The anaphase-promoting complex/cyclosome or APC/C is a key regulator of chromosome segregation and mitotic exit in eukaryotes. It contains at least 11 subunits, most of which are evolutionarily conserved. The most abundant constituents of the vertebrate APC/C are the four structurally related tetratrico-peptide repeat (TPR) subunits, the functions of which are not yet precisely understood. Orthologues of three of the TPR subunits have been identified in Drosophila. We have shown previously that one of the TPR subunits of the Drosophila APC/C, Apc3 (also known as Cdc27 or Mákos), is essential for development, and perturbation of its function results in mitotic cyclin accumulation and metaphase-like arrest. In this study we demonstrate that the Drosophila APC/C associates with a new TPR protein, a genuine orthologue of the vertebrate Apc7 subunit that is not found in yeasts. In addition to this, transgenic flies knocked down for three of the TPR genes Apc6 (Cdc16), Apc7 and Apc8 (Cdc23), by RNA interference were established to investigate their function. Whole-body expression of subunit-specific dsRNA efficiently silences these genes resulting in only residual mRNA concentrations. Apc6/Cdc16 and Apc8/Cdc23 silencing induces developmental delay and causes different pupal lethality. Cytological examination showed that these animals had an elevated level of apoptosis, high mitotic index and delayed or blocked mitosis in a prometaphase-metaphase-like state with overcondensed chromosomes. The arrested neuroblasts contained elevated levels of cyclin B but, surprisingly, cyclin A appeared to be degraded normally. Contrary to the situation for the Apc6/Cdc16 and Apc8/Cdc23 genes, the apparent loss of Apc7 function does not lead to the above abnormalities. Instead, the Apc7 knocked down animals and null mutants are viable and fertile, although they display mild chromosome segregation defects and anaphase delay. Nevertheless, the Apc7 subunit shows synergistic genetic interaction with Apc8/Cdc23 that, together with the phenotypic data, assumes a limited functional role for Apc7. Taken together, these data suggest that the structurally related TPR subunits contribute differently to the function of the anaphase-promoting complex.

Introduction

The ubiquitin-dependent protein degradation system controls mitotic progression by sequential degradation of different regulatory proteins. In this process a pathway of ubiquitin activating (E1), ubiquitin conjugating (E2) and ubiquitin ligase (E3) enzymes catalyse the conjugation of polyubiquitin chains onto substrate proteins, leading to their recognition and degradation by the 26S proteasome. The key component of this proteolytic system is a multi-subunit ubiquitin ligase, the anaphase-promoting complex (APC/C) that provides a platform and specificity for the ubiquitination reactions. The APC/C is active during mitosis and G1 phases of the cell cycle in which, by mediating securin and mitotic cyclin degradation, it triggers the metaphase to anaphase transition and the exit from mitosis (Harper et al., 2002).

The APC/C has been purified and analysed biochemically from yeast, Xenopus and clam egg extracts and from human cells. It turned out to be a large protein complex containing at least 13 stably associated core subunits in yeasts (Passmore, 2004). Orthologues of ten of the yeast subunits could be identified in higher eukaryotes as well, indicating that the structure of APC/C is evolutionarily conserved. Only the Apc9 subunit appears to be specific to unicellular yeasts, whereas the Apc7 subunit has so far been identified only in vertebrates (Harper et al., 2002).

Despite the essential role of APC/C in regulating mitosis, little is known about the function of most of its individual subunits. The presence of structurally related proteins to the cullin homologue Apc2, the RING finger Apc11 and the DOC1 domain containing Apc10 subunits in other E3 ubiquitin ligases suggests direct roles of these subunits in the ubiquitination reaction. Indeed, it was shown that the Apc2 and Apc11 subunits interact with each other (Ohta et al., 1999) and depending on the E2 enzyme, Apc11 alone or together with Apc2 could support the nonspecific transfer of ubiquitin to protein substrates (Gmachl et al., 2000; Leverson et al., 2000; Tang et al., 2001) in vitro. Furthermore, Apc10 was suggested as the processivity factor for the APC/C (Carroll and Morgan, 2002).

Much less is known about the function of other APC/C subunits, such as Apc1, Apc4 and Apc5. The ida gene coding for the Drosophila homologue of Apc5 has been cloned and characterised. Mutant alleles of ida show a characteristic mitotic phenotype suggesting that this subunit controls some sub-functions of the APC/C (Bentley et al., 2002).

The Apc3 (also known as Cdc27 or Mks; hereafter referred to as Apc3/Cdc27/mks), Apc6 (also known as Cdc16; referred to as Apc6/Cdc16), and the Apc8 (also known as Cdc23; referred to as Apc8/Cdc23) subunits constitute a group of structurally related proteins within the APC/C, all of which contain nine to ten copies of the tetratrico-peptide repeat (TPR) motifs in tandem arrays. The TPRs are repeats of 34 amino acid structural motifs with a consensus sequence restricted only to eight residues. There is no invariant residue even within the consensus but amino acids at these positions are conserved in terms of size, hydrophobicity and spacing (Lamb et al., 1995; Blatch and Lässle, 1999). The first X-ray structure of the TPR containing protein phosphatase-5 revealed that each motif forms two α-helices in an antiparallel, helix-turn-helix configuration (Das et al., 1998). The neighbouring motifs are packed in a parallel fashion resulting in the formation of a superhelical structure. TPR motifs are present in functionally divergent proteins and thought to mediate protein-protein interactions and the assembly of multiprotein complexes. It is believed that the TPR subunits of the APC/C form a scaffold-like structure that facilitates the binding of other catalytic subunits, regulatory proteins and substrates (Passmore et al., 2005). TPR subunits are essential for viability in yeasts, and their mutant alleles result in uniform, mitotic G2-M arrest phenotype (Lamb et al., 1994).

In Drosophila, the genes coding for three TPR subunits, Apc3/Cdc27/mks, Apc6/Cdc16 and Apc8/Cdc23 have previously been identified (Deak et al., 2003; Harper et al., 2002; Huang and Raff, 2002). Characterisation of Apc3/Cdc27/mks mutants indicated that this subunit is essential for APC/C function and required for the degradation of both cyclin A and cyclin B (Deak et al., 2003). In another study, GFP-tagged Apc3/Cdc27/Mks and Apc6/Cdc16 subunits localised differentially in living syncytial embryos. Individual depletion of these subunits in cultured Drosophila S2 cells by RNA interference resulted in morphologically distinct mitotic phenotypes. These findings suggest the existence of multiple forms of the APC/C with different functions (Huang and Raff, 2002).

To better understand the role of the TPR subunits in maintaining the structure and function of APC/C, we have initiated a reverse genetic analysis on their homologues in Drosophila melanogaster. First, the putative Drosophila orthologue of the human gene Apc7 was identified through extensive sequence analysis, and then found to associate with the APC/C complex. Following that, transgenic lines were constructed carrying inducible RNA interference (RNAi) constructs specific for the Apc6/Cdc16, Apc7 and Apc8/Cdc23 genes. We show that knocking down these genes results in strikingly different phenotypes that cover viability, cell cycle progression, substrate degradation and apoptosis. These findings suggest that the TPR subunits contribute differentially to different sub-functions of the APC/C.

Results

The CG14444 gene encodes the Drosophila homologue of Apc7

The Drosophila genome project (BDGP) identified a gene designated as CG14444 and localized to the 6C1 region of the X chromosome. EST sequences (LD05103 and LD39177) corresponding to this gene encode a 595 amino acid TPR-containing protein of approximately 68 kDa. Extensive sequence comparisons, using BLAST, gapped BLAST and PSI-BLAST programs (Altschul et al., 1997), showed 30% identity and 51% similarity of the predicted protein along its entire length to the Apc7 subunit of the human APC/C (Yu et al., 1998). Interestingly, predicted proteins from Arabidopsis (accession no. NP 850309) and Apis mellifera (accession no. XP 396165) came up with very similar topology to both the human and Drosophila proteins. In each of these proteins, the number, location and the distribution of the TPR motifs were very similar: they all contained eight TPR motifs in a tandem array at their C terminus and two single repeats toward their N-terminal end (Fig. 1A). Comparison of the ten TPR motifs of the CG14444 gene product to each other showed similarity mostly confined to the consensus sequence of this motif (Fig. 1B). However, sequence comparison of individual motifs to positionally equivalent TPR motifs of related proteins from other species revealed that the conservation among these repeats extends outside of the consensus (Fig. 1C), suggesting functional correspondence. The similarity pattern outside the consensus is specific to individual TPRs. Thus these observations suggest that the CG14444 gene encodes a putative Apc7 protein in Drosophila melanogaster. Likewise, Apc7 orthologues appear to exist in Apis mellifera and Arabidopsis thaliana as well, suggesting that this protein is present in all multicellular eukaryotes. Owing to the high degree of structural and sequential correspondence between CG14444 and the genuine vertebrate Apc7, we designate the Drosophila gene as Apc7 in the rest of this paper.

