25 November 2009. While much attention has been focused on spinal cord processes in amyotrophic lateral sclerosis (ALS), the innate immune system has been quietly going about its own business in the peripheral nervous system (PNS). So says a paper published online this week in PNAS. First author Isaac Chiu and principal investigator Michael Carroll of Harvard Medical School, along with Tom Maniatis of Harvard University and colleagues, report evidence for activation of macrophages outside the central nervous system (CNS) in a mouse model of ALS expressing mutant human superoxide dismutase 1 (SOD1). However, the implications of the macrophage activity are not yet known.
“We have opened up a whole new cell type to study, one that is affecting the neuron at a different level,” Chiu said. Both the innate and adaptive immune systems have received attention in the ALS field recently, but the majority of work has focused on the CNS (see ARF related news story; Henkel et al., 2009; Appel et al., 2009). The current paper provides another example of crosstalk between the nervous and immune systems in ALS, and dovetails with increasing evidence that ALS begins out in the periphery with the dying back of axons (Fischer et al., 2004). It also corroborates previous work linking circulating macrophages and monocytes to the disease (Zhang et al., 2005; Zhang et al., 2006; Zhang et al., 2009).
Chiu was staining spinal cord sections from SOD1-G93A mice for microglia, CNS resident immune cells, when strong labeling in the ventral nerve roots exiting the spinal cord caught his eye. His curiosity piqued, Chiu characterized the nerve roots further with a panel of antibodies. He found immunoreactivity for immune cell markers including CD68, Iba1, CD11c, CD169, and CD11b. Based on their characteristic rounded shape, Chiu concluded the positively stained cells, which were adjacent to axons, were macrophages. Moving farther away from the spinal cord, he found similar macrophages in the sciatic nerve and in the degenerating nerve bundles of the mutant animals. These macrophages first appeared a few weeks before the weight loss that is the first sign of disease in SOD1-G93A mice, and continued to spread as the disease progressed. Macrophages did not accumulate in non-transgenic mice.
To determine where the PNS macrophages came from, Chiu and colleagues irradiated the mutant mice to destroy their bone marrow, the source of circulating immune cells. They then transplanted bone marrow from GFP-expressing animals, so any immune cells newly derived from the bone marrow would glow green. The PNS macrophages were mostly GFP producing, suggesting they infiltrated the tissue from the bloodstream. In contrast, the majority of CNS microglia did not express GFP, suggesting they were derived from the nervous tissue.
The researchers also examined what might recruit monocytes to leave the bloodstream and become macrophages in the PNS. One option was chemokines, and they did find increased mRNA for monocyte chemoattractant protein-1 (MCP-1), which recruits monocytes, in the SOD1-G93A animals compared to non-transgenic mice and those overexpressing wild-type human SOD1. Complement also recruits monocytes, which prompted Chiu and colleagues to cross the SOD1-G93A mice to animals lacking complement C4, a crucial part of the complement cascade. The double mutants evinced fewer activated macrophages in sciatic nerves than single mutant SOD1-G93A mice, suggesting complement also helps to recruit monocytes.
Peripheral macrophages and CNS microglia express many of the same markers, which might muddy interpretation of results, so the scientists looked for differences between the two populations using flow cytometry. Microglia from the spinal cords of SOD1-G93A mice expressed CD11c, CD86, and CD54, while sciatic nerve macrophages from the same mice showed higher levels of MHC class II. Based on the differences in markers and origins, the authors concluded that the two cell types each participate in distinct, separate immune responses.
“My own feeling is that I do not think they are going to be separate and distinct,” said Stanley Appel of The Methodist Neurological Institute in Houston, Texas. “I am not sure that we have heard the end of the crosstalk between the systems.” Chiu concurred that the current data do not preclude the possibility of some interaction between the macrophages and microglia.
The work indicates that PNS innate immunity deserves attention, but leaves open many questions. For one, what are these macrophages doing? In the CNS, scientists have collected evidence that microglia are protective, pathogenic, or maybe just neutral (see ARF related news story on Gowing et al., 2008). “Are [macrophages in the PNS] a secondary response to axonal death…or are they primary in causing some of the degeneration?” Chiu asked.
One way macrophages could cause damage is by promoting inflammation. Chiu did not find increased levels of mRNA for classic pro-inflammatory markers such as TNF-α or IL-6 in the sciatic nerves of SOD1-G93A mice, but noted that without checking a full panel of pro-inflammatory molecules, the evidence against inflammation is not conclusive. Appel suspects that the macrophages likely do more good than harm. And Ben Barres of Stanford University in Palo Alto, California, wrote in an e-mail to ARF that he thinks the macrophages are likely a clean-up crew, called in to gather the components of degenerated axons.
A related question is, Which comes first, the immune response or the axonal dieback? Chiu is working on blocking macrophage activity to find the answer; if the macrophages cause disease, their absence might be of benefit. In addition, he is working with collaborators to look for evidence of peripheral immune responses in human autopsy samples.
One potential implication of the study, Chiu said, is that many experiments focused on microglia could be confounded by the macrophages that often have the same origin and express the same markers. For example, studies with bone marrow transplants (Beers et al., 2006) and myeloid-specific SOD1 deletions (Boillée et al., 2006) have been widely interpreted to show a role for microglia in ALS, but macrophages are also bone marrow-derived myeloid descendants. “What is absolutely clear is that both systems [CNS microglia and PNS macrophages] are involved,” Appel said. “You cannot say that everything is due to one or another.”
The research shows there is crosstalk between the nervous and immune systems, and future work may show crosstalk between the different immune responses. What is needed now, Chiu suggested, is some crosstalk between neuroscientists and immunologists to sort out the role of immunity in ALS. Barres added, “It will be very important in future studies to assess the potential role of PNS immune activation in disease progression.”—Amber Dance.
Chiu IM, Phatnani H, Kuligowski M, Tapia JC, Carrasco MA, Zhang M, Maniatis T, Carroll MC. Activation of innate and humoral immunity in the peripheral nervous system of ALS transgenic mice. PNAS Early Edition. 2009. Abstract