HMS Beagle Letter

Original web address (no longer archived):

Here are selected Letters to the Editor received through feedback.

(Updated August 15, 1997 • Issue 14)

On the Immunologic Tolerance Debate

I have enjoyed reading the self/tolerance debate so far. I disagree that there is an immune system, for the following reasons.

Approximately 700 million years ago, an explosion in the variety of metazoan body forms – a point that doesn't receive much emphasis – coincided with the acquisition of communicating junctions (gap junctions and plasmodesmata). Imagine yourself to be a primitive metazoan. You have recently learned to extend your body plan beyond the cell membrane to encompass a small population of communicating cells. What are your individual cells likely to require of each other? You would hope they would detach themselves when they fall sick and try to trash all their contents should the damage be irreparable; note the hypersensitivity response in plants. You would want remaining cells to replace lost cells, by replication. You would want the remaining cells to gently absorb apoptotic cells and aggressively absorb spilled cytoplasmic messes whenever apoptosis has failed to protect; the latter was provoked by some lethal agent. You would want to extrude or attack any cellular material that is not in junctional communication. In animals, gap junction connection is the most protective, apart, i.e., from a synctium, so note giant multinucleate cells and the synctiotrophoblast.

Well, nothing much has changed in mammals. It is just more sophisticated. This leads to a definition of the grand plan – the raison d'être – of the now ex–immune, morphostatic system. "Anything is welcome in the zygote–derived colony provided it doesn't make a mess or get in the inter– or intracellular way. Preferably, it should be communicating."

Pathogens, by definition, make a mess. The adaptive immune system classifies everything it meets into messy or tidy. Apoptosis is tidy death. Messy death – the spillage of cytoplasm is probably its signature – leads to phagocyte activation, inflammation, and a rejecting adaptive immune response. The adaptive immune system (with its immune networks) allows the baseline phagocytic system to be dampened down, keeping all but a few sentinel phagocytes (APCs) out of the tissues, and vigorously ramps it up on demand. This downregulation of the basal phagocytic system gives the illusion that the adaptive immune system is primarily designed to combat infection, for immune deficiency leads to infective crises.

A protracted sequence of cell lysis events, in the absence of an antigenic caricature that will bring the crisis to a conclusion, will eventually expand a tiny population of self reactive T cells (previously mopped up by widespread physiological apoptosis) into a clonally expanding army of self–reactive T lymphocytes. The only checks on the evolution of self–reactive lymphocytes are the effectiveness of clonal deletion in the thymus, the prior (peripheral) mopping up of self–reactive lymphocytes into tolerance, and quickly focusing on some unusual caricature that will preferentially point aggressive attention at the sick cells.

Rod Langman asked about evidence of new (adult) antigens being rejected. I presume he knows of the experiments on embryonic pituitaries (I can't, without a search, quote site and verse). Animals were hypothesectomised in embryo, deep frozen (note added afterwards: I was wrong! gland transplanted to a holding recipient), then reimplanted in the appropriate syngeneic mature animal, with suitable mock controls showing it was not just an inflammatory reaction. These grafts were promptly rejected.

To suggest that (adaptive immunity–enhanced) autorejection does not occur is the most blinkered comment I have heard. Yes, the frequency of circulating autoantibodies might be relatively low, but not the frequency of cell–mediated autorejection. The seronegative arthritides with all their component disorders (e.g., apthous ulcers, acneiform lesions, iritis, etc.) are all dominantly autorejective disorders. CMI reactivity to myelin basic protein occurs after strokes and brain trauma. Dressler's syndrome is the occasional severe autoimmune sequel to a "massive heart attack" And in burns they are "bursting out of the epithelium". If they were not, I would not have had to postulate on the presence of a fail–safe cut off to the macrophage aggression induced by adaptive immunity. All of this is amplified in two impending articles.

The suggestion that Medawar and Burnet were "wrong" is like saying Newton was wrong. Yes, of course we will all be "wrong" by a subsequent generation's yardstick. But that is the nature of broad paradigms. They are, at best, an approximation and are, at most, useful (current) working analogies. They are "religious beliefs" that should be held only while observational anomalies don't shout that their time is up.

I’m happy with self/nonself discrimination. However, it is not the process envisaged by the lymphocyte brigade. The adaptive immune system cares not one jot about self/nonself discrimination. But since its driving force is influenced by it, it ends up seeming rather the same – but it isn't. I can't go along with the statement that there is no S/NS discrimination in plants and primitive vertebrates. In a forest, tree roots often fuse. This is rarest between different species, more common among different individuals of the same species, and most common in the roots of the same individual. Flowering plants exhibit an exquisitely sophisticated S/NS discrimination in rejecting their own pollen. Many invertebrates have demonstrated a self (individual)/NS discrimination. Mixed mammal embryos cells sort according to tissue rather than species. Sperm and egg fuse according to species. Tnk cells come the closest to a highly specific S(individual)/NS discrimination. Recognition (particularly, for our interest, cell to cell recognition) is a fundamental biological phenomenon.

In the presence of an adaptive immune system, even a tiny number of leaking cells will quickly provoke an aggressive immune response when they sport strange epitopes for which the appropriate lymphocytes have not been previously mopped up into tolerance. Because pathogens need to use mimicry of identity to gain advantage of their hosts, they inevitably pump prime aggressive allo–responses, leading to another illusion that the adaptive immune system is particularly interested in nonself.

There is more detail in my two pending articles (one about to be published, the other accepted for publication in principle, provided I expand the preamble). I am happy to let any interested participants see these provided that the journal's copyright is respected.

Jamie Cunliffe

July 15, 1997