To determine the role of human 14-3-3γ (h14-3-3γ) in cell cycle regulation, we generated transgenic flies with h14-3-3γ cDNA inserted into a P-element vector and driven by a heat shock promoter. The h14-3-3γ transgene copy number was increased by crossing two independent lines as described in the Methods section to generate the HS1433GA/GC line. We first examined the HS1433GA/GC line for molecular evidence of h1433γ expression by performing RT-PCR and Western analyses (Figure 1). The transgenic line HS1433GA/GC showed h14-3-3γ mRNA expression when heat shock was applied to third instar larvae. Applying the heat shock treatment two times produced the most abundant expression of h14-3-3γ (Figure 1A; Lanes 3–5). No h14-3-3γ mRNA was observed in the parental yw^67C2 animals (Figure 1A; Lane 1). However, we did observe faint expression of h14-3-3γ in those transgenic lines without a heat-shock treatment which we interpreted as expression due to leakiness of the heat shock promoter which has also been observed by others 28.

We next optimized the conditions used for induction of h14-3-3γ protein. We found that a one hour heat shock treatment consistently resulted in robust induction of h14-3-3γ protein, whereas with a 30 minute heat-shock the amount of protein expressed was weak (Figures 1B &1C). Importantly, heat shock had no effect on expression of the endogenous Drosophila ε and ζ 14-3-3 genes (data not shown). Recovery time was also an important determinant for maximizing the h14-3-3γ protein expression. We found that h14-3-3γ protein expression was most highly elevated 1–3 hours after a one hour heat shock treatment (Figures 1B; Lane 6 &1C; Lanes 4–5). Neither the exogenous h14-3-3γ protein nor endogenous 14-3-3 protein could be detected in the control flies. This may be caused by the fact that only human 14-3-3γ protein could be detected in the transgenic line by using a pan-specific 14-3-3 antibody, which was raised against human 14-3-3β protein (Figure 1). Since the levels of h14-3-3γ protein declined to near background 4.5 hours after treatment (Figure 1C; Lane 5) all animals exposed to heat shock were examined within 1–3 hours after treatment.

In human lung cancer cell lines, h14-3-3γ may interfere with normal cell cycle progression 23. To determine whether overexpression of 14-3-3γ had any effect on cell proliferation we examined the differentiating neuronal cells in the posterior compartment of the developing eyes in HS1433GA/GC larvae that had been heat shocked. BrdU incorporation was used to identify S-phase cells and immunostaining with anti-Elav antibody marked differentiating neuronal cells. We found BrdU-incorporated cells in the anti-Elav (neuronal cell marker) staining positive regions in HS1433GA/GC imaginal discs after a heat shock treatment (Figure 2C). No replicative cells were seen in the Elav-staining regions of eye discs from animals that were not heat shocked or in the control animals (Figures 2A &2B). The number of replicative cells, determined through BrdU incorporation, in the anti-Elav positive region was quantitated in control and transgenics and the results presented in Table 1. The measurements showed that the average number of S phase cells in the Elav positive region of HS1433GA/GC imaginal discs increased significantly when the larvae were heat shocked compared to animals from the same line but not heat shocked. There was small increase in BrdU-incorporating cells in HS-h1433γ discs without heat shock compared to yw^67C2 control flies. This is probably due to leakiness of the Hsp70 promoter (Figure 1A; Lane2). Our data shows that overexpression of h14-3-3γ promotes abnormal cell proliferation in differentiating tissue and that the effect is dose dependent.

Our experiments with the Hsp70 promoter-driven h14-3-3γ gene suggested that the overexpression of the 14-3-3γ resulted in aberrant cellular proliferation. To confirm our results we made additional flies in which h14-3-3γ expression was specifically targeted to behind of the morphogenetic furrow of posterior compartment of 3^rd-instar eye imaginal discs using a GMR (Glass Multiple Reporter)-Gal4 29 driver. Transgenic UAS-h1433 γ (on 2^nd chromosome, #15D) flies were created as described in the Methods section. Crossing Gal4 flies with UAS-h1433 γ flies induced expression of h1433γ in eye imaginal discs.

To mark the compartment posterior to the morphogenetic furrow in the 3^rd-instar larval eye imaginal discs a GFP (Green Fluorescent Protein) reporter gene was introduced which also responded to the GMR-Gal4 driver. To induce expression of h14-3-3γ, females containing the GMR-Gal4 driver carrying with h14-3-3γ were crossed with male UAS-GFP flies (see Methods section) to create the yw; GMR-Gal4, UAS-h1433γ/UAS-GFP genotype. We first confirmed that h14-3-3γ expression was occurring using RT-PCR and Western blotting to detect expression (Figure 3). A PCR product of the predicted size of human 14-3-3γ mRNA was detected in these larvae indicating the presence of human 14-3-3γ mRNA (Figure 3A; Lane 2), and no product was detected in the control (Figure 3A; Lane 1). Moreover, examination of h14-3-3γ protein levels showed that the exogenous human protein was expressed in the yw; GMR-Gal4, UAS-h1433γ/UAS-nlsGFP transgenics (Figure 3B; Lane 2), and not in the controls.

