Human cytomegalovirus infection modulates DNA base excision repair in fibroblast cells

DNA Repair Mechanisms are Activated in Circulating Lymphocytes of Hospitalized Covid-19 Patients

Purpose

Reactive oxygen species (ROS) are an important part of the inflammatory response during infection but can also promote DNA damage. Due to the sustained inflammation in severe Covid-19, we hypothesized that hospitalized Covid-19 patients would be characterized by increased levels of oxidative DNA damage and dysregulation of the DNA repair machinery.

Patients and Methods

Levels of the oxidative DNA lesion 8-oxoG and levels of base excision repair (BER) proteins were measured in peripheral blood mononuclear cells (PBMC) from patients (8-oxoG, n = 22; BER, n = 17) and healthy controls (n = 10) (Cohort 1). Gene expression related to DNA repair was investigated in two independent cohorts of hospitalized Covid-19 patients (Cohort 1; 15 patents and 5 controls, Cohort 2; 15 patients and 6 controls), and by publicly available datasets.

Results

Patients and healthy controls showed comparable amounts of oxidative DNA damage as assessed by 8-oxoG while levels of several BER proteins were increased in Covid-19 patients, indicating enhanced DNA repair in acute Covid-19 disease. Furthermore, gene expression analysis demonstrated regulation of genes involved in BER and double strand break repair (DSBR) in PBMC of Covid-19 patients and expression level of several DSBR genes correlated with the degree of respiratory failure. Finally, by re-analyzing publicly available data, we found that the pathway Hallmark DNA repair was significantly more regulated in circulating immune cells during Covid-19 compared to influenza virus infection, bacterial pneumonia or acute respiratory infection due to seasonal coronavirus.

Conclusion

Although beneficial by protecting against DNA damage, long-term activation of the DNA repair machinery could also contribute to persistent inflammation, potentially through mechanisms such as the induction of cellular senescence. However, further studies that also include measurements of additional markers of DNA damage are required to determine the role and precise molecular mechanisms for DNA repair in SARS-CoV-2 infection.

Breast cancer and cytomegalovirus

PURPOSE

To determine whether cytomegalovirus is causally associated with breast cancer and whether cytomegalovirus should be categorised as an oncogenic virus.

METHODS

We undertook a review of published epidemiological and laboratory studies, using established causal criteria: Bradford Hill criteria to determine whether cytomegalovirus is associated with breast cancer; and Evans/Mueller criteria to determine whether cytomegalovirus should be categorised as an oncogenic virus.

RESULTS

Although there are inconsistencies in the findings of published epidemiological and laboratory studies, these may be explained by factors such as: differences in timing of blood samples, differences in selection of cases and controls, or high cytomegalovirus seroprevalence among participants in the epidemiological studies; and, in the laboratory studies, differences in sample preparations, age of sample, whether or not paired breast cancer and normal breast tissue samples were used, differences in the tests, primers and/or antibodies used, differences in histological types of breast cancer studied, and/or features of the virus.

CONCLUSIONS

Overall, the results of published studies of cytomegalovirus and breast cancer suggest cytomegalovirus is a causal factor for at least some types of breast cancer. If the evidence for a link between cytomegalovirus and breast cancer continues to strengthen, further research could lead to: targeted screening; therapy using antiviral drugs; and, perhaps, primary prevention of a significant proportion of breast cancer. Vaccination against viruses has already been shown to be effective in preventing cervix and liver cancer; cytomegalovirus vaccines are already under development.

Impact of human cytomegalovirus on glioblastoma cell viability and chemotherapy treatment

The relationship between human cytomegalovirus (HCMV) and tumours has been extensively investigated, mainly in glioblastoma multiforme (GBM), a malignant tumour of the central nervous system with low overall survival rates. Several reports have demonstrated the presence of HCMV in GBM, although typically restricted to a low number of cells, and studies have indicated that viral proteins have the ability to dysregulate cellular processes and increase tumour malignancy. Treatment of GBM involves the use of the chemotherapeutic agents temozolomide (TMZ) and carmustine (bis-chloroethylnitrosourea, BCNU), which lead to the attachment of adducts to the DNA backbone, causing errors during replication and consequent cell death. It is known that HCMV infection can modulate DNA repair pathways, but what effects the virus may exhibit during chemotherapy are unknown. Here we approach this question by analysing HCMV infection and viral protein accumulation in GBM cell lines with different genotypes and their response to TMZ and BCNU in the presence of the virus. We demonstrate that A172, TP365MG and U251MG GBM cells are efficiently infected by both low-passage (TB40E) and high-passage (AD169) HCMV strains. However, the GBM cell lines vary widely in their permissiveness to viral gene expression and exhibit very different patterns of immediate early, early and late protein accumulation. HCMV reduces the viability of permissive GBM cells in a multiplicity-dependent manner in both the absence and presence of TMZ or BNCU. In sum, we demonstrate that GBM cell lines are equally susceptible but differentially permissive to infection by both low- and high-passage strains of HCMV. This observation not only indicates that viral replication is largely controlled by cellular factors in this system, but also provides a possible explanation for why viral gene products are only found in a subset of cells in GBM tumours. Furthermore, we conclude that the virus does not confer increased resistance to chemotherapeutic drugs in various GBM cell lines, but instead reduces tumour cell viability. These results highlight that the oncomodulatory potential of HCMV is not limited to cancer-promoting activities, but also includes adverse effects on tumour cell proliferation or survival.

