GTC
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Genes to Cells (2008) 13, 703-722. doi:10.1111/j.1365-2443.2008.01199.x
© 2008 Blackwell Publishing or its licensors
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kato, Y.
Right arrow Articles by Yamaguchi, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Kato, Y.
Right arrow Articles by Yamaguchi, M.

Characterization of Drosophila G9a in vivo and identification of genetic interactants

 Yasuko Kato1,2, Masaki Kato1,3,a, Makoto Tachibana4, Yoichi Shinkai4 and Masamitsu Yamaguchi1,2,*

1 Department of Applied Biology, and
2 Insect Biomedical Research Center, and
3 Venture Laboratory, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan
4 Experimental Research Center for Infectious Diseases, Institute for Virus Research, Kyoto University, Sakyo-ku, Kyoto 606-8507, Japan


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Experimental procedures
 References
 
In mammals, G9a is a histone H3 lysine 9 (H3-K9)-specific histone methyltransferase (HMTase), known to be essential for murine embryogenesis. It has been reported that Drosophila G9a (dG9a) is a dominant suppressor of position effects of variegation, has HMTase activity in vitro, and is important for Drosophila development. Here we show that dG9a has H3-K9 dimethylation activity in vivo and is important for the recruitment of HP1 in the euchromatic region. Over-expression in eye imaginal discs inhibited the differentiation of pupal ommatidial cells and resulted in abnormal eye morphology (rough eye phenotype) in the adults, although a methylase defective mutant did not demonstrate such effects. These results suggest that HMTase activity of dG9a affects transcription of genes involved in pupal eye formation. The dG9a-induced rough eye phenotype was enhanced by a half-dose reduction of the Polycomb group (PcG) gene. In contrast, mutants for little imaginal discs (lid), encoding histone H3-K4 demethylase, demonstrated suppression of the rough eye phenotype induced by dG9a. Furthermore co-expression of Lid in eye imaginal discs enhanced the rough phenotype induced by dG9a. The results suggest that the function of dG9a is negatively regulated by the PcG complex and positively regulated by Lid in vivo.


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Experimental procedures
 References
 
In eukaryotic cells, the high order chromatin structure is important for epigenetic regulation and control of gene activation and silencing, mitosis and heterochromatin. Histone lysine methylation can be present in mono-, di-, or trimethylation states, and can produce active (H3-K4, H3-K36, H3-K79) or repressive (H3-K9, H3-K27, H4-K20) modifications for gene expression (Fischle et al. 2003; Lachner et al. 2003). Histone methyltransferases (HMTases) have been classified into two types. One group carries the SET domain that is evolutionarily conserved and the other carries the Dot1 domain (Rea et al. 2000; van Leeuwen et al. 2002). G9a is an H3-K9 specific HMTase found in mammals (Tachibana et al. 2001). It contains the SET domain and adjacent cysteine-rich regions (the pre-SET and post-SET domains) and appears responsible for all detectable H3-K9 dimethylation and a significant amount of monomethylation in silent euchromatin, but not in pericentric heterochromatin. G9a knockout mice show early embryonic lethality, and therefore G9a appears to play an important role in transcriptional repression in early mouse development (Tachibana et al. 2002). In recent studies, it was found to be included in a complex containing the transcription factor E2F-6 (Ogawa et al. 2002), or the co-repressor CtBP (Shi et al. 2003). Furthermore, it is reported that G9a is recruited to target genes depending on the promoter context or regulatory elements (Gyory et al. 2004; Nishio et al. 2004; Roopra et al. 2004; Ueda et al. 2006; Nishida et al. 2007) to cause transcriptional repression. In addition, G9a plays a role as a co-activator of nuclear receptors (NRs), collaborating synergistically with CARM1 and other NR co-activators (Lee et al. 2006). G9a functions as a co-repressor or a co-activator is controlled by promoter context and/or regulatory environment.

