaning that vast numbers of different substances can bind to them and can, therefore, be recognised. The definition of an antigen then becomes any substance which the immune system can recognise. In the earlier terminology an antigen was specifically any substance recognised by an antibody; however, we now tend to use the term "antigen" to refer, also, to substances which bind to receptors on cells of the immune system. Antigens tend to be proteins, polysaccharides and, in some cases, lipids. This is because antibodies are generally required to recognise the kind of things which these make up: things like bacterial or viral coats, capsules, cell walls, flagella, fimbrae and toxins. Because these molecules are very large, they may have multiple epitopes. A final, more rounded definition of an antigen, then, is any substance which exhibits one or more epitopes and is, therefore, recognisable by the immune system.

It is worth recognising a distinction, at this point, between antigenicity - the ability to bind to products of the immune response - and immunogenicity - the ability to engender an immune response. Not all antigens are immunogenic, although all immunogenic substances are antigenic, for - if they could not be recognised by the immune system, they could not engender an immune response. This difference is very important. Many of the body's own components have antigens and it is obviously very important that they aren't immunogenic. Unfortunately, in some individuals some of them are, or may become so, and this is what causes autoimmune diseases, such as diabetes - where a self-antigen is recognised and produces an immune response. This is also what is involved in transplant rejection, where a part of another's body is transplanted, recognised as a threat by its antigens, and attacked.

A related problem is allergens which are harmless substances (such as peanuts) which are incorrectly perceived as a threat. They are, then, immunogenic, despite their not being a threat. This occurs as a result of the way paratopes are constructed (more on that later) which is essentially arbitrary. Certainly, the paratope can't tell what it's binding to, it just binds to it. There are mechanisms to subdue the production of low-quality antibodies (which bind poorly to their epitope) and of antibodies which would produce an autoimmune response. However, the construction of paratopes is not guided, the B-cell which produces them is not an intelligence and cannot 'design' an antibody for a particular antigen. It just produces antigens and the immune system relies on the huge variety to recognise as many threats as possible. However, don't get me wrong, as I said above, the impetus of natural selection has furnished us with B-cells which produce antibodies capable of binding to the sort of chemicals which make up (or are produced by) pathogens and there are mechanisms to reduce the risk of an autoimmune response. However, sometimes antibodies will be produced which cause a response to the wrong thing and sometimes things will slip through the net (or, at least, take a long time to be recognised).

There is a lot more to say, and most of it will be said when we turn to the adaptive immune system, to which it is more relevant. Now, however, we need only understand that antibodies bind to antigens. Vast numbers are produced, allowing a vast range of substances to be recognised, allowing a response to be produced to most foreign substances which shouldn't be in the body. There are a number of different responses which can be occasioned, but one of them is the activation of the complement system (which can also be activated by other means). We now return, then, to the complement system.

So, the complement system begins with the activation of C3, which we have already talked about (Inflammation (part 3)). C3 is present in the blood in an inactive state, as we saw earlier, and it needs to be activated as the first stage of the complement system. To reiterate, the complement system is a danger to the body and hence must only be activated in response to a threat. However, it is also a very useful defence. Therefore, the entire basis of the complement system is inactive proteins, which are constantly in circulation, awaiting activation. The genius of the convoluted chain of events I am about to describe is that they ensure that the complement system is only operative when it is required. However, the constant circulation of the components means they can be present in good quantities all over the body, so that the complement system can be readily and quickly activated. The inactive components are harmless and it is their need to be activated that allows them to circulate quite safely until they are needed. The perfect trade-off between a prompt and destructive defence mechanism, and the need to protect the body from its own defences.

C3 is activated by cleavage - the large molecule that is the protein C3 must be split into two smaller proteins - C3a and C3b.

