Portal de Periódicos da CAPES

Drosophila Cdk4 is required for normal growth and is dispensable for cell cycle progression

2000; Springer Nature; Volume: 19; Issue: 17 Linguagem: Inglês

10.1093/emboj/19.17.4533

1460-2075

C Meyer, Henning W. Jacobs, Sanjeev A. Datar, Wei Du, Bruce A. Edgar, Christian F. Lehner,

Ubiquitin and proteasome pathways

Article1 September 2000free access Drosophila Cdk4 is required for normal growth and is dispensable for cell cycle progression Claas A. Meyer Claas A. Meyer Department of Genetics, University of Bayreuth, 95440 Bayreuth, Germany Search for more papers by this author Henning W. Jacobs Henning W. Jacobs Department of Genetics, University of Bayreuth, 95440 Bayreuth, Germany Search for more papers by this author Sanjeev A. Datar Sanjeev A. Datar Division of Basic Sciences, Fred Hutchison Cancer Research Center, Seattle, WA, 98109 USA Search for more papers by this author Wei Du Wei Du Ben May Institute for Cancer Research, University of Chicago, Chicago, IL, 60637 USA Search for more papers by this author Bruce A. Edgar Bruce A. Edgar Division of Basic Sciences, Fred Hutchison Cancer Research Center, Seattle, WA, 98109 USA Search for more papers by this author Christian F. Lehner Corresponding Author Christian F. Lehner Department of Genetics, University of Bayreuth, 95440 Bayreuth, Germany Search for more papers by this author Claas A. Meyer Claas A. Meyer Department of Genetics, University of Bayreuth, 95440 Bayreuth, Germany Search for more papers by this author Henning W. Jacobs Henning W. Jacobs Department of Genetics, University of Bayreuth, 95440 Bayreuth, Germany Search for more papers by this author Sanjeev A. Datar Sanjeev A. Datar Division of Basic Sciences, Fred Hutchison Cancer Research Center, Seattle, WA, 98109 USA Search for more papers by this author Wei Du Wei Du Ben May Institute for Cancer Research, University of Chicago, Chicago, IL, 60637 USA Search for more papers by this author Bruce A. Edgar Bruce A. Edgar Division of Basic Sciences, Fred Hutchison Cancer Research Center, Seattle, WA, 98109 USA Search for more papers by this author Christian F. Lehner Corresponding Author Christian F. Lehner Department of Genetics, University of Bayreuth, 95440 Bayreuth, Germany Search for more papers by this author Author Information Claas A. Meyer1, Henning W. Jacobs1, Sanjeev A. Datar2, Wei Du3, Bruce A. Edgar2 and Christian F. Lehner 1 1Department of Genetics, University of Bayreuth, 95440 Bayreuth, Germany 2Division of Basic Sciences, Fred Hutchison Cancer Research Center, Seattle, WA, 98109 USA 3Ben May Institute for Cancer Research, University of Chicago, Chicago, IL, 60637 USA ‡C.A.Meyer and H.W.Jacobs contributed equally to this work *Corresponding author. E-mail: [email protected] The EMBO Journal (2000)19:4533-4542https://doi.org/10.1093/emboj/19.17.4533 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Complexes of D-type cyclins and cdk4 or 6 are thought to govern progression through the G1 phase of the cell cycle. In Drosophila, single genes for Cyclin D and Cdk4 have been identified, simplifying genetic analysis. Here, we show that Drosophila Cdk4 interacts with Cyclin D and the Rb homolog RBF as expected, but is not absolutely essential. Flies homozygous for null mutations develop to the adult stage and are fertile, although only to a very limited degree. Overexpression of inactive mutant Cdk4, which is able to bind Cyclin D, does not enhance the Cdk4 mutant phenotype, confirming the absence of additional Cyclin D-dependent cdks. Our results indicate, therefore, that progression into and through the cell cycle can occur in the absence of Cdk4. However, the growth of cells and of the organism is reduced in Cdk4 mutants, indicating a role of D-type cyclin-dependent protein kinases in the modulation of growth rates. Introduction D-type cyclins, their kinase partners cdk4 and 6, the INK inhibitors and the kinase substrate retinoblastoma protein (Rb) are all known for their crucial importance in human tumorigenesis (Weinberg, 1995; Sherr, 1996; Sherr and Roberts, 1999). At the cellular level, Rb has been shown to regulate progression through the G1 phase of the mammalian cell