Thursday, March 29, 2007

The Hygiene Hypothesis

The Hygiene Hypothesis: Probably the one idea in Immunology that annoys me the most. It bugs me, drives me nuts, nevertheless it has held and persevered for nearly twenty years. So there must be something to it, and I must shelve my prejudices and look into it.

Let me get my complaints out of the way first, so that you can treat my point of view with the necessary skepticism. On the other hand, maybe I am good person to look into this idea since I am the ultimate skeptic and totally disinclined to believe. Hopefully not too prejudiced, but I’ll leave you to be the judge of that. My biggest peeve with the Hygiene Hypothesis is irrational, it is the name- it is pejorative. There is no nice way to think of it. It is meant to explain the increased incidence of allergic diseases in countries with a “Western” lifestyle, and postulates that decreasing family size and improved hygiene in these countries has led to the increased prevalence of allergic disease. Why? How? Let me try and explain.

At the core of the ideas that eventually constitute the Hygiene Hypothesis and attendant years of research is a set of apparently contradictory observations, best exemplified by this one: farmers have a much lower incidence of hay fever than the rest of the population. Intuitively perhaps, this seems odd. Farmers and people who live on farms are exposed to pollen and other things that give rise to hay fever far more than people who live in cities, so if the tendency to get hay fever was evenly distributed in the population, there should be a similar percentage of farmers who get it as, say, bus drivers. This is not the case, however and the whys and hows have been hotly debated these many years.

There is a really important immunological idea that needs explaining at this point, it is the phenomenon of Th1 and Th2 T cells (which expand as Type 1 helper T cells and type 2 helper T cells, but that is not really necessary to know). Th1 and Th2 T cells as a concept have been around for donkey’s years, and anyone who has ever worked in immunology has brushed up against it. It describes two different subsets of T cells (ha, you would never have figured that one out) and it is a separation into subsets based on function. Specifically, that function is the production of certain small molecules called cytokines.

Cytokines are the e-mail messages of the immune system. Whenever a cell is activated, and this applies to nearly all immune cells, it sends out the message that it is activated by dumping cytokines into its milieu. The particular cytokine- or set of cytokines since immune cells can often produce a whole bunch of cytokines at the same time- dumped out depends on the way in which the cell has become activated. So, if you read about something horribly unjust on the news, you’re activated to be indignant and send indignant activist e-mail to your friends. Inflammatory even. On the other hand, if you’re really annoyed with a colleague’s carping or unprofessional behaviour, you send insistent annoyed e-mail to friends or family. You are irritated in fact. And then a friend sends you seven increasingly frantic messages about the date that didn’t call back and you see a need and send out an instant, powerful, soothing message. A dampening, or regulatory message. These messages have effects on other people, some intended and others unanticipated. These messages also affect people who are near and those who are far; those who are directly involved or hapless bystanders. Cytokines send out messages that initiate war upon an enemy infection, or those that irritate one’s own body to the point of itching or sneezing, or messages of cessation and calm. As required, and as far as blood can circulate. Sometimes they are just sparks, and other things form the flame, of fever for example, but that’s for another day. You get the picture.

Th cells are really good cytokine producers. Th1 cells make so-called Th1 cytokines, which are generally more of the sound-the-charge type. Th2 cells make, well, Th2 cytokines that belong more to the irritant family. The view has been long held that Th1 cytokines are needed to fight infections, and when they go out of control, they cause autoimmune diseases, such as Crohn’s disease. Th2 cytokines on the other hand are thought to be the ones that cause allergic types of reactions, but they are very necessary to clear parasitic infections, like worms living in your gut. Of course the longer something is studied the more ambiguous it becomes and that is not the whole story. Anyway, the key thing is that Th1 cytokines can block the production of Th2 cytokines and vice versa, which is why I have been discussing them as opposites. Depending on the order or amount (or both) in which they are made, either Th1 or Th2 cytokines end up dominating an immune response.

Back to the Hygiene Hypothesis. The idea builds up as follows:
Th2 cytokines are among the primary causes (actually effectors) of allergic disease,
Th1 cytokines suppress the production of Th2 cytokines,
Various microbes in the air, soil, water and environment cause the production of Th1 cytokines
Clean environments contain less microbial “challenges”, therefore less Th1 cytokines are made
Ergo, the Th2 cytokines run amok and you have a higher incidence of allergic disease.

