Home | Archives | About | Login | Submissions | Notify | Contact | Search
 ES Home > Vol. 5, No. 1 > Art. 16

Copyright © 2001 by The Resilience Alliance

The following is the established format for referencing this article:
Walker, B. 2001. Ecosystems and immune systems: useful analogy or stretching a metaphor? Conservation Ecology 5(1): 16. [online] URL: http://www.consecol.org/vol5/iss1/art16/


Ecosystems and Immune Systems: Useful Analogy or Stretching a Metaphor?

Brian Walker

CSIRO Sustainable Ecosystems

KEY WORDS: biological invasion, ecosystems, immune systems, management, metaphor, sustainability.

Published: June 18, 2001


The notion of using the concept of the immune system as a guide to developing long-term sustainable policies for managing ecosystems is appealing. The immune systems we see today have stood the test of time. They are clearly successful systems for dealing with invasions of dangerous and unwanted microorganisms. Janssen (2001) offers an interpretation of the basic elements of an immune system. He also explores the use of this structure to develop an understanding of how to deal with invasive organisms in ecosystems and, more generally, how to design institutions that will promote long-term sustainability. He extends the notion of invasions beyond the usual one of nonhuman species to include other humans, technological innovations, and even cultural invasions (religions and Big Macs), but then restricts the inquiry to ecosystem management, focusing on the institutions that give rise to successful management practices and on the management of biological invasions.

An obvious and immediate concern when attempting to use the immune system of an organism as a model for managing an ecosystem is the essential difference between the two: organisms are evolved, exquisitely integrated, homeostatic entities whose immune systems are managed subconsciously, whereas ecosystems are open, selected combinations of interacting species that are managed very consciously by humans. We need to explore the consequences of these differences before putting the analogy into practice.


In his account of the immune system, Janssen (2001) raises a number of interesting points about how this system functions. It is a distributed system composed of trillions of interacting cells and molecules of different types, each of which has different receptors that respond to particular classes of pathogens and produce local signals of recognition. Although the immune system retains a memory of successful responses, it must also be able to generate a very large diversity of responses on demand, and one of the main mechanisms for achieving this is a pseudo-random process involving the recombination of DNA.

To put Janssen's proposition into operation, we need to look at the immune system as a model, and then determine if it can be applied to ecosystems. First, with regard to the mechanisms in the immune system itself, the following additional points should be noted.

Once a pathogen/antigen gains entry into the body, it is taken up by antigen-presenting cells such as T-cells and B-cells; some of these are fixed (e.g., in the skin), whereas others are mobile. These cells chop up the pathogen into small pieces and present them on the surface of the cell. T-cells recognize only protein-based antigens, but B-cells recognize anything, depending on their shape. The antigens are presented to the circulating immune system, and the T-cells go to the closest node (i.e., grouping of immune cells) in the lymph system. This process is known as "signal priming." If both the T-cells and the B-cells have been primed (i.e., have recognized the antigen as a nonself signal), the immune system is triggered, and proliferation of T-cells and B-cells begins. The fact that both types of cells must be primed prevents the system from being activated to deal with just one piece of foreign pathogen; repeated activation is required to turn the system on.

The immune system is not perfect. If the invading antigen closely enough resembles, or mimics the behavior of, the host cells, it may not be detected and destroyed. Also, some pathogens have the ability to evade immune surveillance. For example, very small molecules are not immunogenic, and some evade attack by "hiding" in cells (e.g., herpes viruses, intracellular parasites) or by mutating (e.g., malaria). Most chronic infections are caused by pathogens that have evaded immune attack.

Once activated, the system stores some of the dividing cells in a preprimed state as "memory," meaning that they are ready to go, but not yet proliferating. The others keep dividing and secreting antibodies that inactivate the replicating antigens. As Janssen so rightly points out, there is not just one type of antibody; depending on whether the antigen is a virus, bacterium, or parasite, a particular antibody production system is switched on. The replicating immune system cells live for only a few weeks, and, as soon as the antigen invasion stops, the levels of these cells, and therefore also of antibodies, decline.