Figure 1.24: A highly simplified graphical representation of C3. This is C3 in its normal, inactivate state. It is a large protein composed of two smaller proteins C3a and C3b. The separation of these from one another is what is achieved by cleavage

These proteins are useful in themselves (see Inflammation (part 3)) - C3b opsonises pathogens and C3a triggers the degranulation of mast cells, such that they release histamine. C3b is also involved in the rest of the cascade and is necessary for bringing about the other effects of the complement system.

So, the trigger for the complement system is the cleavage of C3. This is what all three of the pathways achieve. The classical pathway achieves the cleavage of C3 by producing a particular C3-convertase, which, in this case, is a complex of two proteins - C4b and C2a (the complex of these two is referred to as C4b2a).

Figure 1.25: The C3-convertase C4b2a

There are actually two different molecules which can cleave C3 and activate the complement pathway, and both of them are C3-convertases (this is because C3-convertase is a description - both of them are convertases which cleave C3 - it is not a name). However, the classical pathway makes use of C4b2a to cleave C3. This, then, is what the classical pathway needs to produce, in the first place. The starting point is another complex of proteins called C1-complex. C1-complex is a complex of 5 molecules: 1 × C1q, 2 × C1r and 2 × C1s. This is denoted C1qr²s², which we refer to as C1-complex.

Figure 1.26: C1-complex: C1qr²s²

 
The C1q has good affinity for two of the five different types of antibody (specifically IgM and IgG - more on these later). Recall that antibodies can occasion a number of different responses. The "choice" of response is influenced by which of the five isotypes is involved. Only two are involved in the classical complement pathway. This is noteworthy, as this is one of the ways in which specific, appropriate responses can be produced to specific pathogens. So, C1q binds to antigen:antibody complexes - specifically complexes of antigens and IgM or IgG. C1q can also bind to certain molecules on the surface of some pathogens. In either case, this binding causes a conformational change in C1q, which activates the two C1r molecules - forming a catlytically active molecule referred to as C1r*. This then cleaves and activates the two C1s molecules, forming C1s*.

Figure 1.27: As we've been going along, I've been making self-deprecating remarks about my own diagrams. This isn't just for the purposes of humour or anything like that, I want to stress that they're for illustration only. They're a bit like the tube map - they don't really show what the things they represent actually look like, they use straight lines and bright colours in the hope of providing a useful guide. This time, though, I really need to emphasise this again. I have chosen to leave my diagrams the way they are because I think it is clearer and because I'm bad at drawing. However, C1-complex looks nothing like this. Still, I hope the diagram above gives a flavour for what is going on. The change in the colour of C1q indicates the conformational change in C1q as a result of binding with the antigen:antibody complex. I have chosen to take this as an example and I have chosen to use an IgG antibody purely as an example. As I explained above, there are other possible alternatives

If you're interested in a slightly less stylised and simplified diagram, you may prefer this alternative diagram:

Figure 1.28: An alternative diagram of the same process, showing a more accurate rendering of C1-complex. This is taken from a Wikipedia image and was authored by "Tossh_eng." I have only included the relevant part of a much larger image available here. We will come to the processes shown in the rest of that image in due course

After undergoing this activation, the C1s cleaves another complement component C4. This cleavage produces two smaller molecules - C4a and C4b.

Figure 1.29: Here, C4 is shown in its normal, inactive state, composed of two proteins - C4a and C4b. I have tried to show it moments before it is cleaved by activated C1s. This cleavage, of course, will separate C4a from C4b

C4a, like C3a, is an anaphylatoxin, meaning it stimulates histamine release from mast cells. Meanwhile, C4b binds to the membrane of the pathogen, before binding to our next complement component C2. This causes the activated C1s to cleave C2 into C2a and C2b.

Figure 1.30: Again, I am attempting to show the process moments before the cleavage takes place. You can see that C2 is shown in its inactive state, that it has bound with C4b - which has now bound to the membrane - and that it is about to be cleaved by activated C1s. Note, also, that C4a has now moved away from where this is all taking place

We now have the components for C3-convertase. The C2a binds with the C4b, forming C4b2a.