DNA repair mechanisms and human cytomegalovirus (HCMV) infection

Herpesvirus infections, such as those induced by human cytomegalovirus (HCMV), induce specific DNA damages. DNA damages can lead to cell mutation, death, apoptosis and immune system activation. Various types of DNA damage are repaired through multiple repair pathways, such as base excision, nucleotide excision, homologous recombination and nonhomologous end joining. Changes in the activity of DNA repair proteins during viral infection can cause disturbances in the DNA repair system and change its mechanisms. This report reviews results from studies, assaying a DNA repair system in HCMV infection.

Uracil DNA glycosylase BKRF3 contributes to Epstein-Barr virus DNA replication through physical interactions with proteins in viral DNA replication complex

Epstein-Barr virus (EBV) BKRF3 shares sequence homology with members of the uracil-N-glycosylase (UNG) protein family and has DNA glycosylase activity. Here, we explored how BKRF3 participates in the DNA replication complex and contributes to viral DNA replication. Exogenously expressed Flag-BKRF3 was distributed mostly in the cytoplasm, whereas BKRF3 was translocated into the nucleus and colocalized with the EBV DNA polymerase BALF5 in the replication compartment during EBV lytic replication. The expression level of BKRF3 increased gradually during viral replication, coupled with a decrease of cellular UNG2, suggesting BKRF3 enzyme activity compensates for UNG2 and ensures the fidelity of viral DNA replication. In immunoprecipitation-Western blotting, BKRF3 was coimmuno-precipitated with BALF5, the polymerase processivity factor BMRF1, and the immediate-early transactivator Rta. Coexpression of BMRF1 appeared to facilitate the nuclear targeting of BKRF3 in immunofluorescence staining. Residues 164 to 255 of BKRF3 were required for interaction with Rta and BALF5, whereas residues 81 to 166 of BKRF3 were critical for BMRF1 interaction in glutathione S-transferase (GST) pulldown experiments. Viral DNA replication was defective in cells harboring BKRF3 knockout EBV bacmids. In complementation assays, the catalytic mutant BKRF3(Q90L,D91N) restored viral DNA replication, whereas the leucine loop mutant BKRF3(H213L) only partially rescued viral DNA replication, coupled with a reduced ability to interact with the viral DNA polymerase and Rta. Our data suggest that BKRF3 plays a critical role in viral DNA synthesis predominantly through its interactions with viral proteins in the DNA replication compartment, while its enzymatic activity may be supplementary for uracil DNA glycosylase (UDG) function during virus replication.

IMPORTANCE

Catalytic activities of both cellular UDG UNG2 and viral UDGs contribute to herpesviral DNA replication. To ensure that the enzyme activity executes at the right time and the right place in DNA replication forks, complex formation with other components in the DNA replication machinery provides an important regulation for UDG function. In this study, we provide the mechanism for EBV UDG BKRF3 nuclear targeting and the interacting domains of BKRF3 with viral DNA replication proteins. Through knockout and complementation approaches, we further demonstrate that in addition to UDG activity, the interaction of BKRF3 with viral proteins in the replication compartment is crucial for efficient viral DNA replication.

Modulation of homology-directed repair in T98G glioblastoma cells due to interactions between wildtype p53, Rad51 and HCMV IE1-72

Human cytomegalovirus (HCMV) is a ubiquitous pathogen capable of causing life threatening consequences in neonates and immune-compromised individuals. HCMV inflicts site-specific double strand breaks (DSBs) in the cellular genome. DNA damage infliction raises the corollary question of virus modulation of DNA repair. We recently reported HDR was stimulated in wt human foreskin fibroblasts (HFFs) during fully permissive infection or expression of the HCMV protein IE1-72 (IE72). These studies have been extended into semi-permissive T98G glioblastoma cells. T98Gs encode a mutant p53, which may contribute to their high baseline rate of HDR. We fully expected HCMV infection to increase HDR in T98Gs, similar to its effects in HFFs. Surprisingly in T98Gs HCMV infection, or sole expression of IE72, decreased HDR by two-fold. Transient expression of wt p53 in T98Gs also reduced HDR by two-fold. Dual transient expression of wt p53 and IE72 restored high baseline HDR levels. GST pulldown experiments revealed that both IE72 and wt p53 bound the important HDR protein, Rad51. We conclude that the expression of certain HCMV proteins can modulate HDR in an infected cell, dependent upon p53 status. We propose a model of the protein interactions explaining this behavior.

