Characterization of Enzymatic Activity

In order to characterize the JIL-1 protein, JIL-1-specific polyclonal antibodies were generated against a beta-galactosidase JIL-1 fusion protein as well as a GST-JIL-1 fusion protein in rabbits. On immunoblots of embryo protein extracts (0-6 hr), both antisera detect JIL-1 protein as a doublet or triplet of bands migrating at approximately 150-160 kDa. However, posttranslational modifications of JIL-1 may be developmentally regulated, because JIL-1 protein from later developmental stages and in the S2 cell line is detected as a single band by the antisera (Jin, 1999).

Many kinases have been shown to be regulated by autophosphorylation. To address whether JIL-1 encodes an active kinase and to determine whether JIL-1 could potentially regulate its activity by autophosphorylation, JIL-1 protein was immunopurified from S2 cell tissue culture extracts and the purified JIL-1 protein was tested in an in vitro kinase assay. Immunoprecipitates were incubated in a kinase reaction buffer with radioactive [32P]ATP added. After the incubation, the immunoprecipitates were fractionated by SDS-PAGE, Coomassie blue stained, and dried, and incorporation of radiolabeled phosphate was visualized by autoradiography. Autoradiographs of gel-fractionated kinase assays done with immunoprecipitated JIL-1 reveal a labeled band migrating at the same position as JIL-1, as detected by immunoblot analysis. These results demonstrate that JIL-1 possesses an inherent kinase activity and is able to autophosphorylate in an in vitro kinase assay (Jin, 1999).

Histone H3 phosphorylation has been shown to correlate with activation of gene expression as well as chromatin condensation during mitosis. Therefore whether the nuclear JIL-1 kinase could phosphorylate bovine histone H3 was tested in vitro. The bovine histone H3 NH2-terminal tail is identical to that of its Drosophila homolog except that alanine 31 is substituted for a proline. Kinase assays with either JIL-1- or preimmune antisera immunoprecipitation were performed but with 15 µg of histone H3 included in the reaction. Samples were fractionated by SDS-PAGE, Coomassie blue stained, and autoradiographed. Histone H3 in the JIL-1 immunoprecipitation reaction shows clear labeling after autoradiography. Thus, these experiments suggest that histone H3 can serve as a substrate for the JIL-1 kinase in vitro (Jin, 1999).

The chromodomain protein, Chromator, interacts with JIL-1 kinase and regulates the structure of Drosophila polytene chromosomes

This study generated two new hypomorphic Chro alleles and analyzed the consequences of reduced Chromator protein function on polytene chromosome structure. In Chro71/Chro612 mutants the polytene chromosome arms were coiled and compacted with a disruption and misalignment of band and interband regions and with numerous ectopic contacts connecting non-homologous regions. Furthermore, Chromator co-localizes with the JIL-1 kinase at polytene interband regions and the two proteins interact within the same protein complex. That both proteins are necessary and may function together is supported by the finding that a concomitant reduction in JIL-1 and Chromator function synergistically reduces viability during development. Overlay assays and deletion construct analysis suggest that the interaction between JIL-1 and Chromator is direct and that it is mediated by sequences in the C-terminal domain of Chromator and by the acidic region within the C-terminal domain of JIL-1. Taken together these findings indicate that Chromator and JIL-1 interact in an interband-specific complex that functions to establish or maintain polytene chromosome structure in Drosophila (Rath, 2006).

Chromator was originally identified in a yeast two-hybrid screen as an interaction partner of the putative spindle matrix component, skeletor, and localizes to the spindle and the centrosomes during mitosis (Rath, 2004). Furthermore, functional assays using RNAi-mediated depletion in S2 cells suggest that Chromator directly affects spindle function and chromosome segregation (Rath, 2004). However, localization of Chromator to polytene interbands suggests it also has a functional role in maintaining chromatin structure during interphase. Such a role is supported by the finding (Eggert, 2004) that Chromator (which these workers refer to as Chriz) is found in a protein complex together with the interband-specific zinc-finger protein Z4 (Eggert, 2004; Gortchakov, 2005). That Z4 participates in regulating polytene chromosomal structure is likely because Z4-null mutant chromosomes show a decompaction of chromatin and a loss of a clear band/interband pattern (Eggert, 2004). However, the effect of Chromator on polytene chromosome morphology has been difficult to study because null alleles of Chro die as embryos or first-instar larvae before salivary gland polytene chromosomes can be analyzed (Rath, 2004; Gortchakov, 2005). For this reason an EMS mutagenesis screen was performed that generated two new Chro hypomorphic alleles. The analysis of these alleles shows that impaired Chromator function leads to disorganization and misalignment of band/interband regions resulting in coiling and folding of the polytene chromosomes. In addition, Chromator directly interacts with JIL-1 kinase and the two proteins extensively co-localize at polytene interband regions. Taken together these findings indicate that Chromator and JIL-1 interact in an interband-specific complex that functions to establish or maintain polytene chromosome structure in Drosophila (Rath, 2006).

Although the Drosophila polytene chromosome has served as a widely used model for studying chromatin structure, remarkably little is known about its spatial organization or about the molecular basis for the conjugation of homologous chromatids in the process of polytenization. It has been demonstrated that JIL-1 kinase, which phosphorylates histone H3 Ser10 in interband regions, plays a crucial role in maintaining polytene chromosome structure. In the absence of JIL-1 there is a shortening and folding of the chromosomes with a non-orderly intermixing of euchromatin and the compacted chromatin characteristic of banded regions and there is a striking redistribution of the heterochromatin markers dimethyl H3K9 and HP1 to ectopic chromosome sites. This suggested a model where JIL-1 kinase activity functions to maintain chromosome structure and euchromatic regions by counteracting heterochromatization mediated by histone H3 dimethylation and HP1 recruitment. However, the Chro mutant analysis presented in this study suggests that JIL-1 activity is necessary but not sufficient for maintaining some of these aspects of polytene chromosome morphology and that Chromator function is also required. Nonetheless, it should be noted that although the polytene chromosome phenotypes of JIL-1 and Chro mutants resemble each other with coiled and compacted chromosome arms they are not identical. In contrast to JIL-1 mutant polytene chromosomes, in Chro mutants there is still a clear demarcation between band and interband regions at the ultrastructural level although these bands are severely misaligned. Furthermore, in JIL-1 null mutants the male X chromosome is differentially affected with a 'puffed' appearance whereas in Chro mutants the morphology of the male X chromosome is similar to that of the autosomes. Thus, it is likely that JIL-1 and Chromator control different but related aspects of chromosome morphology within the complex. That both proteins are necessary and may function synergistically is supported by the finding that a concomitant reduction in JIL-1 and Chromator function dramatically reduces viability during development (Rath, 2006).

An important feature of the Chromator protein is the presence of a chromodomain. The function of most chromodomain proteins identified thus far has been related to the establishment or maintenance of a variety of chromatin conformations. For example, HP1 binds to methylated histone H3 and is essential for the assembly of heterochromatin. Thus, it is possible that Chromator through interactions mediated by its chromodomain participates in a complex with JIL-1 that is required for maintaining properly separated and aligned interband regions as well as a more open chromatin configuration. However, loss of JIL-1 or Chromator function also influences the coherence and organization of bands although neither protein is present in these regions. This suggests that JIL-1 and Chromator function may affect the distribution and/or activity of other molecules important for influencing chromatin structure such as boundary elements and/or the molecular machinery regulating heterochromatin formation and spreading. In support of this notion it has recently been demonstrated (Zhang, 2006) that the lethality as well as some of the chromosome morphology defects observed in JIL-1 null or hypomorphic mutant backgrounds may be the result of ectopic histone methyltransferase activity (Rath, 2006).

In addition to the present studies demonstrating an interaction with JIL-1, Chromator has been shown to interact with the spindle matrix protein skeletor (Walker, 2000; Rath, 2004) and with the zinc-finger protein Z4 (Eggert, 2004; Gortchakov, 2005). The interaction with skeletor was first detected in a yeast two-hybrid screen and subsequently confirmed by pull-down assays (Rath, 2004). Immunocytochemical labeling of Drosophila embryos, S2 cells and polytene chromosomes demonstrated that the two proteins show extensive co-localization during the cell cycle although their distributions are not identical (Rath, 2004). During interphase Chromator is localized on chromosomes to interband chromatin regions in a pattern that overlaps that of skeletor. During mitosis both Chromator and skeletor detach from the chromosomes and align together in a spindle-like structure with Chromator additionally being localized to centrosomes that are devoid of skeletor-antibody labeling. The extensive co-localization of the two proteins is compatible with a direct physical interaction between skeletor and Chromator. However, at present it is not known whether such an interaction occurs throughout the cell cycle or is present only at certain stages with additional proteins mediating complex assembly at other stages. The interaction of Chromator with Z4 was identified in co-immunoprecipitation experiments and the two proteins colocalize extensively at interband polytene regions (Eggert, 2004). However, Chromator and Z4 do not appear to associate directly and their chromosomal binding is independent of each other (Gorthakov, 2005). Interestingly, the phenotype of loss of Z4 function is somewhat different from that of loss of JIL-1 or Chromator function. Z4 mutant chromosomes decompact and attain a cloudy appearance when losing their band/interband organization (Eggert, 2004) instead of coiling and shortening as in JIL-1 and Chro loss-of-function mutants. This differential effect on polytene chromosome banding patterns and morphology may reflect that these constituents contribute different activities within one complex or may indicate the presence of more than one molecular assembly, each with different functions. Thus, future studies will be necessary to clarify the interactions of Chromator with interband-specific proteins and its functional role in establishing or maintaining polytene chromosome structure (Rath, 2006).

