Cell Cycle Progression in Monkey Cells Expressing Simian Virus 40 Small t Antigen from Adenovirus Vectors

Abstract

The simian virus 40 small t antigen (small-t) is required for optimal viral replication and transformation, especially during the infection of nondividing cells, suggesting that the function of small-t is to promote cell cycle progression. The mechanism through which small-t promotes cell growth reflects, in part, its binding and inhibition of protein phosphatase 2A (PP2A). The use of recombinant adenoviruses allows small-t expression in a majority of cells in a population, thus providing a convenient source of cells for biochemical analyses. In monkey kidney CV1 cells, small-t expressed from these adenovirus vectors activated the mitogen-activated protein kinase (MAPK) pathway, induced JNK activity, and increased AP-1 DNA-binding activity, all in a PP2A-dependent manner. Expression of small-t also caused an increase in the phosphorylation of the Na⁺/H⁺ antiporter, a mitogen-activated ion exchanger whose activity correlates with its phosphorylation. At least part of the antiporter phosphorylation induced by small-t reflected activation of the MAPK pathway, as suggested by results of assays using a chemical inhibitor of the MAPK-activating kinase, MEK. Finally, small-t expression from adenovirus vectors promoted efficient cell cycle progression by growth-arrested cells. These vectors should facilitate further analysis of effects of small-t on cell cycle mediators.

The small t antigen (small-t) of simian virus 40 (SV40) enhances viral transformation of several nonpermissive murine cell lines (8, 33, 34, 50, 57, 61), is required for induction of focus formation in semipermissive human fibroblasts (11, 42), and also augments viral and cellular DNA replication (13, 56, 58). There is now a substantial body of evidence that this enhancement may be a result of the ability of small-t to promote cell cycle progression. First, the inefficient transformation induced by viruses lacking small-t was improved under conditions which promote cell cycle progression. These conditions include infection of actively growing versus growth-arrested cells, allowing infected cells to undergo one or more rounds of cell division before selection in soft agar, and treatment of infected cells with serum or phorbol ester. SV40-mediated tumorigenesis in vivo is also affected by small-t in that the absence of small-t expression restricts transformation to actively dividing tissues (e.g., lymphoid tissue) (10, 12). The ability of small-t to induce cell cycle progression was more directly supported by experiments with nonpermissive mouse embryo fibroblasts in which wild-type (WT) SV40 was able to induce multiple rounds of cell division whereas mutant viruses lacking small-t induced only a single round of division (23). Finally, growth promotion by small-t was also directly demonstrated in permissive CV1 cells (14, 58).

One of the most important biochemical characteristics of small-t is its association with and inhibition of the cellular serine-threonine protein phosphatase 2A (PP2A) (39, 66, 70, 71). PP2A exists as a heterotrimer of the catalytic C subunit and two regulatory subunits, A and B (15). Small-t, like the B subunits, binds PP2A primarily through interaction with the A subunit (24, 51), although interaction between small-t and the C subunit may further stabilize the small-t/A/C complex (51). The consequence of small-t interaction with PP2A is inhibition of phosphatase activity toward most substrates (54, 58, 59, 70). The importance of PP2A binding on small-t function has been firmly established through analysis of small-t mutants. Point mutation of either of the cysteines at positions 97 and 103, or the interposed proline at position 101, severely diminishes PP2A binding (37), and these mutations decrease viral transformation efficiency in small-t-dependent assays (37, 42). A second region, comprising amino acids 110 to 130 and containing the first cysteine-rich cluster, has also been implicated as important for PP2A binding, and deletion of this region affects the ability of small-t to mediate growth of CV1 cells in low serum (58). Transient transfection of small-t mutants have also established PP2A binding as important for small-t-mediated activation of components of the mitogen-activated protein kinase (MAPK) cascade (58, 69) and some transcriptional events (20, 69).

Another activity of small-t antigen that influences cell growth is not linked to PP2A interaction. Our laboratory has shown that mutations within the region from amino acids 42 to 47 reduce the ability of small-t to transactivate both the adenovirus E2A and mammalian cyclin A promoters and also attenuate small-t-dependent transformation of human and rodent fibroblasts (42). This region is not required for interaction with PP2A, as a bacterially expressed truncated version of small-t that lacks this region binds PP2A as efficiently as full-length small-t (37). However, this region is strongly conserved among the papovaviruses (41) and is related to the DnaJ family of molecular chaperones, known to interact with and regulate certain heat shock proteins (26). Indeed, it was recently shown that small-t can activate the ATPase activity of hsc70 in a manner that is dependent on the integrity of the region from amino acids 42 to 47 (60).