cycle, predominantly by binding to E2F transcription factors, which control a large number of genes involved in cell proliferation and DNA replication (Dyson, 1998). Rb represses expression of E2F target genes by recruiting histone deacetylase activity and by inhibiting the E2F transcriptional activation domain (Harbour et al., 1999). The ability of Rb to block progression through G1 is abolished by Rb hyperphosphorylation, which is initiated by D-type cyclin-dependent kinases during G1 (Sherr, 1994; Sherr and Roberts, 1999). In mammalian cells, the synthesis of D-type cyclins is controlled by extracellular signals. Mitogens induce a rapid accumulation of D-type cyclins. Conversely, antimitogens or withdrawal of mitogens result in a rapid decline. D-type cyclins are therefore thought to function as a functional link between extracellular signals and the cell cycle machinery. Accordingly, D-type cyclin–cdk complexes might be predicted to play an important role in the regulation of cell proliferation during development. In mice, a number of genes of the INK (Serrano et al., 1996; Franklin et al., 1998), cyclin D (Fantl et al., 1995; Sicinski et al., 1995, 1996), cdk4 (Rane et al., 1999; Tsutsui et al., 1999), Rb (Clarke et al., 1992; Jacks et al., 1992; Lee et al., 1992, 1996; Cobrinik et al., 1996), E2F (Field et al., 1996; Yamasaki et al., 1996) pathway have been knocked out. In general, cell proliferation during early embryonic development is normal in the absence of these genes. In addition, mouse embryo fibroblasts derived from homozygous mutants proliferate in cell culture. It is often not clear to what extent functional redundancies explain the absence of severe cell proliferation defects in these mutants, because multiple related pathway components encoded by small gene families are present in mammals. The three mammalian D-type cyclins, for instance, can bind to either cdk4 or cdk6 and the different complexes might have largely overlapping functions (Sherr, 1994). While genetic inactivation of either cyclin D1, D2 or cdk4 does not block cell cycle progression, overexpression of INK inhibitors (Guan et al., 1994; Koh et al., 1995; Lukas et al., 1995b; Medema et al., 1995) and microinjection of antibodies against D-type cyclins (Baldin et al., 1993; Quelle et al., 1993; Lukas et al., 1994, 1995a) have indicated that Rb+ cells fail to progress into S-phase when D-type cyclin–cdk complexes are inhibited. In Drosophila, single genes for Cyclin D and its kinase partner Cdk4 (previously designated Cdk4/6) have been described (Finley et al., 1996; Sauer et al., 1996), as well as in the nematode Caenorhabditis elegans (Park and Krause, 1999). Extensive screening of genomic libraries at low stringency has not revealed additional cdk4-like genes in Drosophila (Sauer et al., 1996). Additional cdk4 homologs also cannot be identified in the large set of Drosophila EST sequences and in the published genome sequence (Adams et al., 2000). Therefore, the analysis of the Cdk4 mutant phenotype, which we describe here, can be expected to provide a definitive answer to the question of whether cdk4 activity represents an obligatory requirement for progression through the G1 phase. In addition, the cell proliferation program of wild-type Drosophila development is very well known and the effects of mutations can be studied at the cellular level within the organism. Our results demonstrate that Cdk4 interacts genetically with the Drosophila Rb family member RBF, as expected from the analyses in mammals. However, Cdk4 is not essential for progression through the cell cycle or for development to the adult stage. Nevertheless, Cdk4 is clearly required for normal fertility and growth of cells and organism. Results Drosophila Cdk4 binds to Cyclin D in vivo Analysis of full-length cDNAs indicated that the characteristic pRb-binding motif LXCXL, which is not encoded by the sequence described previously (Finley et al., 1996), is also present in Drosophila Cyclin D. Co-immunoprecipitation experiments confirmed that Drosophila Cyclin D associates with Cdk4 in vivo (Figure 1). Immunoblotting with a monoclonal antibody revealed the presence of Cyclin D in anti-myc immunoprecipitates