Apc6/Cdc16- and Apc8/Cdc23-specific dsRNA expression causes larval-pupal polyphasic lethality

The inducible transgenic RNAi approach summarized in Fig. 2A was applied to knock down the expression of the Apc6/Cdc16, Apc7 and Apc8/Cdc23 genes. Constructs were made in the pWIZ vector (Lee and Carthew, 2003) in which inverted doublets of cDNA sequences were placed under the control of the GAL4-UAS system, and transgenic lines were established after transformation of isogenic flies. Expression from these constructs leads to the synthesis of RNAs that are predicted to snap back after splicing, and form loopless dsRNAs. Several independent transgenic lines carrying single copy insertions were isolated for each construct and mapped to the major chromosomes. All homozygous viable lines showed comparable results.

To induce expression of transgenes, the RNAi strains were crossed to strains carrying the Act5C-GAL4 driver with a ubiquitous expression of GAL4 throughout development (The FlyBase Consortium, 2003). RT-PCR was used to determine the expression of endogenous TPR subunit-specific mRNAs. As shown on Fig. 2B, induction of all three dsRNA syntheses was able to reduce TPR-specific mRNA levels by ∼90% in early pupae relative to uninduced controls when normalized to the rpL17A calibration control. The da-GAL4 driver was also used for RNAi induction, and provided similar silencing effects to Act5C-GAL4 (data not shown).

To examine how silencing of the TPR subunits affects viability, defined numbers of first instar larvae carrying single copies of Act5C-GAL4 and either UAS-Apc6/Cdc16RNAi or UAS-Apc8/Cdc23RNAi constructs were collected and their survival was monitored by counting the number of third instar larvae and pupae. Induction of Apc6/Cdc16- and Apc8/Cdc23-specific RNAi did not significantly affect survival of embryos and early larvae. However, lethality was observed in the last larval (L3) stage, and all animals died as pupae. In addition to this, the third instar larvae showed a 1- to 2-day delay in puparium formation and frequently these larvae formed melanotic tumours (Table 1A). No melanotic tumours or premature deaths were observed in control animals carrying only a single copy of Act5C-GAL4. The late lethal phenotype is a characteristic feature of mitotic regulatory genes (Gatti and Baker, 1989) and shows that the Apc6/Cdc16 and Apc8/Cdc23 subunits are essential components of the Drosophila APC/C.

Table 1.

Lethal phase of TPR genes knocked down by RNA interference

Genotype L3 Larvae* (%) Prepupal stage P3 (%) Prepupal stage P4(i) (%) Prepupal stage P4(ii) (%) Pupal stage P5(i) (%)
A       
Apc6RNAi  6 + mt   3   –   11   86  
Apc7RNAi  Viable   Viable   Viable   Viable   Viable  
Apc71  Viable   Viable   Viable   Viable   Viable  
Apc8RNAi  13 + mt   20   –   65   15  
B       
Apc6RNA-Apc7RNAi  7 + mt   3   –   15   82  
Apc6RNAi-Apc8RNAi  27 + MT   26   70   4   –  
Apc7RNAi-Apc8RNAi  20 + MT   28   70   2   –  
Genotype L3 Larvae* (%) Prepupal stage P3 (%) Prepupal stage P4(i) (%) Prepupal stage P4(ii) (%) Pupal stage P5(i) (%)
A       
Apc6RNAi  6 + mt   3   –   11   86  
Apc7RNAi  Viable   Viable   Viable   Viable   Viable  
Apc71  Viable   Viable   Viable   Viable   Viable  
Apc8RNAi  13 + mt   20   –   65   15  
B       
Apc6RNA-Apc7RNAi  7 + mt   3   –   15   82  
Apc6RNAi-Apc8RNAi  27 + MT   26   70   4   –  
Apc7RNAi-Apc8RNAi  20 + MT   28   70   2   –  
*

Percentage of all larvae examined. All other values represent percentages of animals reaching at least prepupal stage. 7-10 samples of 100 larvae were collected and analyzed for each genotype. MT and mt, frequent and sporadic occurrence of melanotic tissue in L3 larvae, respectively

Fig. 1.

Distribution and sequence comparisons of TPR motifs in putative Apc7 orthologues. (A) Distribution of TPRs in the human Apc7 subunit (Hs), the Drosophila CG14444 gene product (Dm), and predicted proteins from Arabidopsis thaliana (At; accession no. NP 850309) and Apis mellifera (Am; accession no. XP 396165). The asymmetric rectangular symbols represent single TPR motifs. (B) Sequence alignment of seven TPR motifs of the Drosophila Apc7 protein. The TPR consensus sequence is shown below the alignment. Residues matching the consensus are in bold type and highlighted in grey (also in C). Residues are conserved only at the consensus positions but, even there, there is no invariant position. (C) Sequence comparison of the TPR10 motif from the proteins shown at the top of this figure. Conserved residues outside the consensus are highlighted in yellow. Residue conservation is more extensive both within and outside of the consensus.

Fig. 1.

Distribution and sequence comparisons of TPR motifs in putative Apc7 orthologues. (A) Distribution of TPRs in the human Apc7 subunit (Hs), the Drosophila CG14444 gene product (Dm), and predicted proteins from Arabidopsis thaliana (At; accession no. NP 850309) and Apis mellifera (Am; accession no. XP 396165). The asymmetric rectangular symbols represent single TPR motifs. (B) Sequence alignment of seven TPR motifs of the Drosophila Apc7 protein. The TPR consensus sequence is shown below the alignment. Residues matching the consensus are in bold type and highlighted in grey (also in C). Residues are conserved only at the consensus positions but, even there, there is no invariant position. (C) Sequence comparison of the TPR10 motif from the proteins shown at the top of this figure. Conserved residues outside the consensus are highlighted in yellow. Residue conservation is more extensive both within and outside of the consensus.

Fig. 2.

Gene silencing by RNA interference. (A) Scheme for transgenic RNA interference used in this work to knock down the expression of the TPR subunits of Drosophila APC/C. Inverted exon repeats of these genes (white bars) were cloned into pWIZ transformation vector under the control of the UAS promoter (white circles) and transgenic lines were made. These were crossed to the Act5C-GAL4 driver lines that constitutively expressed the GAL4 transcription activator in all cells throughout development. Progeny that contained both Act5C-GAL4 and UAS transgenic constructs expressed double-stranded hairpin RNAs that triggered gene silencing. (B) Monitoring TPR transcript levels in RNAi induced larvae (I) relative to uninduced controls (C). The rpL17A transcript was used as a calibration control. TPR gene-specific transcript levels appeared significantly lower in all RNAi induced samples.

Fig. 2.

Gene silencing by RNA interference. (A) Scheme for transgenic RNA interference used in this work to knock down the expression of the TPR subunits of Drosophila APC/C. Inverted exon repeats of these genes (white bars) were cloned into pWIZ transformation vector under the control of the UAS promoter (white circles) and transgenic lines were made. These were crossed to the Act5C-GAL4 driver lines that constitutively expressed the GAL4 transcription activator in all cells throughout development. Progeny that contained both Act5C-GAL4 and UAS transgenic constructs expressed double-stranded hairpin RNAs that triggered gene silencing. (B) Monitoring TPR transcript levels in RNAi induced larvae (I) relative to uninduced controls (C). The rpL17A transcript was used as a calibration control. TPR gene-specific transcript levels appeared significantly lower in all RNAi induced samples.

Although the extent of silencing judged by RT-PCR was comparable for both Apc6/Cdc16 and Apc8/Cdc23, a clear distinction could be made in their lethal phenotypes. Whereas pupae from five independent Apc8/Cdc23RNAi lines died at the end of the prepupal metamorphosis in the P4(ii) (moving bubble) stage, animals from all independent Apc6/Cdc16RNAi lines developed further and died during the first part of the phanerocephalic pupal (malpighian tubules migrating) stage of P5(i) (Bainbridge and Bownes, 1981). The Apc8/Cdc23-specific RNAi effect was also more pronounced both in terms of larval lethality and formation of melanotic tumours. However, both Apc6/Cdc16RNAi and Apc8/Cdc23RNAi animals died earlier than the ones homozygous for the mks1 allele of the Apc3/Cdc27/mks gene. The mks1 mutation causes comparable reduction (∼90%) of gene expression compared with the RNAi knock down effect in Apc6/Cdc16RNAi and Apc8/Cdc23RNAi animals, yet these mutants die as late pharate adults with severe rough eye and bristle phenotypes (Deak et al., 2003).