We next examined the eye imaginal discs of yw; GMR-Gal4, UAS-h1433γ/UAS-nlsGFP 3^rd – instar larvae. As in the previous experiments, BrdU incorporation was used to detect replicating cells. GFP expression marked activity of the GMR-Gal4 driver (Figure 4). In Figure 4 panels b and e, which depict BrdU incorporation, the replicating cells marking the second mitotic wave are clearly visible as a band of cells laid out across the center of the imaginal disc. Notably, the apparent width of the SMW is greater in eye discs from flies expressing h14-3-3γ Figure 4B, f). This suggested that there is an increase in the(number of replicating cells within this region. Consistent with this, comparison of the merged images (which showed the relationship between the replicating cells in the second mitotic wave and GFP expression) indicates that replication ceases prior to the onset of GFP expression in control flies (Figure 4A). In contrast, GFP expression invades the band of replicating cells of the second mitotic wave in h14-3-3γ expressing animals (Figures. 4B, d–f). Taken together, these data suggest that h14-3-3γ overexpression may increase the number of proliferative cells or prolong the proliferative phase of cells in the developing fly eye resulting in a wider and more intense band of BrdU incorporating cells.

We and others have suggested that 14-3-3 proteins may regulate entrance into mitosis by regulating activity of the Cdc25 phosphatase 1730, Hence, we sought to determine whether the h14-3-3γ expression had any effect on the onset of mitosis. Consequently we examined the occurrence of mitosis in eye imaginal discs of flies expressing GMR-Gal4-driven h14-3-3γ. Eye imaginal discs were collected from third instar larvae and immunostained for phospho histone H3 (PH3) and the samples examined using confocal microscopy to detect PH3 and GFP (Figure 5). As can be seen mitotic cells are apparent near the morphogenetic furrow of eye imaginal discs from control animals (Figure 5A, b), but that this band of mitotic cells is markedly reduced in eye discs where h14-3-3γ is overexpressed (Figure 5B, e). This suggests that h14-3-3γ can suppress entrance into mitosis in a manner similar to that observed in animal cells 30.

To further examine the mechanism of h14-3-3γ activity we decided to overexpress h14-3-3γ in a genetic background with compromised Cdc25 (String). To this end, we introduced the heat shock inducible h14-3-3γ gene into a genetic background containing the String hypomorphic allele, Stg^9A and then induced h14-3-3γ using the heat shock protocol described previously 31. In the double mutant we found 8.96% (SEM: ± 1.27, 21 out of total 257 flies) of adult eyes had a rough eye phenotype (Table 2). However, in the absence of Stg^9A we found only 1.61% (SEM: ± 1.03, 2 out of total 211 flies) displaying the phenotype. By comparison only 1.39% (3 out of total 191 flies) of the parental background (yw^67C2) flies showed the rough eye phenotype. Hence, presence of the hypomorphic String allele appears to increase sensitivity to the effects of h14-3-3γ suggesting that h14-3-3γ may interact with Cdc25 in cell cycle regulation.

Transgenic flies, yw; UAS-h1433γ and hs-h1433γ lines were generated using a cloned human 14-3-3γ cDNA 23 into P-element vectors such as pUASP 33 and pCaSpeR-hs 34, respectively.

For ubiquitous h14-3-3γ protein expression, we generated two transgenic lines that express gamma using a P-element vector in which the h14-3-3γ gene was driven by the Hsp70 heat shock promoter and could be activated by heat shock. Two independent lines were generated, HS1433GA (on X chromosome) and HS1433GC (on 3^rd chromosome). To increase gene copies (more than 2) of h14-3-3γ progeny, these flies were crossed with each other and selected using eye color as a marker for gene dosage. A stock homozygous for 14-3-3γ on the X and 3^rd chromosome was established from this cross and has remained stable for an extended period of time.

Stock flies with GMR-Gal4 driver were crossed with 8 UAS-h14-3-3γ transgenic lines for initial screening, and we found all lines to be similar. For the experiments in this study, male flies of h14-3-3γ on the 2^nd chromosome (P33#15D) were crossed with GMR-Gal4 (on 2^nd chromosome) female flies to generate recombinant flies for the experiments and scored for eye phenotype. Recombinant flies bearing both UAS-h14-3-3γ and GMR-Gal4 on the second chromosome were balanced with CyO balancer chromosome. Virgin female homozygous GMR-Gal4, UAS-h14-3-3γ flies were selected to cross with homozygous male UAS-GFP (2^nd chromosome) flies (Figures 4 and 5).

Using those increased dosage of hs-1433γ flies,. we crossed on h14-3-3γ transgenes into a String mutant genetic background. String is a homologue of Cdc25, and the allele of Stg^9A (a kind gift from Dr. Patrick O'Farrell) is known to be temperature-sensitive. On the 3^rd chromosome, Stg^9A is balanced with TM3 (third multiple 3) having Sb^' (Stubble) as a dominant marker 35. To examine eye phenotypes in adults expressing h14-3-3γ in Stg^9A heterozygotes, we treated the flies with multiple heat-shocks (a 30 min heat pulse at 37°C every 7 and a half hours) starting from 3^rd-instar larval stage.