The Chriz-Z4 complex recruits JIL-1 to polytene chromosomes, a requirement for interband-specific phosphorylation of H3S10

The conserved band-interband pattern is thought to reflect the looped-domain organization of insect polytene chromosomes. Previously, it has been shown that the chromodomain protein Chriz (Chromator) and the zinc-finger protein Z4 (Putzig) are essentially required for the maintenance of polytene chromosome structure. This study shows that both proteins form a complex that recruits the JIL-1 kinase to polytene chromosomes, enabling local H3S10 phosphorylation of interband nucleosomal histones. Interband targeting domains were identified at the N-terminal regions of Chriz and Z4, and the data suggest partial cooperation of the complex with the BEAF boundary element protein in polytene and diploid cells. Reducing the core component Chriz by RNAi results in destabilization of the complex and a strong reduction of interband-specific histone H3S10 phosphorylation (Gan, 2011).

The chromatin proteins Chriz and Z4 both are ubiquitous and essential proteins that are required for the maintenance of interphase chromosome structure. Chriz and Z4 directly interact by their central and N-terminal domains, respectively. The Z4 N-terminal domain is required for Z4 interband targeting, mediated by Chriz interaction. When overexpressed, it competes the chromosomal binding of the endogenous Z4 protein. The Chriz N-terminal domain is required for interband targeting of the Chriz protein by an as-yet unknown mechanism. Destabilization of the complex results in strong down-regulation of H3S10 phosphorylation on polytene chromosomes (Gan, 2011).

Surprisingly, neither the Chriz chromodomain nor the Z4 zinc-finger region is needed for chromosomal targeting and no interaction partners for these signature protein domains are yet identified. However, both domains provide important functions. Point mutations in the Chriz chromodomain are unable to complement Chriz mutations. Point mutations within the Z4 zinc-finger region are not yet available, but overexpression of the N-terminal Z4 fragment aa 1-516 including the zinc-finger region results in a 'shrunken' interband phenotype, although neither overexpression of the Z4 full-length protein nor of N-Z4 aa 1-240 affect chromosomal structure. This indicates a dominant negative effect for the zinc-finger region in the aa 1-516 construct (Gan, 2011).

Data from coimmunoprecipitation and colocalization in situ suggest that Chriz and Z4 interact with the boundary element binding protein BEAF and the H3S10-specific histone kinase JIL-1. Chriz and JIL-1 directly interact in vitro, but it remains to be shown if the JIL-1-Z4 interaction is also direct or mediated by Chriz protein. Also, direct interaction between Chriz–Z4 and BEAF still has to be demonstrated. Unlike the Chriz-Z4 interaction, neither the BEAF-Z4 nor the JIL-1-Z4 colocalization is 100%. All four proteins are located in many interbands, but in each case there are sites that are occupied by one or the other protein exclusively. The mechanism explaining this promiscuity in interaction is not yet fully understood. However, it has been demonstrated that interband proteins can be shared between different chromatin protein complexes. For instance, CP190 is present in complexes with BEAF or Su(Hw) but not both. Furthermore, CP190 binds to CTCF with ~50% of CP190-CTCF complexes also bound by BEAF (Gan, 2011).

Mutual interactions could be mediated both by cooperation or competition between different factors for local binding. It is worth mentioning that Chriz also shows a significant overlap with CP190 in polytene chromosome binding in situ and in ChIP on chip experiments on Drosophila cell lines. The overlap with both BEAF and CP190, both well known for their activity as boundary element factors, suggests that Chriz and Z4 may be involved in the control of boundaries, a hypothesis that is currently being tested. The qualitative impression of colocalization gained from polytene chromosome staining is confirmed at higher resolution by ChIP on chip data from diploid S2 cells. Although these data still await a more systematic analysis, it is evident that in some regions the match in BEAF/Chriz binding sites is rather convincing but elsewhere it is more moderate. JIL-1 also colocalizes with Chriz but often shows a broader distribution overlapping several Chriz peaks or extending from a Chriz peak into an adjacent region. Strikingly, for the two cases of well-studied interband regions at 3C6-7 and 61C7-8, binding of BEAF, Chriz and JIL-1 is also observed in S2 diploid cells, suggesting that this stretch of open chromatin is conserved in structure between cell lineages. In the case of 3C6-7, the binding may be correlated with Notch transcription in S2 cells. However, there is no Notch transcription in salivary glands. The transcriptional activity at interband 61C7-8 is not known (Gan, 2011).

The effect of reducing the function of Chriz, Z4 and JIL-1 on polytene chromosome structure was studied by the use of hypomorphic alleles and tissue-specific RNAi knockdown. In general, it results in a phenotype of progressive loss of distinct band interband structure, sometimes concomitant with partial condensation or torsional distortion of polytene chromosomes. It has been reported that targeting of active JIL-1 kinase resulted in ectopic histone H3S10 phosphorylation and local chromosome decondensation that was dependent on JIL-1 kinase activity. It was reasoned that a major role of the Chriz-Z4 complex might be the recruitment of JIL-1 to the chromosome that would result in local phosphorylation of nearby nucleosomes, allowing the ≥30 nm chromatin fibre to unfold and form a less condensed interband chromatin. In support of this hypothesis, gland-specific Chriz-RNAi resulted in decreased levels of chromosomally bound Z4 and JIL-1 and in a significant loss of interband specific H3S10 phosphorylation. However, not all chromosomes exhibited the loss of structure phenotype mentioned above. Conceivably, RNAi induction might have been too mild and there was still enough Chriz protein left to provide for the maintenance of chromosome structure. The experiments were done at 22-24°C, a temperature at which the GAL4 inducer is still not at its maximal activity. Raising the temperature to 29°C early on made the inducer more effective but resulted in a tiny salivary gland phenotype that was not amenable to cytological analysis. Using the Sgs4 enhancer active in late 3rd instar at 29°C was not effective since the Chriz protein with its long half-life masked the RNAi effect. A different explanation for the lack of chromosomal phenotype may be that 10%-20% of the JIL-1 binding sites that do not depend on Chriz/Z4 are not affected by Chriz-RNAi and are sufficient to sustain the chromosome structure (Gan, 2011).

In conclusion, this study provides evidence for a chromatin complex with the chromodomain protein Chriz at its core. The complex, which, besides the Z4 protein, may contain the sequence-specific DNA-binding protein BEAF, is required for H3S10 phosphorylation of interphase chromosomes, presumably by local recruitment of the tandem kinase JIL-1. Local binding and activity of the complex on polytene chromosomes may result in decondensation of interband regions and related sites on nonpolytene diploid chromosomes (Gan, 2011).

Sequence-specific targeting of dosage compensation in Drosophila favors an active chromatin context

The Drosophila MSL complex mediates dosage compensation by increasing transcription of the single X chromosome in males approximately two-fold. This is accomplished through recognition of the X chromosome and subsequent acetylation of histone H4K16 on X-linked genes. Initial binding to the X is thought to occur at 'entry sites' that contain a consensus sequence motif ('MSL recognition element' or MRE). However, this motif is only ~2 fold enriched on X, and only a fraction of the motifs on X are initially targeted. This study asked whether chromatin context could distinguish between utilized and non-utilized copies of the motif, by comparing their relative enrichment for histone modifications and chromosomal proteins mapped in the modENCODE project. Through a comparative analysis of the chromatin features in male S2 cells (which contain MSL complex) and female Kc cells (which lack the complex), it was found that the presence of active chromatin modifications, together with an elevated local GC content in the surrounding sequences, has strong predictive value for functional MSL entry sites, independent of MSL binding. These sites were tested for function in Kc cells by RNAi knockdown of Sxl, resulting in induction of MSL complex. Ectopic MSL expression in Kc cells was shown to lead to H4K16 acetylation around these sites and a relative increase in X chromosome transcription. Collectively, these results support a model in which a pre-existing active chromatin environment, coincident with H3K36me3, contributes to MSL entry site selection. The consequences of MSL targeting of the male X chromosome include increase in nucleosome lability, enrichment for H4K16 acetylation and JIL-1 kinase, and depletion of linker histone H1 on active X-linked genes. This analysis can serve as a model for identifying chromatin and local sequence features that may contribute to selection of functional protein binding sites in the genome (Alekseyenko, 2012).

This study considered the roles of chromatin environment and flanking sequence composition in selection of functional binding sites by a sequence-specific protein complex. It is generally not clear whether the chromatin features that are often observed at the binding sites of proteins contribute directly to binding selectivity or are simply a consequence of binding. In the dosage compensation system of the X chromosome in Drosophila, a unique opportunity is presented to address this question because it is possible to compare the chromatin environment of MSL binding sites in female cells, in the absence of the complex, to male cells, where the functional sites are bound. Binding data from an RNAi experiment were used in which a component of the sex determination pathway was knocked down in females to induce dosage compensation. Bioinformatic analysis of a large number of profiles from the modENCODE project suggests that a pre-existing active chromatin context plays a critical role in establishing the initial binding of the MSL complex on the X. The surprising discovery was made that GC content in the DNA surrounding functional binding sites has a characteristic profile (Alekseyenko, 2012).

In summary, the results strongly support a model in which an active chromatin composition helps define the initial entry sites selected by the MSL complex. Functional MSL binding results in increased lability of local nucleosomal composition, and H4K16 acetylation and JIL-1 binding along the bodies of virtually all active X-linked genes. This work provides key insights into the order of events leading to dosage compensation in Drosophila, and can also serve as a model for using genome-wide data sets to understand how sequence-specific factors find their ultimate targets (Alekseyenko, 2012).