isolated from extracts of Cdk4-myc-expressing embryos. In contrast, Cyclin D was not detected in immunoprecipitates from Cdk1-myc or Cdk2-myc extracts. Moreover, while Cyclins A, B, B3 and E were clearly present in either of these two latter immunoprecipitates, they were not detected in Cdk4-myc immunoprecipitates. Our results indicate therefore that Cyclin D binds specifically to Cdk4. Figure 1.Cyclin D but not Dacapo is bound to Drosophila Cdk4 in vivo. Immunoprecipitates (IP) were isolated with anti-myc antibodies from extracts of embryos expressing either myc-tagged Cdk1 (Cdk1-myc), Cdk2 (Cdk2-myc), Cdk4 (Cdk4-myc) or mutant Cdk4 (Cdk4D175N-myc) and analyzed in immunoblot experiments (IB) with antibodies against the myc epitope (myc), Cyclin D (CycD), Cyclin E (CycE), Cyclin A (CycA), Cyclin B (CycB), Cyclin B3 (CycB3) or Dacapo (DAP). Immunoblot signals resulting from the binding of secondary antibodies to mouse immunoglobulin used for immunoprecipitation are indicated by dots, while arrowheads mark signals reflecting the presence of Cyclin D and Cyclin B. Two independent experiments are shown in the panels on the right and left sides. Download figure Download PowerPoint Mammalian cdk4 and cdk6 bind to INK- and CIP/KIP-type cdk inhibitors. While INK inhibitors have not been identified in Drosophila, the dacapo gene has been shown to encode a CIP/KIP-type inhibitor, which binds to Cyclin E–Cdk2 complexes (De Nooij et al., 1996; Lane et al., 1996). While Dacapo was readily observed in Cdk2-myc immunoprecipitates as expected, it could not be detected in Cdk4-myc immunoprecipitates (Figure 1). Cdk4 is required for normal growth and fertility but is not essential for cell cycle progression By mobilizing a P element [l(2)s4639] we isolated an intragenic deletion Cdk43 eliminating essential kinase domains (Figure 2). Surprisingly, homozygous Cdk43 progeny from heterozygous parents developed into adult flies which eclosed without a developmental delay. While the fertility of homozygous females was severely reduced (see Supplementary material available at The EMBO Journal Online for a more detailed description of the fertility defects), occasional progeny could be obtained even from homozygous parents demonstrating that Cdk4 is not absolutely essential for cell proliferation or development to the adult stage. Figure 2.Molecular characterization of wild-type and mutant Cdk4 alleles. (A) Cdk4 exon sequences are indicated by boxes. Filled boxes indicate translated regions. Putative transcriptional start sites are indicated by arrows and the structure of alternative transcripts is illustrated. Triangles indicate insertion sites of P elements. While the line EP(2)0844 carries a P element in the 5′ region, the following lines have insertions clustered around the position indicated by the second triangle: l(2)05428, l(2)s4639, l(2)k06503, EP(2)2192, EP(2)2358. The size and position of the intragenic deletion in Cdk43 resulting from an imprecise excision of l(2)s4639 is indicated by the black bar. The positions of the primers 1 and 2 used for the PCR experiment (B) are indicated by the open arrows. The regions used as probes 3 and 4 for the Southern blot (C) are indicated by boxes and relevant BamHI restriction sites are indicated. (B) Genomic DNA from wild-type (+/+), Cdk43/CyO (+/−) and Cdk43 (−/−) flies was analyzed by PCR for the presence of Cdk4 sequences. The primers 1 and 2 (A) result in the amplification of a 2.1 kb fragment from the wild-type allele and a 0.3 kb fragment from the Cdk43 allele. (C) Genomic DNA from Cdk43/CyO and Cdk43 flies was digested with BamHI and analyzed on Southern blots using Cdk4 fragments 3 and 4 (A) as probes. (D) 0–2 h embryo extracts from either wild type (+/+) or Cdk43 (−/−) were analyzed for the presence of Cdk4 by immunoblotting with anti-Cdk4 antisera (CDK4). Immunoblotting with anti-tubulin (TUB) was used as a loading control. Download figure Download PowerPoint Interestingly, homozygous Cdk43 flies were found to be significantly smaller than wild-type flies (Table I; Figure 3). Size reduction affected all aspects of the flies proportionally. The weight of Cdk43 homozygotes was found to be ∼20% lower than that of