Apc6/Cdc16- and Apc8/Cdc23-specific gene silencing results in metaphase-like arrest with overcondensed chromosomes

To determine whether Apc6/Cdc16 or Apc8/Cdc23 silencing affects cell cycle progression, we examined orcein-stained brain squash preparations of RNAi induced third instar larvae for mitotic defects. In many mitotic cells, the chromosomes showed up as dot-like structures instead of the characteristic rod-like wild-type chromosomes, and frequently they appeared either scattered all over the cells or congregated at the metaphase plate (Fig. 3E-G,I-K). The chromosome overcondensation indicated that the cells were delayed or arrested in mitosis, as chromosomes continue condensation during this time. Accordingly, the proportion of cells in mitosis was significantly higher, and this leads to about double the mitotic index (MI) of the wild type (Table 2). Most of the mitotic cells were in a prometaphase-metaphase-like state, and at the same time, the number of cells in ana- and telophases stayed relatively low. This is reflected in the two- and three-times higher metaphase-anaphase ratios (M:A) in the Apc6/Cdc16 and Apc8/Cdc23 RNAi preparations, respectively (Table 2), and suggests that loss of function of these subunits leads to metaphase-like delay or arrest.

Table 2.

Mitotic phenotypes of TPR genes knocked down by RNA interference

Genotype Preparations/fields* Mitotic index (MI) Metaphase:anaphase ratio Chromosomes at metaphase plate (%) Polyploid (%)
A       
Canton S  4/60   1.7±0.3   1.9±0.3   2.7   0  
Apc6RNAi  4/60   3.2±0.3   4.2±0.5   11.0   4.6±1.8  
Apc7RNAi  6/70   2.1±0.3   1.4±0.4   2.2   0  
Apc8RNAi  4/50   3.5±0.4   5.2±0.3   9.7   8.1±0.9  
B       
Apc6RNAi-Apc7RNAi  5/90   3.5±0.3   5.5±0.5   10.0   3.0±0.8  
Apc6RNAi-Apc8RNAi  5/50   4.4±0.4   6.6±0.8   11.6   7.7±3.0  
Apc7RNAi-Apc8RNAi  4/40   4.2±0.4   6.3±0.5   9.3   6.2±2.1  
Genotype Preparations/fields* Mitotic index (MI) Metaphase:anaphase ratio Chromosomes at metaphase plate (%) Polyploid (%)
A       
Canton S  4/60   1.7±0.3   1.9±0.3   2.7   0  
Apc6RNAi  4/60   3.2±0.3   4.2±0.5   11.0   4.6±1.8  
Apc7RNAi  6/70   2.1±0.3   1.4±0.4   2.2   0  
Apc8RNAi  4/50   3.5±0.4   5.2±0.3   9.7   8.1±0.9  
B       
Apc6RNAi-Apc7RNAi  5/90   3.5±0.3   5.5±0.5   10.0   3.0±0.8  
Apc6RNAi-Apc8RNAi  5/50   4.4±0.4   6.6±0.8   11.6   7.7±3.0  
Apc7RNAi-Apc8RNAi  4/40   4.2±0.4   6.3±0.5   9.3   6.2±2.1  
*

The number of preparations examined and microscopic fields screened for mitotic cells. On average, there were about 100-200 cells per field

Mitotic index is given as the average number of mitotic cells in an optical field±s.d.

This is percentage of mitotic cells ± s.d.

The scattered distribution of chromosomes in RNAi induced Apc6/Cdc16 and Apc8/Cdc23 cells were very similar to the phenotype caused by the Apc3/Cdc27/mks, Apc5/ida and Apc2/mr mutations (Deak et al., 2003; Bentley et al., 2002; Reed and Orr-Weaver, 1997).

However, in about 10% of Apc6/Cdc16 and Apc8/Cdc23 RNAi induced mitotic cells the chromosomes appeared to congregate precisely to the metaphase plate (Table 2 and Fig. 3F,J). This true metaphase stage was quite rare or indeed could not be detected in Apc3/Cdc27/mks, Apc5/ida and Apc2/mr mutants, and was significantly higher than that seen in the wild type (2-3%). The chromosomes at the metaphase plate were also overcondensed, indicating that the delay or arrest was maintained during that stage as well.

Fig. 3.

Silencing of the Apc6/Cdc16, Apc7 and Apc8/Cdc23 genes by RNAi results in mitotic abnormalities. Wild-type mitotic cells in prometaphase (A), metaphase (B) and anaphase (C). Neuroblast cells from both Apc6/Cdc16 (E-G) and Apc8/Cdc23 (I-L) knocked down larvae show metaphase-like arrest with overcondensed chromosomes. The chromosomes in most of the arrested cells appear scattered (E,G,I,K), in about 10% of mitotic cells the chromosomes are locked at the metaphase plate (F,J). Some degree of chromosome overcondensation could also be observed in anaphase figures (L) together with irregular chromosome segregation. Some cells appear polyploid (G,K), and they invariantly show chromosome overcondensation. Overcondensed chromosomes sometimes appear to be in the process of decondensation (G). Induction of Apc7-specific RNAi causes a mild mitotic phenotype. Neuroblast cells from Apc7RNAi larvae show signs of uneven chromosome condensation and telomeric fusions (D) and occasionally chromosome bridges in anaphase (H).

Fig. 3.

Silencing of the Apc6/Cdc16, Apc7 and Apc8/Cdc23 genes by RNAi results in mitotic abnormalities. Wild-type mitotic cells in prometaphase (A), metaphase (B) and anaphase (C). Neuroblast cells from both Apc6/Cdc16 (E-G) and Apc8/Cdc23 (I-L) knocked down larvae show metaphase-like arrest with overcondensed chromosomes. The chromosomes in most of the arrested cells appear scattered (E,G,I,K), in about 10% of mitotic cells the chromosomes are locked at the metaphase plate (F,J). Some degree of chromosome overcondensation could also be observed in anaphase figures (L) together with irregular chromosome segregation. Some cells appear polyploid (G,K), and they invariantly show chromosome overcondensation. Overcondensed chromosomes sometimes appear to be in the process of decondensation (G). Induction of Apc7-specific RNAi causes a mild mitotic phenotype. Neuroblast cells from Apc7RNAi larvae show signs of uneven chromosome condensation and telomeric fusions (D) and occasionally chromosome bridges in anaphase (H).

In some of the arrested cells the chromosomes appeared to undergo decondensation before attempting to complete anaphase and cytokinesis (Fig. 3G). It has been suggested that arrested cells could revert to interphase by decondensing their chromosomes (Gatti and Baker, 1989). Following DNA replication, such cells become tetraploid, or with the recurrence of events, polyploid. A proportion of Apc6/Cdc16 and Apc8/Cdc23 RNAi cells were polyploid, most frequently with highly overcondensed tetraploid or less frequently octaploid or higher ploidy chromosome complements (Table 2 and Fig. 3G,K).

In addition to metaphase, irregular chromosome behaviour could be detected in anaphase as well, which included lagging chromosomes and chromosome bridges (Fig. 3L).

Apc6/Cdc16 and Apc8/Cdc23 RNAi lines arrest with high levels of cyclin B but with apparent destruction of cyclin A

One of the main functions of APC/C is to aid the degradation of mitotic cyclins. Previously, we have shown that mutations (mks1 and mksL7123) in the Apc3/Cdc27/mks gene stabilized both cyclin A and B in the mitotically arrested cells (Deak et al., 2003) (M.P. and P.D., unpublished). We were interested to follow mitotic cyclin levels in RNAi induced neuroblasts by immunostaining with polyclonal antibodies raised against cyclin A and cyclin B. In wild-type cells, cyclin A could be detected in prophase and early prometaphase (Fig. 4A-D), but it became undetectable around metaphase and early anaphase (Fig. 4E-H). Cyclin B degradation follows that of cyclin A, so its level declines rapidly in metaphase, and at and after chromosome segregation (Fig. 5A-H). As expected, both cyclin A- and cyclin B-specific antisera gave strong staining in the RNAi induced Apc6/Cdc16 and Apc8/Cdc23 preparations. Closer inspection revealed that the cells strongly stained by cyclin A-specific antibody gave rather weak and somewhat diffuse DNA signal, indicative of being in prophase or early prometaphase. The majority of cells (80%) arrested in late prometaphase or in metaphase were not stained by anti-cyclin A antibody above background level (Fig. 4I-L,M-P). However, more than 80% of cells that stained intensely with cyclin B-specific antibody also showed bright staining of DNA at the metaphase plate and a spindle indicative of being in metaphase arrest with overcondensed or polyploid sets of chromosomes (Fig. 5I-L,M-P). Thus, it appears that in terms of cyclin A degradation, the Apc6/Cdc16 and Apc8/Cdc23 mitotic phenotypes differ from that of Apc3/Cdc27/mks mutants, where both cyclin A and cyclin B accumulate (Deak et al., 2003). The apparent degradation of cyclin A and the accumulation of cyclin B in cells with diminished Apc6/Cdc16 and Apc8/Cdc23 functions indicate that these subunits may have lesser roles in cyclin A removal but are required for cyclin B degradation.