The results support roles for local chromatin environment and flanking GC content in discrimination of true target sites of the MSL dosage compensation complex. The model (see Model for binding site selection by a chromatin associated factor) depicts the GC content and active chromatin marks surrounding MREs in female Kc cells that predict binding by MSL complex in male S2 or BG3 cells (or after MSL induction in female Kc cells). MREs that do not pre-exist in a favorable environment are not bound by MSL complex and thus are non-functional. Definition of the favorable chromatin features that pre-exist factor binding may be a general tool, in addition to DNA motif analysis, for prediction of functional binding sites(Alekseyenko, 2012).

Protein Interactions

In order to determine the relationship of JIL-1 localization with respect to the MSL dosage compensation complex proteins, chromosomal squashes prepared from larval salivary glands were double labeled with JIL-1 antisera and with either MSL1, MSL2, or MSL3 antibody. JIL-1 colocalizes with MSL1 along the entire length of the male X chromosome except at the tip. In all cases examined, the banding patterns themselves overlapped. Identical results were obtained in double labelings with JIL-1 and MSL2 and MSL3 antibodies. In contrast with JIL-1 neither MSL1, MSL2, nor MSL3 are found at significant levels on the autosomes (Jin, 2000).

The observed colocalization of JIL-1 begs the question of whether JIL-1 would also be inappropriately upregulated on the female X chromosomes in the presence of ectopic MSL2. This has been demonstrated to be the case for other proteins in the MSL complex. Chromosomal squashes prepared from salivary glands of third instar female larvae that ectopically express MSL2 in a heterozygous msl1 or msl3 background and thus have assembled MSL complexes on their paired X chromosomes were immunostained with JIL-1 antibody. When the MSL complex is targeted to the female X chromosomes, it is accompanied by a concomitant increase of JIL-1 localization. That this increase is dependent on the presence of the MSL complex is supported by JIL-1 immunostainings of female polytene chromosome squashes ectopically expressing MSL2 in an msl1/msl1 or msl3/msl3 homozygous background incapable of MSL complex assembly due to absence of the MSL1 or MSL3 subunit, respectively. In these females JIL-1 is no longer upregulated on the paired X chromosomes but now is found distributed on both X chromosomes and throughout all of the autosomes in a pattern consistent with its normal wild-type localization. These results suggest that the upregulation of JIL-1 on the male X chromosome is directly correlated with the presence of the MSL complex. There are two possible scenarios that can account for these observations: either JIL-1 is an integral component of the MSL complex itself or the activity of the MSL complex directs the upregulation of JIL-1 to specific sites on the X chromosome (Jin, 2000).

To address whether JIL-1 may associate with the MSL dosage compensation complex, coimmunoprecipitation experiments were performed designed to test for molecular interactions. Protein extracts from the Drosophila S2 cell line that expresses both the MSL complex as well as JIL-1 were used. To facilitate the cross-immunoprecipitation experiments, a V5-epitope-tagged full-length JIL-1 construct was introduced into stably transfected S2 cells. The V5-tagged JIL-1 fusion protein can be detected in Western blot analysis and retains its kinase activity. For immunoprecipitation experiments, proteins were extracted from the transfected S2 cells, fractionated on SDS-PAGE after immunopreciptation, Western blotted, and probed with the appropriate antibodies. Anti-MSL1, anti-MSL2, and anti-MSL3 are each able to coimmunoprecipitate V5-tagged JIL-1 as shown on Western blots probed with anti-V5 antibody. Anti-V5 monoclonal antibody is able to coimmunoprecipitate MSL1 and MSL2 as well as MSL3 from S2 cells that are stably expressing V5-tagged JIL-1, but not from control untransfected S2 cells that do not express the V5-JIL-1 fusion protein. Endogenous JIL-1 can be immunoprecipitated by MSL1 antibody and, conversely, MSL1 can be immunoprecipitated by JIL-1 antibody from untransfected S2 cell lysate. These results demonstrate a molecular interaction of JIL-1 with the MSL complex. However, since immunoprecipitation of any one of the MSL complex proteins will co-immunopreciptation the other proteins, it does not indicate whether any of these proteins interact directly with JIL-1 (Jin, 2000).

To further characterize which region(s) of the JIL-1 kinase is likely to interact with other components of the MSL dosage compensation complex, GST-JIL-1 fusion proteins covering different domains of JIL-1 were generated including a nearly full-length JIL-1 construct (GST-JIL-1), the NH2-terminal domain (GST-NTD), the first kinase domain (GST-KDI), the second kinase domain (GST-KDII), and the COOH-terminal domain (GST-CTD). These GST-JIL-1 fusion proteins or a GST-only control were coupled with glutathione agarose beads, incubated with S2 lysate, washed, fractionated by SDS-PAGE, and analyzed by Western blot analysis using antibodies specific for MSL1 or MSL3. Whereas the GST-only control shows no pull-down activity for either MSL protein, GST-JIL-1 is able to pull down both MSL1 and MSL3. Furthermore, this pull-down activity is specific to the kinase domains of JIL-1: both GST-KDI and GST-KDII are able to pull down both MSL1 and MSL3, but neither GST-NTD nor GST-CTD shows such activity. These results suggest that the direct molecular interaction of JIL-1 with the MSL complex is mediated by the kinase domains and not the NH2- or COOH-terminal sequences (Jin, 2000).

To analyze the function of the chromosomal kinase JIL-1, an allelic series of hypomorphic and null mutations was generated. JIL-1 is an essential kinase for viability, and reduced levels of JIL-1 kinase activity lead to a global change in chromatin structure. In JIL-1 hypomorphs, euchromatic regions of polytene chromosomes are severely reduced and the chromosome arms condensed. This is correlated with decreased levels of histone H3 Ser10 phosphorylation. These levels can be restored by a JIL-1 transgene placing JIL-1 directly in the pathway mediating histone H3 phosphorylation. A model is proposed where JIL-1 kinase activity is required for maintaining the structure of the more open chromatin regions that facilitate gene transcription (Wang, 2001).

The original EP(3)3657 line in conjunction with the newly generated JIL-1z60 and JIL-1z2 lines constitute an allelic series of JIL-1 hypomorphic and null mutations. This allows an anaysis of the effects of decreasing levels of JIL-1 protein on viability and male to female sex ratios. In order to measure and compare viability, homozygous pupae were collected from heterozygous crosses of the three alleles and their eclosion rates were determined. The homozygous mutant pupae could be readily identified because they did not display the Tubby marker carried on the balancer third chromosome. It was found that homozygous EP(3)3657 larvae, which have only one tenth of the level of JIL-1 protein normally found in wild-type, have an eclosion rate of 81%. However, as the level of JIL-1 protein further decreases to about 3% in JIL-1z60/JIL-1z60 larvae, the eclosion rate reduces to only 0.5%. Moreover, the few homozygous JIL-1z60/JIL-1z60 animals surviving to adulthood are unable to produce offspring and most die shortly after eclosion. The eclosion rate for the null allele JIL-1z2/JIL-1z2 larvae was 0%. These results strongly suggest that JIL-1 is an essential kinase for viability (Wang, 2001).

Eclosed EP(3)3657/EP(3)3657 adults are fertile and able to produce offspring, and thus embryos from homozygous parents can be analyzed for the effect of reduced levels of JIL-1 on embryonic development. The hatch rate of such embryos is only 4%, showing a significant decrease below the 83% observed in wild-type. Whereas reduced eclosion levels are observed for both males and females, male viability is more severely affected in all cases of lowered JIL-1 expression. For example, in EP(3)3657/EP(3)3657 homozygous offspring from heterozygous mothers that provide maternal levels of JIL-1 protein during early development due to the mother's wild-type allele, the number of males eclosing was 73% that of females. In adults eclosing from crosses of homozygous parents and thus developing without the increased maternal levels of JIL-1, the percentage of males relative to females was 48%. Further reduction is observed in the severe JIL-1z60/JIL-1z60 hypomorph, which gives rise to only 32% the expected number of males relative to females. The male to female sex ratio in EP(3)3657 flies can be rescued to near wild-type ratio by the JIL-1- GFP transgene. When this transgene is introduced into these flies, the male to female sex ratio recovers from 48% to 97% (Wang, 2001).

The phenotypic consequences of reduced levels of JIL-1 kinase were investigated in embryos from EP(3)3657 homozygous parents by labeling chromatin with Hoechst and microtubules with anti-tubulin antibody. The average expression level of JIL-1 kinase in these mutant animals is reduced to about one tenth that of wild-type. A range of phenotypes was observed from embryos appearing wild-type with regularly spaced nuclei to embryos where chromatin structure had completely disintegrated. In intermediate phenotypes, nuclei in various stages of fragmentation were still discernible. The variable penetrance and range of phenotypes are likely to be a result of different levels of JIL-1 expression in individual embryos. Some embryos have enough JIL-1 to carry them through embryogenesis as reflected in the 4% hatching rate, whereas others are below the threshold for maintaining JIL-1 kinase function. In embryos double labeled with Hoechst and anti-tubulin antibody, centrosomes were often observed to be separated from the nuclear remnants, and in other cases, the nuclear fragmentation would lead to aberrant and misaligned tubulin spindles. These data suggest that reduced levels of JIL-1 kinase lead to a disintegration of nuclear and chromatin structure during embryonic development (Wang, 2001).

