- Top of page
It's a wormy world. All natural vertebrate populations are subject to infection and re-infection with helminth parasites (Stoll 1947). Even in humans, around one billion people in developing nations are infected by one or several of a range of helminth parasites (Lustigman et al. 2012). Infection by worms is therefore the norm and is reflected in vertebrate immune responses. Thus, there is probably little point in generating an inflammatory response to clear every last worm, with ensuing collateral damage to our own tissue, when rapid re-infection from the environment by another worm is pretty much assured. Instead, the vertebrate immune system modifies its response to worms, controlling (but not always clearing) these infections and at the same time limiting damage to host tissue caused by inflammatory immune responses (Jackson et al. 2009). The immune system, however, has to fight battles on several fronts and, while fighting a prolonged war of attrition against helminth parasites, it also has to protect against periodic invasion by bacteria, where a rapid response to kill invading microbes before they spread is essential (Fig. 1). In this issue of Molecular Ecology, Friberg et al. (2013) ask what effect worm infections have on a host's ability to mount antimicrobial responses.
Figure 1. Helminths generally produce chronic infections that elicit immune responses characterized by both the activation of T helper type 2 (Th2) cells and the production of regulatory responses, such as the cytokines transforming growth factor beta (TGF-β) and interleukin-10 (IL-10) and regulatory T helper cells (TREG). In combination, this immunological phenotype is often called a ‘modified’ Th2 response. Bacterial infections are recognized by pattern recognition receptors such as the Toll-like receptors (TLRs) that are able to detect bacterial molecules such as lipopolysaccharides (by TLR-4), flagellins (by TLR-5) and unmethylated CpG (by TLR-9) and generate a rapid inflammatory response [characterized by tumour necrosis factor (TNF)-α production], particularly at the site of infection. As indicated by the dashed line, these contrasting responses have the potential to interact, especially in animals that spend much of their life harbouring chronic helminth infections that may have systemic anti-inflammatory effects.
Download figure to PowerPoint
Friberg et al. (2013) begin their study in laboratory mice, where the immunological reagents exist to make mechanistic insights into helminth/bacterial co-infections in a controlled environment. They focus on Toll-like receptors (TLRs), which provide a first line of immune defence by recognizing microbial infection, triggering a rapid inflammatory response at the site of infection and can also help develop subsequent, acquired responses (Medzhitov 2001). Infection with the nematode Heligmosomoides polygyrus modified these responses but in a counter-intuitive fashion. One might expect worm infection to reduce TLR-mediated responses, consistent with a role of chronic helminth infections in down-regulating inflammatory responses to limit tissue damage. Instead, TLR-2, TLR-4 and TLR-9-mediated cytokine responses tended to be elevated in worm infections, including that of tumour necrosis factor (TNF)-α, which is a potent pro-inflammatory cytokine. Potential explanations are that helminth infection changes the bacterial community composition of the gut flora or that helminth infection damages the gut wall and forces the host to defend itself against gut bacteria.
Examining systemic effects of infection for immune responses within a whole animal (as opposed to cell culture) is challenging because immune responses are dynamic and variable. Having given an infective dose of nematodes, both the number of worms and the immune responses to these worms change through time. Friberg et al. (2013) also find that the dynamics of TLR-mediated responses differ somewhat depending on the mouse strain used and on whether worm infections are delivered as a single pulse or split across multiple small doses (a so-called ‘trickle’ infection), which may be more representation of infection in the field (Paterson et al. 2008). Varying the mode of infection and the host genetic background is important because the approach taken by much of the immunological literature is to try to eliminate variation by infecting a single host genotype with a single infective dose, which makes it difficult to generalize immunological results to the field.
If examining immune responses in laboratory animals is difficult, what are the prospects for examining immune responses in a natural population? From a mechanistic perspective, the prospects would seem hopeless. The (already complex) dynamics of an immune response through time will be compounded by immunological variation among hosts in their pathogen exposure, age, nutrition and so forth that are found in natural populations. However, from an ecological perspective, quantifying the magnitude of variation in TLR-mediated responses and identifying associations between immune responses and measurable variables such as macroparasite burden help to define what groups of individuals are most vulnerable to bacterial infection (Pedersen & Babayan 2011). Friberg et al. (2013) therefore performed an immuno-epidemiological study on a natural population of wood mice (Apodemus sylvaticus) and found significant associations between H. polygyrus and ectoparasites on TLR2-mediated TNF-α production. The result, however, is a complex one in that the direction of the effect switched sharply between the 2 years in which the population was sampled. The cause of this switch is unclear, but may reflect qualitatively or quantitatively different pathogen exposures between the 2 years. Certainly, these data highlight that immune responses in the field are context-dependent, even if the nature of that context remains elusive.
Friberg et al.'s (2013) results, and a small but growing body of similar studies (Abolins et al. 2011; Boysen et al. 2011; Jackson et al. 2011), show both the potential and the difficulty in analysing immunological responses in the natural environment. In an earlier review in Molecular Ecology, Pedersen & Babayan (2011) make a compelling case to study the ecological context of immune responses in the wild. Ultimately, immune responses should enhance individual fitness, but it is not necessarily clear what constitutes a ‘good’ immune response. In particular, individuals in the wild may not have the luxury of deciding whether to tolerate a chronic worm infection or to clear an acute bacterial infection; the prevalence of co-infection means that often they have to do both, often under conditions of nutritional stress (Pedersen & Babayan 2011). We are now only starting to bridge the gap between laboratory immunology and the ecology of natural populations. The technical barriers are readily apparent, including a lack of immunological reagents for non-model species and the difficulty of sampling individuals and of measuring pathogen infection. Probably, a greater barrier, however, is linguistic; immunologists and ecologists speak very different languages. There is a steep learning curve for any ecologist trying to pick out what immune parameters they should try to measure in their system and how to interpret these. Equally, immunologists are unused to dealing with variation among individuals as anything other than a nuisance. However, there are great rewards for both ecologists and immunologists in understanding the sources of immunological variation in the field. It is therefore to be hoped that a common language can be developed between laboratory immunology and field ecology and that Molecular Ecology will be at the forefront of bringing these two fields together.