Apc7 is a nonessential protein that associates with the Drosophila APC/C

As Fig. 2B demonstrates, RNA interference caused significant reduction in the expression of the Apc7 gene. This effect was confirmed independently by using quantitative fluorescent real-time PCR that showed greatly delayed amplification in Apc7RNAi samples compared with controls (data not shown). The delayed amplification is consistent with lower concentration of Apc7 transcripts in those samples. Although the degree of gene silencing was at least as effective as in the Apc6/Cdc16 and Apc8/Cdc23 RNAi lines, the Apc7RNAi animals developed without any delay or abnormalities and hatched as fertile adults (Table 2).

Orcein-stained brain squash preparations from Apc7RNAi third instar larvae lack any sign of prometaphase or metaphase-like arrest characteristic of the other TPR subunit mutants. Instead, chromosomes appeared somewhat more tangled and exhibited signs of uneven condensation together with frequent chromosome bridges in anaphase (Fig. 3D,H). The number of cells in mitosis remained low but the proportion of the anaphase and telophase figures approximated the number of cells in prometaphase and metaphase, resulting in a mitotic index that was similar to that of wild type, and a metaphase-anaphase ratio (M:A) that was lower than in wild type (Table 2). The lower M:A ratio may indicate some delay during anaphase in the Apc7RNAi cells.

In order to further support the nonessential nature of the Apc7 gene, a null mutant allele (Apc71) was isolated by imprecise excision of the P{EPgy2}EY11333 element localized 63 bp distally from the 3′ end of the gene. The remobilization generated a 1576 bp deletion extending from the insertion site toward the 5′ end of Apc7. This internal deletion removed about two-thirds of the Apc7 gene, including the functionally important eight tandem TPR repeats (Vodermaier et al., 2003). The Apc71 homozygotes appeared viable and fertile, and could be maintained as a homozygous stock. Their mitotic phenotype was indistinguishable from that of Apc7RNAi lines. To determine the role of Apc7 in cyclin degradation, we monitored cyclin levels in Apc7RNAi and Apc71 neuroblast preparations by immunostaining. We could not detect abnormal changes in the level and localisation of either cyclin A or cyclin B in mitotic cells and their turnover appeared similar to that in wild type (data not shown).

To see whether Apc7 is a genuine component of the Drosophila APC/C, we examined physical interaction between Apc7 and universal APC/C subunits Apc8/Cdc23 or Apc3/Mks by co-transfection of S2 cells and affinity chromatography. FLAG-tagged Apc7 (F7) was co-expressed in S2 cells with either haemagglutinin (HA) epitope-tagged Apc8/Cdc23 (H8) or Apc3/Mks (H3), and then FLAG-Apc7 and its associated proteins were pulled down from cell extracts with an anti-FLAG affinity column. Following repeated washing, bound proteins were eluted with an excess of free FLAG peptide, and detected by western analysis using monoclonal anti-FLAG or anti-HA antibodies. Fig. 6 shows that both HA-Apc3/Mks (lanes 7 and 8) and HA-Apc8 (lanes 3 and 4) were retained on the column with FLAG-Apc7 under washing condition of 0.15 M NaCl (in 20 mM Tris), but there was no detectable retention without Apc7 (Fig. 6, lanes 9 and 10). However, in a reciprocal experiment, when FLAG-Apc8 was co-expressed with HA-Apc7, and FLAG-Apc8-associated proteins were pulled down, only a faint band of HA-Apc7 could be detected (Fig. 6, lane 6), indicating a weak association of HA-Apc7 with FLAG-Apc8. The expression levels of the different N-terminal fusion constructs were comparable in these experiments as judged from western blots of total protein extracts (lanes 1 and 2). These data show that at least a fraction of Apc7 is clearly associated with the APC/C in Drosophila, but perhaps not as a core subunit. Possible interpretations of the low-yield co-purification of Apc7 in the FLAG-Apc8 pull-down complex is that Apc7 either interacts only with certain forms of the APC/C complex or it interacts only transiently with the APC/C.

Fig. 4.

Cyclin A is degraded in RNAi silenced Apc6/Cdc16 and Apc8/Cdc23 cells. Columns of images show tubulin (green and second column), DNA (blue and third column) and cyclin A (red and fourth column) localisations in mitotic cells. Cyclin A staining is visible in wild-type prophase or prometaphase (A-D) but it is undetectable in most of the wild-type metaphase (E-H) cells (n=12). Similarly to this, in the majority (80%) of both Apc6/Cdc16 (I-L) and Apc8/Cdc23 (M-P) RNAi induced cells (n=19 and 26, respectively), there is no detectable cyclin A at the onset of metaphase, indicating that cyclin A degradation is not affected in these lines.

Fig. 4.

Cyclin A is degraded in RNAi silenced Apc6/Cdc16 and Apc8/Cdc23 cells. Columns of images show tubulin (green and second column), DNA (blue and third column) and cyclin A (red and fourth column) localisations in mitotic cells. Cyclin A staining is visible in wild-type prophase or prometaphase (A-D) but it is undetectable in most of the wild-type metaphase (E-H) cells (n=12). Similarly to this, in the majority (80%) of both Apc6/Cdc16 (I-L) and Apc8/Cdc23 (M-P) RNAi induced cells (n=19 and 26, respectively), there is no detectable cyclin A at the onset of metaphase, indicating that cyclin A degradation is not affected in these lines.

Apc8/Cdc23RNAi show synergism with Apc6/Cdc16RNAi and Apc7RNAi

TPR subunits are thought to be involved in protein-protein interactions to form large protein complexes (Blatch and Lässle, 1999), and were suggested to interact with each other within the budding yeast APC/C (Lamb et al., 1994). Since genetic interaction tests of double mutants are powerful techniques to investigate functional relationships either between genes or their protein products, we applied this to our RNAi lines. Transgenic strains carrying two RNAi constructs in different combinations were generated and crossed to Act5C-GAL4 partners to induce gene silencing. Again, defined number of first instar larvae carrying single copies of Act5C-GAL4 and pair-wise combinations of the UAS-Apc6/Cdc16RNAi, UAS-Apc7RNAi or UAS-Apc8/Cdc23RNAi constructs were collected and their lethal phase and mitotic phenotypes were determined.

Similarly to single RNAi lines, the embryonic and early larval developments were not affected in any of the double RNAi combinations. However, as Table 1 illustrates, the late larval lethality and the number of larvae with melanotic tumours were greatly increased in the Apc6/Cdc16RNAi-Apc8/Cdc23RNAi (27%) and Apc7RNAi-Apc8/Cdc23RNAi (20%) combinations compared to Apc6/Cdc16RNAi (6%), Apc7RNAi (viable) or Apc8/Cdc23RNAi (13%) alone. The Apc6/Cdc16RNAi-Apc8/Cdc23RNAi and the Apc7RNAi-Apc8/Cdc23RNAi double mutants also showed a definite shift in pupal lethality toward earlier stages compared to the single RNAi lines (Table 1).

The range of mitotic phenotypes found in the Apc6/Cdc16RNAi-Apc8/Cdc23RNAi and the Apc7RNAi-Apc8/Cdc23RNAi double RNAi lines was essentially the same as that found in the Apc6/Cdc16RNAi and Apc8/Cdc23RNAi lines alone; however, the MI of these double RNAi combinations were significantly higher (4.4 and 4.2, respectively; P<0.05) than that of Apc6/Cdc16RNAi (3.2), Apc7RNAi (2.1) or Apc8/Cdc23RNAi (3.5) alone (Table 2). Taken together, the increased L3 lethality, the shift of lethal phase and the more pronounced mitotic arrests phenotype of the double RNAi lines suggest a firm synergistic interaction between Apc8/Cdc23 and Apc6/Cdc16 as well as Apc8/Cdc23 and Apc7, thus implicating that these subunits contribute independently to the same function of the APC/C.