The consequences of loss of JIL-1 on polytene chromosome structure in interphase nuclei was assessed. Chromosomal squashes prepared from either wild-type or homozygous hypomorphic EP(3)3657, JIL-1z60, or null JIL-1z2 larvae were fixed and labeled with Hoechst to visualize the DNA, anti-MSL2 antibody to identify the male X chromosome, and, in some cases, with anti-JIL-1 antibody. Labeling with anti-MSL2 antibody revealed that MSL2 protein still localizes to the X chromosome in all three JIL-1 mutant alleles, indicating that JIL-1 is not necessary for targeting of the MSL complex to the male X chromosome. Identical results were obtained using anti-MSL1, -MSL3, or histone H4Ac16 antibodies, confirming this observation. However, chromosome morphology in both males and females is markedly affected. Whereas wild-type polytene chromosomes show extended arms with a regular pattern of Hoechst-stained bands, this pattern, while relatively normal in the hypomorphic EP(3)3657 mutant animals, becomes severely perturbed in strong JIL-1z60 hypomorphs and the null JIL-1z2 larvae. In these latter preparations, the euchromatic interband regions are largely absent and the chromosome arms are highly condensed. Thus, these results suggest that the JIL-1 kinase is involved in both males and females in establishing or maintaining the more open chromatin structure found in the gene-active interband regions that comprise less tightly packed euchromatin (Wang, 2001).

Although all of the chromosomes from JIL-1 mutant animals display abnormalities, perturbation of the male X chromosome is relatively more severe than that of the autosomes. This can be observed in the weaker hypomorphic phenotype from EP(3)3657 preparations where although the autosomes are only subtly affected, the male X chromosome is significantly shorter and has lost a large degree of its banding pattern. In the strong JIL-1z60 hypomorph or the null JIL-1z2 mutant, the male X chromosome is even more condensed with no remaining observable banding pattern or structure. That the reduction of JIL-1 protein level is responsible for the defects observed in the homozygous animals is further supported by rescue experiments in which transgenic JIL-1-GFP is introduced into JIL-1z2/JIL-1z2 animals. Chromosomes from these animals now appear essentially wild-type including the male X chromosome; JIL-1 antigenicity is restored and upregulated on the X chromosome as detected by JIL-1 antibody. Identical results were observed in rescue experiments employing a full-length JIL-1 transgene which did not contain the GFP moiety (Wang, 2001).

The upregulation of JIL-1 on the male X chromosome in conjunction with its ability to phosphorylate histone H3 Ser10 in vitro led to an examination of the question of whether higher levels of phosphorylated histone H3 Ser10 (pH3S10) are also present on the male X. Conventional polytene chromosome fixation and squash techniques lead to inconsistent banding patterns. It was reasoned that the highly acidic fixation conditions of the conventional squash protocol might be interfering with either antibody performance or antigen stabilization during fixation. Therefore, a modified whole-mount staining technique was developed for salivary glands that gently compress nuclei beneath a coverslip before fixation in a standard paraformaldehyde/PBS solution with a physiological pH. Although the overall resolution of the bands is inferior to the normal squash technique, it does allow visualization of the chromosomes suitable for analysis. Such salivary gland preparations were double labeled with antibodies to JIL-1 and pH3S10 as well as with antibodies to JIL-1 and phosphoacetylated histone H3 (pH3S10Ac14). JIL-1 protein is upregulated on the male X chromosome. This upregulation is concomitant with an upregulation of both pH3S10 and pH3S10Ac14 labeling on the male X chromosome as compared to the autosomes. Furthermore, it is evident that the staining pattern of the antibodies in wild-type animals overlap as indicated by a predominantly yellow banding pattern. This labeling pattern was consistently observed in different experiments using different lots of pH3S10 antibody from two different companies. Such an upregulation was not observed in the female, and in homozygous JIL-1z2/JIL-1z2 polytene chromosomes; neither JIL-1 nor pH3S10 or pH3S10Ac14 labeling was detectable. These data suggest that levels of phosphorylated histone H3 Ser10 are increased on the male X chromosome in a pattern overlapping with that found for the JIL-1 kinase. However, it is of interest to note that Western blot analysis does not indicate higher overall levels of pH3S10 in males than females (Wang, 2001).

Recent studies have revealed a tight correlation between histone H3 Ser10 phosphorylation and proper chromosome condensation and segregation during mitosis. This mitotic phosphorylation of histone H3 is governed by the lpl1/aurora kinase in budding yeast and nematodes and by the NIMA kinase in Aspergillus. Thus, different kinases or more than one kinase may serve this function in different organisms. This raises the question whether JIL-1 regulates mitotic histone H3 Ser10 phosphorylation in Drosophila. To address this issue, pH3S10 levels were analyzed in null JIL-1z2/JIL-1z2 larval neuroblast mitotic chromosomes. In the null JIL-1 background, pH3S10 is not observed in interphase nuclei, but is enriched on the mitotic chromosomes at a level comparable to wild-type. Therefore, loss of JIL-1 activity does not appear to alter the mitotic phosphorylation of histone H3 Ser10 in larval neuroblasts (Wang, 2001).

Since the interphase upregulation of pH3S10 phosphorylation levels correlates with JIL-1 kinase localization and since the majority of larval cells are in interphase at any one given time, whether pH3S10 phosphorylation levels were decreased in JIL-1 hypomorphs was examined, as would be predicted if JIL-1 were involved in this process. Levels of phosphorylated histone H3 Ser10 were determined by immunoblot analysis of larval protein lysates from wild-type or homozygous JIL-1 mutant (EP(3)3657, JIL-1z60, JIL-1z2) animals which were fractionated and probed with anti-pH3S10, anti-tubulin, anti-lamin, and anti-histone H3 antibodies. All of the JIL-1 mutants show lower levels of pH3S10 than observed in wild-type larvae. Furthermore, the level of reduction of pH3S10 corresponded directly to the severity of the JIL-1 allele, with the null JIL-1z2/JIL-1z2 allele showing the lowest level of pH3S10 phosphorylation and EP(3)3657/EP(3)3657, the weaker of the two hypomorphs showing higher pH3S10 levels than the strong JIL-1z60/JIL-1z60 hypomorph. In contrast, the levels of control proteins such as histone H3, tubulin, and lamin were roughly equivalent to wild-type levels in all three mutant lines. Since introduction of the JIL-1-GFP transgene on the second chromosome of JIL-1z2/JIL-1z2 animals rescued the chromosomal defects, it was of interest to determine whether there was also a corresponding restoration of pH3S10 levels in these animals. Western blots of larval protein lysates from wild-type, JIL-1z2/JIL-1z2, or JIL-1z2/JIL-1z2 larvae carrying the JIL-1-GFP transgene were probed with anti-pH3S10 antibody or anti-histone H3 total protein antibody. In the presence of the JIL-1-GFP transgene, pH3S10 levels are restored to essentially wild-type levels (Wang, 2001).

Thus, in a JIL-1 null mutant allele that shows no detectable JIL-1 kinase, the level of histone H3 Ser10 phosphorylation is reduced to about 5% of wild-type levels. These results suggest the existence of another kinase that can phosphorylate histone H3 Ser10 and are consistent with the recent identification of Aurora B kinase as the likely mitotic H3 Ser10 kinase in Drosophila (Giet, 2001). However, these results suggest that JIL-1 is the predominant kinase regulating the phosphorylation state of this residue at interphase. This is further supported by findings that histone H3 Ser10 phosphorylation is upregulated on the male X chromosome in a pattern similar to that of the JIL-1 kinase and that the loss of histone H3 Ser10 phosphorylation as well as aberrant chromosome structure found in JIL-1 mutants can be rescued by the presence of a JIL-1 transgene. Thus, taken together these results demonstrate that JIL-1 is in the pathway mediating histone H3 Ser10 phosphorylation and that JIL-1 kinase activity is required for the maintenance of normal chromosome architecture in Drosophila at interphase (Wang, 2001).

In embryos with dividing nuclei, a reduction in JIL-1 kinase activity leads to nuclear fragmentation and to dispersion of centrosomes and deformed mitotic spindles. However, this phenotype is thought to be a consequence of altered chromatin structure that occurs during interphase such as that observed in polytene chromosomes. The aberrant chromosome structure interferes with proper mitotic condensation and segregation and ultimately leads to chromatin disintegration. The findings that a low level of histone H3 Ser10 phosphorylation persists in JIL-1z2 homozygous flies, that mitotic chromosomes in neuroblasts from JIL-1 null mutant larvae show high levels of histone H3 Ser10 phosphorylation, and that the Drosophila Aurora B kinase phosphorylates histone H3 Ser10 during mitosis (Giet, 2001) further support the notion that JIL-1 is not a mitotic histone H3 Ser10 kinase in Drosophila, but rather is important for maintenance of chromatin structure at interphase. It is becoming clear that multiple kinases can phosphorylate the Ser10 residue on histone H3 and that this single histone modification can elicit diverse cellular responses. What determines which of multiple pathways are activated may be influenced by the presence of additional histone tail modifications that, used in a combinatorial fashion, mediate context-dependent signaling (Wang, 2001 and references therein).

