Assistant Professor University of Florida, United States
The Mycobacterium bovis BCG vaccine is primarily used to prevent tuberculosis (TB) infections but has wide-ranging immunogenic effects. It is also used to treat certain types of bladder cancer and has been investigated as a general immune stimulant to protect against other diseases such as COVID-19. One of its most notable properties is its ability to induce trained immunity, a memory-like response in innate immune cells such as macrophages that are not traditionally associated with immunological memory. These trained cells generally show increased activation when exposed to a stimulus unrelated to the original training stimulus, identified by responses such as higher secretion of IL-6 and TNF-α, a shift towards glycolytic metabolism, and an accumulation of histone 3 lysine 4 methylation (H3K4me). It has been established that these changes are generated through epigenetic modification, but the mechanisms behind the modifications are still largely unknown.
Deeper understanding of the driving forces behind BCG-induced changes in cell behavior will identify novel ways to leverage its immunogenicity. This will have diverse impacts, including improved vaccine design for TB and non-TB diseases, potential cancer therapies, and new molecular targets to either enhance or suppress immunogenicity. Identifying phosphorylation sites in the proteome will indicate kinases involved, furthering our understanding of the molecular network involved in the trained immunity response.
Materials and Methods: RAW 264.7 murine macrophages were infected with BCG at a multiplicity of infection of 4, then rinsed with PBS after 30 minutes to remove excess bacteria. They were lysed with modified ice-cold RIPA buffer and treated with Benzonase endonuclease. The resulting proteins were denatured with dithiothreitol and alkylated with iodoacetamide, then digested overnight with trypsin. The digestion was quenched with trifluoroacetic acid and desalted with C18 centrifugation. Eluted peptides were enriched for phosphorylation with a TiO2 phosphopeptide enrichment kit and analyzed in a Q Exactive mass spectrometer. MaxQuant was used to search the Mus musculus reference phosphoproteome from UniProt, then data was further processed in Perseus. The identified molecules were not filtered by subcellular localization or function. Significance was determined by a Student’s t-test p-value less than or equal to 0.05.
Results, Conclusions, and Discussions: Phosphorylation of 5 molecules was identified as significantly altered between infected and non-infected macrophages. All 5 proteins are involved in epigenetic regulation or DNA interaction, indicating the high epigenetic specificity of BCG-related cell activity.
Histone lysine deacetylase (HDAC) complexes were found to be differentially phosphorylated. These complexes indiscriminately remove acetyl groups from histones, and deacetylation is linked to decreased gene transcription. There are numerous subunits in each complex, but HDAC1 and HDAC2 are the common catalytic subunits.
Phosphorylation of HDAC1 at S393 and HDAC2 at S394 were reduced in infected cells. This modification enhances their deacetylase activity.
SDS3 phosphorylation at S234 and S236 was increased in infected cells. This specific modification has not been studied but is at the end of the N-terminal disordered domain adjacent to the Sin3-interaction domain (amino acids 188-226). Phosphorylation of S228 in the disordered region increases SDS3 dimerization, sequestering it and preventing formation of the Sin3/HDAC complex. This would likely be reflected by phosphorylation of S234 and S236.
A second finding is the involvement of HMGN1, a highly charged protein with unclear activities.
HMGN1 phosphorylation was increased at S87, corresponding to human S89. This modification potentially increases DNA binding activity, allowing transcribed genes to remain accessible.
Finally, activity of a histone methyltransferase complex (MLL1/KMT2A) was increased. This complex methylates H3K4, which is a characteristic finding of trained immunity.
ASH2L phosphorylation was decreased in infected cells at serine residues 426, 529, and 618. Only phosphorylation of S618 has been reported in phosphoproteomics studies, corresponding to S623 in human ASH2L. This phosphorylation is within the RBBP5-binding domain, which would likely inhibit formation and activity of the MLL1 histone methyltransferase complex. Removing this modification would allow MLL1 to methylate H3K4.
Lowered activity of HDAC catalytic molecules combined with inhibited ability to form HDAC complexes implies that histones would remain acetylated, and DNA would remain accessible. DNA accessibility is also supported by phosphorylation of HMGN1. These findings also support the mechanism of H3K4me addition in trained immunity. As a whole, this data not only supports existing theories of trained immunity, but implicates exact pathways.