Fig. 5.

Cyclin B is not degraded in RNAi silenced Apc6/Cdc16 and Apc8/Cdc23 cells. Colours are the same as on Fig. 4. In wild-type cells (n=10) the cyclin B level is high in prophase and prometaphase (A-D) and it starts to diminish at or after the onset of metaphase (E-H). Cyclin B staining is quite pronounced in more than 80% of RNAi silenced Apc6/Cdc16 (I-L) and Apc8/Cdc23 (M-P) cells (n=20 and 17, respectively) arrested in metaphase indicating that cyclin B is not degraded.

Fig. 5.

Cyclin B is not degraded in RNAi silenced Apc6/Cdc16 and Apc8/Cdc23 cells. Colours are the same as on Fig. 4. In wild-type cells (n=10) the cyclin B level is high in prophase and prometaphase (A-D) and it starts to diminish at or after the onset of metaphase (E-H). Cyclin B staining is quite pronounced in more than 80% of RNAi silenced Apc6/Cdc16 (I-L) and Apc8/Cdc23 (M-P) cells (n=20 and 17, respectively) arrested in metaphase indicating that cyclin B is not degraded.

The phenotype of the third double RNAi line, Apc6/Cdc16RNAi-Apc7RNAi appeared different from the ones discussed above. The lethal phase and mitotic phenotype were similar to those of Apc6/Cdc16RNAi alone (Table 1), the hallmark of epistatic interaction. However, as one of the lines, Apc7RNAi, is viable with mild mitotic phenotype, one would also expect the double mutant phenotype to reflect the stronger mutant if these gene products act independently, therefore no obvious interaction can be established between Apc6/Cdc16 and Apc7.

Apc6/Cdc16- and Apc8/Cdc23-specific RNA interference induces apoptosis

In addition to mitotic cells, we observed many pycnotic nuclei in the orcein-stained preparations (Fig. 7A,B). The cells were rounded, relatively small and their nuclei showed strong but uneven staining. Pycnotic nuclei characteristic of apoptotic cells could be found sporadically in wild-type tissues as well, but their frequency increased by three- to fourfold in Apc6/Cdc16 (P<0.005) and five to sixfold in Apc8/Cdc23 (P<0.005) knocked down preparations (Fig. 7E), suggesting strong induction of apoptosis. To confirm that cell death was induced in these RNAi lines, we stained larval brains with Acridine Orange. As Fig. 7C,D illustrates, the Acridine Orange accumulation was much more intensive in the Apc8/Cdc23 RNAi induced brains than in the control. The Apc6/Cdc16 preparations yielded similar results to Apc8/Cdc23.

Fig. 6.

FLAG affinity chromatography of FLAG-tagged Apc7 and Apc8 complexes. (A) Western blots of total protein extracts from S2 cells transfected with HA-tagged Apc8 (H8, lane 1) or FLAG-tagged Apc7 (F7, lane 2) plasmids. Strip 1 was reacted with anti-HA (αH, lane 1), and strip 2 with anti-FLAG (αF, lane 2) monoclonal antibodies. (B) Western blots of proteins affinity-purified on anti-FLAG-M2 antibody beads from S2 cells co-transfected with FLAG-Apc7 and HA-Apc8 (F7-H8, lanes 3 and 4), FLAG-Apc8 and HA-Apc7 (F8-H7, lanes 5 and 6) or FLAG-Apc7 and HA-Apc3 (F7-H3, lanes 7 and 8) plasmids. Blots were reacted with anti-FLAG (lanes 3, 5 and 7) or anti-HA (lanes 4, 6 and 8) monoclonal antibodies. The immunoreactive bands labelled with an asterisk are proteolytic degradation products of Apc8. (C) Blots of anti-FLAG-M2 column-bound (lane 10) and unbound (flow-through, lane 9) proteins from S2 cells transfected with HA-Apc8 (H8) plasmid and incubated with anti-HA monoclonal antibody. Even after heavily overloading the anti-FLAG column, no nonspecific binding of HA-Apc8 could be detected. Similarly, HA-Apc3 did not bind to the anti-FLAG-M2 column (data not shown) E, eluate; FT, flow-through.

Fig. 6.

FLAG affinity chromatography of FLAG-tagged Apc7 and Apc8 complexes. (A) Western blots of total protein extracts from S2 cells transfected with HA-tagged Apc8 (H8, lane 1) or FLAG-tagged Apc7 (F7, lane 2) plasmids. Strip 1 was reacted with anti-HA (αH, lane 1), and strip 2 with anti-FLAG (αF, lane 2) monoclonal antibodies. (B) Western blots of proteins affinity-purified on anti-FLAG-M2 antibody beads from S2 cells co-transfected with FLAG-Apc7 and HA-Apc8 (F7-H8, lanes 3 and 4), FLAG-Apc8 and HA-Apc7 (F8-H7, lanes 5 and 6) or FLAG-Apc7 and HA-Apc3 (F7-H3, lanes 7 and 8) plasmids. Blots were reacted with anti-FLAG (lanes 3, 5 and 7) or anti-HA (lanes 4, 6 and 8) monoclonal antibodies. The immunoreactive bands labelled with an asterisk are proteolytic degradation products of Apc8. (C) Blots of anti-FLAG-M2 column-bound (lane 10) and unbound (flow-through, lane 9) proteins from S2 cells transfected with HA-Apc8 (H8) plasmid and incubated with anti-HA monoclonal antibody. Even after heavily overloading the anti-FLAG column, no nonspecific binding of HA-Apc8 could be detected. Similarly, HA-Apc3 did not bind to the anti-FLAG-M2 column (data not shown) E, eluate; FT, flow-through.

Fig. 7.

Loss of both Apc6/Cdc16 and Apc8/Cdc23 functions induces apoptosis. Orcein stained preparations from brains of Apc6/Cdc16 (A) and Apc8/Cdc23 (B) RNAi L3 larvae show small, rounded cells usually in pairs with intense but uneven nucleolar staining resembling apoptotic cells. Acridine Orange staining further indicates the high incidence of dying cells in brains of RNAi larvae (D, only an Apc8/Cdc23 RNAi sample is shown) relative to wild-type larvae (C). (E) Bar graph showing the apoptotic index in neuroblast preparations of the respective genotypes. The apoptotic index is given as the mean number of apoptotic cells in a microscope field.

Fig. 7.

Loss of both Apc6/Cdc16 and Apc8/Cdc23 functions induces apoptosis. Orcein stained preparations from brains of Apc6/Cdc16 (A) and Apc8/Cdc23 (B) RNAi L3 larvae show small, rounded cells usually in pairs with intense but uneven nucleolar staining resembling apoptotic cells. Acridine Orange staining further indicates the high incidence of dying cells in brains of RNAi larvae (D, only an Apc8/Cdc23 RNAi sample is shown) relative to wild-type larvae (C). (E) Bar graph showing the apoptotic index in neuroblast preparations of the respective genotypes. The apoptotic index is given as the mean number of apoptotic cells in a microscope field.

The pycnotic nuclei appeared as doublets in neighbouring cells (though their partition into two cells was not always obvious in our preparations; Fig. 7A,B). This implied that the induction of apoptosis occurred immediately after completing mitosis in those cells that managed to escape the mitotic arrest.

Discussion

The APC/C is essential for the ubiquitin-dependent degradation of cell cycle regulatory proteins. The complex multisubunit structure of APC/C facilitates its intimate involvement in the formation of substrate-ubiquitin conjugates, and thus determines substrate specificity of the whole process. More than half of the total mass of the APC/C consists of tandem arrays of TPR motifs present in several evolutionarily conserved subunits. Little is known about the specific functions of the TPR subunits, although a structural role to form a scaffold-like core of the APC/C has been suggested (Passmore et al., 2005). The TPR subunits in budding yeast are coded by the Cdc27, Cdc16 and Cdc23 genes that are all essential with uniform loss-of-function mitotic arrest phenotypes. We show here that in addition to the orthologues of these subunits, a gene is present in the Drosophila genome that codes for a protein closely related to the Apc7 subunit of the vertebrate APC/C. Our data support the nomination of this protein as the Drosophila Apc7 homologue. Firstly, its primary sequence appears to be conserved, including the number, location and distribution of its TPR motifs. The TPR motifs in the Drosophila protein are more related to the TPR motifs of the human Apc7 subunit than to the motifs of other Drosophila TPR proteins, suggesting functional correspondence. Secondly, the putative Apc7 protein is associated with universal subunits of the Drosophila APC/C. Our affinity purification data suggest that this association is either weak, or occurs only in certain forms of the APC/C complex. The existence of multiple forms of the APC/C has previously been suggested (Huang and Raff, 2002), and APC/C dimers were reported in budding yeast (Passmore et al., 2005). Alternatively, it is also possible that Apc7 is not a core subunit but a transiently interacting nonessential partner of the Drosophila APC/C.