The significant decrease in male viability beyond that observed in females argues strongly for a specific role of JIL-1 in dosage compensation in males that may be separate from its function in maintenance of global chromatin structure. Dosage compensation results in a 2-fold hypertranscription of the male's single X chromosome relative to the female's two X chromosomes. The level of H4Ac16 is increased on the male X chromosome as a consequence of MSL chromatin remodeling complex activity, and this targeted acetylation has been directly linked to transcriptional activation. Notably, levels of phosphorylated histone H3 Ser10 as well as the double H3 Ser10Ac14 modifications are also upregulated on the male X, in agreement with the model that enhanced transcription may require a combined signaling via both acetylation and phosphorylation motifs. However, that the male X continues to be hyperacetylated at histone H4Ac16 in the absence of JIL-1 indicates that JIL-1's role in dosage compensation is not likely to be effected via regulation of the MSL complex's histone acetyltransferase activity (Wang, 2001).

The JIL-1 kinase interacts with lamin Dm0 and regulates nuclear lamina morphology of Drosophila nurse cells

A yeast two-hybrid screen was used to identify lamin Dmo as an interaction partner for the nuclear JIL-1 kinase. This molecular interaction was confirmed by GST-fusion protein pull-down assays and by co-immunoprecipitation experiments. Using deletion construct analysis a predicted globular domain of the basic region of the COOH-terminal domain of JIL-1 was shown to be sufficient for mediating the molecular interactions with lamin Dmo. A reciprocal analysis with truncated lamin Dmo constructs showed that the interaction with JIL-1 required sequences in the tail domain of lamin Dmo that include the Ig-like fold. Further support for a molecular interaction between JIL-1 and lamin Dmo in vivo was provided by genetic interaction assays. Nuclear positioning and lamina morphology were abnormal in JIL-1 mutant egg chambers. The most common phenotypes observed were abnormal nurse cell nuclear lamina protrusions through the ring canals near the oocyte, as well as dispersed and mislocalized lamin throughout the egg chamber. These phenotypes were completely rescued by a full-length JIL-1 transgenic construct. Thus, these results suggest that the JIL-1 kinase is required to maintain nuclear morphology and integrity of nurse cells during oogenesis and that this function may be linked to molecular interactions with lamin Dmo (Bao, 2005).

The results from the yeast two-hybrid interaction assays suggest that the interaction between JIL-1 and lamin Dmo is direct. However, JIL-1 is localized to euchromatic regions of chromosomes and lamin Dmo is mainly a component of the inner nuclear membrane raising the question of how this interaction occurs. Recently it has become clear that lamins and associated proteins in the nuclear envelope are involved in several nuclear activities apart from providing a barrier between the nucleoplasm and the cytoplasm. One of these functions of the nuclear lamina is to serve as a scaffold that provides attachment sites for interphase chromatin directly or indirectly regulating many nuclear activities such as DNA replication and transcription, nuclear and chromatin organization, cell development and differentiation, nuclear migration, and apoptosis. In Drosophila, it has been shown that direct interactions between the tail domain of lamin Dmo and histone H2A and H2B may mediate the attachment of chromosomes to the nuclear lamina. Interestingly, the early embryonic nuclear lamina protein YA (Young Arrest), which is a lamin Dmo binding protein, when ectopically expressed in larval salivary gland cells, associates with interband regions of polytene chromosomes. Thus, there is considerable evidence for direct interactions of lamins with chromatin associated proteins such as JIL-1. Furthermore, lamins have also been found in the nuclear interior and the possibility remains that there may be a hitherto undetected soluble pool of JIL-1 that potentially could provide additional avenues for direct interactions (Bao, 2005).

To determine whether disruptions in nuclear lamina organization could be detected in JIL-1 mutant backgrounds, fixed ovaries, embryos, imaginal discs and polytene salivary glands labeled with lamin Dmo antibody were examined. Abnormal lamin Dmo distribution was observed only in ovaries of JIL-1z2/JIL-1h9 flies. One phenotype which was found in about 5% of mutant egg chambers was dispersed and mislocalized lamin was found throughout the egg chamber. This is not likely to be a consequence of apoptotic events because lamins are degraded by proteolysis during apoptosis and do not show accumulation. The phenotype may therefore reflect a destabilization of the integrity of the nuclear lamina leading to lamin Dmo dispersal. Thus, these experiments may provide evidence that the stability of the nuclear lamina in Drosophila egg chambers depends on JIL-1 kinase activity and phosphorylation of lamin Dmo. Unfortunately, this hypothesis cannot be tested at the present time because of a lack of a functional in vitro JIL-1 kinase assay. The other phenotype observed in JIL-1 mutant egg chambers with high penetrance (42.8%) is abnormally positioned nurse cell nuclei which extended nuclear lamina protrusions through the ring canals near the oocyte. It is not clear how this phenotype arises. However, several morphogenetic processes such as anterior-posterior/dorso-ventral axis formation as well as cell and nuclear migration during oogenesis require reciprocal cell signaling between germline, oocyte and nurse cells, and somatic follicle cells. In JIL-1 mutant backgrounds cell signaling pathways that normally prevent nurse cell nuclei from responding to posterior migration signals may be downregulated, resulting in a posterior dislocalization towards the oocyte. It has been shown that the nuclear lamina is involved in regulating nuclear migration in the developing eye through interactions of the lamin Dmo-binding protein Klarsicht with the microtubule organizing center. Furthermore, Bicaudal-D, a dynein-interacting protein required for control of nuclear migration and cytoskeletal organization in oogenesis has been shown to interact with lamin Dmo in yeast two-hybrid assays. Thus, dynamic local interactions of cytoskeleton-associated motor proteins linked to lamin Dmo may be capable of providing the forces necessary for generating the observed deformations of the nuclear lamina. Nuclear lamins are generally considered to provide stiffness and incompressibility to the nuclear envelope suggesting that the aberrations in nuclear morphology observed here may be linked to a weakening of the nuclear lamina. However, the present experiments cannot distinguish between the possibilities that JIL-1 may be involved in nuclear deformation by regulating nuclear lamina cytoskeletal interactions via direct modulation of lamin Dmo or indirectly by modulating a signal transduction pathway, or both (Bao, 2005).

These results suggest that JIL-1 kinase is required to maintain nuclear morphology and integrity of nurse cells during oogenesis. It has recently been shown that some lamin Dmo interactions occur only during early development, indicating that special properties of the nuclear lamina may be required for regulating nuclear processes and morphology at specific developmental stages. For example, the lamin Dmo binding protein YA is expressed only in ovaries and pre-gastrulation embryos and is required for the interaction between chromatin and the nuclear envelope during early embryogenesis. Previously, it has been shown that the interaction between JIL-1 and Lola zf5, a splice variant of the complex lola locus encoding multiple different transcription factors, is developmentally regulated and restricted to early embryogenesis as well. Thus, it will be informative in future experiments to further explore the interaction between JIL-1 and lamin Dmo to clarify how this interaction contributes to nuclear lamina function in development (Bao, 2005).

Phosphorylation of histone H3 at Ser10 by JIL-1 facilitates RNA polymerase II release from promoter-proximal pausing in Drosophila

The Drosophila JIL-1 kinase is known to phosphorylate histone H3 at Ser10 (H3S10) during interphase. This modification is associated with transcriptional activation, but its function is not well understood. Evidence is presented suggesting that JIl-1-mediated H3S10 phosphorylation is dependent on chromatin remodeling by the brahma complex and is required during early transcription elongation to release RNA polymerase II (Pol II) from promoter-proximal pausing. JIL-1 localizes to transcriptionally active regions and is required for activation of the E75A ecdysone-responsive and hsp70 heat-shock genes. The heat-shock transcription factor, the promoter-paused form of Pol II (Pol IIoser5), and the pausing factor DSIF (DRB sensitivity-inducing factor) are still present at the hsp70 loci in JIL-1-null mutants, whereas levels of the elongating form of Pol II (Pol IIoser2) and the P-TEFb kinase are dramatically reduced. These observations suggest that phosphorylation of H3S10 takes place after transcription initiation but prior to recruitment of P-TEFb and productive elongation. Western analyses of global levels of both forms of Pol II further suggest that JIL-1 plays a general role in early elongation of a broad range of genes. Taken together, the results introduce H3S10 phosphorylation by JIL-1 as a hallmark of early transcription elongation in Drosophila (Ivaldi, 2007).

The eukaryotic cell packages its DNA wrapped around histone proteins to form nucleosomes, the basic units of chromatin. These nucleosomes assemble into higher-order chromatin structures through which the transcription machinery must navigate each time it is signaled to transcribe. Mechanisms have consequently evolved to maintain a flexible chromatin state that can readily respond to intrinsic and extrinsic stimuli and accordingly modulate gene expression. Most prominently, histone-modifying enzymes can methylate, acetylate, and phosphorylate various amino acid residues of histone N termini, thereby changing their affinity for different transcriptional regulators. ATP-dependent chromatin remodeling complexes can also be recruited to alter the position and accessibility of the nucleosome. The binding of specific transcription factors triggers a cascade of events during which these diverse chromatin modulators work in concert to allow the RNA polymerase II (Pol II) machinery to bind target genes, initiate transcription, and elongate the messenger RNA (mRNA). These regulators maintain tight control of transcription throughout the elongation process by continuously communicating with the C-terminal domain (CTD) of the largest subunit of Pol II (Ivaldi, 2007 and references therein).