If the immune system can be used as a model for managing ecosystems, then, based on the discussion above, the key features to include are a distributed system, both stationary and mobile components for detection, a double trigger system for turning it on, different kinds of responses depending on the type of problem, a "memory" component to ensure that successful solutions are retained, the ability to allocate large amounts of resources to solving the problem when needed (an inadequate response results in the evolution of more resistant microbes and makes matters worse for the species as a whole), compartmentalization of effort (moving or returning to sites of infection to maximize the response), and isolating the response (e.g., boils).

Is such a model a useful one for managing ecosystems? Is the analogy Janssen (2001) draws between the production and management of antibodies by the immune system and institutions for managing the ecosystem, i.e., rules and operating procedures, a valid one? Of the examples he suggests, it is easier to use this analogy to examine invasions by organisms (because the situations are very similar conceptually), so I will concentrate on that. When thinking about using the immune system as the basic design for a system to manage invasive species, there are several important points to consider.

Although invasive species enter ecosystems, the entity to be managed is actually a country or nation. The institutional framework for management is always a mixture of state and private arrangements. As Janssen states in his discussion on limitations, people can plan proactively based on other information, not just on the memory stored by the system that deals with invasives. For example, rather than spreading our monitoring efforts equally throughout an ecosystem, we can determine in advance the more likely points of entry and take extra precautions there. The immune system does this, too, to some extent. Natural Killer (NK) cells do not need a memory trigger and act proactively against certain invasions (although invading organisms have evolved evasive mechanisms). Also, as previously noted, the body can localize the immune response and preferentially raise an antibody response at sites where infection occurred previously, such as the nasal membranes or the genital tract. In this regard, the immune response depends on how the antigen is introduced; the same antigen elicits very different responses when it is inhaled and when it is injected through the skin.

In the context of ecological management, the use of both fixed and mobile detection would translate into the use of, for example, both sentry animals and surveillance visits for detecting designated (i.e., already recognized) animal diseases. Most countries already do this for indigenous wildlife, although not for invasions of new plant or animal species. At the time of writing, fire ants had been discovered on a number of properties in the vicinity of the port of Brisbane, Australia; presumably they arrived undetected several years previously. Fire ants are most definitely "nonself" from an Australian perspective, and they do not fall into the category of beneficial nonself, as do most of our crop and all our livestock species. Our surveillance effort was clearly inadequate. Could we have learned something from the immune system model?

After detecting the invasion, the immune system needs a double confirmation and repeated stimulation before it is activated, although the NK cells may already be on the move. It makes a huge effort to minimize the waste that would result from attacking nonproliferating pathogens or self-proteins. Examples of diseases caused by misdirected attacks on harmless, nonself antigens or the body's own tissues include asthma, arthritis, and lupus. Even worse, the immune system risks attacking beneficial or essential antigens, as in the case of food allergies, if dampening mechanisms are not present. As suggested above, mistakes about nonself attacks are unlikely to happen in ecosystem management, because we know which species do not belong. However, we do not know which invasive species are likely to pose a serious problem (i.e., become proliferating pests). For the most efficient response, the immune system relies on a double confirmation before it swings into action, although there are exceptions: the immune system responds to certain pathogens (e.g., some bacteria) after only a single stimulation. The comparative records of successful natural disease control in animals and of invasive species in ecosystems, and their subsequent control costs in countries all over the world, suggest that our institutional mechanisms for detecting and initiating control of invasive species are less effective than those of the immune system in controlling invasive diseases.

This last point raises the important issue of cost, an area in which the use of the immune system as a model might be misleading. The immune system is costly. Activating and operating it take a great deal of energy; only the maintenance of brain functions requires more. The evolved immune systems we see today are the result of many millions of generations of selection. The less successful models developed along the way did not survive, to no one's regret. However, it is difficult, if not impossible, to estimate what proportion of their energy evolving species devoted to their immune systems as opposed to other requirements for survival and reproduction, such as acquiring food, attracting a mate, rearing offspring, etc. This is complicated by the fact that energy expenditure varies greatly over time. The body has no choice but to expend a certain amount of energy on surveillance by producing both fixed and circulating cells. This energy expenditure rises dramatically as soon as infection is detected, and the success of the response depends on the availability and allocation of energy to meet these requirements (Buttgereit et al. 2000).