RT-PCR was performed using total RNA extracted from adult flies or 3^rd-instar larvae. For the total RNA extraction, a FastRNA Pro Green kit (Qbiogene) was used and followed the manufacturer's instructions. For Reverse Transcription (RT), a mixture of Oligo-dT, dNTPs (10 mM), RNA (5 μg), and DEPC- ddH2O was incubated for 5 min at 6°C. The RT contents were collected at the bottom by centrifuging, and 5 × buffer, 0.1 M DTT and RNAse inhibitor were added. After they were incubated for 2 minutes at 42°C, 1 μl Superscript II RT was added. For PCR reactions, a mixture of 41 μl Platinum Supermix (Invitrogen), 3 μl DMSO, 10 mM dNTP, DNA Digest 1 (1 μg of Total RNA, 10× DNAase free buffer, DNAase, ddH2O up to 10 μl) and 10 μM each forward and reverse primer for 1433γ cDNA were added (Forward, CTGAATGAGCCACTGTCGAA; Reverse, CACACAGCCTCCAACTCCTT). Drosophila ribosomal protein 49 encoding gene (dRP49) was used as a loading control for the PCR reactions. The primer sequences to flank dRP49 were 5'-GTGTATTCCGACCACGTTACA (RP49-antisense) and 5'-TCCTACCAGCTTCAAGATGAC (RP49-sense).

Cell lyses were performed on ice by using 3^rd-instar larval imaginal discs in RIPA buffer. Total protein 50 μg per lane was loaded on a SDS/PAGE gel. The protein bands were transferred onto Nitrocellulous membranes (BioTrace NT, Pall Corporation) in transfer buffer (89.3 g glycine, 19.3 g Tris, 1.6 L Methanol, ddH₂O up to 8 L) for overnight on a Transphor Unit (Amersham Biosciences, Cat # 80-6205-97). Then, a primary antibody, mouse anti-14-3-3β antibody (pan-specific antibody detecting all human 14-3-3 isoforms, Santa Cruz, Cat # SC1657), was used at a 1:100 dilution at 4°C for overnight. A goat anti-mouse secondary antibody was diluted 1:5000 in PBST (1 × PBS + 0.1% Tween 20) buffer with 5% Non-Fat Dry Milk, and the membrane was incubated in the secondary antibody for 2 hrs at room temperature. Using an ECL detection kit (Pierce), specific protein bands were detected onto X-ray films (Kodak) using an autoradiographic machine (Konica SRX-101A). The membranes were stripped in 0.1 M NaOH for 5 min at room temperature and treated with a loading control anti-beta-actin antibody(AbCam).

Mature 3^rd-instar larvae were collected from controls (Wild Types) and 1433γ overexpressing animals. Heat-shock treatment was performed for 1 hr at 37°C in a water bath. After the heat-shock treatment, 1 hr of recovery time was given. For fixation, imaginal discs were dissected and immediately treated with 5% paraformaldehyde for 30 min. Tissues were washed twice for 10 min. in PBST (1 × PBS, 0.1% Triton ×-100). For a blocking step, 1% Normal Goat Serum (NGS) was added to PBST and treated for 30 min. A neuronal cell marker, rat anti-elav antibody was diluted 1:200. An anti-pH3 (phosphorylated Histone H3 at Serine 10, Upstate Biotechnology) antibody was used at 1:250 dilution.

The primary antibodies were added and incubated for overnight at 4°C. The antibody rat anti-elav was obtained from the Developmental Studies Hybridoma Bank at the University of Iowa. FITC- or TRITC-labeled secondary antibodies (KPL) were treated for 2 hr at room temperature. After washing in PBS-T and PBS-BT (PBST + 0.5% BSA), the tissues were mounted in Mowiol mounting medium (Calbiochem, Cat # 475904).

The BrdU (Bromodeoxyuridine) incorporation was performed in 1 × PBS with 5 μg/ml BrdU for 1 hr at room temperature with gentle shaking on Nutator. Imaginal discs were fixed in 5% paraformaldehyde for 30 min, and washed for 5 min 3× in PBST (1 × PBS, 0.3% Triton ×-100). Then, the tissues were treated with 2 N HCl for 30 min and neutralized for 2 min in 100 mM Borax (Sigma). A primary antibody, mouse anti-BrdU (Becton Dickinson) was used at a 1:20 dilution in a mixture of PBST and 5% NGS (Vector Labs). The primary antibody incubation was done at 4°C overnight. A goat-anti mouse TRITC-labeled secondary antibody (Jackson ImmunoResearch) was used at a 1:200 dilution and the imaginal discs were treated for 2 hrs at room temperature. The discs were mounted using Vectashield mounting medium (Vector Labs).

Confocal images were collected by using a Confocal Microscope (Nikon Eclipse E800). For screening of the immunostaining, a fluorescence microscope (Nikon Eclipse E800) with X-cite 120 (Fluorescence Illumination Systems) was also utilized. The microscopic images were analyzed by using an Adobe Photoshop 7 software.