An early system used to study growth promotion by small-t is its ability to confer resistance to growth arrest elicited by the methylxanthine compound theophylline (52), a known inhibitor of cyclic AMP phosphodiesterase. Surprisingly, theophylline-mediated growth arrest correlates not with increased levels of intracellular cyclic AMP (47) but rather with inhibition of the Na⁺/H⁺ antiporter (38). The antiporter, or Na⁺/H⁺ exchanger (NHE), represents a family of mitogen-responsive ion transporters and regulators of intracellular pH (5, 16, 72). It is activated rapidly after stimulation by many, diverse growth factors, and its activation correlates with an increase in phosphorylation (6). Recent evidence suggests that phosphorylation of the antiporter may be regulated, directly or indirectly, by the MAPKs (1, 4, 63, 67).

In this study, we used recombinant adenovirus vectors expressing WT small-t (Adt) or various point mutants to further investigate the biochemical effects of small-t expression in permissive cells. The use of these adenoviruses has several advantages over other previously used systems, not least of which is the ability to infect virtually all cells in a population. These recombinant viruses allow the expression of small-t, in the absence of the SV40 large-T, in very short periods of time, making it easier to determine that effects seen after infection are directly attributable to the expression of small-t. We demonstrate that expression of small-t during adenovirus infection can elicit established small-t effects, including activation of the MAPK pathway and the AP-1 transcription factor, and induction of cell cycle progression. Furthermore, we show that small-t expression can affect the phosphorylation state of the Na⁺/H⁺ antiporter, an effect dependent on small-t inhibition of PP2A and, in part, activation of the MAPK pathway. Our studies underscore the potential value of these recombinant adenoviruses in exploring additional effects of small-t antigen on signal transduction and cell cycle pathways.

MATERIALS AND METHODS

Cell culture.

African green monkey kidney (CV1) cells and human helper 293 cells were maintained in Dulbecco modified Eagle (DME) medium containing 10% fetal bovine serum (FBS). WT SV40 and the small-t mutant dl888 were propagated and titered in CV1 cells, while helper 293 cells were used to grow recombinant adenoviruses. In both cases, stocks were initiated from low-multiplicity infections. Construction of the recombinant adenoviruses was described previously (42).

Immunoprecipitation of the NHE.

Infected cells were labeled with 1 mCi of [³²P]orthophosphate per ml for 6 h in phosphate-free DME (Adt infections) or overnight in phosphate-free DME containing 50 μM sodium phosphate (SV40 infections). Labeled cells were washed three times and then used to prepare membranes. Washed cell pellets were resuspended in 0.5 ml of hypotonic buffer (20 mM Tris [pH 7.4], 50 mM NaCl) and allowed to swell on ice for 20 min. The cell suspension was passed through a 25-gauge needle 10 times, and nuclei were removed by centrifugation at 2,000 rpm for 10 min at 4°C. The supernate was transferred to an SW50.1 tube (Beckman Industries), brought to a final volume of 4.8 ml with hypotonic buffer, and centrifuged at 40,000 rpm for 2 h at 15°C. The supernate was aspirated, and the membrane pellet was solubilized in 400 to 500 μl of NHE lysis buffer (50 mM HEPES [pH 7.4], 150 mM NaCl, 3 mM KCl, 12.5 mM sodium pyrophosphate, 5 mM EDTA, 10 mM ATP, 1% Thesit [Boehringer Mannheim], 1 mM phenylmethylsulfonyl fluoride, 1 μg each of aprotinin, leupeptin, and pepstatin per ml) for 1 h on ice. After gentle vortexing for 15 s, the solution was transferred to a microcentrifuge tube, and insoluble material was removed by centrifugation. Soluble proteins were reacted with rabbit polyclonal antibody to the NHE (7) and formalin-fixed Staphylococcus aureus Cowan. Immunoprecipitates were washed six to eight times with NHE lysis buffer and then boiled in Laemmli sample buffer before polyacrylamide gel electrophoresis (PAGE) on sodium dodecyl sulfate (SDS)–7.5% polyacrylamide PAGE gels. Quantitation of images was done either by phosphorimaging or by scanning of films with a Molecular Dynamics SI densitometer.

PD98059 (MEK inhibitor).

PD98059 (New England Biolabs) was used at a concentration of 50 μM from a 50 mM stock in dimethyl sulfoxide (DMSO). For cells treated with vehicle alone, 1 μl of DMSO per ml was added to the medium.

Western blotting.