HCMV-infected cells maintain efficient nucleotide excision repair of the viral genome while abrogating repair of the host genome

Many viruses subvert the host cell's ability to mount and complete various DNA damage responses (DDRs) after infection. HCMV infection of permissive fibroblasts activates host DDRs at the time of viral deposition and during replication, but the DDRs remain uncompleted without arrest or apoptosis. We believe this was in part due to partitioning of the damage response and double strand break repair components. After extraction of soluble proteins, the localization of these components fell into three groups: specifically associated with the viral replication centers (RCs), diffused throughout the nucleoplasm and excluded from the RCs. Others have shown that cells are incapable of processing exogenously introduced damage after infection. We hypothesized that the inability of the cells to process damage might be due to the differential association of repair components within the RCs and, in turn, potentially preferential repair of the viral genome and compromised repair of the host genome. To test this hypothesis we used multiple strategies to examine repair of UV-induced DNA damage in mock and virus-infected fibroblasts. Comet assays indicated that repair was initiated, but was not completed in infected cells. Quantitative analysis of immunofluorescent localization of cyclobutane pyrimidine dimers (CPDs) revealed that after 24 h of repair, CPDs were significantly reduced in viral DNA, but not significantly changed in the infected host DNA. To further quantitate CPD repair, we developed a novel dual-color Southern protocol allowing visualization of host and viral DNA simultaneously. Combining this Southern methodology with a CPD-specific T4 endonuclease V alkaline agarose assay to quantitate repair of adducts, we found efficient repair of CPDs from the viral DNA but not host cellular DNA. Our data confirm that NER functions in HCMV-infected cells and almost exclusively repairs the viral genome to the detriment of the host's genome.

Human cytomegalovirus (HCMV) is a leading cause of birth defects. This may be due in part to this virus' ability to inflict specific damage to its host's DNA, combined with the disruption of an infected cell's ability to repair damage. Earlier studies found that components of the cell's repair machinery were differentially associated with the HCMV viral replication centers in the nucleus. Experiments here extend this observation to include components of the machinery involved in UV lesion repair. We hypothesized that association of components of the DNA repair machinery within the viral replication centers could favor the repair of viral DNA, but more importantly, be detrimental to the repair of cellular DNA. Infected cells were irradiated and examined for repair by three different methods. In the course of this study, we developed a new technique allowing simultaneous evaluation of both the viral and host genomes in an infected cell. These experiments found rapid, selective removal of UV lesions from the viral and not the cellular DNA within infected cells. Our results indicate the differential association of certain cellular repair proteins with this virus may have far-reaching implications in the disease pathogenesis of HCMV infection.

The chromatin remodeling factor SMARCB1 forms a complex with human cytomegalovirus proteins UL114 and UL44

BACKGROUND

Human cytomegalovirus (HCMV) uracil DNA glycosylase, UL114, is required for efficient viral DNA replication. Presumably, UL114 functions as a structural partner to other factors of the DNA-replication machinery and not as a DNA repair protein. UL114 binds UL44 (HCMV processivity factor) and UL54 (HCMV-DNA-polymerase). In the present study we have searched for cellular partners of UL114.

METHODOLOGY/PRINCIPAL FINDINGS

In a yeast two-hybrid screen SMARCB1, a factor of the SWI/SNF chromatin remodeling complex, was found to be an interacting partner of UL114. This interaction was confirmed in vitro by co-immunoprecipitation and pull-down. Immunofluorescence microscopy revealed that SMARCB1 along with BRG-1, BAF170 and BAF155, which are the core SWI/SNF components required for efficient chromatin remodeling, were present in virus replication foci 24-48 hours post infection (hpi). Furthermore a direct interaction was also demonstrated for SMARCB1 and UL44.

CONCLUSIONS/SIGNIFICANCE

The core SWI/SNF factors required for efficient chromatin remodeling are present in the HCMV replication foci throughout infection. The proteins UL44 and UL114 interact with SMARCB1 and may participate in the recruitment of the SWI/SNF complex to the chromatinized virus DNA. Thus, the presence of the SWI/SNF chromatin remodeling complex in replication foci and its association with UL114 and with UL44 might imply its involvement in different DNA transactions.