In Drosophila, H3-K9 mono-, di- and trimethylation is enriched in pericentromeric heterochromatin including the fourth chromosome in the polytene chromosome (Schotta et al. 2002; Ebert et al. 2004). SU(VAR)3–9 was first identified H3-K9 HMTase (Czermin et al. 2001). In Su(var)3–9 null mutants, H3-K9 di- and trimethylation disappear from the chromocentre of the salivary gland chromosome but not from the heterochromatic fourth chromosome (Ebert et al. 2004). Recently, other two HMTases in Drosophila have been identified. One of them, Drosophila SETDB1/egg (dSETDB1) was involved in H3-K9 trimethylation in ovary and was required for oogenesis at early stages of egg chamber formation (Clough et al. 2006). Furthermore, dSETDB1 is required for gene silencing of the fourth chromosome (Seum et al. 2007; Tzeng et al. 2007). Another HMTase, Drosophila G9a (dG9a) appears to be a functional homologue of human G9a (Mis et al. 2006; Stabell et al. 2006). dG9a appears to have H3-K9 methyltransferase activity in vivo and functions as a dominant suppressor of position effect of variegation (Mis et al. 2006), although, in vitro, dG9a methylates not only H3 but also H4 (Stabell et al. 2006). dG9a appears to play a role in the Ecdysone signaling pathway (Stabell et al. 2006).

Here we further explored dG9a functions in vivo. Immunodetection with anti-dG9a antibody revealed endogenous dG9a to be enriched in interband and telomeric regions on polytene chromosomes along with nucleoplasm of the salivary gland cell nuclei. Over-expression or knockdown of dG9a in transgenic fly lines induced either lethality or abnormal morphology in adult wings or eyes. Furthermore, in nuclei of dG9a knockdown salivary glands, the localization of HP1 in euchromatic region was decreased and Phosphorylated RNA Polymerase II (Pol II) signals were increased. These data suggest that the euchromatic H3-K9 methylation catalyzed by dG9a plays an important role in organization of the higher-order chromatin structure and transcriptional repression. It is well known that the Polycomb group (PcG) proteins of Drosophila are chromatin components that maintain stable and heritable repression of many target genes during development. The polycomb repressive complex 2 (PRC2) including E(Z) mediates the mono-, di-, and tri-methylation of H3-K27. Polycomb (PC), which is a subunit of PRC1, binds to the methylated H3-K27 via its chromodomain (Schuettengruber et al. 2007). We therefore here examined possible genetic interaction between dG9a and PcG genes. The rough eye phenotype induced by dG9a was enhanced by mutations in genes encoding components of PRC1, suggesting that these PcG genes negatively regulate the dG9a function. In contrast, mutants of little imaginal discs (lid), encoding trithorax-group (trxG) histone H3-K4 demethylase, suppressed the rough eye phenotype induced by dG9a, suggesting that Lid positively regulates the dG9a function.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Experimental procedures
 References
 
Localization of dG9a in polytene chromosome

The anti-dG9a antibody raised in rabbits was affinity-purified with Sepharose conjugated with GST-dG9a fusion protein. In order to confirm the specificity, we carried out a Western immunoblot analysis using extracts from Drosophila cultured S2 cells transfected with or without the plasmids pAct5c-GAL4 and pUAS-FLAG-dG9a (Fig. 1A). The anti-dG9a antibody detected a band of 200 kDa in S2 cell extracts that was not detectable with control IgG (Fig. 1A, lanes 1–3) and this immuno-reactive band was increased in extracts of S2 cells expressing the FLAG-dG9a fusion protein (Fig. 1A, lane 4). The 200 kDa band was also detected with anti-FLAG antibody (Fig. 1A, lane 6). The estimated size of the band is comparable with the predicted molecular weight from the primary amino acid sequence (181 kDa) and no other immuno-reactive band was detected with the anti-dG9a antibody. These results indicate that the anti-dG9a antibody specifically reacts with the dG9a protein.