heterozygotes (Table I). The wing area of Cdk43 homozygotes was found to be ∼10% lower than that of heterozygotes. As each cell is known to secrete one hair during wing development, we determined the hair density in a defined region of the adult wings as a measure of cell size. The results indicated that the cells in wings of Cdk43 homozygotes are slightly larger compared with heterozygotes (Table I). The values determined for the total wing size and for the cell size were used to extrapolate the total cell number present in the wing. These estimations indicated that the wings of Cdk43 homozygotes contained fewer cells than those of heterozygotes (Table I). Figure 3.Cdk4 is required for normal growth of organism and organs. (A–F) Adult male flies (A–C) and wings (D–F) are shown. The region boxed in (D) was used to determine cell densities (see Table I). Genotypes were +/Cdk43 (Cdk4+), Cdk43/Cdk43 (Cdk43) and Cdk43/Cdk43; da-GAL4, UAS-Cdk4 III.1 (Cdk43; UAS-Cdk4). (G) Males heterozygous for a UAS transgene (UAS-Cdk4 III.1, UAS-Cdk4D175N-myc III.1, UAS-Cdk4D175N-myc III.7 or UAS-Cdk1 III.1) were crossed with females homozygous for da-GAL4 (black bars) or with control females (white bars). The UAS transgene therefore was only present in 50% of the progeny from these crosses. The weight of progeny with or without UAS transgene was determined and the weight ratio (+UAS transgene/−UAS transgene) was calculated. Black bars represent the weight ratio in flies carrying da-GAL4 (+); white bars indicate the weight ratio in flies lacking da-GAL4 (−). (H and I) Cdk43/CyO; UAS transgene/+ males were crossed with Cdk43/CyO; da-GAL4/da-GAL4 females. Thus, only 50% of the progeny expressed the UAS transgene (UAS-Cdk4 III.1, UAS-Cdk4-myc III.4, UAS-Cdk4D175N-myc III.7, UAS-Cdk4K55M-myc III.2, UAS-Cdk1-myc III.1 or UAS-Cdk1 III.1). The weight of progeny that either expressed the UAS transgene or not was determined and the weight ratio (+UAS transgene/−UAS transgene) was calculated. Weight ratios were determined for male (H) and female (I) progeny. White bars indicate the weight ratio in homozygous Cdk43/Cdk43 progeny (−); black bars in heterozygous Cdk43/CyO siblings (+). The significant P values as determined by a t-test are indicated by asterisks (*P <0.05; **P <0.01). Download figure Download PowerPoint Table 1. The effect of Cdk4 on the size of cells, organs and organism Crossa Genotype Weightb (mg) Wing areac (mm2) Cell sized (μm2) Cell numbere 1 Cdk43/Cdk43 0.69 1.19 137.9 8573 2 Cdk43/+ 0.84 1.27 130.3 9687 3 Cdk43/Cdk43; da-GAL4/+ 0.68 1.14 146.5 7729 Cdk43/Cdk43; da-GAL4/UAS-Cdk4 III.1 0.94* 1.33** 137.9*** 9617 4 Cdk43/Cdk43; da-GAL4/+ 0.73 1.16 147.5 7811 Cdk43/Cdk43; da-GAL4/UAS-Cdk4D175N-mycIII.7 0.77 1.17 143.2 8114 5 Cdk43/Cdk43; da-GAL4/+ 0.73 1.15 149.5 7641 Cdk43/Cdk43; da-GAL4/UAS-Cdk4D175N-mycIII.4 0.77 1.22 145.6 8328 a The crosses that resulted in progeny with the different genotypes analyzed are described in Materials and methods. As the growth of flies is very sensitive to growth conditions, comparisons between sibling progeny of a cross raised in the same bottle are most accurate. The values given in Table I are from experiments that were independent of those used for the calculation of the weight ratios (Figure 3G and H). b The average weight of male flies was determined as described in Materials and methods. Standard deviations (SDs) were smaller than 0.04 mg, except for the Cdk43/Cdk43; da-GAL4/+ flies in cross 3 where SD was 0.06. c The area of 10 wings obtained from 10 different males was measured and averaged, except for cross 5 where only five wings were analyzed. SDs were <0.03 mm2 except for cross 2 where the SD was 0.05 mm2. d An average cell size was determined as described in Materials and methods by analyzing the same wings that were also used for the wing area measurements. SDs were <5 μm2. e Total cell numbers present on one side of the wing blade were estimated by dividing the wing area by the cell size. Note that cell density is not uniform throughout the wing blade and that cell size was determined in a region that has a relatively low cell density. Comparison with sibling flies with t-test results in P <10−3. Comparison with