Thirdly, the putative Drosophila Apc7 gene shows synergistic genetic interaction with the Apc8/Cdc23 gene that also implicates a functional relationship. Although the idea was that Apc7 is only a vertebrate-specific subunit of the APC/C (Yu et al., 1998), it is now shown to be also present in species from plants and insects, suggesting that this protein became associated with the APC/C in most multicellular organisms.

We have previously shown that one of the TPR genes in Drosophila, Apc3/Cdc27/mks, is essential and involved in both cyclin A and cyclin B degradation (Deak et al., 2003). To address the function of the other TPR proteins, we established transgenic RNAi lines to reduce Apc6/Cdc16-, Apc7- and Apc8/Cdc23-specific gene expression in intact animals. Knocking down the expression of Apc6/Cdc16 or Apc8/Cdc23 genes caused lethality, indicating that as in yeast, these genes are essential in Drosophila as well. Furthermore, Apc6/Cdc16RNAi and Apc8/Cdc23RNAi cells show mitotic phenotypes with high mitotic index, overcondensed chromosomes and metaphase-like arrest that is very similar to the mr, mks1 and ida mutants that code for the Apc2, Apc3/Cdc27/Mks and Apc5 subunits of the Drosophila APC/C, respectively (Reed and Orr-Weaver, 1997; Deak et al., 2003; Bentley et al., 2002). This hallmark phenotype supports the assumption that the Apc6/Cdc16 and Apc8/Cdc23 genes code for genuine, functionally conserved APC/C subunits in Drosophila.

Unlike in yeast, however, the phenotypes of the TPR mutants are not uniform, since beyond common characteristics, they show significant differences. Whereas the Apc7RNAi and Apc71 individuals are viable, mks1 mutants die as pharate adults, Apc6/Cdc16RNAi animals die as P5(i) early pupae, and Apc8/Cdc23RNAi animals die as P4(ii) prepupae. Differences can also be seen in the ability to degrade APC/C substrates, such as the mitotic cyclins. In this regard, removal of Apc7 has no effect, depletion of Apc6/Cdc16 and Apc8/Cdc23 appear to affect mainly cyclin B, whereas loss of Apc3/Cdc27/Mks affects both cyclin A and B degradation. These observations suggest different requirements of the TPR subunits in mitotic cyclin degradation. However, it should be noted that these differences could also be due to different inactivation efficiencies, despite the fact that the residual expression levels of the genes appear to be very similar in all TPR mutants as judged by RT-PCR (Fig. 2B) (Deak et al., 2003). In this case depletion of individual TPR subunits may affect the affinity of substrate binding, or the efficiency of the ubiquitylation reaction leading to an APC/C activity level at which cyclin A degradation is still sufficient, whereas cyclin B degradation is not.

The metaphase-like arrest of Apc6/Cdc16RNAi and Apc8/Cdc23RNAi cells is unique in that they frequently show de facto metaphase arrest with overcondensed chromosomes aligned precisely at the metaphase plate. This correlates with the ability of these cells to degrade cyclin A. This is in stark contrast with mks1 cells that accumulate cyclin A and do not, or seldom, form literal metaphase figures. These observations suggest that cyclin A degradation is required in Drosophila for proper chromosome alignment at metaphase. This is consistent with the delay in chromosome alignment reported for the overexpression of cyclin A in HeLa and PtK1 cells (den Elzen and Pines, 2001) and overexpression of stable cyclin AS in Drosophila embryos (Parry et al., 2003). In Apc6/Cdc16RNAi and Apc8/Cdc23RNAi cells, degradation of cyclin A facilitates chromosome alignment, whereas the imposed mitotic arrest prolongs metaphase and hence the higher number of cells with overcondensed but precisely aligned chromosomes. The chromosomes of cells escaping from or passing this phase become scattered similarly to mks1 arrested cells. This appearance is the result of uncoordinated, oscillating poleward movements of chromosomes caused by the accumulation of cyclin B (Parry and O'Farrell, 2001). Since a high level of cyclin A has a similar effect on chromosome movements in the arrested cells as cyclin B (den Elzen and Pines, 2001), the accumulation of both cyclin A and B prevents chromosome alignment in mks1 cells.

The diverse phenotypic features of the TPR mutants are somewhat surprising and appear to be inconsistent with the predicted role of TPR subunits to form a central scaffold-like structure to mediate APC/C assembly (Vodermaier et al., 2003; Passmore et al., 2005). One would expect such a scaffold to become unstable or collapse when its components are removed one by one or in double mutant combinations, leading to inactive APC/C, but we could not detect that in our RNAi lines. Instead, all single and double TPRRNAi lines had slightly smaller than normal larval brains and imaginal discs that indicate reasonable mitotic activity and the existence of an APC/C with at least residual functions. Since the TPR subunits show extensive structural similarities, it is conceivable that a missing subunit can be substituted by another one. This could theoretically explain the relative stability of the scaffold structure but it would not account for the different phenotypes of the TPR subunits. Therefore, the data presented here better fit an alternative model of APC/C that is centred on the largest subunit, the Apc1/Shtd (Shattered in Drosophila), as a scaffold (Thornton et al., 2006). In this model, the Apc1/Shtd provides a platform to which all the other subunits bind by forming functional subcomplexes. Consistently with this model, the loss-of-function Apc1/shtd Drosophila mutants express an extremely strong mitotic phenotype including larval lethality, lack of imaginal discs, strong metaphase arrest with no or rare anaphases, and very high frequency of polyploidy (M.P. and P.D., unpublished). The mitotic phenotype of Apc1/shtd mutants appears stronger than all the other known APC/C subunit mutants, and RNAi lines presented here, suggesting complete loss of APC/C structure and function. Taken all data together, our view is that the TPR subunits may form functional subcomplexes of the APC/C that could bind to the Apc1/Shtd scaffold and most probably be involved in activator and substrate binding.

Removal of most of the TPR subunits seems to have a direct effect on cell viability as well. This is apparent from the significant increase in the number of apoptotic cells in the mitotically active larval brains of Apc6/Cdc16RNAi and Apc8/Cdc23RNAi (Fig. 7), or mks1 animals (Deak et al., 2003). These perturbations of APC/C function cause stabilization of cyclin B, thus leading to prolonged activation of the mitotic kinase, Cdc2, at the metaphase to anaphase progression and beyond, when its activity must normally plummet to allow mitosis to proceed. A plausible explanation for the high level of apoptosis is that the extended active state of Cdc2 during late mitosis and mitotic exit could induce cell death. High Cdc2 activity has been found in apoptotic conditions and shown to be required for the induction of this process (Fotedar et al., 1995; Konishi et al., 2002). Since the molecular details of Cdc2 action are not fully known, further investigation is required to link cell cycle regulation to the cell death machinery.

Materials and Methods

Fly stocks

Fly stocks were cultured at 25°C on standard Drosophila food. A w1118 isogenic stock was used as wild type (Ryder et al., 2004). All genetic markers used are described by Lindsley and Zimm (Lindsley and Zimm, 1992). Stocks were obtained from the Bloomington Stock Center unless stated otherwise. For phenotypic analysis, the chromosomes carrying a transgenic construct were balanced over CyO, ActGFP or T(2;3)TSTL, CyO: TM6B, Tb.

The Apc71 mutant was established after imprecise excision of the P{EPgy2}EY11333 element localized 63 bp downstream of the predicted stop codon of CG14444 (Bellen et al., 2004). Five lines were recovered carrying internal deletions of this gene. The largest deletion removed 1576 bp, about two-thirds of the gene, starting at the insertion site and extending toward the 5′ end of the Apc7 gene. The insertion site is marked by a 24 bp long remnant of the original P{EPgy2}EY11333 element. The resulting truncated Apc7 protein lacks two-thirds of its mass including the eight tandem TPR repeats. This mutation was designated as Apc71 and used in this work as a knocked out allele of the Apc7 gene.