There is a lot of evidence in favour of the hypothesis and a fair amount against it. The interesting thing about the Hygiene Hypothesis is that most of the data for or against comes from epidemiological studies. Epidemiology is the statistical study of diseases in populations, and is by its nature observational while Immunology is the mechanistic study of disease and is primarily an experimental science. The reason for this strong reliance on epidemiological methods, in my opinion, is that it has been tremendously hard to show experimentally that normal every day environmental stimuli induce Th1 cytokines. Exceptions exist, such a group that has collected stable dust in Switzerland and is going to use that to determine if there is any immunologic response to it. (Stable dust! There’s also a study that compares the effects of cats and dogs as opposed to farm animals during pregnancy.) Most of the data come, however, from observations of large cohorts of people in different environments.

Some of these studies have come out with really cool results. One idea that seems to have a lot of validity is that persistent, heavy worm infections protect you from developing asthma, or more precisely atopic disease, which is more or less allergic skin irritations. This is graded however, and you need to have a really massive worm infection to get this protection. Heavy but not massive worm infections do not really protect, but they do not harm (as far as asthma is concerned anyway) and light worm infections, as might be found under hygienic conditions, actually mean you may get more asthma. Worms induce strong Th2 responses, so this may seem like we’re really on to something. Here’s the rub: the same societies that have a higher incidence of Th2-type allergic diseases also have a higher incidence of Th1-mediated autoimmune diseases like Crohn’s disease.

There is a lot more data for or against, but this example is clear and well-characterized, so I'll just stick to the case of the worms.

So Th1 and Th2 do not provide the entire explanation. Here’s one that I found a lot more satisfying. Take the case of massive worm infections; a person with a massive worm infection has usually mounted an intensely strong immune response to it (Th2, but that isn’t really material). The body has very effective mechanisms in place to ensure precisely that immune responses don’t run amok and eventually destroy everything in their paths. These mechanisms, dampening or regulatory mechanisms, kick in when there is a really strong immune response and turn it off irrespective of the nature or outcome of the immune response. So highly worm-infected people have probably turned down their immune response quite a bit, and since allergy is an immune response, it is turned down as well. Therefore people who have much less worm infections have revved up and ready immune systems and may therefore be more prone to develop allergies or autoimmune diseases. This mechanism feels very plausible to me. Does that mean I have been converted to the Hygiene Hypothesis? Not entirely, and these are my reasons.

Allergic diseases abound in less hygienic environments, South Asians and Africans do get asthma, eczema etc. Of course, the hygiene hypothesis doesn’t claim to explain all asthma incidence, merely increased incidence in certain societies.

The other objection is a technical one. Most of the epidemiological studies use “atopy” or skin irritation (oversimplified, but) as their readout since asthma may take a long time to come up as actual wheezing symptomatic asthma. And asthma patients are usually heavily medicated, so all observation are much more difficult to interpret. However, atopy is not asthma, and may not even be a prelude to asthma, so in a way, they are all using apples to measure orange growth. There are some studies that go much deeper though, but in general, with epidemiological data, one need many many numbers of people and much statistical significance before one can have confidence in the data. Time will tell.

This is linked to the one above: For every study that has a certain result, there is one that has the opposite. That worries me, because I wonder if we are not inventing a correlation when there isn’t one?

And one can go on, genetics, economics etc. I think the crux of the matter is that it is not as simple as the original Th1 vs. Th2 idea, nor is it an explanation of all relevant phenomena. However the Hygiene Hypothesis does seem to have some validity and one needs to wait and see. And change the name!

Papers read (All reviews I’m afraid):

Yazdanbaksh M., et al Allergy, Parasites and the Hygiene Hypothesis. 2002. Science. Volume 296. p 490.

Vercelli D., Mechanisms of the Hygiene Hypothesis-Molecular or Otherwise. 2006. Current Opinion in Immunology. Volume 18. p 733.

Liu AH., and Leung DYM., Renaissance of the Hygiene Hypothesis. 2006. Journal of Allergy and Clinical Immunology. Volume 117. p 1063.

Ramsey CD., and Celedon JC., The Hygiene Hypothesis and Asthma. 2005. Current Opinion in Pulmonary Medicine. Volume 11. p 14.

Tuesday, March 6, 2007

"The Three Es" Hypothesis of Tumor Immune Surveillance

Cancer Immunoediting: The three Es Hypothesis

Okay, here is post number two, and I am truly sorry for the delay, I have been working (yes really!). Also, in response to comments from dearly valued readers, I am going to try and make this less technical. Which should be easy, considering that I am going to try and explain the concept I read in an amazing review. So, no data parsing. Whew.