sibling flies with t-test results in P <10−10. Comparison with sibling flies with t-test results in P <10−2. To demonstrate that the decrease in weight, wing size and cell numbers observed in Cdk43 homozygotes is caused by loss of Cdk4 function and not by potential second site mutations on the chromosome, we expressed UAS-Cdk4 ubiquitously in Cdk43 homozygotes with the help of da-GAL4. This prevented the reduction in weight, wing size and cell numbers (Table I; Figure 3). Analogous ubiquitous expression of mutant Cdk4D175N-myc had essentially no effect in Cdk43 homozygotes (Table I). The D175N mutation affects an aspartate residue which is conserved in all protein kinases and known to be required for the phosphotransfer reaction (van den Heuvel and Harlow, 1993). Overexpression of cdk1 and cdk2 with analogous mutations results in dominant-negative inhibition of the corresponding endogenous cdks in mammalian cells. Mutant Cdk4D175N-myc protein binds to Cyclin D with the same efficiency as Cdk4-myc (Figure 1) and thus is expected to act in a dominant-negative manner. While Cdk4D175N-myc expression did not reduce the weight of Cdk43 mutants, it resulted in a slight but significant weight decrease in Cdk4+ flies (Figure 3G). In additional experiments, we compared directly the effects of Cdk4D175N-myc expression on the weight of Cdk43 homo- and heterozygotes that had developed in the same bottle (Figure 3H and I). These experiments clearly confirmed that UAS-Cdk4D175N-myc reduces fly weight when expressed in Cdk43 heterozygotes, while it had at most the opposite effect in Cdk43 homozygotes (Figure 3H and I). Moreover, this same differential effect on Cdk43 homo- and heterozygotes was also observed with a UAS-Cdk4K55M-myc transgene (Figure 3H and I), which carried another mutation known to abolish kinase activity but not D-type cyclin binding, when introduced into mammalian cdk4 (Kato and Sherr, 1993). The finding that both Cdk4D175N-myc and Cdk4K55M-myc expression mimicked the Cdk43 phenotype when expressed in Cdk4+ flies suggests that they act in a dominant-negative manner. The fact that they have no effect in Cdk43 flies argues strongly that Cdk4 is the only Cyclin D-dependent cdk in Drosophila. We emphasize that this conclusion is complicated but not invalidated by the possibility that Cdk4 might not only act as a protein kinase but also by titrating putative INK inhibitors as described for mammalian cdk4/6. We point out that the Drosophila genome sequence (Adams et al., 2000) has not revealed INK orthologs. To analyze the effects of loss of Cdk4 function on cell growth and cell cycle progression, we compared the behavior of homozygous Cdk43 and their wild-type sister clones induced by mitotic recombination (Figure 4). Sixty-seven hours after clone induction, the area covered by Cdk43 clones in imaginal wing discs was found to be only about two-thirds of the area covered by sister clones (Figure 4A). Analysis by flow cytometry revealed a significant difference neither in cell size nor in the cell cycle profile (Figure 4B). In addition, we did not observe pyknotic nuclei in Cdk43 mutant clones, suggesting that cell death did not play a significant role in the reduced growth of Cdk43 mutant clones. These data indicate that Cdk43 cells grow slowly and that their cell cycle is lengthened by a proportionate increase in the G1, S and G2 phases. Figure 4.Cdk4 is required for normal growth of imaginal disc cells. (A) Twin spots of homozygous Cdk43 clones and homozygous Cdk4+ clones were induced in a Cdk43/Cdk4+ background by mitotic recombination at 48 h AED. Wing imaginal discs were fixed at 115 h AED and the size of Cdk43 (black) and Cdk4+ clones (gray) was determined. The sizes measured for clone pairs are indicated by bars ordered according to the size of the Cdk4+/Cdk4+ clones. Values for median and average clone area are indicated. (B) Cdk43 clones lacking UAS-GFP were induced by mitotic recombination at 48 h AED. Wing imaginal discs were dissected and dissociated for FACS analysis at 96 h AED. Green fluorescent protein (GFP) fluorescence intensity and DNA content were analyzed as well as forward scatter (FCS) as a measure of cell size. Cells were