The CTD of Pol II consists of a heptad repeat (Tyr-Ser-Pro-Thr-Ser-Pro-Ser) that is conserved from yeast to humans. It integrates transcriptional events by interacting with distinct regulatory proteins that recognize different patterns of CTD phosphorylation. When Pol II is first recruited to the promoter as part of the preinitiation complex, its CTD is hypophosphorylated. After Pol II disengages from the promoter, the CTD becomes phosphorylated at Ser5 (Pol IIoser5) by TFIIH, a general transcription factor that is part of the Pol II machinery. As part of an early elongation complex, Pol II progresses 20-40 base pairs (bp) downstream from the promoter. It then pauses in a process referred to as promoter-proximal pausing to allow for capping of the nascent mRNA. DRB sensitivity-inducing factor (DSIF, Spt5) and negative elongation factor (NELF) cooperate to repress transcription elongation and maintain this pause. Pol II is released once the P-TEFb kinase is recruited to relieve the negative effects of DSIF and NELF and phosphorylate the CTD at Ser2 (Pol IIoser2), marking the onset of productive elongation. The various transcriptional steps are associated with distinct histone modifications and chromatin remodeling complexes. Set1, the enzyme responsible for methylating Lys4 of histone H3 (H3K4) in Saccharomyces cerevisiae, is known to physically associate with the CTD of Pol II when it is phosphorylated at Ser5. At the same time, trimethylation of H3K4 has been found concentrated at the 5' end of transcribed genes. Methylation of Lys36 of H3 (H3K36), on the other hand, is associated with a later step in elongation; this mark accumulates further downstream from the promoter and associates with the CTD when phosphorylated at Ser2. Other modifications, such as lysine acetylation, arginine methylation, and serine phosphorylation, have also been associated with activation of gene expression. Of interest, phosphorylation of histone H3 at the Ser10 residue (H3S10) has been shown to be important for activation of transcription in yeast, Drosophila, and mammalian cells, but its precise role in this process is not well understood (Ivaldi, 2007 and references therein).

Several studies have suggested an important role for H3S10 phosphorylation in specific transcriptional responses to signaling stimuli. The yeast Snf1 kinase phosphorylates H3S10 upon activation of the INO1 gene. In mammalian fibroblasts, rapid phosphorylation of histone H3 concomitant with activation of immediate-early (IE) response genes takes place when cells are treated with growth factors and various stress-inducing agents. Further, Coffin-Lowry syndrome is characterized by impaired transcriptional activation of the c-fos gene and a loss of EGF-induced phosphorylation of histone H3S10. Treatment of immature rat ovarian granulosa cells with follicle-stimulating hormone produces rapid H3S10 phosphorylation in a PKA-dependent manner, suggesting a role for histone phosphorylation in cellular differentiation. Additionally, H3S10 phosphorylation follows the stimulation of the suprachiasmatic nucleus of rats with light and activation of hippocampal neurons. It further appears to play a central role during cytokine-induced gene expression mediated by IkappaB kinase α (IKK-α). What remains unclear from these studies is whether H3S10 phosphorylation is limited to mediating signal transduction events or whether it plays a more general role in the activation of gene expression in vertebrates (Ivaldi, 2007).

Studies in Drosophila suggest that this modification may be required for the transcription of most genes in this organism. Using the heat-shock response as a model system, it has been established that H3S10 phosphorylation patterns parallel those of active genes. Drosophila responds to a rise in temperature by rapidly increasing the transcription of heat-shock genes while repressing genes expressed previously. Before heat shock, phosphorylated H3S10 localizes to euchromatic regions of polytene chromosomes and colocalizes with Pol II. After heat shock, this modification redistributes to the active heat-shock loci and disappears from the rest of the chromosome, where genes are now repressed (Nowak, 2000; Ivaldi, 2007).

Despite these observations, the precise role of H3 phosphorylation in gene activation remains elusive. The mammalian MSK1 and MSK2 kinases, among others, have been shown to be responsible for H3S10 phosphorylation associated with transcription. The Drosophila homolog of MSK1/2, the JIL-1 threonine/serine kinase, has been shown to phosphorylate H3S10 in vitro. H3S10 phosphorylation levels in vivo are dramatically reduced in JIL-1z2-null mutants. The JIL-1 protein localizes to interband regions of polytene chromosomes and is found up-regulated on the male X chromosome. Furthermore, the JIL-1z2 allele enhances the phenotype of trx-G mutations. These data indirectly suggest that JIL-1-mediated H3S10 phosphorylation plays an important role in transcriptional activation (Ivaldi, 2007).

This study further characterizes the role of JIL-1-mediated H3S10 phosphorylation in transcription. JIL-1 is required for the transcription of the majority of, if not all, Drosophila genes. Mechanistic analyses place the phosphorylation event subsequent to transcription initiation but prior to productive elongation; JIL-1 plays an integral role in the release of Pol II from promoter-proximal pausing. The data therefore highlight H3S10 phosphorylation as a novel hallmark of early productive elongation in Drosophila (Ivaldi, 2007).

These results establish H3S10 phosphorylation by JIL-1 as a key event during early elongation of transcription in Drosophila. JIL-1 appears to interact with the transcription machinery at most or all actively transcribed regions on Drosophila polytene chromosomes, including active ecdysone and heat-shock genes. At the same time, expression levels of the hsp70 and E75A genes are decreased in JIL-1-null mutants. Importantly, when JIL-1 is mutated, a global decrease in the phosphorylation levels of elongating RNA polymerase II is observed, suggesting that JIL-1 is required for transcription of the majority of genes (Ivaldi, 2007).

The results further elucidate the timing of H3S10 phosphorylation within the framework of the cascade of events that lead to activation of transcription in eukaryotes. Phosphorylation of H3S10 is not required for transcription factor recruitment, since loss of JIL-1 does not affect binding of HSF at the hsp70 genes after heat shock. Also, H3S10 phosphorylation is dependent on BRM chromatin remodeling, which is required genome-wide prior to the recruitment of Pol II. Transcription initiation can take place independently of JIL-1, as shown by the normal levels of Pol IIoser5 and H3K4 methylation in JIL-1z2 mutants, indicating that the chromatin environment in the absence of JIL-1 is still suitable for transcription initiation. However, productive elongation is impaired in these mutants, as is evident by the decrease in Pol IIoser2 levels. These findings introduce H3S10 phosphorylation as a new component of an increasingly complex chromatin environment that is required at the onset of transcription elongation in Drosophila, suggesting a role for JIL-1 in the release of Pol II from promoter-proximal pausing and facilitation of early elongation. Specifically, in JIL-1 mutants, P-TEFb is not detected at the induced hsp70 genes while levels of DSIF are maintained. In the absence of P-TEFb, neither DSIF nor Pol II can be phosphorylated, which is sufficient to block productive elongation. It is likely that Pol II arrests in a paused state and cannot elongate in these mutants. It is also possible that Pol II continues to elongate but is unable to communicate with the proper mRNA processing machinery, which is normally contingent on Ser2 phosphorylation of its CTD (Orphanides, 2002). In this case, the mRNA would be produced but quickly degraded, leading to the transcription defects observed in the Northern analyses. Further work is needed to distinguish between these two possibilities (Ivaldi, 2007).

Although JIL-1 is required for transcription, its presence is not sufficient to ensure gene activation, since JIL-1 is present at all previously transcribed genes that are silenced after heat shock, whereas phosphorylated H3S10 is found exclusively at the transcriptionally active heat-shock genes (Nowak, 2000). Nevertheless, recruitment of JIL-1 to the hsp70 gene is transcription dependent. One possibility is that JIL-1 can exist in both active and inactive states. Once recruited to activate a gene, it may eventually be repressed by inactivation rather than disassociation. Alternatively, the net levels of phosphorylated H3S10 could result from a delicate balance between kinase and phosphatase activities. It has been proposed previously that phosphatase 2A (PP2A) plays a major role in transcription-dependent H3S10 phosphorylation (Nowak, 2003). Therefore, even if JIL-1 is actively maintained at silent genes, its action may be counterbalanced by PP2A. Further studies are required to shed light on how JIL-1 activity can be regulated to affect transcription (Ivaldi, 2007).

In vertebrates, phosphorylation of H3S10 seems to be limited to transcription activation of specific genes in the context of particular signal transduction pathways. In fact, activation of the hsp70 genes by different stressors in mammalian cells is associated with distinct signaling pathways that are not always linked to H3S10 phosphorylation. Contrary to the Drosophila response, heat shock elicits histone H4 acetylation instead of H3S10 phosphorylation at the hsp70 loci in mouse fibroblasts. In contrast, both H3S10 phosphorylation and H4 acetylation are detected at the hsp70 genes upon arsenite treatment of the same cells (Thomson. 2004). Therefore, mammals appear to have more diverse mechanisms of transcription activation and may partially rely on H3S10 phosphorylation in a context-dependent manner. In yeast, substituting the H3 Ser10 for an Ala prevents the recruitment of the TATA-binding protein to the INO1 and GAL1 gene promoters, suggesting that H3S10 phosphorylation is required for the assembly of the preinitiation complex. It would be interesting to explore the significance of this apparent diversity across species (Ivaldi, 2007 and references therein).

The results presented in this study shed light on the mechanism of transcription regulation by H3S10 phosphorylation. It has been recently shown that H3S10 phosphorylation antagonizes the binding of the heterochromatin protein HP1 to histone H3 methylated in Lys9 (H3K9) during mitosis in mammalian cells. It was consequently proposed that JIL-1 maintains chromosome structure in Drosophila by counteracting heterochromatin formation and preventing its spreading into euchromatin. This model for JIL-1 activity could explain a lack of transcription in JIL-1z2 mutants, since any ectopic heterochromatin would make the DNA inaccessible to the Pol II machinery. However, contrary to such a prediction, the current results show that heat-shock puffs are still formed in JIL-1z2mutants, and transcription factors and the Pol II machinery retain the ability to bind despite the disruption of chromatin structure. Furthermore, transcription can be initiated, as is evident by the phosphorylation of Pol II at Ser5. This requires several components of the core transcription machinery and the procession of Pol II a few bases downstream from the promoter. These results suggest that, rather than contribute to global chromosome structure, JIL-1-mediated H3S10 phosphorylation may be required to maintain a local chromatin environment that serves as a platform for the recruitment of P-TEFb and the consequent release of Pol II from promoter-proximal pausing (Ivaldi, 2007).

