Dectin-1 ligands produce distinct training phenotypes in human monocytes through differential activation of signaling networks

Cells of the innate immune system retain memory of prior exposures through a process known as innate immune training. β-glucan, a Dectin-1 ligand purified from the Candida albicans cell wall, has been one of the most widely utilized ligands for inducing innate immune training. However, many Dectin-1 ligands exist, and it is not known whether these all produce the same phenotype. Using a well-established in vitro model of innate immune training, we compared two commercially available Dectin-1 agonists, zymosan and depleted zymosan, with the gold standard β-glucan in the literature. We found that depleted zymosan, a β-glucan purified from Saccharomyces cerevisiae cell wall through alkali treatment, produced near identical effects as C. albicans β-glucan. However, untreated zymosan produced a distinct training effect from β-glucans at both the transcript and cytokine level. Training with zymosan diminished, rather than potentiated, induction of cytokines such as TNF and IL-6. Zymosan activated NFκB and AP-1 transcription factors more strongly than β-glucans. The addition of the toll-like receptor (TLR) ligand Pam3CSK4 was sufficient to convert the training effect of β-glucans to a phenotype resembling zymosan. We conclude that differential activation of TLR signaling pathways determines the phenotype of innate immune training induced by Dectin-1 ligands.