Cancer and the immune system have been co-existing for centuries, which is one of the most remarkable-and challenging-aspects of cancer biology. Basically, cancers occur when some cells in the body transform (that is the technical term) and become malignant, upon which they are no longer subject to the basic social rules that govern other, normal or untransformed, cells. Such as, stick to your own tissues, stop growing when told to, do not invade the rest of the body, do not divert the body’s blood vessels to give yourself nourishment and so on. Transformation of normal cells into cancerous ones can occur for a variety of reasons, such as radiation, carcinogenic chemicals, infection with an “oncogenic” virus that specializes in transformation of cells or a random, spontaneous mutation that results in a transformed cell with its attendant anti-social behaviour. This is really simplified and I but brush the surface, but you get the picture: a random event, often a mutation in the body’s own cells can lead to cancer, as both radiation and many carcinogenic chemicals cause cancer by causing mutations.

Here’s the rub: mutations occur frequently. Frequently! When cells divide, they duplicate their DNA and make mistakes while doing so, resulting in mutations. DNA damaging agents, like chemicals and radiation, cause hundreds of mutations, many of which are not productive, in that the cell’s DNA repair machinery kicks in and fixes it (That awesome machinery is the subject of a whole new discussion some day, immunology or not). Even so, transforming events are not rare, and that means we should get cancers more often that we do, and the fact that we don’t has led to much speculation on why. One of the earliest hypotheses on the role of the immune system in cancer, the “immunosurveillance” hypothesis proposed that, in immunocompetent hosts, the immunes system is patrolling the body for sign of transformed cells and eliminates them upon detection, and its only rare ones that get free to cause cancer.

Side note: Immunology, more than other fields of Biology, has the habit of coining terms that are more than likely to send you spell check into a hopeless tizzy. It’s a simple strategy really, just add the prefix immuno- to any English word and you have a whole new term. Immunosurveillance therefore means “surveillance by the immune system” and immunocompetent means “in possession of a competent immune system”.

The biggest problem with the idea of immunosurveillance is that it is, in the words of its proponent Lewis Thomas, “it cannot be shown to exist in experimental animals”. To quote the authors of the review I refer to below, “…how could such an invisible process be experimentally revealed?” If the immune system does actually patrol at steady state and eliminate transformed cells, how do you demonstrate it? The short answer is, using immunodeficient (see!) mice, made possible by sophisticated modern genetic techniques. These immunodeficient mice developed tumors much more frequently than their immunocompetent counterparts. In the interests of full disclosure, there was one study that directly contradicted this one but was then refuted by four groups. Anyway, so Robert Schreiber’s group found evidence for immunosurveillance.

The “three Es hypothesis” comes from this idea, and has been proposed by Schreiber and colleagues. They extend the concept of immunosurveillance to one of immunoediting, which is not the most intuitive title, but bear with me. The idea is that immunosurveillance is one phase of a more comprehensive process-immunoediting-which can be broken up into three component phases: elimination, equilibrium and escape. In the elimination phase, immune cells recognize and eliminate transformed cells. Generally, this is sufficient. After, or simultaneous with, the elimination phase, is the equilibrium phase in which the tumor cells and the immune system exist in an equilibrium of inaction: the tumor doesn’t grow and the immune system doesn’t attack it. This can continue for years, and the individual remains cancer free. Then, some tumors escape. This last phase occurs when a tumor mutates sufficiently to evade elimination by the immune system and grows out. The interesting thing that Schreiber and colleagues propose is that the cues the tumor receives during the elimination and equilibrium phases drive the evolution of escapes: that if the immune system fires bullets, tumor cells learn to make Kevlar and escape. The result is cancer.

This was a really good paper to read, and it is a really interesting idea. There is good data to support the existence of the elimination and escape phases, but the equilibrium phase remains hypothetical and will be hard to demonstrate. The basic idea is logical; at least it seems so to me, neat and believable. However, it unlikely to be the whole story. The hugest caveat with studying tumor immunology is the fact that tumor cells or basically ones own cells that have gone berserk. The immune system has extensive, elaborate machinery in place to ensure precisely that it doesn’t attack its own host body. How then is it supposed to tell that tumor cells are foreign? This is the million-dollar question in tumor immunology and for all the promise that tumor immunotherapy has- after all what could be better than using one’s own cells to combat cancer? -It is unanswered.

Paper referred to:

Dunn G.P., et al. The Immunobiology of Cancer Immunosurveillance and Immunoediting. Immunity. 2004. Volume 21, p137-148.