KSHV LANA acetylation-selective acidic domain reader sequence mediates virus persistence

Abstract

Viruses modulate biochemical cellular pathways to permit infection. A recently described mechanism mediates selective protein interactions between acidic domain readers and unacetylated, lysine-rich regions, opposite of bromodomain function. Kaposi’s sarcoma (KS)-associated herpesvirus (KSHV) is tightly linked with KS, primary effusion lymphoma, and multicentric Castleman’s disease. KSHV latently infects cells, and its genome persists as a multicopy, extrachromosomal episome. During latency, KSHV expresses a small subset of genes, including the latency-associated nuclear antigen (LANA), which mediates viral episome persistence. Here we show that LANA contains two tandem, partially overlapping, acidic domain sequences homologous to the SET oncoprotein acidic domain reader. This domain selectively interacts with unacetylated p53, as evidenced by reduced LANA interaction after overexpression of CBP, which acetylates p53, or with an acetylation mimicking carboxyl-terminal domain p53 mutant. Conversely, the interaction of LANA with an acetylation-deficient p53 mutant is enhanced. Significantly, KSHV LANA mutants lacking the acidic domain reader sequence are deficient for establishment of latency and persistent infection. This deficiency was confirmed under physiological conditions, on infection of mice with a murine gammaherpesvirus 68 chimera expressing LANA, where the virus was highly deficient in establishing latent infection in germinal center B cells. Therefore, LANA’s acidic domain reader is critical for viral latency. These results implicate an acetylation-dependent mechanism mediating KSHV persistence and expand the role of acidic domain readers.

Antibody epitope profiling of the KSHV LANA protein using VirScan

“…Functional motifs and empirically determined domains of LANA were annotated. These include the Histone 2A/B binding domain (the chromosome binding motif), nuclear localization signals (NLS) [ 32 – 34 ], the SUMO-interacting motif (SIM) [ 33 , 35 ], the Elongin B and C box (BC-box) [ 33 , 36 ], and the proline- and the aspartic-glutamic acid-rich regions within the N-terminal region, along with the glutamine-glutamic acid-rich region and the Cullin 5 box (Cul-Box) within the DNA binding domain (DBD) on the C-terminus [ 33 , 36 ]. Finally, results from a B-cell epitope predictor were included using default parameters (BepiPred-2.0) [ 14 ], where the epitope prediction scores are represented in probabilities mapped on the alignments as a heatmap.…”

Conquering the Host: Determinants of Pathogenesis Learned from Murine Gammaherpesvirus 68

“…mLANA also carries an E3 ligase domain essential for GC B cell expansion (54). Conservation between mLANA and kLANA facilitated studies examining precise domains of kLANA crucial for in vivo function (55)(56)(57) and revealed that kLANA and mLANA are functionally interchangeable for episome maintenance (58,59).…”

Deciphering the PTM codes of the tumor suppressor p53

“…Intriguingly, the same recognition of unacetylated p53 by the acidic domain reader has been further demonstrated during the virus infection process. Latency-associated nuclear antigen, which is encoded by Kaposi’s sarcoma (KS)-associated herpesvirus (KSHV) and contains two tandem acidic domains, selectively interacts with unacetylated p53 and facilitates KSHV latency establishment ( Juillard et al, 2020 ).…”