Figure 1
View larger version (26K):
[in this window]
[in a new window]

  		Figure 1  Subcellular localization of dG9a in salivary glands. (A) Western blot analysis of Drosophila S2 cell extracts. The blots were probed with control rabbit IgG, anti-dG9a ({alpha}-dG9a), anti-FLAG ({alpha}-FLAG) and anti-{alpha}-tubulin ({alpha}-{alpha}-tubulin) antibodies, {alpha}-tubulin being used as a loading control. (*) nonspecific band detected with {alpha}-FLAG. (B) Immunostaining of a whole mount salivary gland from a third instar larva. The salivary gland was immunostained with {alpha}-dG9a (left). DNA was stained with propidium iodide (PI) (middle). Merged images of {alpha}-dG9a and PI staining (right). (C) Higher magnification images of salivary gland nuclei. {alpha}-dG9a signals (left), DNA staining with PI (middle) and the merged image (right) are shown. Arrowheads show chromocenters.

 
It is reported that dG9a localizes to euchromatic regions of the salivary gland polytene chromosomes in a discrete banding pattern, but not to chromocentres (Stabell et al. 2006). To further examine subcellular localization of dG9a, we carried out immunostaining of third instar larval salivary glands (Fig. 1B). Strong dG9a signals were observed in salivary gland cell nuclei, but none were detected in the cytoplasm (Fig. 1C). Merged images between dG9a signals and DNA staining indicate dG9a to localize both on polytene chromosomes and in the nucleoplasm. No dG9a signals were detected on the specific chromosomal region, located very close to the nuclear envelope, showing dense PI-staining signals (Fig. 1C, arrowhead) and very likely having a chromocentre nature. Immunostaining of polytene chromosome spreads using both anti-dG9a and anti-HP1 antibodies, HP1 localizing predominantly on chromocentres (James et al. 1989) (Fig. 2A–F), demonstrated dG9a signals in euchromatic sites on the polytene chromosomes (Fig. 2A–C), but not on the heterochromatic chromocentre and the fourth chromosome where strong HP1 signals were detected (Fig. 2D–F,P–R). dG9a signals were detected in nearly 200 sites of the euchromatic region of polytene chromosomes and no strict correlation were observed between the sites and band/interband regions, although some enrichment of dG9a in interband regions was observed (Fig. 2A–C,J–O). These immunolocalization data suggest that dG9a regulates chromatin structure and transcription in euchromatic regions. In addition, telomeric localization of dG9a was also observed, (Fig. 2J–O), suggesting some additional roles of dG9a in maintaining chromatin structure of telomere regions.


Figure 2
View larger version (37K):
[in this window]
[in a new window]

  		Figure 2  Distribution of dG9a protein on polytene chromosomes. (A–C, J–O) Polytene chromosomes of third instar larvae were immunostained with anti-dG9a (A, J, M). DNA (red) was stained with DAPI (B, K, N). Merged images of dG9a and DNA signals (C, L, O). dG9a (green) binds to nearly 200 sites on polytene chromosomes. (J–L) High magnification images of the telomere region of the X chromosome of another spread. (M–O) High magnification images of the telomere region of the 2 L chromosome of another spread. (J–O) Arrowheads show the telomere regions. (D–F, P–R) Polytene chromosomes of third instar larvae were immunostained with {alpha}-dG9a (D, P) and {alpha}-HP1 (E, Q). Merged images of dG9a and HP1 signals (F, R). (P–R) High magnification images of the heterochromatic chromocentre and the IV chromosome region. dG9a (green) is not associated with the heterochromatic chromocentre and the IV chromosome. HP1 (red) signals mark the chromocenters (arrows) and the IV chromosomes (arrowheads, A–F, P–R). Polytene chromosomes from third instar larvae were immunostained with antibodies for dG9a (G, S) and active RNA polymerase II (Pol II) (H, T). Merged images of dG9a and Pol II signals (I, U). (S–U) High magnification images of {alpha}-dG9a and {alpha}-Pol II signals on polytene chromosome from another spread. dG9a is not colocalized with active Pol II. Arrowheads show strong Pol II signals.