Cells were washed twice with ice-cold phosphate-buffered saline, scraped in a volume of 1 ml of phosphate-buffered saline, and pelleted by brief (20-s) centrifugation at room temperature in a microcentrifuge. The supernatant was aspirated, and the volume of the pellet was estimated. Cells were lysed in cold lysis buffer (50 mM Tris [pH 7.4], 200 mM NaCl, 1 mM EDTA, 1 mM EGTA, 1 mM dithiothreitol, 2% glycerol, 0.5% Nonidet P-40) containing protease and phosphatase inhibitors (0.5 mM phenylmethylsulfonyl fluoride, 10 μg each of leupeptin, pepstatin, and aprotinin per ml, 1 mM NaF, 1 mM sodium orthovanadate). After vigorous vortexing for 15 s, extracts were incubated on ice for 10 to 15 min, and then insoluble material was removed. Equal amounts of total protein were loaded onto SDS-polyacrylamide gels (Laemmli), separated by PAGE, and then transferred to nitrocellulose. After incubation with appropriate primary and secondary antibodies, proteins were visualized by using enhanced chemiluminescence reagents (Pierce Chemical).

The following primary antibodies were obtained from Santa Cruz Biotechnology (specificity and typical dilutions are given in parentheses): sc-154 (ERK2; 1:1,000); sc-474 (JNK1; 1:1,000); sc-093 (ERK1; 1:1,000); sc-437 (MEKK; 1:750); sc-44 (c-Jun; 1:500); and sc-052 (c-Fos; 1:500). Transduction Laboratories antibody 17020/L3 (MEK1; 1:2,000) was also used. The monoclonal antibody PAb419 (22) was used at a dilution of 1:20 for detection of small-t.

Immunoprecipitation kinase (IP Kinase) assays.

Cells were extracted in cold lysis buffer containing 0.1% SDS. Equal amounts of protein (typically 400 to 800 μg) were used in all experiments, and extracts were incubated with saturating amounts of the antibodies described above. Immunoprecipitates were collected by using protein A-agarose (Santa Cruz Biotechnology) and then washed once in lysis buffer, twice in 100 mM Tris (pH 8.6)–500 mM LiCl, and once in kinase buffer (50 mM HEPES [pH 7.4], 10 mM MgCl₂, 10 mM MnCl₂, 1 mM DTT). Immunoprecipitates were resuspended in 40 μl of kinase buffer containing approximately 50 μg of the appropriate substrate per ml and 15 μCi of [γ-³²P]ATP (3,000 Ci/mmol, Amersham) and then incubated at 25°C for 20 min. Reactions were stopped by the addition of SDS sample buffer, and the mixture was boiled for 5 min. Substrates included myelin basic protein (MBP; Sigma) to assay MAPK, histidine-tagged kinase-inactive p42 ERK1 for MEK (MAPK^K/R) (48), histidine-tagged, kinase-inactive MEK (His₆-MEK) (29, 30) for MEKK and Raf, and glutathione S-transferase (GST)–Jun^1-135 fusion protein (a gift from Richard Pestell, Albert Einstein College of Medicine) for JNK1.

Electrophoretic mobility shift assays.

Nuclear extracts were made essentially as described previously (55). Protein concentration was determined by the Bradford method (Bio-Rad). A double-stranded, end-labeled oligonucleotide (5′-TCGAGA TGAGTCA GGTGA TGAGTCA GC-3′) containing the two tandem AP-1 sites (boldface) from the c-jun promoter was used as a probe. For binding reactions, 3 to 5 μg of nuclear extract was incubated with 1 to 5 fmol (50,000 to 70,000 cpm) of labeled probe and 5 μg of poly(dI-dC) (Boehringer Mannheim). Complexes were separated by using nondenaturing 5% acrylamide. For supershift analysis, 2 μl of anti-c-Jun antibody (Santa Cruz sc-44X) was added to diluted extracts and incubated for 45 min on ice prior to the addition of the labeled probe–poly(dI-dC) mix.

Cell cycle analysis by flow cytometry.

Confluent CV1 cells were removed from 6-cm-diameter tissue culture plates by trypsinization and pelleted by centrifugation for 10 min at 2,000 rpm. At 30 h postinfection (p.i.), cells were trypsinized and collected by centrifugation; then cells were lysed and stained in a solution containing 4 mM sodium citrate (pH 7.8), 100 μg of propidium iodide (PI) per ml, 0.5 mg of RNase per ml, 0.1% Triton X-100, and 3% polyethylene glycol 8000. After 20 min at 37°C, an equal volume of hypertonic salt solution (0.35 M sodium chloride, 100 μg of PI, 0.1% Triton X-100, and 3% polyethylene glycol 8000) was added. Stained nuclei were held overnight at 4°C, passed through nylon filters, and then analyzed on a Becton Dickinson flow cytometer using CellQuest software.