Transgenic RNA-interference lines

To make transgenic double-stranded RNA-interference constructs, PCR primer pairs of 5′-ACTCTAGATTGCCTGTCTGGTGGAAAAC-3′ and 5′-ATTCTAGACTTGCGCAGCGAATGACC-3′; 5′-ACTCTAGAGCTCTCGCCCAAATGTTCA-3′ and 5′-ATTCTAGAGAGGCGTATCGTCCAGTTGC-3′; and 5′-ACTCTAGAACGCGCCCTGAAACTGAAT-3′ and 5′-ATTCTAGATCGACGCATGCCCTGAAT-3′ were used to amplify 656 bp, 917 bp and 910 bp sequences corresponding to exon 5, exon 3 and exon 2 and 3 of the Drosophila Cdc16, Apc7 and Cdc23 transcripts, respectively (XbaI sites are underlined). The PCR products were cloned into the pWIZ vector (a gift from Richard Carthew, Northwestern University, USA), in opposite orientation on both sides of the white intron present in the vector (Lee and Carthew, 2003). The obtained gene-specific RNAi constructs were designated as P{UST-Cdc16RNAi}, P{UST-Apc7RNAi} and P{UST-Cdc23RNAi}, and all constructs were verified before being processed for transformation. Following P-element-mediated germline transformation (Spradling and Rubin, 1982) of w1118 isogenic flies, at least five independent transgenic lines were obtained for each gene. To test RNAi effects, homozygous transgenic flies carrying subunit-specific RNAi constructs (such as P{UST-Cdc16RNAi}) were crossed to Act5C-GAL4 females which expressed the Gal4 transcription factor under the control of the actin5C promoter. The progeny of these crosses were examined for subunit-specific gene expression as well as lethal and mitotic phenotypes.

RT-PCR

Total RNA was isolated from young pupae using the Tri Reagent extraction kit (Sigma). The RNA samples were DNase treated with RQ1 RNase-Free DNase (Promega). cDNA was synthesized from 5 μg total RNA as template with M-MuLV reverse transcriptase (Fermentas, Vilnius, Lithuania) and random hexanucleotide primers (Fermentas, Vilnius, Lithuania). Two primers complementary to the 4. and 5. exons of Apc6, to the 2. and 3. exons of Apc7 and to the 1. and 3. exons of Apc8 were used in 20 cycle PCR amplifications. 20 cycle PCRs were performed using rpL17A primers (rpL17A upper, 5′-GTGATGAACTGTGCCGACAA-3′; rpL17A lower, 5′-CCTTCATTTCGCCCTTGTTG-3′) on the same cDNA templates to serve as calibration controls. RT-PCR products were separated by agarose gel electrophoresis.

Cytology

Brains from third instar larvae were dissected in a large drop of 0.7% NaCl solution then transferred to a drop of 45% acetic acid on a microscope slide. After 10-15 seconds, the liquid was absorbed with a piece of filter paper and replaced with a drop of 3% orcein (Gurr's 23282) in 45% acetic acid. After 3-5 minutes staining, the orcein was removed and the brains were destained in a drop of 65% acetic acid for 5-10 seconds. Following that, the destaining solution was replaced with a drop of 3% orcein in 65% acetic acid. The drop was immediately covered with a coverslip, wrapped up in filter paper and squashed firmly for 10-15 seconds. The preparations were examined and photographed under phase-contrast optics using an Olympus BX51 upright microscope with a UPLFLN 100XO2PH plan-apochromat objective. A DP 70 digital colour camera was used to take images. Mitotic index (MI) was defined as the number of mitotic cells on an optical field. On average, 10-20 optical fields per preparation were scored. The mean number of cells at different stages of mitosis was determined for each single or double mutant line and compared using Student's t-test.

Quantification of cell death was performed by observing the morphology of orcein-stained cells, and scoring for apoptotic cells with pycnotic nuclei. The mean number of apoptotic cells was determined and used for further statistical analysis. Acridine Orange staining was performed by dissecting third instar larval brains in a drop of PBS. The brains were transferred into a drop of 1.6 μg/ml Acridine Orange (C.I. 46005, Molar Chemicals) solution and incubated for 5 minutes in the dark, then rinsed in PBS. Following that, the brains were transferred to a drop of PBS on a microscope slide, covered with a coverslip and sealed using nail polish. Before transferring the brains, two Sellotape cushion were made on the slides. The preparations were examined using an Olympus FV1000 confocal microscope within 15-20 minutes from dissection.

Epitope constructs and affinity chromatography

Plasmid constructs to express N-terminal FLAG and haemagglutinin (HA) epitope-tagged proteins in S2 cells under the control of the Actin-5C promoter were constructed using the Drosophila Gateway Vector System (http://www.ciwemb.edu/labs/murphy/Gateway%20vectors.html). First, full-length cDNAs of Apc3/Cdc27/mks, Apc7 and Apc8/Cdc23 were cloned into the appropriate entry (pENTR) vectors in frame. Following that, the ORFs of these genes were recombined either into the pAFW or pAHW destination vectors using Gateway LR clonase II Enzyme Mix (Invitrogen). The pAFW vector contains three FLAG whereas the pAHW vector contains three HA epitope tags 5′ of the Gateway cassette. Successful recombinant clones were selected based on their resistance to ampicillin and lack of ccdB toxicity. All clones were verified by sequencing before transfection. The N-terminal fusion constructs were designated in this paper as follows: pAFW-Apc7 as FLAG-Apc7 or F7; pAFW-Apc8 as FLAG-Apc8 or F8; pAHW-Apc3 as HA-Apc3 or H3; pAHW-Apc7 as HA-Apc7 or H7 and pAHW-Apc8 as HA-Apc8 or H8.

Schneider 2 (S2) tissue culture cells were either transfected with 2.5 μg plasmid DNA of FLAG-Apc7, HA-Apc7, HA-Apc3, FLAG-Apc8 or HA-Apc8 alone or co-transfected with 2.5+2.5 μg FLAG-Apc7+HA-Apc3, FLAG-Apc7+HA-Apc8 or FLAG-Apc8+HA-Apc7 double combinations using 15 μl Cellfectin (Invitrogen) in serum-free S2 cell medium (Sigma). For preparation of total protein extracts, transfected cells were dissolved in SDS sample buffer 48 hours after transfection. For affinity chromatography, cells were washed in PBS 48 hours after transfection and cell extracts were prepared by incubating the cells in 25 mM Hepes pH 7.6, 10 mM KCl, 5 mM MgCl2, 0.1 mM EDTA, 0.5 mM EGTA, 1 mM PMSF and 0.2% NP40. After 10 minutes incubation on ice, an equal volume of cytoplasmic dilution buffer (25 mM Hepes pH 7.6, 10 mM KCl, 5 mM MgCl2, 0.1 mM EDTA, 0.5 mM EGTA, 1 mM PMSF and 450 mM NaCl) was added, the nuclei were removed by centrifugation (3 minutes at 10,000 g) and FLAG-tagged proteins were affinity-purified on anti-FLAG-M2-agarose beads (Sigma-Aldrich). After 2 hours of incubation at 4°C, anti-FLAG-M2-agarose beads were washed four times with five volumes of 20 mM Tris-HCl pH 7.5, 150 mM NaCl and bound proteins were eluted with an excess of FLAG peptides. Total protein extracts or affinity-purified samples were fractionated on 8% SDS-PAGE and blotted to PVDF membrane strips. Western blots were reacted with anti-FLAG or anti-HA monoclonal antibodies.

Immunostaining

Third instar larval brains (n=3-6) were dissected in PBS, fixed in ice-cold methanol for 20 minutes, washed three times for 5 minutes in PBS and than permeabilised in PBS containing 0.3% Triton X-100 for 10 minutes at room temperature and washed again in PBS containing 1% BSA for 40 minutes at room temperature. The brains were incubated with primary antibodies diluted to an appropriate concentration in PBS containing 1% BSA overnight in a humid chamber. They were than washed four times for 15 minutes each with PBST (PBS supplemented with 0.1% Triton X-100) containing 1% BSA at room temperature before being incubated with appropriate secondary antibodies (Jackson) diluted in PBST containing 1% BSA for 2 hours at room temperature in the dark. The preparations were finally washed twice for 15 minutes in PBST and two times 15 minutes in PBS. DNA was stained with TOTO-3 (Molecular Probes) before the specimens were mounted in Vectashield.

The primary antibodies were Rb270 rabbit anti-cyclin A antibody (Whitfield et al., 1990), Rb271 rabbit anti-cyclin B antibody (Whitfield et al., 1990) and rat monoclonal anti-α-tubulin antibody, YL1/2 (Kilmartin at al., 1982). The preparations were examined using an Olympus FV1000 confocal microscope. For comparing cyclin levels, images of metaphase cells were acquired using identical settings and cyclin immunostaining was determined by visual inspection of the images.