 
To further explore the potential role of dG9a in gene expression, we compared dG9a distribution on polytene chromosomes with that of Pol II that marks transcriptional elongation (Fig. 2G–I,M). The serine 5 of the C-terminal domain of the largest subunit of Pol II is phosphorylated after promoter clearance and during early step of elongation, while the serine 2 is phosphorylated at later stages of transcription elongation (Komarnitsky et al. 2000). The anti-Pol II antibody (H14) recognizes phosphorylated serine 5. No significant overlap between dG9a and Pol II signals on polytene chromosome was observed. Although we can not conclude that dG9a indeed plays a role in transcriptional repression from these data, the observation suggests that dG9a mediates transcriptional repression through modification of chromatin.

Over-expression of dG9a induced dimethylation of H3-K9

Previous studies in vivo and in vitro suggested that dG9a has H3-K9 methyltransferase activity (Mis et al. 2006; Stabell et al. 2006). For further confirming this in different way, we established nine independent transgenic fly lines carrying UAS-FLAG-dG9a and nine independent transgenic fly lines carrying UAS-FLAG-dG9a{Delta}1532–1538 lacking the catalytic core motif of the SET domain and therefore lacking HMTase activity (Table 1). These transgenic flies were crossed with Sg-GAL4 driver strains to specifically express dG9a in the salivary glands. In Drosophila polytene chromosomes, H3-K9 dimethylation mainly localizes at the chromocentre (Schotta et al. 2002). If dG9a has H3-K9 dimethylase activity in vivo, the pattern of H3-K9 dimethylation in nuclei should be changed when dG9a is over-expressed. The overall size of salivary glands over-expressing dG9a was much smaller than that of wild-type salivary glands. Moreover, polytene chromosomes of salivary glands over-expressing dG9a were fragile and not suitable for preparing polytene chromosome spreads. These properties were not observed in salivary glands expressing dG9a{Delta}1532–1538. We therefore examined the pattern of H3-K9 dimethylation in the salivary gland whole nuclei of dG9a-over-expressing flies.


View this table:
[in this window]
[in a new window]

  		Table 1  Transformants carrying the wild-type dG9a cDNA and its deletion derivative dG9a{Delta}1532–1538
 
In control flies (Sg-GAL4), dimethylation of H3-K9 was detected exclusively on the chromocentre, apparent as a single spot per nucleus (Fig. 3A, 2meH3-K9). In flies over-expressing dG9a (Sg-GAL4 > FLAG-dG9a), H3-K9 dimethylation signals were detected as several distinct spots on the chromosomes in addition to the chromocentre (Fig. 3B, 2meH3-K9). In contrast, in the case of the expression of dG9a{Delta}1532–1538 in salivary glands, no ectopic H3-K9 dimethylation signals were observed (Fig. 3C). Quantitative studies with deconvolution microscopy revealed that dimethyl H3-K9 signals in nuclei of salivary glands over-expressing dG9a were increased to 1.9-fold in compared with that of wild-type (Fig. 3D). In contrast no significant change was observed in level of dimethyl H3-K9 signals in nuclei of salivary glands over-expressing dG9a{Delta}1532–1538. These data indicate that enzymatically active dG9a can induce histone H3-K9 dimethylation in euchromatic regions in vivo.


