Scand J Immunol, Volume 54, Issue1/2
"This is first of a series of essays describing the Danger model of immunity. In this first essay, I will describe the model itself and the historical context in which it sits. Later essays will show how the model deals with a wide variety of immunological phenomena, covering aspects in which the immune system seems to do things right (like responses to infections and nonresponses to foetuses) as well as some in which it appears to get it wrong (like the lack of response to most tumours and the occasional presence of responses to self components that result in autoimmune disease). Along the way, we will cover such topics as transplantation immunity, tumour immunity, the need for adjuvants, the maternal/foetal relationship, self tolerance during puberty and metamorphosis, graft versus host disease, autoimmune disease versus ‘useful’ auto–recognition, responses to parasites, and the control of effector class."
First, I need to state, 'with unabashed advocacy', that my beliefs lie firmly in the 'Danger model' ballpark. I do not think, left to my own devices, that I would ever have chosen that precise analogy but I note that Polly's views have moved so that Danger is now closer to 'the induction of host damage' (or 'mess' as I abbreviated it). Much of my opinion can be found in 4 published papers (all referenced in [Ref:1]) and more of it can be easily found on the WWW.
Polly raises the point of 'evolutionary consequences'. At this point, it is worth noting that models formulated in languages evolved dominantly for social interaction, must resort to the transient use of teleologies. It is useful to assign (temporary) purpose to any evolved characteristic and then to go back and iron out these teleologies once we can envisage how natural selection led, by chance evolution, to the final attribute. The glaring teleology of 'self/nonself' concepts may have been to attribute them with the 'purpose' of finding and eliminating foreign organisms. I will resurrect this point a little later.
Whenever we look back in time and consider what has led to the so called 'scientific revolutions' it has nearly always been that a discipline gets swept up a blind alley on a presumption. Revolutions occur because 'obviously obvious' assumptions are made and they are not recognised as such (until they are pointed out and then they become glaringly obvious). The false assumption dams up progress until, at last, (by analogy) the dam can no longer hold and a sudden flood sweeps across the landscape until a normal steady flow of progress is restored.
So, I would like to challenge, absolutely, the security of the 'obviously obvious' assumption that the immune system (henceforth the IS) has evolved to find, kill and eliminate foreign organisms. It is beyond dispute that the system favours the survival of healthy self cells and everything else tends to degenerate. However, simply look at the evolutionarily old alternative complement system. All biological surfaces get C3b deposited on them. Healthy self (species) cells are protected by the inhibition of the aggressive consequences of this deposition. The system encourages the degeneration of ALL biological material unless it is actively protected by complement inhibitors. Antibodies simply amplify complement deposition on those biological surfaces that have previously been associated with disorder (as judged by APCs, particularly dendritic cells). Inhibitor protection for self cells still applies but it is now harder to assert. So that sheds a different light on the SPECIFIC attack on foreign organisms. The same can be said of Tc and Th1 activation. The first identifies SELF cells that have an MHC/peptide signature (Class I) that was previously (closely) associated with uncontrolled cell shutdown (or death). Similarly, Th1 activity just amplifies and accelerates the accumulation of phagocytes at the site of APCs displaying MHC/peptide signatures (Class II) that were previously encountered in an inflammatory context. Again, there is no direct attack on the foreign organisms by Th1 cells. However, they do encourage an inhospitable, local, inflammatory environment.
So that leads us back to 'evolutionary consequences'. Internal cell surveillance (note the teleology! – there are more to come), elective suicide to sanitize the cells contents, the clearance of tissue debris by APCs or phagocytes and the encouragement of regeneration by phagocytes are all ancient processes that were there long, long, long, long before the emergence of lymphocytes. The dominant result of aggressive lymphocyte activation is to enhance the activity of the innate IS in the immediate vicinity of an epitope/paratope interaction. All this is a sort of mess/non–mess discrimination or order/disorder discrimination. Basically, pathogenic organisms are given a hard time by the IS because they gave local tissue and the IS a hard time on their first encounter. But only because the IS recognises some peculiar signature characteristic of that 'hard time', not necessarily an antigen of the organism. However, as the foreign organism is more likely to have strange epitopes than are self tissues (i.e, not previously encountered) it is usually an antigen peculiar to this organism that is selected as the peg on which to hang a future inflammation promoting response.
Now, talking about pathogens, the self/non–self supporters tend to talk in terms of self/pathogen discrimination. In analysing what discrimination pairs are possible it is important to choose clear opposites. The following are admissible:
And so we could go on. Each may contribute to a greater or lesser extent. But, what is inadmissible, I submit, is a crossed pair like self vs pathogen discrimination. As a pathogen is any organism OR SUBSTANCE that leads to tissue damage, then when the self vs non–self supporters use this pair they are, surely, in danger of scoring an own goal in favour of the 'Danger' supporters.
And so that leads to PRRs (pattern recognition receptors). What worries me about this is that simple commensals have no great ability to invade an animal body until it dies. Pathogens, by definition, are honed to invade and mess up the colony of cells that constitute an animal. They evolve sophisticated evasion and invasion techniques. So why are gram negative pathogens so provocative about the way they display LPS? David Lo's team unearthed an enlightening quote. "The gram negative bacteria .. display LPS endotoxin in their walls, and these macromolecules are read by our tissues as the very worst of bad news. When we sense LPS, we are likely to turn on every defence at our disposal; we will bomb, defoliate, blockade, seal off, and destroy all the tissues in the area. All this seems unnecessary, panic–driven. It is, basically, a response to propaganda." (Lewis Thomas, The Lives of a Cell).
PRRs are more likely to be (as Polly Matzinger's team have suggested) a deliberate stimulation of receptors 'designed' to recognise tissue damage (probably cytoplasmic/mitochondrial markers) and provoke such excessive auto–aggression that the system has little choice but to switch off locally to prevent piecemeal auto–destruction. That leaves the pathogen with a glut of debris on which to flourish.
(Ref:1) Cunliffe, J. Med Hypotheses 1999. 52:213–9