Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 510, 205-215

Infected T98G Glioblastoma Cells Support Human Cytomegalovirus Reactivation From Latency

Affiliations

Infected T98G Glioblastoma Cells Support Human Cytomegalovirus Reactivation From Latency

Shuang Cheng et al. Virology.

Abstract

T98G cells have been shown to support long-term human cytomegalovirus (HCMV) genome maintenance without infectious virus release. However, it remains unclear whether these viral genomes could be reactivated. To address this question, a recombinant HCMV (rHCMV) containing a GFP gene was used to infect T98G cells, and the infected cells absent of infectious virus production were designated T98G-LrV. Upon dibutyryl cAMP plus IBMX (cAMP/IBMX) treatment, a serial of phenomena were observed, including GFP signal increase, viral genome replication, lytic genes expression and infectious viruses release, indicating the reactivation of HCMV in T98G-LrV cells from a latent status. Mechanistically, HCMV reactivation in the T98G-LrV cells induced by cAMP/IBMX was associated with the PKA-CREB signaling pathway. These results demonstrate that HCMV was latent in T98G-LrV cells and could be reactivated. The T98G-LrV cells represent an effective model for investigating the mechanisms of HCMV reactivation from latency in the context of neural cells.

Keywords: Human cytomegalovirus; Latent cell model of brain origin; Latent infection; Reactivation; T98G cells.

Figures

Fig. 1.
Fig. 1.. GFP is a reliable marker for monitoring rHCMV infection in T98G cells.
T98G cells cultured on coverslips were infected with rHCMV at an MOI of 10. At 7 days post infection (dpi), cells were analyzed for expression of GFP and late protein pp28 by IFA. Representative images from three independent experiments are shown. Scale bar: 50 µm.
Fig. 2.
Fig. 2.. Establishment of latently infected T98G cells with rHCMV (T98G-LrV).
T98G cells were infected with rHCMV at an MOI of 10 and passaged every 7 days. Supernatants and cells of each passage were harvested before subculture. (A) Infectious virus. HFFs monolayers were inoculated with the supernatants (upper panel) or the infected cells (frozen-thawed cells, lower panel). GFP-positive foci were observed 14 days post inoculation. Scale bar: 100 µm. (B) Cell-free virus titer. Virus titers in the supernatants were determined by plaque forming assay. (C) Viral DNA. Cellular DNA was extracted from the infected cells of each passage, and viral genomic DNA was examined by nested PCR based on IE1 gene. Shown are the representative results of the first (Outer) and nested (Inner) amplification of PCR, respectively. (D) Cell-associated viral DNA and GFP-positive cells. Viral genome copy numbers in the infected cells were determined by qPCR. Serial dilution of T98G cells was isolated for DNA and subjected to qPCR for determination the GAPDH copy number of each cell. Viral genome copies were calculated as “copies/cell” with adjusting to GAPDH. GFP-positive cells at each passage was analyzed by FACS. (E) Transcription of representative viral genes. Total RNA was extracted from each set of the passaged cells, reversely transcribed to cDNA, and assayed for the indicated genes by PCR. The mRNA without reverse transcription (RT) were assayed for UL44 as control to exclude viral DNA contamination.
Fig. 3.
Fig. 3.. cAMP/IBMX reactivates latent HCMV in T98G-LrV cells.
T98G-LrV cells were treated with DMSO or cAMP/IMBX for 30 days. (A) GFP signal in the treated T98G-LrV cells. GFP signal in cells was observed using a Nikon A1 confocal microscope. The numbers of GFP foci were counted and the fold changes compared to the DMSO control were calculated. Representative images (upper panel, left) and the fold change (upper panel, right) from 3 independent experiments are shown. The magnified images of the indicated sites are shown in a1–3 and b1–3, respectively. Scale bar: 500 µm. ** p < 0.01. (B) Infectious virions release. The cell-free supernatants from the treated T98G-LrV cells were harvested, and inoculated onto HFFs monolayers for 14 days to detect infectious virions. Scale bar: 100 µm.
Fig. 4.
Fig. 4.. cAMP/IBMX increases GFP signal.
T98G-LrV cells were treated with DMSO or cAMP/IBMX for 30 days to detect the GFP signal. (A) Flow cytometry analysis of GFP-positive cells. Representative results are shown from 3 independent experiments. (B-C) High content analysis of GFP-positive cells. Representative high content images and the average GFP positive cells per well from 10 independent replicates are shown. *** p < 0.001.
Fig. 5.
Fig. 5.. cAMP/IBMX initiates HCMV replication.
T98G-LrV cells were treated with DMSO or cAMP/IBMX for about 30 days. Cells and supernatants were collected for following assays. (A) Viral replication centers. The cells were plated onto coverslips and examined the expression of GFP and UL44 by IFA. Scale bar: 10 µm. (B) Viral genome copy number. Total DNA was extracted from the collected cells, and viral genome copy number per cell was determined by qPCR. *** p < 0.001. (C) Transcription levels of the representative lytic viral genes. RNAs were extracted from the collected cells followed by RT-PCR, and relative transcription levels of the indicated viral genes were assessed by qPCR with GAPDH as the reference gene. (D) Viral proteins expression. The collected cells were lysed and subjected to western blotting for examination of the indicated viral proteins. Actin serves as the loading control. (E) Effect of PAA on IE1 transcription. T98G-LrV cells were pretreated with PAA for 12 days, then cultured for additional 8 days in the presence of DMSO or cAMP/IBMX. Total RNA was extracted, and IE1 transcription level was quantified by qPCR. The relative level of IE1 was calibrated to that in the DMSO treated cells in the presence of PAA, with GAPDH as the reference gene, and the uninfected T98G cells as the mock.
Fig. 6.
Fig. 6.. cAMP/IBMX increases infectious viruses.
(A) Cell-free virus titer. Supernatant samples were harvested at the indicated times and cell-free virus was titrated by plaque forming assay. The results were presented as mean ± SD from 3 biological replicates. (B-C) Cell-associated virus. T98G-LrV cells were treated with DMSO or cAMP/IBMX for 30 days. Then intact cells or sonicated cells were cocultured with HFFs monolayers in 6-well plates. GFP signals were observed at 10 days post coculture and the numbers of GFP plaque per well were counted. Representative images of GFP plaques and the average results of GFP plaque presented as mean ± SD from 3 biological replicates are shown. Scale bar: 100 µm. * p < 0.05; *** p < 0.001.
Fig. 7.
Fig. 7.. HCMV reactivation in T98G-LrV cells is associated with the PKA-CREB signaling pathway.
T98G-LrV cells were pretreated with H89 for 1 h, and then treated with cAMP/IBMX in the presence of H89 for 8 days. Medium was changed every 4 days. Supernatants and cells were collected for the subsequent assays. (A) Cell-free virus titer. Supernatants were subjected to the plaque forming assay to determine the cell-free virus. The results were presented as mean ± SD from 3 biological replicates. (B) Relative level of IE1. Total RNA was extracted from the collected cells, and relative transcription level of IE1 was examined by qPCR with GAPDH as the reference gene. *** p < 0.001. (C) Protein levels of IE1 and representative components of PKA-CREB signaling pathway. Cell lysates were prepared and equal amount of proteins were subjected to western blotting. The protein levels were quantitated by densitometry using Image J software (National Institutes of Health) and presented as the relative level to that of DMSO control after normalization to actin.

Similar articles

See all similar articles

Cited by 2 PubMed Central articles

Publication types

Feedback