Figure 3
View larger version (49K):
[in this window]
[in a new window]

  		Figure 3  In vivo HMTase activity of dG9a. (A–C) Immunostaining of third instar larval salivary glands with {alpha}-H3-K9me2 (2me, red) and {alpha}-FLAG (green) antibodies. DNA was stained with DAPI (blue). Merged images of FLAG, H3-K9me2 and DAPI signals. (A) Sg-GAL4/yw; +, (B) Sg-GAL4/+; UAS-FLAG-dG9a/+, (C) Sg-GAL4/+; UAS-FLAG-dG9a{Delta}1532–1538/+. (D) Quantification of H3-K9me2 fluorescence. The {alpha}-H3-K9me2 staining of salivary glands was analyzed by deconvolution microscopy. Means of five independent measurements are shown, error bars represent SD. Intensity units were relative to yw that was scored as 1. yw (Sg-GAL4/yw; +), dG9a (Sg-GAL4/+; UAS-FLAG-dG9a/+), dG9a{Delta} (Sg-GAL4/+; UAS-FLAG-dG9a{Delta}1532–1538/+).

 
The localization and distribution patters of FLAG-dG9a, DNA and H3-K9 dimethylation signals appear to differ cell by cell to some extent (Fig. 3B). These differences likely represent differences in spatial pattern of signals in each nucleus, although we cannot rule out the possibility that the difference represent cells in different cell cycle phases. Further analyses are necessary to address these points.

Knockdown of dG9a in salivary glands affected distribution of HP1 and phosphorylated RNA polymerase II in euchromatic regions

We also investigated the function of dG9a in vivo by RNA interference (RNAi), establishing transgenic fly lines carrying UAS-5'IR-dG9a or UAS-3'IR-dG9a which express double-stranded RNAs (dsRNAs) targeted for different regions of dG9a mRNA (Table 2). To deplete dG9a, we crossed the transgenic line with an Act25-GAL4 driver that expresses GAL4 mainly in imaginal discs (Yoshida et al. 2004). The results of semi-quantitative RT-PCR using total RNA from larval progeny showed a half reduction of dG9a mRNA level in dG9a-knockdown flies compared with control flies (Fig. 4A). Furthermore, immunoblot analysis revealed the dG9a protein to be decreased to undetectable levels in extracts of the dG9a-knockdown flies (Fig. 4B). All progeny showed lethality in early pupal stages. Furthermore, lethality was observed at the larval stage when UAS-5’IR-dG9a flies were crossed with tubP-GAL4 flies expressing GAL4 ubiquitously throughout development (Table 3). The transgenic lines carrying UAS-5'IR-dG9a and UAS-3'IR-dG9a suppressed the phenotype induced by the over-expression of dG9a (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2  Transformants carrying IR-dG9a
 

Figure 4
View larger version (38K):
[in this window]
[in a new window]

  		Figure 4 Knockdown of dG9a decreased HP1 signals and increased phosphorylated Pol II signals in euchromatic regions on polytene chromosomes. (A) Reduction of dG9a mRNA levels in larval tissues. rp49 mRNA was used as an internal control. yw (yw/+; Act25-GAL4/+), RNAi (yw/+; Act25-GAL4/UAS-5'IR-dG9a). PCR results with the same primers in a dilution series (1, 4–1 : 2 dilution, 5–1 : 4 dilution; 3, 6–1 : 8 dilution) of the input cDNA to assess the exponential phase are shown. (B) Western blot analysis of extracts from third instar larvae. The blots were probed with {alpha}-dG9a, {alpha}-HP1 and {alpha}-{alpha}–tubulin. The {alpha}-tubulin was used as a control. yw (yw/+; Act25-GAL4/+), RNAi (yw/+; Act25-GAL4/UAS-5'IR-dG9a). (C, D) Immunostaining of salivary glands of third instar larvae with {alpha}-HP1 (green) and {alpha}-dG9a (red). DNA was stained with DAPI (blue). Merged images of {alpha}-dG9a, {alpha}-HP1 and DAPI signals. (C) Sg-GAL4/+; +, (D) Sg-GAL4/+; UAS-5'IR-dG9a/+. (E, F) Immunostaining of polytene chromosomes of third instar larvae with {alpha}-active Pol II. Pol II signals (green) were increased dramatically on interband regions of dG9a RNAi polytene chromosome. DNA was stained with DAPI (red). (E) Sg-GAL4/+; +, (F) Sg-GAL4/+; UAS-5'IR-dG9a/+. Merge, merged images of {alpha}-Pol II and DAPI signals.