Acknowledgements

We wish to thank Katalin Udvardy for her contribution in carrying out the S2 cell transfection experiments. We are grateful to Mary Lilly and Edward Wojcik for discussions and comments on the manuscript. This work was supported by a Hungarian Scientific Research Fund (OTKA) grant awarded to P.D. (T037322).

References

Altschul, S. F., Madden, T. L., Schaffer, A. A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D. J. (
1997
). Gapped BLAST and PSI-BLAST: a new generation of protein database search programs.
Nucleic Acids Res.
25
,
3389
-3402.
Bainbridge, S. P. and Bownes, M. (
1981
). Staging the metamorphosis of Drosophila melanogaster.
J. Embryol. Exp. Morphol.
66
,
57
-80.
Bellen, H. J., Levis, R. W., Liao, G., He, Y., Carlson, J. W., Tsang, G., Evans-Holm, M., Hiesinger, P. R., Schulze, K. L., Rubin, G. M. et al. (
2004
). The BDGP gene disruption project: single transposon insertions associated with 40% of Drosophila genes.
Genetics
167
,
761
-781.
Bentley, A. M., Williams, B. C., Goldberg, M. L. and Andres, A. J. (
2002
). Phenotypic characterization of Drosophila ida mutants: defining the role of APC5 in cell cycle progression.
J. Cell Sci.
115
,
949
-961.
Blatch, G. L. and Lässle, M. (
1999
). The tetratricopeptide repeat: a structural motif mediating protein-protein interactions.
BioEssays
21
,
932
-939.
Carroll, C. W. and Morgan, D. O. (
2002
). The Doc1 subunit is a processivity factor for the anaphase-promoting complex.
Nat. Cell Biol.
4
,
880
-887.
Das, A. K., Cohen, P. W. and Barford, D. (
1998
). The structure of the tetratricopeptide repeats of protein phosphatase 5: implications for TPR-mediated protein-protein interactions.
EMBO J.
17
,
1192
-1199.
Deak, P., Donaldson, M. and Glover, D. M. (
2003
). Mutations in makos, a Drosophila gene encoding the Cdc27 subunit of the anaphase promoting complex, enhance centrosomal defects in polo and are suppressed by mutations in twins/aar, which encodes a regulatory subunit of PP2A.
J. Cell Sci.
116
,
4147
-4158.
den Elzen, N. and Pines, J. (
2001
). Cyclin A is destroyed in prometaphase and can delay chromosome alignment and anaphase.
J. Cell Biol.
153
,
121
-136.
Fotedar, R., Flatt, J., Gupta, S., Margolis, R. L., Fitzgerald, P., Messier, H. and Fotedar, A. (
1995
). Activation-induced T-cell death is cell cycle dependent and regulated by cyclin B.
Mol. Cell. Biol.
15
,
932
-942.
Gatti, M. and Baker, B. S. (
1989
). Genes controlling essential cell-cycle functions in Drosophila melanogaster.
Genes Dev.
3
,
438
-453.
Gmachl, M., Gieffers, C., Podtelejnikov, A. V., Mann, M. and Peters, J. M. (
2000
). The RING-H2 finger protein APC11 and the E2 enzyme UBC4 are sufficient to ubiquitinate substrates of the anaphase-promoting complex.
Proc. Natl. Acad. Sci. USA
97
,
8973
-8978.
Harper, J. W., Burton, J. L. and Solomon, M. J. (
2002
). The anaphase-promoting complex: it's not just for mitosis any more.
Genes Dev.
16
,
2179
-2206.
Huang, J. Y. and Raff, J. W. (
2002
). The dynamic localisation of the Drosophila APC/C: evidence for the existence of multiple complexes that perform distinct functions and are differentially localised.
J. Cell Sci.
115
,
2847
-2856.
Kilmartin, J. V., Wright, B. and Milstein, C. (
1982
). Rat monoclonal antitubulin antibodies derived by using a new nonsecreting rat cell line.
J. Cell Biol.
93
,
576
-582.
Konishi, Y., Lehtinen, M., Donovan, N. and Bonni, A. (
2002
). Cdc2 phosphorylation of BAD links the cell cycle to the cell death machinery.
Mol. Cell
9
,
1005
-1016.
Lamb, J. R., Michaud, W. A., Sikorski, R. S. and Hieter, P. A. (
1994
). Cdc16p, Cdc23p and Cdc27p form a complex essential for mitosis.
EMBO J.
13
,
4321
-4328.
Lamb, J. R., Tugendreich, S. and Hieter, P. (
1995
). Tetratrico peptide repeat interactions: to TPR or not to TPR?
Trends Biochem. Sci.
20
,
257
-259.
Lindsley, D. L. and Zimm, G. G. (
1992
).
The Genome of Drosophila melanogaster
. San Diego: Academic Press.
Lee, Y. S. and Carthew, R. W. (
2003
). Making a better RNAi vector for Drosophila: use of intron spacers.
Methods
30
,
322
-329.
Leverson, J. D., Joazeiro, C. A., Page, A. M., Huang, H., Hieter, P. and Hunter, T. (
2000
). The APC11 RING-H2 finger mediates E2-dependent ubiquitination.
Mol. Biol. Cell
11
,
2315
-2325.
Ohta, T., Michel, J. J., Schottelius, A. J. and Xiong, Y. (
1999
). ROC1, a homolog of APC11, represents a family of cullin partners with an associated ubiquitin ligase activity.
Mol. Cell
3
,
535
-541.
Parry, D. H. and O'Farrell, P. H. (
2001
). The schedule of destruction of three mitotic cyclins can dictate the timing of events during exit from mitosis.
Curr. Biol.
11
,
671
-683.
Parry, D. H., Hickson, G. R. and O'Farrell, P. H. (
2003
). Cyclin B destruction triggers changes in kinetochore behavior essential for successful anaphase.
Curr. Biol.
13
,
647
-653.
Passmore, L. A. (
2004
). The anaphase-promoting complex (APC): the sum of its parts?
Biochem. Soc. Trans.
32
,
724
-727.
Passmore, L. A., Booth, C. R., Venien-Bryan, C., Ludtke, S. J., Fioretto, C., Johnson, L. N., Chiu, W. and Barford, D. (
2005
). Structural analysis of the anaphase-promoting complex reveals multiple active sites and insights into polyubiquitylation.
Mol. Cell
20
,
855
-866.
Reed, B. H. and Orr-Weaver, T. L. (
1997
). The Drosophila gene morula inhibits mitotic functions in the endo cell cycle and the mitotic cell cycle.
Development
124
,
3543
-3553.
Ryder, E., Blows, F., Ashburner, M., Bautista-Llacer, R., Coulson, D., Drummond, J., Webster, J., Gubb, D., Gunton, N., Johnson, G. et al. (
2004
). The DrosDel collection: a set of P-element insertions for generating custom chromosomal aberrations in Drosophila melanogaster.
Genetics
167
,
797
-813.
Spradling, A. C. and Rubin, G. M. (
1982
). Transposition of cloned P elements into Drosophila germ line chromosomes.
Science
218
,
341
-347.
Tang, Z., Li, B., Bharadwaj, R., Zhu, H., Ozkan, E., Hakala, K., Deisenhofer, J. and Yu, H. (
2001
). APC2 Cullin protein and APC11 RING protein comprise the minimal ubiquitin ligase module of the anaphase-promoting complex.
Mol. Biol. Cell
12
,
3839
-3851.
The FlyBase Consortium (
2003
). The FlyBase database of the Drosophila genome projects and community literature.
Nucleic Acids Res.
31
,
172
-175.
Thornton, B. R., Ng, T. M., Matyskiela, M. E., Carroll, C. W., Morgan, D. O. and Toczyski, D. P. (
2006
). An architectural map of the anaphase-promoting complex.
Genes Dev.
20
,
449
-460.
Vodermaier, H. C., Gieffers, C., Maurer-Stroh, S., Eisenhaber, F. and Peters, J. M. (
2003
). TPR subunits of the anaphase-promoting complex mediate binding to the activator protein CDH1.
Curr. Biol.
13
,
1459
-1468.
Whitfield, W. G., Gonzalez, C., Maldonado-Codina, G. and Glover, D. M. (
1990
). The A- and B-type cyclins of Drosophila are accumulated and destroyed in temporally distinct events that define separable phases of the G2-M transition.
EMBO J.
9
,
2563
-2572.
Yu, H., Peters, J. M., King, R. W., Page, A. M., Hieter, P. and Kirschner, M. W. (
1998
). Identification of a cullin homology region in a subunit of the anaphase-promoting complex.
Science
279
,
1219
-1222.