A myeloid progenitor cell line capable of supporting human cytomegalovirus latency and reactivation, resulting in infectious progeny

Abstract

Human cytomegalovirus (HCMV) is a herpesvirus that establishes a lifelong, latent infection within a host. At times when the immune system is compromised, the virus undergoes a lytic reactivation producing infectious progeny. The identification and understanding of the biological mechanisms underlying HCMV latency and reactivation are not completely defined. To this end, we have developed a tractable in vitro model system to investigate these phases of viral infection using a clonal population of myeloid progenitor cells (Kasumi-3 cells). Infection of these cells results in maintenance of the viral genome with restricted viral RNA expression that is reversed with the addition of the phorbol ester 12-O-tetradecanoylphorbol-13-acetate (TPA, also known as PMA). Additionally, a latent viral transcript (LUNA) is expressed at times where viral lytic transcription is suppressed. Infected Kasumi-3 cells initiate production of infectious virus following TPA treatment, which requires cell-to-cell contact for efficient transfer of virus to other cell types. Importantly, lytically infected fibroblast, endothelial, or epithelial cells can transfer virus to Kasumi-3 cells, which fail to initiate lytic replication until stimulated with TPA. Finally, inflammatory cytokines, in addition to the pharmacological agent TPA, are sufficient for transcription of immediate-early (IE) genes following latent infection. Taken together, our findings argue that the Kasumi-3 cell line is a tractable in vitro model system with which to study HCMV latency and reactivation.

Fig 1

HCMV infection of Kasumi-3 cells results in approximately 11% infectivity rate. Kasumi-3 cells were mock (A) or HCMV (B) infected with TB40/Ewt-GFP at an MOI of 10 PFU/cell. Infected cells were gated first on viability and then gated for eGFP expression.

Fig 2

HCMV-infected Kasumi-3 cells maintain viral genomes with restricted viral gene expression. Viral RNA (black bars) and viral DNA (white bars) were assayed over the indicated times for UL123. RNA transcripts were normalized to cellular GAPDH (glyceraldehyde-3-phosphate dehydrogenase), and DNA was normalized to cellular MDM2. Each sample was performed in triplicate. AU, arbitrary units.

Fig 3

Repression of HCMV transcripts is released by TPA treatment. Kasumi-3 cells were infected with TB40/Ewt-GFP in the presence or absence of TPA. RNA was harvested at the indicated time points, and representative transcripts from each class of viral gene expression were assessed. UL123 and UL122 (IE, top panel), UL44 and UL54 (E, middle panel), and UL99 and UL83 (L, bottom panel) transcripts were increased with the addition of TPA to the infected cultures. Each sample was performed in triplicate and normalized to cellular GAPDH. AU, arbitrary units.

Fig 4

The latent transcript LUNA is expressed while UL123 transcription is repressed, which is reversed with the addition of TPA. Infected Kasumi-3 cells were sorted 2 dpi and cultured for an additional 18 days, at which point cells were cultured for an additional 2 days with or without TPA. Cells were then harvested, and total RNA was collected. LUNA-specific primers were used in the RT reaction to subsequently quantify the LUNA transcript. To quantify UL123, total cDNA was reverse transcribed using random hexamers. Each sample was analyzed in triplicate. AU, arbitrary units.

Fig 5

TPA-induced reactivation of HCMV-infected Kasumi-3 cells results in the production of infectious viral particles. (A) Extracellular viral production of HCMV-infected Kasumi-3 cells is increased with TPA treatment. Cell-free viral particles were harvested from TB40/Ewt-GFP-infected cells cultured in the presence or absence of TPA. Encapsidated viral DNA was assessed by qPCR using primers directed at UL123. (B and C) To determine if the viral particles were infectious, infected Kasumi-3 cells were cultured in the presence or absence of TPA for 6 days, after which the cells were washed with PBS and cocultured with HFFs for 2 days. Kasumi-3 cells were then removed from the cocultures, the HFFs were washed twice with PBS, and the HFFs were cultured for an additional 4 days. (B) Fluorescence microscopy reveals eGFP-positive plaques in the HFF monolayer cocultured with TPA-reactivated Kasumi-3-infected cells. (C) HFF-associated DNA was assayed for viral genomes using primers directed at UL44. Each sample was normalized to cellular MDM2 and assessed in triplicate. AU, arbitrary units.

Fig 6

Frequency of reactivation in infected Kasumi-3 cultures in the presence and absence of TPA. Limiting dilution (ELDA) using fibroblasts reveals 131 infected Kasumi-3 cells treated with TPA are required for 1 plaque compared to untreated infected cells, which require 349 cells to generate 1 plaque (P < 0.0005). Black lines represent cultures treated with TPA; gray lines represent cultures not treated with TPA. Solid lines represent the trend line estimating the active cell frequency; dashed lines represent the 90% confidence interval. Circles represent data points of the log proportion of negative cultures; upside-down triangles represent data with zero negative response.

Fig 7

PAA treatment of infected Kasumi-3 cells limits spontaneous reactivation of HCMV. (A) Infected Kasumi-3 cells were sorted 2 dpi and cultured for 18 days, at which time the cells were cultured for an additional 2 days in the presence or absence of TPA. Cells were then harvested for IFA using an antibody to the late protein pUL99 and DAPI. (B) Infected Kasumi-3 cells that did not receive TPA treatment show a significantly less number of pUL99-positive cells than those that received TPA (P < 0.05). Additionally, the fluorescent intensity of pUL99 observed in the infected cells that were not treated with TPA did not vary between the infected cultures treated with TPA (panel A). (C) Kasumi-3 cells were infected and sorted as described above and then cultured for 18 days in the presence or absence of PAA. These cells were then cultured for an additional 2 days in the presence or absence of TPA. Total RNA was harvested and quantified by RT-qPCR for UL123 expression. Each sample was normalized to cellular GAPDH and analyzed in triplicate. AU, arbitrary units.

Fig 8

Cell-to-cell contact is required for infected endothelial, epithelial, and fibroblast cells to transfer virus to Kasumi-3 cells. Primary HUVECs (endothelial cells) (A), ARPE19 cells (epithelial cells) (B), or HFFs (fibroblasts) (C) were infected and then cocultured with Kasumi-3 cells for 5 days. Kasumi-3 cells were removed from the coculture and cultured in the presence or absence of TPA for 2 days. Kasumi-3 cell-associated DNA was harvested to quantify viral genomes by qPCR using primers recognizing UL99. Samples were normalized to cellular MDM2 and analyzed in triplicate. AU, arbitrary units.

Fig 9

Inflammatory cytokines increase HCMV IE transcription. (A) Either TNF-α or IL-1β treatment induces UL123 gene expression in infected Kasumi-3 cells. Cells were treated with either cytokine for 48 h, after which cells were harvested for RT-qPCR for UL123. (B) TNF-α-induced IE gene expression as early as 2 h posttreatment. Infected Kasumi-3 cells (10 dpi) were treated with TNF-α, IL-1β, or TPA for the indicated times. Cells were collected for RT-qPCR for UL123. In all cases, samples were analyzed in triplicate with primers toward UL123. Each sample was normalized to cellular GAPDH and then normalized to the NT control. NT, no treatment; AU, arbitrary units.