License to mutate: AID link to transcription elongation

AID mutates the antibody genes and, as a side-effect, a few hundred other genes; but most of the genome is in fact spared. We have found that, even though AID is widely associated to chromatin, its ability to mutate a given gene depends on a specific domain of AID that couples the enzyme to transcription elongation and thereby licenses it for mutagenesis.

Methot et al. Nat. Commun. 9 (1):1248, (2018).

Mechanisms regulating AID activity via subcellular localization.

We have identified mechanisms that regulate AID subcellular localization. AID requires active nuclear import because a mechanism of cytoplasmic retention prevents its diffusion into the nucleus. This mechanism depends on the eEF1A translation factor. The combined action of these mechanisms strike a balance that results in antibody gene diversification, while limiting the off target effects of AID. We have also identified drugs inhibiting the interaction between AID and eEF1A. This is the first pharmacological approach to increase AID function, which can have biotechnological and perhaps clinical applications.

Methot et al. J. Exp. Med. 212, 581–596 (2015) ; Patenaude et al. Nat. Struct. Mol. Biol. 16, 517–527 (2009).

A pathway stabilizing AID protein that can be exploited to decrease its activity.

We discovered that immature AID is protected from degradation by an HSP90/DNAJA1 molecular chaperoning pathway in the cytoplasms. This pathway directly impacts the efficiency of somatic hypermutation and class switching, as well as the oncogenic activity of AID. We exploited these findings to describe pharmacological approaches to reduce AID activity in vitro and in vivo; by either using inhibitors of HSP90 or inhibitors that prevent DNAJA1 farnesylation, which is necessary for the interaction with AID.

Montamat-Sicotte et al. Eur J Immunol 45, 2365-76 (2015) ; Orthwein et al. EMBO J 31, 679-91 (2012) ; Orthwein et al. J. Exp. Med. 207, 2751-65 (2010).

The interplay between AID and DNA repair in antibody responses and cancer.

We showed that the C-terminus of AID links DNA damage by AID to DNA repair by non-homologous end joining during class switch recombination. Our results provide an explanation to an autosomal dominant immunodeficiency syndrome in patients expressing truncated forms of AID, which results highly detrimental for cell proliferation and fitness.

We have also a long interest in uracil DNA N-glycosylase (UNG), the major enzyme recognizing the Uracils made by AID. We have shown that UNG is critical for an efficient antibody response against acute antigenic exposure. A backup pathway mediated by MSH2/6 allows generating switched antibodies in the long term but the kinetics of isotype switching by MSH2/6 are too slow for an effective acute immune response.

Zahn et al. Proc Natl Acad Sci U S A.111, E988-E997 (2014); Cortizas et al. Cortizas et al. J immunol. 191, 5751-63 (2013); Zahn et al. J. Immunol. 190, 5949–5960 (2013).

In collaboration with Dr Ramiro Verdun (U. of Miami) we have also shown that UNG protects the telomeres of B cells from AID-induced damage. In the absence of UNG, MSH2/6-dependent damage leads to sudden telomere loss. We demonstrated that UNG protects the fitness of normal and cancer B cells in vivo, in the germinal center as well as in B cell neoplasms using mouse models. We also find that UNG, despite being a presumed tumor suppressor, is seldom lost in human B cell neoplasms that express AID. Since inhibiting UNG in B cell lymphoma cells that express AID can reduce their proliferation, we posit that UNG could be a novel drug target. Inhibiting UNG would eliminate lymphoma or leukemia clones expressing AID, which are associated with disease progression and drug resistance.

Safavi et al, NAR Cancer (2020); Cortizas et al. J Exp Med. 213, 2459-2472 (2016).

Protein arginine methyl transferases (PRMT) catalyze the post translational methylation of arginine residues in cellular proteins, thereby regulating their function. PRMTs are emerging as critical in cancer and normal biology of different tissues. Their role in B cells is little known.

We showed that PRMT5 is major regulator of B cell biology, being required for B cell development in the bone marrow, B cell activation after the first antigen encounter, and for adequate GC B cell dynamics. Specific PRMT5 deficiency in GC B cells results in altered GC dynamics and prevents the generation of antigen specific antibodies. PRMT5 effects are surely pleiotropic but at least in part, it regulates B cel function via transcriptional regulation and enforcing mRNA splicing fidelity.

Litzler et al. Nature Communications, 2019, 10(1):22.

We have also found that the uracil DNA glycosylase UNG protects the fitness of germinal center B cells, by protecting it from the genotoxic activity of AID, notably at the telomeres. The same effect is afforded to B cell lymphoma and leukemia that express AID. Thus, while UNG, as DNA repair enzyme, can and would normally be considered a tumor suppressor, in the context of a cancer cell that expresses AID it becomes a tumor enabler.

Safavi et al. NAR Cancer 2020, doi: 10.1093/narcan/zcaa019