 

View this table:
[in this window]
[in a new window]

  		Table 3  Summary of phenotype of each transgenic flies with each GAL4 driver lines
 
It has been noted that H3-K9 methylation results in binding of HP1 via its chromodomain (Bannister et al. 2001; Lachner et al. 2001). In homozygous loss-of-function Su(var)3–9 larvae, HP1 associated with chromocentre heterochromatin was dramatically reduced (Schotta et al. 2002). In order to examine whether the localization of HP1 is dependent on dG9a, we examined HP1 localization in salivary glands of dG9a-knockdown flies (Fig. 4D). In salivary glands of control flies (Sg-GAL4), HP1 signals were detected in overall nuclei with the highest signals on chromocenters (Fig. 4C). In salivary glands of dG9a-knockdown flies (Sg-GAL4 > UAS-5'IR-dG9a), HP1 signals were detected exclusively on the chromocentres, and no signals were detected in euchromatic regions of the chromosomes (Fig. 4D, HP1). Decreased dG9a levels in the salivary glands of dG9a-knockdown flies were confirmed by anti-dG9a immunostaining (Fig. 4C–D, dG9a). No difference in the level of HP1 was observed between control and dG9a knockdown flies as shown in Western blots (Fig. 4B). These results strongly suggest that the binding of HP1 in euchromatic regions is dependent on H3-K9 methylation catalyzed by dG9a.

Next, to examine the effect of dG9a knock down on the transcriptional activation, we compared the level and distribution of phosphorylated Pol II on polytene chromosome from wild-type and that of dG9a RNAi line (Fig. 4E,F). Phosphorylated Pol II signals on the interband region of RNAi polytene chromosomes were dramatically increased relative to wild-type. These data suggest that dG9a plays a negative role in gene expression on euchromatic region.

Ectopic expression of dG9a in eye imaginal discs does not inhibit cell cycle progression but interferes with cell differentiation of pupal retinae

In GMR-GAL4, the promoter carrying transcription factor Glass-binding sites can be used for the expression of GAL4 in the region within and posterior to the morphogenetic furrow (MF) (Moses et al. 1991). Over-expression of FLAG-dG9a by the GMR-GAL4 driver in the eye disc resulted in abnormal eye morphology with a rough appearance in the adults (Table 3 and Fig. 5A b). In these flies, the ommatidia lacked normal shape and were fused. Loss of pigment cells and increased or decreased bristles were also observed (Fig. 5A e,h). Over-expression of FLAG-dG9a{Delta}1532–1538 lacking the catalytic core motif of the SET domain did not induce such an eye phenotype (Fig. 5A c,f,i), suggesting that the observed effect is due to the HMTase activity of dG9a. Immunostaining of the eye imaginal discs from GMR-GAL4 > FLAG-dG9a flies or GMR-GAL4 > FLAG-dG9a{Delta}1532–1538 flies with anti-FLAG antibody confirmed FLAG-dG9a or FLAG-dG9a{Delta}1532–1538 expression in the region within and posterior to the furrow, at apparently equivalent levels (Fig. 5A j–l). Similar levels of FLAG-dG9a and FLAG-dG9a{Delta}1532–1538 expression in transgenic flies were further confirmed by Western blot analyses of extracts from hs-GAL4 > FLAG-dG9a or hs-GAL4 > FLAG-dG9a{Delta}1532–1538 flies (Fig. 5B). These results indicate that HMTase activity of dG9a is responsible for the rough eye phenotype.


