Stress Hormones Bring Birds, Pathogens and Mosquitoes Together.

Do stress hormones, such as corticosterone, enhance bird susceptibility to mosquitoes in ways that enhance rates of co-infection? Does this then enhance pathogen emergence?

Interactions between pathogens are often mediated by the immune system of the host [1]. Two recent experimental papers suggest that interactions between the immune system and corticosterone are important in mediating synergistic interactions between pathogens that enhance rates of co-infection in birds. As part of a long-term study of the effect of the bacterial pathogen Mycoplasma gallisepticum on the house finch Haemorhous mexicanus, Dana Hawley and her team investigated the extent to which stress hormones that are important for energy mobilization and regulation of the immune system are influenced by experimental infection with M. gallisepticum [2]. They found that corticosterone levels increased after house finches were infected with M. gallisepticum, decreased again to pre-infection levels once the infection was cleared, and that individuals with greater disease severity had the highest corticosterone concentrations. Trying to identify factors that influence West Nile virus transmission, Lynn Martin and his team started from the premise that stress hormones might represent a key link between individual levels of infection, population levels of parasite transmission, and zoonotic disease risk. They experimentally manipulated zebra finch Taeniopygia guttata stress hormones by implanting corticosterone-filled silastic tubules and examined subsequent feeding preferences, feeding success, and productivity of mosquito vectors [3]. Despite performing more frequent defensive behaviors against mosquitoes, birds with elevated stress hormone concentrations were approximately twice as likely to be fed on by mosquitoes compared with control birds.

If we combine these apparently unrelated studies with those carried out by Sylvain Gandon’s group on avian malaria in domestic canaries Serinus canaria [4] and with an older experiment by Applegate on avian malaria in house sparrows (Passer domesticus) 5, 6 things become quite interesting. Cornet et al. [4] asked to what extent variable relapse rates of birds with chronic Plasmodium infections would represent a plastic transmission strategy used by the parasite in fluctuating environments. They showed that, in domestic canaries with chronic Plasmodium relictum infections, the parasite responded to the birds being bitten by mosquitoes by increasing its parasitemia and, hence, increasing its transmission probability. By contrast, while various authors have suggested that stress-related hormonal changes would increase hematozoan parasitemia 7, 8, the effect of corticosterone on parasitemia had already been tested experimentally by Applegate, 45 years ago! While attempting to understand what drives the spring increase in P. relictum infection intensity in house sparrows with a chronic infection, Applegate [6] treated the birds with corticosterone; this resulted in an increase in Plasmodium parasitemia: a higher proportion of blood films had demonstrable parasites, and a higher proportion of erythrocytes was infected than in the control group. In a follow-up experiment, Applegate and Beaudoin [6] demonstrated that corticosterone and not gonadotropin caused the spring relapse in avian malaria.

In Figure 1 , we summarize these findings by linking the observation that Mycoplasma gallisepticum infection causes an increase in corticosterone level (Figure 1A) making the bird more atractive to mosquitoes (Figure 1B), and increasing Plasmodium parasitemia directly (Figure 1C) and indirectly (Figure 1D). If the presence of Plasmodium in a bird were to increase the disease severity caused by Mycoplasma gallispeticum, then we would have a positive feedback loop. If this were confirmed experimentally, this would imply that the transmission success of both pathogens is enhanced by prior infection with the other pathogen. Could co-infection have contributed to the emergence and subsequent evoluton of virulence of M. gallisepticum 9, 10? At first glance, co-infection cannot be beneficial to either parasite or to the host; if all else is equal, co-infected hosts should die faster than hosts with a single infection because they are experiencing two sources of pathology. Alternatively, the presence of two pathogens may reduce the immunopathological effects of an overworked immune system and this may allow each pathogen to persist for longer in co-infected hosts. This could enhance either the transmission success of one or both pathogens or the colonization of a new host species by a novel pathogen.

Figure 1

The Role of Corticosterone When a Bird Is Co-infected with the Bacterium Mycoplasma gallisepticum and the Protozoan Plasmodium. Arrows show causal relationships proven experimentally. Infection with M. gallisepticum causes an increase in corticosterone level (A) [2], making the bird more attractive to mosquitoes (B) [3], and causing an increase in Plasmodium parasitemia directly (C) 5, 6 and indirectly (D) [4]. The question mark represents the hypothesis that Plasmodium infection in a bird increases the disease severity caused by M. gallisepticum, which would result in a positive feedback loop.

Prior infection with one pathogen (the malaria parasite Plasmodium) may facilitate later infection with, and transmission of, another completely unrelated pathogen, the bacterium M. gallisepticum. A similar situation is the example of Babesia microti (a malaria-like parasite that infects red blood cells and is the cause of babesiosis) and Borrelia burgdorferi, a bacterium that causes Lyme disease [11]. Diuk-Wasser and colleagues showed how Babesia only successfully establishes in populations in which Lyme disease is already present. More subtly, these examples suggest that prior infections, such as malaria, or parasitic helminths, help facilitate the emergence of pathogens, such as Ebola, Nipah, Hendra, severe acute respiratory syndrome (SARS, or Zika virus. They also underline the previously underexplored potential role that co-infection interacting with host endocrine stress might have in mediating the emergence of novel pathogens. For example, pregnant mothers who are often immunologically stressed and have major upheavals in their endocrine system are more susceptible to mosquito bites [12]. Was this a significant factor in the emergence of Zika virus? To date, the role of endocrine and nutritional processes in the within-host dynamics of pathogens has been essentially ignored. However, it may prove to be an important missing link in many future studies of the emergence and subsequent coevolutionary dynamics of many host–parasite systems.