RESULTS

Recombinant adenoviruses used to study small-t.

To directly assess functions of small-t in the absence of large-T, we used defective adenoviruses that expressed WT or mutant small-t under the control of the cytomegalovirus promoter (42, 62). By 8 to 12 h after infection with Adt, levels of small-t present were similar to levels found in SV40-infected CV1 cells at 48 h p.i., a time that is about midway through the permissive infection (Fig. 1A). Adenoviruses expressing mutant alleles of small-t were also constructed. Ad97 and Ad103 express small-t with Ser in place of the Cys residues at positions 97 and 103, respectively. Both of these forms of small-t have greatly reduced ability to bind and inhibit PP2A (37). Ad43/45 expresses a small-t in which Pro43 and Lys45 are replaced by Leu and Asp, respectively. Sequences in this region of small-t have no effect on PP2A binding but reduce DnaJ-like activity. All mutant forms of small-t were stable and were expressed at about the same level as the WT protein (Fig. 1B), consistent with earlier findings with these mutant proteins (37, 42).

FIG. 1.

Expression of WT and mutant small-t by adenovirus vectors. (A) CV1 cells were infected with Adt (20 PFU/cell) and then extracted after 8 or 12 h. For comparison, CV1 cells were infected with SV40 (WT) or the small-t deletion mutant dl888 (DL) (each at 20 PFU/cell) and then extracted after 48 h, the peak time in SV40 productive infection. Levels of small-t in these extracts were determined by analyzing 50 μg of total protein on SDS-gels, followed by transfer to nitrocellulose and detection of small-t with monoclonal antibody PAb419 and peroxidase-conjugated rabbit anti-mouse serum. (B) CV1 cells were infected for 8 h with Adt, Ad43/45, Ad97, or Ad103 (each at 20 PFU/cell), and then extracts were made and used to compare levels of WT and mutant proteins by Western blot analysis. Bacterially expressed small-t (Bact. t) is shown as a marker.

Phosphorylation of the Na⁺/H⁺ antiporter during SV40 infection.

To determine the effect of small-t on antiporter phosphorylation during SV40 infection, the antiporter was immunoprecipitated from whole-cell extracts of CV1 cells that had been infected with various multiplicities of the small-t deletion mutant dl888 (T⁺/t⁻) or WT SV40. Infected cells were labeled with [³²P]orthophosphate from 36 to 48 h p.i. In the experiment shown in Fig. 2A, the level of antiporter phosphorylation in dl888-infected cells was similar to the basal phosphorylation seen in mock-infected cells, although a slight increase in antiporter phosphorylation could be detected at higher multiplicities. In cells infected with WT SV40, which expresses small-t, the level of antiporter phosphorylation was consistently higher than that seen in mock-infected cells. Thus, the presence of small-t results in phosphorylation of the antiporter to a level that is not attainable by large-T alone. The increase in phosphorylation was not attributable to increased antiporter expression in WT-infected cells, as both dl888- and WT-infected cells had similar levels of antiporter protein, detected by Western blot analysis (Fig. 2A, inset). Phosphorylation of the antiporter in response to serum is shown as a positive control but cannot be directly compared to those of infected cells because of differences in the labeling periods.

FIG. 2.

Small-t increases phosphorylation of the Na⁺/H⁺ antiporter. (A) Confluent CV1 cells were serum starved for 48 h and then treated with mock lysate (M) or infected with 50, 20, or 5 PFU of WT SV40 or dl888 per cell. Cells were labeled from 36 to 48 h p.i. with [³²P]orthophosphate, and then the antiporter was solubilized from partially purified membranes and immunoprecipitated as described in Materials and Methods. As a positive control, cells were stimulated with 10% serum for 30 min (S⁺) and then processed for immunoprecipitation. Levels of phosphorylation in this control cannot be directly compared to those in infected cell extracts because of differences in labeling periods. A Western blot for total antiporter protein in WT- and dl888 (DL)-infected cell extracts is shown in the inset to the right. The arrow indicates the position of the antiporter at 115 kDa in the immunoprecipitates and the Western blot. (B) Serum-depleted CV1 cells were infected with various recombinant adenoviruses (each at 20 PFU/cell) for 2 h and then incubated in ³²P-containing medium until 8 h p.i. The Na⁺/H⁺ antiporter was immunoprecipitated from partially purified membranes as described in Materials and Methods. Levels of ³²P-labeled antiporter in Adt-infected cells were compared to those found following mock infection (M), infection with the control virus AdE4 (E4), or infection with Ad97, which expresses the C97S mutant small-t. As a positive control, cells were stimulated with 10% serum for 30 min (S⁺).

