The adenovirus e3 promoter is sensitive to activation signals in human T cells

Abstract

The group C adenoviruses typically cause acute respiratory disease in young children. In addition, a persistent phase of infection has been observed in which virus may be shed for years without producing overt pathology. Our laboratory recently reported that group C adenovirus DNA can be found in tonsil and adenoid T lymphocytes from the majority of pediatric donors (C. T. Garnett, D. Erdman, W. Xu, and L. R. Gooding, J. Virol. 76:10608-10616, 2002). This finding suggests that immune evasion strategies of human adenoviruses may be directed, in part, toward protection of persistently or latently infected T lymphocytes. Many of the adenoviral gene products implicated in prevention of immune destruction of virus-infected cells are encoded within the E3 transcription unit. In this study, the E3 promoter was evaluated for sensitivity to T-cell activation signals by using a promoter reporter plasmid. Indeed, this promoter is extremely sensitive to T-cell activation, with phorbol myristate acetate (PMA) plus ionomycin increasing E3-directed transcription 100-fold. By comparison, in the same cells E1A expression leads to a 5.5-fold increase in transcription from the E3 promoter. In contrast to induction by E1A, activation by PMA plus ionomycin requires the two E3 NF-kappaB binding sites. Interestingly, expression of E1A inhibits induction of the E3 promoter in response to T-cell activation while increasing E3 promoter activity in unactivated cells. Collectively, these data suggest that the E3 promoter may have evolved the capacity to respond to T-cell activation in the absence of E1A expression and may act to upregulate antiapoptotic gene expression in order to promote survival of persistently infected T lymphocytes.

FIG. 1.

Structure of the adenovirus E3 promoter. Numbers indicate nucleotide positions on the adenovirus type 2 genome. Sequences required for E1A induction of E3 (E1A-RE) are described in the text.

FIG. 2.

Activation of the adenovirus type 2 E3 promoter reporter by T-cell stimulation. CEM and Jurkat cells were transfected with 50 or 5 μg (respectively) of the wild-type E3 reporter plasmid, pE3-luc. At 24 h after transfection, cells were treated with PMA (16 nM), ionomycin (0.75 μM), PMA plus ionomycin, or plate-bound anti-CD3 (UCHT1, 10 μg/ml), and luciferase activity was determined after 12 h. Error bars were determined from the averages of three independent transfection experiments.

FIG. 3.

Transcription factor binding sites required for activation of the E3 promoter. (A) CEM cells were transfected with E3 promoter reporter plasmids containing the indicated mutations and treated with PMA plus ionomycin. Cells were harvested at various time points, and luciferase activity was determined. Results are the averages of three independent transfection experiments. (B) HeLa cells were transfected with 5 μg of the indicated E3 promoter reporter plasmids together with or without 1 μg pCMV-E1A; cells were harvested, and luciferase activity determined 48 h posttransfection. Results are the averages of two independent transfection experiments. Induction by E1A is shown. (C) HeLa cells were transfected with 5 μg of the indicated E3 promoter reporter plasmid and treated with TNF (500 U/ml); cells were harvested, and luciferase activity was determined after 12 h. Induction by TNF is shown.

FIG. 4.

NF-κB from activated CEM T cells binds to both E3 κB sites. Nuclear extracts from unactivated or PMA-ionomycin (Iono)-activated CEM cells were incubated with double-stranded oligonucleotides comprising either the 5′ or the 3′ E3 NF-κB binding sites as detailed in Materials and Methods. Unlabeled 5′ or 3′ E3 NF-κB (“specific”) or HLA-DRA oligonucleotide (36) (“non-specific”) were used as specificity controls.

FIG. 5.

Inhibition of E3 promoter reporter activation in T cells by E1A. CEM cells were cotransfected with 50 μg of pE3-luc and increasing amounts of pCMV-E1A (or pN1-EGFP [Clontech] as a control). Transfected cells were split and left untreated or treated with PMA plus ionomycin. Cells were harvested after 4 or 10 h of treatment, and reporter activity was determined. Results are the averages of three independent transfection experiments.