Phagocytes containing a disease-promoting Toll-like receptor/Nod ligand are present in the brain during demyelinating disease in primates

Abstract

Recent studies claim a central role for Toll-like receptor (TLR) ligands in stimulating autoimmune disease by activation of antigen-presenting cells in the target organ, but it is unclear if and how TLR ligands reach target organs. Most evidence comes from rodent models, and it is uncertain whether this principle holds in primates. Here we identify which cells contain peptidoglycan (PGN) in multiple sclerosis brain and in two nonhuman primate experimental autoimmune encephalomyelitis (EAE) models with different disease courses: acute (rhesus monkey) versus chronic disease (marmoset). Because persistence of TLR ligands in the central nervous system might be consequential for disease progression, we also determined the expression of two major PGN-degrading enzymes, ie, lysozyme and N-acetylmuramyl-l-alanine amidase. Distinct phagocyte subsets, including granulocytes, macrophages, and dendritic cells, contained PGN in the brain and coexpressed the inflammatory cytokine interleukin-12. The number of phagocytes carrying PGN increased in acute and chronic EAE compared with control animals, with the highest number of PGN-containing cells in acute EAE brain. Lytic enzymes were scarcely expressed in monkey and multiple sclerosis brain, favoring PGN persistence. PGN stimulated interleukin-12p70 release by leukocytes from all three primate species. The presence of PGN in the inflamed brain may have major implications because TLR2/Nod ligation potentially promotes inflammation and disease progression.

Figure 1

PGN-containing cells are present in rhesus monkey and marmoset brain. We assessed whether EAE development is associated with elevated numbers of PGN-containing cells in the brain and whether the number of PGN-containing cells differs in brain tissue from rhesus monkeys with acute EAE compared with marmoset monkeys with chronic EAE. Left column, rhesus brain tissue; right column, marmoset brain tissue. Significantly higher numbers of cells in infiltrates of marmoset EAE brain tissue (D) contain PGN when compared with control brain tissue. C: Many PGN-containing infiltrates are present in rhesus EAE brain tissue. Significantly elevated numbers of PGN-containing cells are present in the parenchyma of rhesus (E) but not in marmoset (F) EAE brain tissue, compared with control brain tissue. Many cells in infiltrates of rhesus EAE brain tissue (A) (animal 6) and a modest cell number in infiltrates of marmoset EAE brain tissue (animal 10) contain PGN (B, D). Difference in frequency of PGN-containing cells likely reflects the distinct CNS pathologies in marmoset versus rhesus, with high abundance of neutrophils in rhesus monkeys with EAE. Note in E the maximum number of PGN-containing cells in infiltrates of marmoset monkeys is demonstrated, whereas for rhesus monkey brain tissue the maximum number of PGN-containing cells in infiltrates is depicted in Table 3. Scale bars = 100 μm.

Figure 2

Restricted NAMLAA expression in MS and monkey EAE-affected brain. A and B: In rhesus EAE brain tissue (animal 6), many perivascular infiltrates are present with a moderate to high number of NAMLAA-containing cells. C: In marmoset EAE brain tissue (animal 10) some NAMLAA-containing cells are localized near blood vessels. D: Foamy macrophages in active MS brain lesions (sample 97-160) express NAMLAA. E: Expression of NAMLAA is restricted to a certain subpopulation of foamy macrophages. F: Occasionally, perivascular cells also express NAMLAA in MS brain tissue (sample 00-082). Scale bars = 100 μm.

Figure 3

PGN-containing cells in rhesus monkey and marmoset brain express CD11b. Double staining of brain tissue of rhesus monkeys (A, top) shows that 84 to 87% of PGN-containing cells (green) express CD11b (red). In brain tissue of marmoset monkeys (B, bottom), 88 to 100% of the PGN-containing cells expressed CD11b.

Figure 4

S. aureus-derived sPGN induces IL-12p70 production by marmoset and rhesus monkey APCs. To demonstrate the functional potential of highly purified PGN to induce proinflammatory cytokines, PBMCs from human (n = 1), marmoset (n = 2), or rhesus monkeys (n = 2) were stimulated with different concentrations of sPGN for 21 hours. Supernatants were harvested, and IL-12p70 was determined by enzyme-linked immunosorbent assay.

Figure 5

IL12p40/p70 is expressed by PGN-containing cells in the brain of rhesus and marmoset monkeys. Double staining for PGN and IL-12p40/p70 of both rhesus and marmoset brain tissue (A and B, top and bottom panels, respectively) shows that 95 to 100% of PGN-containing cells (green) express IL-12p40/p70 (red).

Figure 6

Proinflammatory PGN as a co-factor in MS and EAE. This figure models the concepts and findings of this study in relation to key references supporting the numbered items. The four relevant anatomical compartments are shown from top to bottom. Sources of PGN are the normal mucosal flora, notably from the gut (1) and infection (2), mimicked by adjuvant administration in EAE induction (3). The uptake of gut compounds by DC extending protrusions throughout the gut epithelium has been demonstrated recently by several groups. Studies by us and others have demonstrated the presence of PGN in secondary lymphoid organs (4), including within DCs, as well as the proinflammatory action in human spleen. Migration of APC-containing PGN through the circulation (5) has also been documented. Our current and previous studies have confirmed the presence of PGN in the CNS of primates with MS and EAE (6). Finally, it has been demonstrated in vitro that TLR/Nod ligands stimulate microglia in vitro to produce inflammatory cytokines.