The success, if not the survival, of a nation in today's world may well be influenced by its ability to detect and manage undesirable weeds and animal pests, but it is also influenced by many other factors requiring allocations of effort and resources, such as the need to remain economically competitive. If the relative significance of managing invasions vs. other requirements were the same in species and nations, it might be feasible to use the immune system as a model. It is likely, however, that the proportion of its energy/budget that the system (nation, region, enterprise) can afford to spend on these activities is different from the amount that evolving species spend on their immune systems. I thought it might be informative to get estimates of these relative costs, in a species and a nation, but discovered that the information is not available (at least not in the time at my disposal). All those approached, however, including immunologists and government bureaucrats, expressed great interest in the numbers and suggested that it would be a very useful project! The allusion to instances of body failure due to very high demands by overactive immune systems (Buttgereit et al. 2000) suggests that, when necessary, species spend massively on controlling infection because they cannot afford not to. The consequences of invasions by unwanted species into nations, dramatic as they have sometimes been, are seldom as crucial; no nation, to my knowledge, has "died" as the result of an invasion by a species other than fellow humans (an interesting variant on nonself). If we extend our discussion of the analogy to ecosystem management in general, then the disappearances of many ancient societies that have been attributed to ecosystem collapse (Tainter 1988) suggest that the significance of the effort devoted to institutions for maintaining ecosystem resilience is of the same order as that for controlling diseases in populations of species.


To use the immune system as an analogy for ecosystem management, we must take into account the relative costs of the two systems, or the relative amounts that we can afford to spend on them. Although the energy cost of the immune system was certainly a factor in the net evolutionary selection process, what sort of system would have evolved if the consequences of its high costs had been more far reaching and shorter term, as they probably are in today's world of economic competition? Would systems of this type perhaps have been more appropriate models? Past evolutionary pressures have bequeathed our immune systems many redundancies, and a system geared only to today's world would be different from the one we have.

This discussion has focused on the mammalian immune system. Lower organisms use less sophisticated mechanisms and have also survived. Frogs, for example, rely more on a barrier system based on skin secretions, and fish have yet another model. The picture that emerges is one of a very complex system that is constantly adapting and adding to its mechanisms in response to new situations. It is not easy to reduce it to a few simple rules, for the success of the system seems bound to multiple interactions—the devil is very much in the details. Perhaps, however, a thorough comparison of several such complex systems—the immune systems of a few, quite different organisms and the pest management systems of a few, quite different countries—might provide some illuminating insights into how we could improve the latter.


Responses to this article are invited. If accepted for publication, your response will be hyperlinked to the article. To submit a comment, follow this link. To read comments already accepted, follow this link.


I am indebted to Chris Hardy for information and a stimulating discussion on immune system processes and to Jenny Langridge for a diligent, if fruitless, search for information on relative expenditures.


Buttgereit, F., G.-R. Burmester, and M. D. Brand. 2000. Bioenergetics of immune functions: fundamental and therapeutic aspects. Immunology Today 21: 192-199.

Janssen, M. A. 2001. An immune system perspective on ecosystem management. Conservation Ecology 5(1): 13 [online] URL: http://www.consecol.org/vol5/iss1/art13.

Tainter, J. A. 1988. The collapse of complex societies. Cambridge University Press, Cambridge, UK.

Address of Correspondent:
Brian Walker
CSIRO Sustainable Ecosystems
P.O. Box 284
Canberra ACT 2601 Australia
Phone: 61-02-6242-1740
Fax: 61-02-6241-1742

Home | Archives | About | Login | Submissions | Notify | Contact | Search