Phosphorylation of the Na⁺/H⁺ antiporter during Adt infection.

To determine if small-t could affect antiporter phosphorylation in the absence of large-T, serum-starved CV1 cells were infected with Adt, control virus AdE4, or Ad97 and labeled with [³²P]orthophosphate throughout the 8-h infection period; then the antiporter was immunoprecipitated from partially purified membranes. As shown in Fig. 2B, cells infected with AdE4 showed little phosphorylation of the antiporter above the basal level seen in mock-infected cells. Antiporter isolated from Adt-infected cells, however, showed a dramatic increase in phosphorylation. The ability of small-t to affect antiporter phosphorylation depended on the ability of small-t to bind to and inhibit PP2A, as little or no increase in phosphorylation was seen during infection with Ad97. Again, the response to serum is shown as a positive control but cannot be directly compared to responses of Adt-infected cells because of differences in the labeling periods.

Activation of MAPK by Adt infection.

The increase in antiporter phosphorylation during Adt infection is likely to reflect activation of growth-regulated kinases by small-t. A key enzyme involved in cell activation is MAPK, a phosphoprotein and known substrate for PP2A in vitro. Increased MAPK activity has been observed in cells 72 h after transfection with a small-t-expressing plasmid (58). To determine if expression of small-t during Adt infection could activate MAPK, we infected serum-starved CV1 cells for 8 or 24 h with either AdE4 or Adt and determined the MAPK activity in the infected cells by an IP kinase assay (Fig. 3A). The MAPK activity in cells infected with AdE4 for 8 h was identical to mock levels, but significant activation occurred in Adt infections. In the example shown here, scanning densitometry showed a 3.2-fold increased phosphorylated MBP at 8 h p.i. Activation of MAPK by small-t was seen as early as 5 h p.i. (data not shown) and was still apparent at 24 h p.i., although levels of MAPK activity had declined in both Adt- and AdE4-infected populations. Activity of MAPK at 8 h p.i. was similar to that found 15 min following addition of 10% serum (Fig. 4B).

FIG. 3.

Activation of MAPK and JNK in Adt-infected cells. CV1 cells were mock infected (M) or infected with recombinant adenoviruses (each at 20 PFU/cell) and then extracted at various times p.i. to determine activities of the intracellular kinases, MAPK and JNK. (A) Cells were infected with AdE4 (E4) or Adt (t) for 8 or 24 h prior to extraction. MAPK was immunoprecipitated as described in Materials and Methods and then assayed with MBP as a substrate. (B) Cells were infected with adenoviruses that express mutant small-t for 8 h and then extracted and immunoprecipitated to determine MAPK activity as described for panel A. As a positive control, cells were stimulated with 10% serum for 15 min before analysis (S⁺). (C) Infected or serum-stimulated cells were extracted and immunoprecipitated with antibody to JNK, which was then assayed with GST-Jun^1-135 as a substrate.

FIG. 4.

Effects of small-t antigen on MEK, Raf-1, and MEKK. CV1 cells were mock infected (M) or infected with recombinant adenoviruses (each at 20 PFU/cell) and then extracted at 8 h p.i. (A) Extracts were immunoprecipitated with antibody to MEK, and then precipitates were analyzed with kinase-inactive MAPK (MAPK^K/R) as a substrate. As a positive control, cells were stimulated with 10% serum for 10 min before extraction (S¹⁰). (B) MEK activity was inhibited by incubating infected cells in the presence of 50 μM PD98059 throughout the 8-h infection period. The effect of this treatment on MAPK activity was determined by immunoprecipitation of MAPK for analysis of its kinase activity. (C) IP kinase assays were performed with antibodies to Raf-1 or MEKK, and precipitates were assayed with His-MEK as the substrate. The asterisk denotes the position of autophosphorylated Raf-1.

Adenoviruses that expressed mutant forms of small-t were also assayed for the ability to induce MAPK activity (Fig. 3B). The double-point mutations at positions 43 and 45 did not affect PP2A binding or inhibition, and expression of this mutant protein resulted in a fourfold increase in MAPK activity, similar to that elicited by WT small-t. In contrast, mutations at position 97 or 103, which greatly reduce binding to PP2A, abolished the ability of small-t to activate MAPK. Thus, activation of MAPK by small-t antigen requires inhibition of PP2A.