It has become increasingly evident that transcription elongation is a rate-limiting step of gene expression that requires tight regulation. It was reported recently that the majority of gene promoters in human embryonic stem cells are occupied by a promoter-proximally paused Pol II, poised for productive elongation (Guenther, 2007). This suggests that the expression of these genes is predominantly regulated at the level of Pol II release rather than during preinitiation. The exact mechanism of P-TEFb recruitment, a key step in this process, remains to be determined. Several transcription regulators have been shown to recruit P-TEFb, but this is the first evidence of a histone modification required precisely at the timing of recruitment (Ivaldi, 2007).

The exact contribution of H3S10 phosphorylation to P-TEFb recruitment remains open to further investigation. Recent reports have shown that the ubiquitous protein 14-3-3 binds to H3 only when phosphorylated at Ser10, and this interaction could provide a mechanistic link between H3S10 phosphorylation and P-TEFb (Macdonald, 2005). It is possible that 14-3-3 interacts with P-TEFb directly or indirectly through other transcription regulators that are known to recruit it. Alternatively, 14-3-3 is known to interact with many chromatin-related proteins, thus providing another avenue to manipulate the local chromatin environment to support P-TEFb recruitment and early elongation. Further analyses will be necessary to test these hypotheses and clarify the role and mechanism of regulation of JIL-1 and H3S10 phosphorylation in gene expression (Ivaldi, 2007).

Global analysis of the relationship between JIL-1 kinase and transcription

The ubiquitous tandem kinase JIL-1 is essential for Drosophila development. Its role in defining decondensed domains of larval polytene chromosomes is well established, but its involvement in transcription regulation has remained controversial. For a first comprehensive molecular characterisation of JIL-1, a high-resolution, chromosome-wide interaction profile of the kinase in Drosophila cells was generated, and its role in transcription determined. JIL-1 binds active genes along their entire length. The presence of the kinase is not proportional to average transcription levels or polymerase density. Comparison of JIL-1 association with elongating RNA polymerase and a variety of histone modifications suggests two distinct targeting principles. A basal level of JIL-1 binding can be defined that correlates best with the methylation of histone H3 at lysine 36, a mark that is placed co-transcriptionally. The additional acetylation of H4K16 defines a second state characterised by approximately twofold elevated JIL-1 levels, which is particularly prominent on the dosage-compensated male X chromosome. Phosphorylation of the histone H3 N-terminus by JIL-1 in vitro is compatible with other tail modifications. In vivo, phosphorylation of H3 at serine 10, together with acetylation at lysine 14, creates a composite histone mark that is enriched at JIL-1 binding regions. Its depletion by RNA interference leads to a modest, but significant, decrease of transcription from the male X chromosome. Collectively, the results suggest that JIL-1 participates in a complex histone modification network that characterises active, decondensed chromatin. It is hypothesised that one specific role of JIL-1 may be to reinforce, rather than to establish, the status of active chromatin through the phosphorylation of histone H3 at serine 10 (Regnard, 2011).

The genome-wide mapping of JIL-1 and ePol in male SL2 cells together with the reporter assay in flies suggest that JIL-1 is part of a network of factors that collectively define the state of active chromatin. Taken together with previous genetic analyses, these data lead to a consideration of a role for JIL-1 not in the establishment of active chromatin, but in the reinforcement of the active state, independently of the extend of transcription (Regnard, 2011).

Although JIL-1 binds the bodies of most active gene and is recruited to a reporter gene upon activation, the degree of binding does not correlate well with ePol occupancy. In this respect JIL-1 belongs to a class of several other active chromatin components whose presence in chromatin does not scale with ePol, such as H3K36me3, the DCC subunit MSL1, or H4K16ac mark. By contrast, H3K4me2 levels are proportional to ePol levels. Conceivably, JIL-1 marks active chromatin independent of whether it is currently transcribed or not. A basal level of JIL-1 binding correlates strongly with H3K36me3, an elongation marker that is placed co-transcriptionally by ePol-associated dSet2 (Larschan, 2007). The turnover time of this modification is not known and hence it is possible that H3K36me3 remains on chromatin between pulses of transcription (Regnard, 2011).

Most recently, high-resolution mapping of 56 chromatin components using Dam-ID revealed five types of chromatin in Drosophila KC cells (Filion, 2010). A colour code was used to illustrate those types. The two types of active chromatin, 'red' and 'yellow', differ in that active genes in red chromatin tend to be depleted in H3K36me3 and MRG15, which are abundant constituents of yellow chromatin. Monitoring the relative distributions of ePol and JIL-1 in those types of chromatin it was found that whereas ePol is equally distributed in red and yellow chromatin, JIL-1 is clearly mostly found in yellow chromatin. This correlation again highlights the relationship between JIL-1 and H3K36me3. The precise link between H3K36me3 and JIL-1 recruitment remains to be explored (Regnard, 2011).

A second targeting principle is evident from the fact that JIL-1 levels are approximately twofold increased on the male X chromosome, where due to the action of the DCC H4K16ac levels are high. The situation is complex since the recruitment of the DCC to transcribed genes on the X chromosome is promoted by the potential interaction of the DCC subunit MSL3 with H3K36me3 (Larschan, 2007). Because JIL-1 binding is insensitive to different transcription rates on autosomes, it is not plausible that a presumed twofold increase in transcription on the X chromosome is directly responsible for this elevated association. Likewise, it is not thought that direct interactions between JIL-1 and the DCC can explain the targeting. Although some interactions have been observed in vitro between JIL-1, MSL1 and MSL3, a quantitative co-purification of endogenous JIL-1 with the DCC was never documented. Indeed, activation of a reporter gene in female flies through recruitment of Gal4-MOF is sufficient to recruit JIL-1 to male levels on the reporter gene and on the adjacent Cortactin gene. Altogether, the hypothesis is favored that the enrichment of JIL-1 on X-linked genes is a consequence of a modulating feature of dosage-compensated chromatin. The genome-wide distribution of JIL-1 indeed correlates best with the combination of the two chromatin modifications, H3K36me3 and H4K16ac, on the X-chromosome suggesting that JIL-1 recruitment occurs downstream of those two histone marks. Interestingly, the correlation still holds for autosomal probes, although autosomal H4K16ac is at least in part placed at the 5' end of genes by alternative MOF complexes. JIL-1 recruitment has also been linked to the H4K12ac mark, which is placed by the GCN5-containing ATAC complex. However, it has recently been found that ATAC also contains an acetyltransferase with specificity for H4K16 (atac2), which may contribute to JIL-1 recruitment (Regnard, 2011).

In the cases of very strong transcription activation by high levels of Gal4 activator, no accumulation was seen of JIL-1 to endogenous levels. Conversely, JIL-1 is also not enriched at developmental puffs on polytene chromosomes where robust transcription takes place according to the strong ePol staining. The massive heat-induced activation leads to extensive chromatin decondensation (puffing) on polytene chromosomes, which is probably accompanied by nucleosome loss (Regnard, 2011).

In summary, the idea is favored that the recruitment of JIL-1 relies on several, partly redundant features of active chromatin. JIL-1 harbors an H3-tail binding domain in its C-terminus and another determinant of chromatin targeting in its N-terminal domain, but it does not contain any domain known to directly bind modified histone tails. Its recruitment to active chromatin is, therefore, most probably indirect. No histone modification or any other chromatin-associated feature is known to be enriched on the male X chromosome except for JIL-1 and the JIL-1-dependent phospho-acetyl mark. An appealing model for JIL-1 targeting is that its general recruitment mode (correlating with H3K36me3) would be the same on all chromosomes. The additional presence of H4K16ac might enhance the access of JIL-1 to the first targeting principle, since it has been shown that H4K16ac prevents the folding of the nucleosomal fibre into more compact structures (Regnard, 2011).

It has been observed that histone H3S10ph can enhance acetylation of histone H3K14 and inhibit methylation of histone H3K9. The biochemical analysis of JIL-1 kinase showed that a number of other modifications of the H3 N-terminus are compatible with the phosphorylation by JIL-1. Therefore, JIL-1 phosphorylation is compatible with the prior existence of other modifications, consistent with a downstream function of JIL-1 (Regnard, 2011).

The data suggest that in Drosophila the function of H3S10ph may at least in part differ from that in other organisms, where it is implicated in the fast and transient induction of promoters in response to various inducers. In Drosophila, JIL-1 is not particularly enriched at promoters, but associates with genes along their entire length. This distribution is more compatible with a role in fine-tuning transcriptional elongation. The effects of JIL-1 may - at least in part - be mediated by 14-3-3 proteins, which are able to recognize the interphase H3S10ph mark (Regnard, 2011).

