If you caught a cold this past summer, it’s possible that the culprit was the adenovirus. Data from the Seattle Flu Study indicated that King County saw an uptick in adenovirus cases this summer compared to previous summers. Among other symptoms, the adenovirus can cause pink eye and respiratory illnesses in people of all ages.
In a new study published in PLOS Pathogens, researchers from Fred Hutch and the University of Washington report that an adenovirus histone-like protein represses the host cell’s innate immune response to promote infection. Edward Arnold, a PhD candidate in the Avgousti lab, led the study.
“This work characterizes the interaction of a viral histone-like protein, adenovirus protein VII, with a key inflammatory regulatory molecule, HMGB1,” said Dr. Daphne Avgousti, an Assistant Professor in the Human Biology Division at Fred Hutch and the senior author of the study. To compact their genomes, adenoviruses encode a histone-like protein called protein VII. During infection of the host cell, protein VII interacts with the HMGB1 (High mobility group box 1) protein. HMGB1 plays important roles both inside and outside the cell: it functions as a transcription co-factor inside the cell, and as an alarmin to signal infection outside the cell. “Because HMGB1 is a critical player in inflammation, we sought to define the interaction with protein VII to determine how and why the virus targets this protein,” explained Dr. Avgousti.
HMGB1 consists of three domains: the A-box, B-box, and the C-terminal tail. To determine which of these domain binds to protein VII, the researchers from the Avgousti lab collaborated with Dr. Monica Guo’s lab at the University of Washington to perform a bacterial two-hybrid analysis. In this assay, the interaction between protein VII and HMGB1 activates the reporter gene, resulting in red bacterial colonies. The data revealed that protein VII directly binds to the A-box of HMGB1.
The researchers then generated human cell lines expressing different domains of HMGB1 to interrogate the interaction of protein VII and HMGB1 during infection. When they infected the cells with human adenovirus type 5, they observed that the different constructs resulted in three different localization patterns: a diffusely nuclear pattern, ring-like pattern and a punctate pattern. The full-length HMGB1 construct, for example, had an even distribution of all three patterns, but the constructs containing the A-box had an overabundance of the ring-like and punctate patterns. Moreover, they found that the changes in the localization patterns resulted from differences in binding chromatin.
Given that the release of HMGB1 from cells is critical for extracellular inflammatory signaling, the authors hypothesized that protein VII might make HMGB1 insoluble to prevent its release. They used mild detergent treatment to separate cells into soluble and insoluble fractions. The soluble fraction contains nuclear and cytoplasmic proteins, while the insoluble fraction is made up of chromatin-bound proteins and histones. They found that during infection, protein VII is present in both the soluble and insoluble fractions, but that this distribution is not random: the mature form of protein VII was predominantly soluble, whereas the precursor form was mostly insoluble.
The researchers also applied this fractionation method to the HMGB1 constructs, and found that all HMGB1 constructs are soluble in the absence of infection. During infection, however, full-length HMGB1 and constructs containing the A-box were insoluble, while the constructs lacking the A-box were soluble. This indicated that protein VII binds the A-box of HMGB1 during infection to make HMGB1 insoluble, thereby sequestering HMGB1 in the host chromatin and preventing its release.
The authors again utilized the bacterial two-hybrid analysis to determine that the interaction between protein VII and HMGB1 is not dependent on DNA binding. They also found that in human cell lines, protein VII mislocalizes and sequesters HMGB1 in chromatin independent of DNA binding, as well as post-translational modifications on protein VII.
The authors hypothesized that protein VII might alter HMGB1’s role as a transcription co-factor to repress the host’s immune response. The authors induced expression of protein VII in human cells, then treated the cells with interferons to induce the expression of downstream interferon-stimulated genes. They found that although protein VII did not affect the downstream expression of interferon-stimulated genes, it significantly decreased the expression of interferon beta in the presence of HMGB1. The loss of HMGB1 was able to partially, but not fully, rescue interferon beta expression, suggesting that other factors, such as HMGB2, may be implicated in this process.
“We were surprised to find that protein VII inhibits the induction of interferon transcription in an HMGB1-dependent manner. This is surprising because HMGB1 is well-known to be an extracellular signaling molecule, but its function in the nucleus and on transcription is more enigmatic,” said Dr. Avgousti. “The next steps are to define whether this inhibition of interferon expression is conserved across different species of adenovirus, that is, when did it evolve, and how protein VII selectively targets interferon and not other genes.”
This work was supported by the National Institutes of Health, University of Washington, and Fred Hutchinson Cancer Center.
The Fred Hutch/University of Washington/Seattle Children’s Cancer Consortium member Dr. Daphne Avgousti contributed to this work.
Arnold EA, Kaai RJ, Leung K, Brinkley MR, Kelnhofer-Millevolte LE, Guo MS, Avgousti DC. 2023. Adenovirus protein VII binds the A-box of HMGB1 to repress interferon responses. PLoS Pathog. 19(9):e1011633.