Based on the similarity between regulation of JNK and MAPK activity, it seemed possible that small-t could activate the JNK pathway in addition to the MAPK pathway. JNK activity was determined by using an IP kinase assay with GST-Jun^1-135 as a substrate. JNK activity was activated but less dramatically than MAPK by Adt infection (Fig. 3C). The 2-fold activation of JNK by small-t and the reduced response to serum (1.5-fold) were consistent and seen in at least three separate experiments.

Activation of MEK by small-t.

MAPK is activated by phosphorylation on both Thr and Tyr residues by the dual-specificity kinase MEK, which is also a substrate for PP2A (2, 21, 73). Thus, the ability of small-t to inhibit PP2A may result in activation of MEK, which in turn would activate MAPK. Using the adenovirus vectors described above, we tested the abilities of WT and mutant small-t antigens to activate MEK. MEK activity in infected cells was tested 8 h p.i. by IP kinase assay using a kinase-inactive form of MAPK purified from bacteria as a substrate. As with MAPK, MEK was activated by WT small-t as well as the double-point mutant Ad43/45 (Fig. 4A). No activation was seen in cells infected with control virus (AdE4) or either of the viruses expressing the PP2A-binding mutants of small-t (Ad97 and Ad103).

Inhibition of PP2A by small-t could activate MAPK directly by inhibiting its dephosphorylation or indirectly by activating MEK. To determine which mechanism is responsible, the effect of PD98059, a synthetic inhibitor of MEK activation (3, 18), was determined. PD98059 inhibits activation of MEK by binding unphosphorylated MEK and preventing its phosphorylation. The inhibitory effect of PD98059 is relatively specific for MEK, as several other enzymes, most notably SEK, were unaffected by the drug (3, 18, 40). DME containing either 50 μM PD98059 or the vehicle DMSO was added to cells at the end of the infection period and was present until the cells were extracted at 8 h p.i. Data were collected and analyzed by phosphorimaging. In the example shown in Fig. 4B, MAPK activity was increased 3.3-fold in cells infected with Adt in the presence of DMSO alone. In contrast, cells maintained in PD98059-containing medium showed lower basal levels of MAPK, and these were not increased by Adt infection. The concentration of PD98059 used and length of treatment were not toxic to CV1 cells, as pretreatment of cells with 50 μM PD98059 for 6 h followed by extensive washing did not prevent subsequent activation of MAPK by Adt infection (data not shown). These studies suggest that the increased MAPK detected in Adt-infected cells requires MEK activity and that MAPK is not directly activated by the presence of small-t.

Involvement of MEK in antiporter phosphorylation.

To determine whether activation of MAPK was required for small-t to increase antiporter phosphorylation, mock- or Adt-infected cells were labeled with [³²P]orthophosphate in the presence or absence of 50 μM PD98059. The ability of Adt to increase antiporter phosphorylation was reduced 60 to 70% in cells treated with 50 μM PD98059 (Fig. 5), suggesting that the increase of antiporter phosphorylation by small-t is carried out, at least in part, through activation of MEK and the MAPK pathway.

FIG. 5.

Inhibition of MEK decreases small-t-mediated antiporter phosphorylation. Confluent, serum-starved CV1 cells were infected with recombinant adenoviruses (each at 20 PFU/ml) and then placed in medium containing [³²P]phosphate (1 mCi/m) in the presence or absence of 50 μM PD98059. At 8 h p.i., cells were extracted and the antiporter was immunoprecipitated from partially purified membranes as described in Materials and Methods. Immunoprecipitated proteins were separated by SDS-PAGE and visualized on a Molecular Dynamics Phosphorimager SI. The radioactivity in the band corresponding to the antiporter was quantitated for each sample and compared to radioactivity present in control infections in which uninfected cells were incubated with or without PD98059.

Small-t does not activate Raf-1 or MEKK.

Upstream activators of MAPK and JNK are Raf-1 and MEKK, respectively. These enzymes were tested by IP kinase assays with appropriate antibodies and with His₆-MEK as a substrate for either enzyme. Analysis was done by phosphorimaging. Autophosphorylation of Raf-1, manifest by a second radiolabeled band migrating above the substrate band, occurred during these assays. Serum stimulation induced Raf-1 kinase activity, but no activation was seen 8 h after Adt infection (Fig. 4C). This finding, along with the observation that inhibition of MEK ablates MAPK activation by small-t, suggests that small-t activates the MAPK kinase cascade primarily through activation of MEK. MEKK was also unaffected by small-t expression (Fig. 4C), suggesting that the activation of JNK by small-t is through an effector downstream of MEKK.