Intracellular imaging of host-pathogen interactions using combined CARS and two-photon fluorescence microscopies

Intracellular imaging is a key tool in the investigation of host-pathogen interactions. Advances in this area are particularly sought to understand the effect of viral infection processes on the host cell and its metabolic functions including those cases where host cell lipid metabolism is modulated as a result of infection. We demonstrate the use of combined coherent anti-Stokes Raman scattering (CARS) and two-photon fluorescence microscopies to image fibroblast cells infected by cytomegalovirus. CARS is used to image the host cell membrane, lipid droplets and morphology of the nucleus. Cell nuclei are found to expand during infection, approximately doubling in area. Some cells also show accumulations of lipid droplets at the nuclear periphery. Using a genetically modified virus strain expressing the green fluorescent protein also enables two-photon imaging of the same cells to reveal the location, nature and extent of viral protein expression.

Characterization of human cytomegalovirus uracil DNA glycosylase (UL114) and its interaction with polymerase processivity factor (UL44)

Here, we report the molecular characterization of the human cytomegalovirus uracil DNA glycosylase (UNG) UL114. Purified UL114 was shown to be a DNA glycosylase, which removes uracil from double-stranded and single-stranded DNA. However, kinetic analysis has shown that viral UNG removed uracil more slowly compared with the core form of human UNG (Delta84hUNG), which has a catalytic efficiency (k(cat)/K(M)) 350- to 650-fold higher than that of UL114. Furthermore, UL114 showed a maximum level of DNA glycosylase activity at equimolar concentrations of the viral polymerase processivity factor UL44. Next, UL114 was coprecipitated with DNA immobilized to magnetic beads only in the presence of UL44, suggesting that UL44 facilitated the loading of UL114 on DNA. Moreover, mutant analysis demonstrated that the C-terminal part of UL44 (residues 291-433) is important for the interplay with UL114. Immunofluorescence microscopy revealed that UL44 and UL114 colocalized in numerous small punctuate foci at the immediate-early (5 and 8 hpi) phases of infection and that these foci grew in size throughout the infection. Furthermore, coimmunoprecipitation assays with cellular extracts of infected cells confirmed that UL44 associated with UL114. Finally, the nuclear concentration of UL114 was estimated to be 5- to 10-fold higher than that of UL44 in infected cells, which indicated a UL44-independent role of UL114. In summary, our data have demonstrated a catalytically inefficient viral UNG that was highly enriched in viral replication foci, thus supporting an important role of UL114 in replication rather than repair of the viral genome.

Characterization of the uracil-DNA glycosylase activity of Epstein-Barr virus BKRF3 and its role in lytic viral DNA replication

Uracil-DNA glycosylases (UDGs) of the uracil-N-glycosylase (UNG) family are the primary DNA repair enzymes responsible for removal of inappropriate uracil from DNA. Recent studies further suggest that the nuclear human UNG2 and the UDGs of large DNA viruses may coordinate with their DNA polymerase accessory factors to enhance DNA replication. Based on its amino acid sequence, the putative UDG of Epstein-Barr virus (EBV), BKRF3, belongs to the UNG family of proteins, and it was demonstrated previously to enhance oriLyt-dependent DNA replication in a cotransfection replication assay. However, the expression and enzyme activity of EBV BKRF3 have not yet been characterized. In this study, His-BKRF3 was expressed in bacteria and purified for biochemical analysis. Similar to the case for the Escherichia coli and human UNG enzymes, His-BKRF3 excised uracil from single-stranded DNA more efficiently than from double-stranded DNA and was inhibited by the purified bacteriophage PBS1 inhibitor Ugi. In addition, BKRF3 was able to complement an E. coli ung mutant in rifampin and nalidixic acid resistance mutator assays. The expression kinetics and subcellular localization of BKRF3 products were detected in EBV-positive lymphoid and epithelial cells by using BKRF3-specific mouse antibodies. Expression of BKRF3 is regulated mainly by the immediate-early transcription activator Rta. The efficiency of EBV lytic DNA replication was slightly affected by BKRF3 small interfering RNA (siRNA), whereas cellular UNG2 siRNA or inhibition of cellular and viral UNG activities by expressing Ugi repressed EBV lytic DNA replication. Taking these results together, we demonstrate the UNG activity of BKRF3 in vitro and in vivo and suggest that UNGs may participate in DNA replication or repair and thereby promote efficient production of viral DNA.