Interaction of an adenovirus E3 14.7-kilodalton protein with a novel tumor necrosis factor alpha-inducible cellular protein containing leucine zipper domains

Abstract

Early region 3 (E3) of group C human adenoviruses (Ad) encodes several inhibitors of tumor necrosis factor alpha (TNF-alpha) cytolysis, including an E3 14.7-kDa protein (E3-14.7K) and a heterodimer containing two polypeptides of 10.4 and 14.5 kDa. To understand the mechanism by which the viral proteins inhibit TNF-alpha functions, the E3-14.7K protein was used to screen a HeLa cell cDNA library to search for interacting proteins in the yeast two-hybrid system. A novel protein containing multiple leucine zipper domains without any significant homology with any known protein was identified and has been named FIP-2 (for 14.7K-interacting protein). FIP-2 interacted with E3-14.7K both in vitro and in vivo. It colocalized with Ad E3-14.7K in the cytoplasm, especially near the nuclear membrane, and caused redistribution of the viral protein. FIP-2 by itself does not cause cell death; however, it can reverse the protective effect of E3-14.7K on cell killing induced by overexpression of the intracellular domain of the 55-kDa TNF receptor or by RIP, a death protein involved in the TNF-alpha and Fas apoptosis pathways. Deletion analysis indicates that the reversal effect of FIP-2 depends on its interaction with E3-14.7K. Three major mRNA forms of FIP-2 have been detected in multiple human tissues, and expression of the transcripts was induced by TNF-alpha treatment in a time-dependent manner in two different cell lines. FIP-2 has consensus sequences for several potential posttranslational modifications. These data suggest that FIP-2 is one of the cellular targets for Ad E3-14.7K and that its mechanism of affecting cell death involves the TNF receptor, RIP, or a downstream molecule affected by either of these two molecules.

FIG. 1

Expression of FIP-2 mRNA in human tissues. A blot of mRNAs obtained from eight human organs as indicated was purchased from CloneTech and hybridized under stringent conditions as described by the supplier. The FIP-2 cDNA used as probe was labeled with [³²P]dCTP (Amersham). (The top band of approximately 7.5 kb was also visible in some of the other organs as well as skeletal muscle on the original autoradiograph.) Numbers on the left are sizes in kilobases.

FIG. 2

Polypeptide and cDNA sequences of FIP-2. The cDNA sequences were derived from clones from the yeast two-hybrid screening and RACE studies. Polypeptide sequences were deduced by using conventional genetic codons through a computerized program. Three different splicing forms are shown as follows: form I, unspliced; form II, the italicized sequence upstream of the first methionine is spliced out; and form III, the underlined sequences are spliced out. Forms I and II probably utilize the same start codon (bases 328 to 330), while form III utilizes the start codon at amino acid 58 (double-underlined Met at bases 499 to 501). The 5′ ends of three clones identified in the yeast two-hybrid screening are identified by arrows. The amino acids in the putative leucine zipper domains are boxed.

FIG. 2

Polypeptide and cDNA sequences of FIP-2. The cDNA sequences were derived from clones from the yeast two-hybrid screening and RACE studies. Polypeptide sequences were deduced by using conventional genetic codons through a computerized program. Three different splicing forms are shown as follows: form I, unspliced; form II, the italicized sequence upstream of the first methionine is spliced out; and form III, the underlined sequences are spliced out. Forms I and II probably utilize the same start codon (bases 328 to 330), while form III utilizes the start codon at amino acid 58 (double-underlined Met at bases 499 to 501). The 5′ ends of three clones identified in the yeast two-hybrid screening are identified by arrows. The amino acids in the putative leucine zipper domains are boxed.

FIG. 2

Polypeptide and cDNA sequences of FIP-2. The cDNA sequences were derived from clones from the yeast two-hybrid screening and RACE studies. Polypeptide sequences were deduced by using conventional genetic codons through a computerized program. Three different splicing forms are shown as follows: form I, unspliced; form II, the italicized sequence upstream of the first methionine is spliced out; and form III, the underlined sequences are spliced out. Forms I and II probably utilize the same start codon (bases 328 to 330), while form III utilizes the start codon at amino acid 58 (double-underlined Met at bases 499 to 501). The 5′ ends of three clones identified in the yeast two-hybrid screening are identified by arrows. The amino acids in the putative leucine zipper domains are boxed.