PP2A binding by small-t antigen is required for induction of AP-1.

The recombinant adenoviruses were also used to examine more downstream effects that often result from activation of the MAPK pathway. It was of interest to examine levels of the c-Fos protein because of a recent report that c-fos transcription may actually be decreased by small-t antigen (68). Contrary to this report, cells infected with Adt showed increased c-Fos protein levels at times that paralleled activation of MAPK (Fig. 6A).

FIG. 6.

Induction of c-Fos protein levels and AP-1 activity in Adt-infected cells. Confluent, serum-starved CV1 cells were mock infected (M) or infected with Adt or AdE4 for 5 or 8 h and then used to prepare nuclear extracts. (A) Levels of c-Fos protein were determined by Western blotting and compared to levels found in cells that had been stimulated with 10% serum for 60 (S⁶⁰) or 90 (S⁹⁰) min. (B) Nuclear extracts were prepared and incubated with a 27-residue oligonucleotide that contained two tandem AP-1 sites from the c-jun promoter as described in Materials and Methods. Extracts of cells infected with Adt and AdE4 were compared to extracts of cells that had been serum stimulated for 30 min (S⁺). In lane α, antibody to c-Jun was shown to disrupt complex formation.

It has also been reported that small-t can increase AP-1-driven transcription in microinjected cells, especially when plasmids that encode small-t are coinjected with plasmids that express MAPK itself. Serum-starved CV1 cells had low basal levels of AP-1 binding activity which increased four- to fivefold 30 min after serum stimulation (Fig. 6B). The identity of the shifted complex was confirmed by preincubating extracts with an anti-c-Jun antibody which disrupts AP-1 binding and by competition of probe binding with unlabeled AP-1 oligonucleotide (data not shown). A dramatic increase in AP-1 binding activity was seen 8 h after Adt infection. Some of this increase may be due to binding of adenovirus virions or to a cellular response to adenovirus infection, as a slight increase was also seen after AdE4 infection. However, the increase in AP-1 binding activity in response to small-t expression was far greater than that seen in response to AdE4 infection or to serum stimulation.

Adenoviruses expressing mutant small-t were also assayed for the ability to induce AP-1 binding. The increase elicited by WT small-t was abrogated by mutation of Cys97 or Cys103, both of which affect PP2A binding (not shown). This finding is in agreement with results of microinjection studies that suggested that PP2A binding by small-t was required to drive expression from AP-1-responsive promoters (20). The double mutation at residues 43 and 45 had no effect on the induction of AP-1 DNA binding activity (not shown).

PP2A binding, but not the DnaJ region, of small-t is required for induction of cell proliferation.

To determine whether infection with Adt could result in active cell cycle progression, cells were analyzed by flow cytometry (Fig. 7). It is difficult to completely arrest the growth of CV1 cells. In these experiments, cells were grown to confluence and then held in serum-free medium for 48 h before infection. Mock-infected cultures treated in this fashion showed significant arrest, with the majority of cells showing a single peak of DNA fluorescence typical of cells in G₀/G₁. In contrast, by 30 h after infection with WT Adt, over half the infected cells had exited G₁. Growth induction required the PP2A binding ability of small-t. Cells infected with Ad103 did not progress through the cell cycle, and the proportion of cells remaining in G₁ (79%) was nearly identical to that seen in uninfected cells (82%). Mutations in the DnaJ domain did not affect this small-t function, and Ad43/45 was fully competent for induction of cell cycle progression.

FIG. 7.

Induction of cell cycle progression by small-t. CV1 cells were either mock infected or infected with 20 PFU of Adt, Ad43/45, or Ad103 per cell. At 30 h p.i., cells were trypsinized, and nuclei were prepared for flow cytometric analysis as described in Materials and Methods. The fluorescent dye PI binds DNA in proportion to its concentration. In these profiles, cells with resting, diploid DNA content (G₁) were found in the tallest peak on the left side of each tracing and the percentage of cells in G₁ is shown for each population.