Figure 1.31: You may recognise the C3-convertase C4b2a from before

C4b2a cleaves C3 into C3a and C3b. We know that C3a is an anaphylatoxin and that C3b opsonises pathogens. C3b also binds with C4b2a, giving C4b2a3b - C5-convertase.

Figure 1.32: This time I have chosen to show unactivated C3 moments before it is cleaved on the left. On the right is the result of this cleavage - C3a goes off to fulfil its role as an anaphylatoxin, while C3b binds with the C3-convertase (C4b2a) to form C5-convertase (C4b2a3b)

C5 is another complement component and - you will be amazed to hear - is cleaved by C5-convertase into C5a and C5b. In another turn-up for the books, C5a is an anaphylatoxin, while C5b is involved in the membrane attack complex (Inflammation (part 3)).

Figure 1.33

C5b binds with C6, forming a complex. This complex then binds with C7, forming a complex which binds with C8. This complex catalyses the polymerisation of C9 and these polymers also bind with this same complex, giving C5b-C6-C7-C8-C9{n}. This is the membrane attack complex. When C7 binds to the C5bC6 complex it alters the configuration of the proteins and exposes a hydrophobic site of the C7 (i.e. a site which is repelled by water). The cell membrane of the pathogens affected by the MAC are composed of phospholipid bilayers which have a hydrophobic tail. They also have a hydrophilic head, which has good affinity for water. Since the hydrophobic tails are repelled by water, they aggregate when exposed to it, forming the lipid bilayer cell membrane. This consists of a two-layered sheet with all of the tails in the centre and all of the heads on the outside. This is highly impermeable to water-soluble molecules and allows the cell to regulate the passage of substances into and out of it. The hydrophobic site of the C7 is able to insert into this bilayer.

When the C8 binds to the complex of C5b, C6 and C7, a hydrophobic area is also exposed, which can also insert into the bilayer. The hydrophobic area of C8 (C8 alpha-gamma - C8 is composed of a complex of C8-beta and C8 alpha-gamma) also polymerises C9. Hydrophobic sites of C9 are similarly exposed and associate with the lipid bilayer. Brilliantly, the C9 has this external face which is hydrophobic and a hydrophilic internal face, allowing water to enter the cell. When enough of these pores are opened, the cell is simply flooded with water and, ultimately, falls apart.

Figure 1.34: A very highly stylised diagram of the MAC, giving a broad idea of how it works and the proteins involved. I would imagine there would be more units of C9 and it should not be presumed that the proteins would join together in this way. This is merely the principle

This, then, is the complement system. This is what it does and how it does it. Well, it's one way it can do what it does. There are two other pathways and we continue onto the alternative pathway. The alternative pathway is based on the properties of C3. C3 is a little unstable in aqueous solutions and this means that it undergoes occasional hydrolysis. Hydrolysis is a simple process where water cleaves chemical bonds. In this case, it is an internal thioester bond that is cleaved. A thioester has the functional group C-S-CO-C, i.e. a radical bonded to a sulfur atom, bonded to a carbon atom, itself double bonded to an oxygen atom and bonded to a radical. The generic structural formula, then, looks like this:

Figure 1.35: The structural formula of a thioester. (Image courtesy of Ben Mills and "Vladsinger," via Wikipedia)

Anyway, this thioester bond is able to be cleaved by hydrolysis, forming C3(H₂O):

Figure 1.36: A very unrealistic representation of C3(H₂O), purely for diagrammatic purposes

This is constantly happening spontaneously, but at a low level. The cleavage brings about a change in the shape of the molecule, which means that it can now bind with a plasma protein called factor B:

Figure 1.37

After this has happened, another protein called factor D can cleave factor B into Ba and Bb. The Bb remains bound to the C3(H₂O), forming C3(H₂O)Bb:

Figure 1.38: The Factor B is cleaved by Factor D

This is a fluid-phase C3-convertase, which is capable of cleaving C3 into C3a and C3b, in order to activate the complement system. Because this is constantly taking place, the process needs to be regulated. The C3(H₂O) and the C3b that is produced by C3-convertase are inactivated by factor H and factor I (examples of complement control proteins), preventing the spontaneous activation of the entire complement system.