FIG. 3

Intracellular colocalization of FIP-2 with Ad E3-14.7K. FIP-2 was cloned behind the cytomegalovirus promoter and coexpressed as a fusion protein with a T7 tag in the murine C3HA cell line constitutively expressing Ad E3-14.7K (16). (A) Ad E3-14.7K was visualized with a polyclonal rabbit antibody which recognized the viral protein either alone as a cytoplasmic protein (∗) or in cells also cotransfected with T7-FIP-2 (∗∗). (B) FIP-2 was visualized on identical cells with antibody to T7 (∗∗). The distribution of Ad E3-14.7K within individual cells in panel A was different in cells expressing E3-14.7K alone (∗) or overexpressing FIP-2 (∗∗) together with E3-14.7K, which appeared in bead-like perinuclear structures (arrows). (C) The colocalization of FIP-2 and Ad E3-14.7K is also highlighted by the yellow color, resulting from the convergence of the rhodamine image of panel A plus the fluorescence image of panel B. Bar, 10 μm.

FIG. 4

In vitro and in vivo interaction between FIP-2-and E3-14.7K. (A) The in vitro interaction of radiolabeled FIP-2 with the Ad E3-14.7K–GST fusion protein or with GST alone was assayed as described in Materials and Methods. The amounts of FIP-2 absorbed and eluted from Ad E3-14.7K (GST-14.7) or from the GST protein alone as a negative control are shown after SDS-PAGE. (B) The in vivo interaction between E3-14.7K and FIP-2 is shown by coimmunoprecipitation from extracts of 293 cells transiently transfected with plasmids expressing T7-tagged FIP-2 (Δ134) and FLAG-tagged E3-14.7K proteins. Twenty hours after transfection, the cells were harvested and lysed. Cleared lysates were subjected to immunoprecipitation with either anti-FLAG (specific for E3-14.7K) or anti-gp19 (nonspecific) monoclonal antibody. The immunoprecipitates were analyzed by SDS-PAGE followed by Western blotting. FIP-2 was detected by anti-T7 monoclonal antibody.

FIG. 5

FIP-2 reverses the protective effect of E3-14.7K on TNF receptor-induced cytolysis. Human embryonic kidney 293 cells in six-well plates were transfected with the following plasmids: (A) pcDNA-FLAG-14.7; (B) pcDNA-FLAG-TR55; (C) pcDNA-T7-FIP2; (D) pcDNA-FLAG-14.7; (E) pcDNA-TR55; (F) pcDNA-FLAG-14.7 plus pcDNA-TR55; (G) pcDNA-FLAG-14.7K plus pcDNA-TR55 plus pcDNA T7 FIP-2; (H) pcDNA-FIP-2 CΔ346 plus pcDNA-FLAG-14.7 plus pcDNA-TR55; (I) pcDNA FIP-2Δ134 plus pcDNA-FLAG-14.7 plus pcDNA-TR55; (J) pcDNA FIP-2Δ268 plus pcDNA-FLAG-14.7 plus pcDNA-TR55; (K) pcDNA-FIP-2Δ395 plus pcDNA FLAG-14.7 plus pcDNA TR55. All of these cells were cotransformed with a plasmid expressing the GFP gene (see Materials and Methods). Twenty-four hours after transfection, the cells were observed with a fluorescence microscope and photographed with a fluorescein isothiocyanate filter. Panels A and B were photographed through a 40× objective. The arrows in panel A point to transfected normal cells stained with GFP. In panel B, the arrow shows a rounded cell with cytoplasmic blebbing, and the arrowheads show completely disintegrated cells. Both morphologies are typical for apoptosis. Panels C to K were photographed through a 20× objective. Panels C, D, F, and H show morphologies that are normal or nearly normal. The other panels (E, G, I, J, and K) demonstrate various degrees of apoptosis. The percentages of cells in the panels that appeared normal after each transfection were as follows: C, 97%; D, 92%; E, 6%; F, 81%; G, 4%; H, 87%; I, 11%; J, 9%; and K, 14%.

FIG. 6

FIP-2 binding to E3-14.7K correlates with FIP-2 reversal of E3-14.7K inhibition of TR55 cytolysis. Deletion mutants of FIP-2 were cloned in appropriate vectors, and the ability of each mutant to interact with E3-14.7K in the yeast two-hybrid system (β-galactosidase [β-gal] activity) was compared with the reversal of the E3-14.7K inhibition on TR55 killing. +, full activity; −, no activity detected (see Materials and Methods); LZ, leucine zipper.

FIG. 7

Induction of FIP-2 expression by TNF-α. Human embryonic kidney 293 cells and adenocarcinoma MCF-7 cells were treated with TNF-α as indicated in Materials and Methods. The purified total RNAs were used in RPAs with FIP-2 as a probe. The upper band was expected from sequence data, but the lower band probably represents an alternative splicing form of FIP-2. Human glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as an internal control.