DISCUSSION

Interaction with PP2A is a principal mechanism through which SV40 small-t exerts its biological effects. While small-t can reduce PP2A activity, potential modification of substrate selectivity and/or intracellular localization may also be an important element in small-t function, as it is believed to be for various B subunits (17, 25). The identification of in vivo targets affected by small-t expression is crucial to understanding its helper function. Our work along these lines was greatly facilitated by the use of adenovirus vectors expressing WT and mutant small-t. Adenovirus vectors have become increasingly useful biological tools because of their high virus yields, allowing infection of large numbers of target cells, and rapid expression of recombinant proteins. In this report, we have documented the small-t-dependent increase in phosphorylation of the Na⁺/H⁺ antiporter in cells infected for 48 h with SV40. With recombinant adenoviruses, such phosphorylation is apparent within 8 h after infection. Similarly, we have confirmed findings of others for transfected cells that small-t, through its PP2A-binding activity, results in increased MAPK and MEK activities (58). We have extended these studies to show modest increases in JNK1 as well. The overall timing of our experiments reinforces the belief that small-t is an immediate effector and that cascades that develop only over long periods may not be responsible for increased kinase activities.

While key to the growth process, activation of the MAPK pathway alone does not necessarily lead to cell division, especially when kinase activity is sustained rather than transient (32). Thus, other intracellular targets must be considered as necessary to drive cell cycle progression in response to small-t. The present work implicates the Na⁺/H⁺ antiporter as an excellent candidate in this regard. A significant literature has shown that activation of the Na⁺/H⁺ antiporter, resulting in cytoplasmic alkalinization, is a prerequisite for the activation of quiescent cells, and early studies suggested that such alkalinization may even be sufficient for S-phase progression (9, 36). Certainly, cell growth and DNA synthesis are restricted below a certain threshold intracellular pH (28, 43). In addition, disruption of antiporter function, through mutation (44, 49) or pharmacological inhibition (28, 64), precludes the growth of normal cells at neutral to acidic pH (44), prevents mitogenic signal-induced gene expression (64), and presents cell cycle progression (28).

The Na⁺/H⁺ antiporter is likely to be a key target of small-t’s effects on PP2A in permissive cells, i.e., cells that support viral DNA replication. This possibility is suggested by the ability of small-t to overcome growth-inhibitory effects of theophylline, an inhibitor of antiporter function. Other potential targets in permissive cells include the viral large-T itself, a molecule that is both positively and negatively regulated by phosphorylation and is a known substrate of PP2A (19, 54). Interestingly, PP2A was shown to stimulate SV40 DNA replication in a purified in vitro system (65). While this at first seemed contradictory, because small-t enhances viral DNA replication in vivo yet inhibits PP2A in vitro, it was argued that G₁ progression driven by small-t could far outweigh any inhibitory effects (13). Alternatively, small-t might target PP2A to specific substrates that play a role in regulating DNA replication in vivo. Evidence for altered intracellular localization of PP2A by small-t was suggested recently in studies of microtubule-associated enzyme (59).

It will be equally critical to identify cellular targets of PP2A modified by small-t in nonpermissive systems. Given the wealth of activities stimulated by large-T, it is remarkable that small-t plays such a significant role in many nonpermissive cell systems. It is likely that the importance of small-t in these systems will be found to be attributable to its ability to regulate biochemical targets over which large-T has no direct control, such as MAPK and the Na⁺/H⁺ antiporter. Although not formally demonstrated, activation of the Na⁺/H⁺ antiporter is likely to have effects in nonpermissive cells that are as pronounced as those in permissive cells. It would be even more important to demonstrate that such activation was required for cell cycle progression or, ultimately, transformation of a particular nonpermissive cell type.

Effects of small-t explored in this study focus on early events in cell cycle progression, an area in which large-T has no apparent mechanism of action. It is interesting to speculate that the cellular response to small-t expression might mimic the normal steps followed in progression through the cell cycle and thereby balance potentially apoptotic signals generated by the effects of large-T. For example, such apoptotic signals might ensue in a cell that has encountered enhanced E2F activity resulting from large-T sequestration of Rb but has failed to follow a normal sequence of earlier, i.e., pre-S-phase, events in cellular activation (27, 45).

The joint requirement for small-t and large-T is well established in several nonpermissive cell systems. For example, neither small-t nor large-T alone stimulates cell cycle reentry by density-arrested human diploid fibroblasts, and both viral proteins must be present for cell cycle progression to occur (53). In a second system, induction of anchorage-independent growth in rat F111 cells requires both large-T and small-t (37). It has recently been established that mitogenic activation of the MAPK pathway is inefficient in nonadherent cells (31, 35, 46). Thus, it will be of interest to determine the ability of small-t to induce known biochemical effects (e.g., activation of MAPK, Na⁺/H⁺ antiporter, or other enzymes) in nonadherent conditions or density arrest of cell types such as primary human fibroblasts. The use of adenovirus constructs that individually express the large-T and small-t, and various mutants of each protein, will be a valuable method for analyzing specific cellular functions altered in growth promotion and the initial stages of transformation in these more strictly growth controlled cell types.