The depletion of JIL-1 has a clear but small effect on overall transcription. The fact that JIL-1 binds gene bodies rather that promoters and the the binding is independent of actual transcription rates leads to the speculation that the kinase may influence transcription efficiency only very indirectly by reinforcing the active chromatin state - once established by co-activators - by protecting it from neighbouring repressive chromatin. This could become important when moderately active genes are not constantly transcribed. Highly transcribed genes experience nucleosome depletion and therefore do not provide anchoring sites for silencing factors. Moderately transcribed genes, however, may experience periods without active elongation. These 'gaps' bear the risk of placement of silencing marks, most notable H3K9me2. In this context JIL-1 may safeguard active genes (marked by H3K36me3 as a sign of recent transcriptional activity). This function may be achieved by phosphorylation of H3S10, which prevents the recognition of HP1 by occlusion of H3K9me2 but possibly also by phosphorylation of other substrates, like Su(var)3,9. In addition JIL-1 might have a scaffolding function for other factors, like lamins. The observation that genes sensitive to the depletion of JIL-1 tend to have less JIL-1 bound both on the X chromosome and on autosomes also support a model where JIL-1 is associated to active chromatin domains to prevent the spreading of more repressive structures. It is speculated that the role of JIL-1 in maintaining the balance between heterochromatin and euchromatin is not only true at the hetero/euchromatin boundary, but also within euchromatin at the level of active chromatin domains (Regnard, 2011).

Genome-wide phosphoacetylation of histone H3 at Drosophila enhancers and promoters

Transcription regulation is mediated by enhancers that bind sequence-specific transcription factors, which in turn interact with the promoters of the genes they control. This study shows that the JIL-1 kinase is present at both enhancers and promoters of ecdysone-induced Drosophila genes, where it phosphorylates the Ser10 and Ser28 residues of histone H3. JIL-1 is also required for CREB binding protein (CBP)-mediated acetylation of Lys27, a well-characterized mark of active enhancers. The presence of these proteins at enhancers and promoters of ecdysone-induced genes results in the establishment of the H3K9acS10ph and H3K27acS28ph marks at both regulatory sequences. These modifications are necessary for the recruitment of 14-3-3, a scaffolding protein capable of facilitating interactions between two simultaneously bound proteins. Chromatin conformation capture assays indicate that interaction between the enhancer and the promoter is dependent on the presence of JIL-1, 14-3-3, and CBP. Genome-wide analyses extend these conclusions to most Drosophila genes, showing that the presence of JIL-1, H3K9acS10ph, and H3K27acS28ph is a general feature of enhancers and promoters in this organism (Kellner, 2012).

Activation of transcription in higher eukaryotes requires the interaction between transcription factors bound to distal enhancers and proteins present at the promoter. Recent findings indicate that enhancers contain a variety of histone modifications that change during the establishment of specific cell lineages suggesting that these sequences may play a more complex role in transcription than previously thought. Given the presence of common as well as specific histone marks at enhancers and promoters, it is tempting to speculate that epigenetic modifications at these sequences serve to integrate various cellular signals required to converge in order to activate gene expression. Results described in this study support this hypothesis, demonstrating that the proteins that carry out these histone modifications are necessary to establish enhancer-promoter contacts and activate transcription of ecdysone-inducible genes (Kellner, 2012).

The execution of this process in Drosophila requires the recruitment of JIL-1 by mechanisms that are not well understood. Although the direct involvement of JIL-1 in the transcription process has been brought into question due to the failure to observe recruitment of JIL-1 to heat shock genes in polytene chromosomes, results presented in this study clearly indicate that JIL-1 affects transcription at different steps in the transcription cycle. At the promoter region, phosphorylation of H3S10 by JIL-1 results in the recruitment of 14-3-3 and, subsequently, histone acetyltransferases Elp3 and MOF (Karam, 2010). This study found that JIL-1 is also able to phosphorylate H3S28 at both promoters the enhancers. The establishment of the H3K9acS10ph and H3K27acS28ph modifications correlates with the recruitment of 14-3-3 to enhancers and promoters of ecdysone-induced genes. 14-3-3 has been implicated in numerous cellular processes, where it functions as a scaffold protein). 14-3-3 is found as dimers and multimers; each monomer is capable of binding two targets and can mediate and stabilize interactions between two phosphoproteins. Additionally, acetylation facilitates the dimerization of 14-3-3 molecules and their ability to bind certain substrates. Binding assays have demonstrated that 14-3-3 interacts weakly with H3 tail peptides phosphorylated at S10 and S28, but strong binding is detected if the peptide is both phosphorylated and acetylated on the neighboring lysine residues. Given the ability of 14-3-3 to serve as a scaffold for large protein complexes, its demonstrated interactions with H3K9acS10ph and H3K27acS28ph and the presence of these two modifications at enhancers and promoters, it is possible that contacts between these two sequences are stabilized by 14-3-3. This hypothesis is supported by 3C experiments indicating that induction of transcription of the Eip75B gene is accompanied by strong enhancer-promoter interactions. These interactions are lost in JIL-1, CBP, and 14-3-3 knockdown cells. Since these proteins act several steps downstream from transcription factor binding in the pathway leading to enhancer-promoter contacts, and loss of these proteins results in the abolishment of these contacts, it appears that these proteins, rather than specific transcription factors, may be responsible for enhancer promoter interactions at the ecdysone-inducible genes (Kellner, 2012).

Genome-wide studies using ChIP-seq clearly show the presence of JIL-1, H3K9acS10ph, and H3K27acS28ph at enhancers and promoters of most Drosophila genes. There is a clear correlation between the amount of JIL-1, H3K9acS10ph, and H3K27acS28ph at promoters and the level of transcripts associated with the gene. These three marks are also present at enhancers defined by the occurrence of H3K4me1 and H3K27ac, suggesting that the JIL-1 kinase is a regulator of histone dynamics at enhancers and promoters genome-wide. JIL-1, H3K9acS10ph, and H3K27acS28ph are found at low levels at enhancers before activation, which then increase in intensity and drop in baseline when found in combination with H3K27ac, a mark of active enhancers. These conclusions are different from those previously published examining the role of JIL-1 in transcription and dosage compensation (Regnard, 2011). This study concluded that JIL-1 binds active genes along their entire length and that the levels of JIL-1 are not associated with levels of transcription. The differences in the conclusions may be due to the different cell lines used -- male S2 cells versus female Kc cells -- and the emphasis of the analysis by Regnard on the expression of dosage-compensated genes in the male X-chromosome, which may contain JIL-1 throughout the genes as a consequence of their regulation at the elongation step. In addition, the study by Regnard used ChIP-chip on custom tiling arrays of the X chromosome plus cDNA arrays containing the whole genome. This strategy may bias the conclusions and suggest the presence of JIL-1 in the coding region of genes rather than at enhancers and promoters (Kellner, 2012).

Results presented in this study extend the previous list of histone modifications characteristic of active enhancers to include H3K9acS10ph and H3K27acS28ph. Enhancers tend to be cell type-specific and are determined during differentiation with the characteristic H3K4me1 modification. It is unclear how these regions are designated before activation and what keeps them in a poised state ready for activation upon receiving the proper signal from the cell. It is tempting to speculate that the presence of JIL-1 at enhancers prior to activation might play a role in maintaining the enhancer in this poised state. An important question for future studies is the mechanistic significance of the looping between enhancers and promoters in order to achieve transcription activation. One interesting possibility is that various signaling pathways in the cell contribute to building epigenetic signatures at enhancers and promoters in the form of histone acetylation and/or phosphorylation of various Lys/Ser/Thr residues. Acetylation marks at enhancers and promoters may then cooperate to recruit BRD4 (FS(1)H in Drosophila), which contains two bromodomains each able to recognize two different acetylated Lys residues. The requirement for acetylation of histones at enhancers and promoters in order to recruit Brd4 would ensure that several different signaling events have taken place before recruitment of P-TEFb by BRD4 can release RNAPII into productive elongation (Kellner, 2012).

Histone H3S10 phosphorylation by the JIL-1 kinase in pericentric heterochromatin and on the fourth chromosome creates a composite H3S10phK9me2 epigenetic mark

The JIL-1 kinase mainly localizes to euchromatic interband regions of polytene chromosomes and is the kinase responsible for histone H3S10 phosphorylation at interphase in Drosophila. However, recent findings raised the possibility that the binding of some H3S10ph antibodies may be occluded by the H3K9me2 mark obscuring some H3S10 phosphorylation sites. Therefore, this study has characterized an antibody to the epigenetic H3S10phK9me2 double mark as well as three commercially available H3S10ph antibodies. The results showed that for some H3S10ph antibodies their labeling indeed can be occluded by the concomitant presence of the H3K9me2 mark. Furthermore, it was demonstrated that the double H3S10phK9me2 mark is present in pericentric heterochromatin as well as on the fourth chromosome of wild-type polytene chromosomes but not in preparations from JIL-1 or Su(var)3-9 null larvae. Su(var)3-9 is a methyltransferase mediating H3K9 dimethylation. Furthermore, the H3S10phK9me2 labeling overlapped with that of the non-occluded H3S10ph antibodies as well as with H3K9me2 antibody labeling. Interestingly, when a Lac-I-Su(var)3-9 transgene is overexpressed, it upregulates H3K9me2 dimethylation on the chromosome arms creating extensive ectopic H3S10phK9me2 marks suggesting that the H3K9 dimethylation occurred at euchromatic H3S10ph sites. This is further supported by the finding that under these conditions euchromatic H3S10ph labeling by the occluded antibodies was abolished. Thus, these findings indicate a novel role for the JIL-1 kinase in epigenetic regulation of heterochromatin in the context of the chromocenter and fourth chromosome by creating a composite H3S10phK9me2 mark together with the Su(var)3-9 methyltransferase (Wang, 2014).

JIL-1: Biological Overview | Evolutionary Homologs | Developmental Biology | Effects of Mutation | References

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.