However, naturally, if there are pathogens present, the activation of the complement system is desirable. In this case, it is often true that the C3 will be cleaved by hydrolysis caused by a reaction with a hydroxyl or amino group of a cell surface molecule on a pathogen. This will leave C3b bound to the surface. Alternatively, the C3(H₂O)Bb discussed earlier may cleave C3 and produce C3b, which will then bind to the surface of a pathogen. Once the C3b has bound to a pathogen, it is protected from being inactivated by Factor H.

In either case, C3b - now bound to the surface of a pathogen - can bind with factor B, forming C3bB. The factor B can then be cleaved by factor D, leaving us with the C3bBb complex bound to the surface of the pathogen. This is very similar to what happened before:

Figure 1.39

C3bBb is an unstable molecule; however it can bind with factor P (also called properdin), forming C3bBbP:

Figure 1.40

This is a stable C3-convertase, which can cleave C3 and produce more C3b, which binds to the same cell surface. Gradually, the amount of C3b will build up, "decorating" the cell and opsonising it. Many pathogens have no way of preventing this; however, the body's own cells possess complement regulatory proteins, such as CD35, CD46, CD55 and CD59. These inhibit this activity, preventing the body's own cells from activating the complement system. Furthermore, factor H can deactivate C3b on self cells, which is achieved by binding to glycosaminoglycans (GAGs). These are present on self cells, but not present on many pathogens. Therefore, the alternative complement pathway is halted when activated C3b binds to self cells, but not to pathogens.

Now, assuming the C3 convertase C3bBbP is bound to a pathogen, which needs to be eliminated, it will now be able to bind to another molecule of C3b which has been produced. This forms C3bBbC3b (sometimes shortened to C3b₂Bb) - the C5-convertase of the alternative pathway. This cleaves C5 into C5a and C5b:

Figure 1.41: Again I show unactivated C5 moments before it is cleaved on theleft. On the right is the result of this cleavage

C5b can then bind with C6 and, eventually, the MAC will be formed and will be able to destroy pathogens by lysis.

To summarise, then, the alternative pathway is constantly being activated. However, if there are no pathogens around, it is quickly halted by complement control proteins. If, on the other hand, there are pathogens present, the cascade is not halted and the full complement system is initiated and takes effect. Once the pathogens are destroyed, there is nothing to sustain the complement system and it is halted.

Finally, we move on to the lectin pathway, which is very similar to the classical pathway. However, rather than C1-complex, the lectin pathway begins with a protein called mannose-binding lectin (MBL). Alternatively, a group of molecules called ficolins may also initiate this pathway. MBL binds with sugars, such as mannose and glucose, which are present on the surface of many pathogens, including salmonella and listeria. MBL is present in the blood in a complex with MASPs (MBL-associated serine proteases). There are three such MASPs, conveniently named MASP-1, MASP-2 and MASP-3. When MBL, in such a complex, binds with an appropriate sugar (say, for example, mannose), the MASP-1 and MASP-2 are activated and begin to cleave C4 and C2. Thus, we get C4a, C4b, C2a and C2b, whence we get C4b2a, whence we get C3a and C3b, whence we get the rest of the complement system. A similar process occurs with ficolins.

This, then, is the complement system - the second of the two major components of the innate immune system. As I highlighted above, the complement system does "interface" with the adaptive immune system and can be initiated by that system. It is, however, a much older and more primitive defence. Its response is also non-specific and the alternative pathway can be activated "automatically" and is not reliant upon antibodies. The key point to stress, though, is that it does not adapt over the course of the lifetime of an individual

¹Some paratopes can bind to more than one epitope