gated as indicated on the left panel and gray traces reflect Cdk43 mutant cells (GFP-negative) and black traces cells with either one or two Cdk4+ copies (GFP-positive) in the panels illustrating essentially indistinguishable cell cycle profiles and cell size distribution. Download figure Download PowerPoint Cdk4 interacts with RBF and Cyclin E–Cdk2 In addition to D-type cyclin–cdk complexes, cyclin E–cdk2 has also been implicated in the regulation of cell cycle progression through the G1 phase. The presence of Cyclin E–Cdk2 might explain the relatively mild phenotype observed in flies lacking Cdk4 function. To evaluate this notion, we studied the effects of heterozygosity for Cyclin E and Cdk2 mutations in Cdk4 mutants. While heterozygosity for mutations in Cyclin E is readily tolerated in Cdk43 heterozygotes, it resulted in complete lethality in Cdk43 homozygotes. Similarly, heterozygosity for mutations in Cdk2 resulted in an almost complete lethality in Cdk43 homozygotes, while it has no effect in Cdk43 heterozygotes. In contrast, mutations in Cyclin A, Cyclin B, Cyclin B3 and Cdk1 had no effect on the survival of Cdk43 homozygotes. These observations demonstrate that Cdk4 mutants are particularly sensitive to reduction in Cyclin E–Cdk2 levels. Vertebrate D-type cyclin–cdk complexes inhibit Rb function. The defects observed in Cdk4 mutants, therefore, might reflect increased Rb function. To evaluate this notion, we studied the effects of heterozygosity for mutations in the gene encoding the Drosophila Rb family member (RBF) on the fertility and size of Cdk4 mutant females. For these experiments we used two putative null alleles, RBF11 and RBF14. We developed a PCR assay to monitor the presence of the RBF11 allele (Figure 5A). Cdk4 mutant females heterozygous for RBF11 (RBF11/+; Cdk43/Cdk43) were found to be significantly more fertile than sibling females without RBF11 (+/+; Cdk43/Cdk43) (Figure 5, compare B with C). In the experiment with RBF14, we were unable to monitor the presence of this mutation. Based on our crossing scheme, however, one half of the Cdk4 mutant females was expected to be heterozygous for RBF14 (RBF14/+; Cdk43/Cdk43), while the other half was expected to have two functional RBF gene copies (+/+; Cdk43/Cdk43). As in the RBF11 experiment, we also observed an increased fertility in ∼50% of the Cdk4 mutant females in the RBF14 experiment (Figure 5D). These 50% females with increased fertility, therefore, presumably correspond to the RBF14 heterozygotes. We conclude that the fertility of Cdk4 mutant females is increased by a reduction in the copy number of functional RBF genes from two to one. Similarly, heterozygosity for mutations in RBF was found to increase the weight of Cdk4 mutant females (data not shown). The antagonistic activities of Cdk4 and RBF could also be demonstrated in experiments involving overexpression using the UAS-GAL4 system (Datar et al., 2000). Figure 5.Reduction of RBF gene dose suppresses fertility defects of mutant Cdk4 females. (A) The presence of the RBF11 allele was monitored by a PCR assay. A 650 bp fragment is amplified exclusively when RBF11 is present. (B and C) Sibling Cdk43 mutant females, which were either +/+ (B) or RBF11/+ (C), were crossed individually to wild-type males and the number of progeny developing to the adult stage was determined. (D) Sibling Cdk43 mutant females, which were either +/+ or RBF14/+, were crossed individually to wild-type males and the number of progeny developing to the adult stage was determined. The presence of RBF14 could not be monitored. However, ∼50% of the females were found to have significantly higher numbers of adult progeny than Cdk43 mutant females and are therefore presumably RBF14/+. Download figure Download PowerPoint D-type cyclin complexes are thought to be the major Rb kinase in mammalian cells (Geng et al., 1999). Therefore it was of interest to compare Rb kinase activity in wild-type and Cdk4 mutant embryos. Unfortunately, our antibodies failed to precipitate Rb kinase activity even from wild-type embryo extracts. Moreover, in contrast to vertebrate Rb, Drosophila RBF does not change its apparent electrophoretic mobility upon phosphorylation (Du et al., 1996). As also observed with mammalian Rb (Khandjian and Tremblay, 1992), we were unable to focus Drosophila RBF on two-dimensional gels. Thus, using da-GAL4 we expressed mouse Rb in Drosophila embryos from a UAS transgene (UAS-mRb). Immunoblotting with an antibody against Rb clearly revealed two forms with different electrophoretic mobility (Figure 6A, lane 2). Phosphatase treatment converted the lower into the higher mobility form (Figure 6, lane 7). Conversely, co-expression of UAS-Cdk4 and UAS-Cyclin D resulted in a relative increase in the lower mobility form (Figure 6, lane 3). These observations suggested that Cyclin D–Cdk4 might in principle function as Rb kinase. However, we failed to detect a decrease in the abundance of the low mobility form in extracts from embryos lacking both maternal and zygotic Cdk4 function (Figure 6, lane 9). It appears therefore that kinases other than Cyclin D–Cdk4 can phosphorylate mRb. In fact, experiments involving expression of either UAS-Cyclin E or UAS-dacapo suggested that Cyclin E–Cdk2 is a major Rb kinase in Drosophila embryos. UAS-Cyclin E resulted in a strong enrichment of the lower mobility form (Figure 6, lane 5). In contrast, UAS-dacapo resulted in a severe reduction of the lower mobility form (Figure 6, lane 4). Figure 6.Rb kinase in Cdk4 mutants. Extracts from embryos with UAS-mRb (lane 1), da-GAL4 and UAS-mRb (lane 2), da-GAL4, UAS-mRb, UAS-Cyclin D and UAS-Cdk4 (lane 3), da-GAL4, UAS-mRb, UAS-dacapo (lane 4) and da-GAL4, UAS-mRb and UAS-Cyclin E (lane 5), were analyzed by immunoblotting with anti-Rb. Alternatively, mRb was immunoprecipitated from da-GAL4, UAS-mRb embryo extracts and incubated without phosphatase (lane 6) or with phosphatase either in the absence (lane 7) or presence (lane 8) of phosphatase inhibitors before immunoblotting with anti-Rb. To analyze the effect of Cdk4 on mRB phosphorylation, we analyzed extracts from embryos collected from a cross of Cdk43; UAS-mRb females with either Cdk43; da-GAL4 males (lane 9) or da-GAL4 males (lane 10) by immunoblotting with anti-Rb. Download figure Download PowerPoint Discussion D-type cyclin–cdk complexes are thought to control specifically progression through G1 in response to extracellular growth factors. Surprisingly, we find that Drosophila Cdk4, which encodes the kinase partner of Drosophila Cyclin D, is not an absolutely essential gene. Cdk4 mutant flies are smaller than normal and almost infertile. However, some progeny can develop into adults in the complete absence of both maternal and zygotic Cdk4 function. Moreover, we demonstrate that imaginal disc cells lacking Cdk4 function are characterized by a reduced cellular growth rate, an increased cell cycle duration with a proportional extension during G1, S and G2 and a normal cell size. Rather than promoting the progression specifically through G1 in response to cellular growth, therefore, D-type cyclin–cdk complexes appear primarily to stimulate cellular growth in Drosophila. This notion is further supported by our data from overexpression analyses (Datar et al., 2000). The discussion here is focused on Cdk4 loss-of-function mutations and their genetic interactions, while the growth-regulatory role of Cyclin D–Cdk4 is discussed in detail in Datar et al. (2000). Progression through the cell division cycle in the absence of Cdk4 It is not yet clear to what extent the presence of the highly related cdk6 gene explains the fact that cdk4 is not required for development to the adult stage in mice (Rane et al., 1999; Tsutsui et al., 1999). In contrast, Drosophila does not appear to have a Cdk6 gene (Sauer et al., 1996; Adams et al., 2000). Moreover, our experiments involving overexpression of dominant-negative Cdk4 proteins (Cdk4D175N, Cdk4K55M) argue strongly against the idea that Cdk4 null mutants are rescued by functionally redundant D-type cyclin–cdk complexes. Overexpression of these mutant kinases in Cdk4+ flies mimics a Cdk4 loss-of-function phenotype, while it does not at all enhance the mutant phenotype in Cdk4 null mutants. Just like Cdk4, Drosophila Cyclin D should not be an essential gene, if Cyclin D functions exclusively in a complex with Cdk4. Indeed, near-sat