Figure 5
View larger version (51K):
[in this window]
[in a new window]

  		Figure 5  Over-expression of dG9a induces a rough eye phenotype. (A) Images of adult eyes taken by scanning electron microscopy (a–f) and light microscopy (g–i) of adult compound eyes. (j–l) Immunostaining of eye imaginal discs with {alpha}-FLAG. Scale bars show 50 µM (a–c, j–l), 14.2 µM (d–f). GMR-GAL4/+; +, (b, e) GMR-GAL4/+; UAS-FLAG-dG9a/+, (c, f) GMR-GAL4/+; UAS-FLAG dG9a{Delta}1532–1538/+. The flies were developed at 28 °C. (B) Western blot analysis of third instar larval protein extracts from yw/+; hs-GAL4/+(yw), yw/+; hs-GAL4/UAS-FLAG-dG9a (dG9a) and yw/+; hs-GAL4/UAS-FLAG-dG9a{Delta}1532–1538 (dG9a{Delta}) flies. The blots were probed with {alpha}–FLAG and {alpha}-{alpha}-tubulin. The third instar larvae were heat-shocked for 1 h at 37 °C, and then allowed to recover for 2 h at 25 °C before preparing extracts. The {alpha}-tubulin was used as a control.

 
We focused on analysis of the mechanism that directs the rough eye phenotype induced by dG9a over-expression, since cell proliferation and differentiation during Drosophila eye development have been well characterized. To determine if dG9a over-expression has any effect on apoptosis, we crossed flies expressing FLAG-dG9a with flies expressing a broad specificity Caspase inhibitor, encoded by the baculovirus Autographa californica, p35 (Crook et al. 1995) and anti-apoptotic proteins, Drosophila IAP, DIAP1 and DIAP2 (Hay et al. 1995). If apoptosis mediated by Caspase is induced by over-expression of FLAG-dG9a, the rough eye phenotype should be suppressed by the expression of these apoptosis inhibitors. However no such suppressive effect on the rough eye phenotype was observed (Fig. 6A). Furthermore, we carried out immunostaining of eye imaginal discs with anti-activated Caspase-3 antibody. No significant difference in the activated Caspase-3 signals between eye imaginal discs of control and dG9a over-expressing flies was observed (Fig. 6B). The data thus indicate that ectopic apoptosis is not induced in eye imaginal discs of dG9a over-expressing flies.


Figure 6
View larger version (55K):
[in this window]
[in a new window]

  		Figure 6 Over-expression of dG9a does not induce ectopic cell death and does not affect S-phase progression. (A–B) dG9a does not induce apoptosis. (A) Scanning electron micrographs of adult compound eyes. (a) GMR-GAL4/+; UAS-FLAG-dG9a/+, (b) GMR-GAL4/+; UAS-p35/+; UAS-FLAG-dG9a/+, (c) GMR-GAL4/+; GMR-DIAP1/+; UAS-FLAG-dG9a/+, (d) GMR-GAL4/+; UAS-FLAG-dG9a/GMR-DIAP2. Scale bars show 50.0 µM (a–d). (B) Detection of cleaved Caspase-3 to visualize cell death. (b, d) Eye imaginal discs were immunostained with {alpha}-cleaved Caspase-3. (a, c) DNA was stained with DAPI. (a) GMR-GAL4/+; +. (b) GMR-GAL4/+; UAS-FLAG-dG9a/+. Arrowheads show morphogenetic furrow. (C) BrdU incorporation to visualize S-phase cells. (a) GMR-GAL4/+; +. (b) GMR-GAL4/+; UAS-FLAG-dG9a/+. Arrowheads indicate morphogenetic furrows.

 
Cells in front of the MF proliferate asynchronously, while those on the MF are arrested synchronously at G1 phase. A part of cells behind the MF leave the cell cycle and differentiate into photoreceptors of the adult ommatidium, others undergoing one more cell division. This cell cycle is a final synchronous round, and produces an S-phase band in the eye discs. Therefore, no mitotic cells behind the S-phase band are seen. To examine the effect of dG9a over-expression on DNA synthesis in the eye imaginal discs, we visualized S-phase cells with 5-Bromo-2'-deoxyuridine (BrdU) incorporation. In eye imaginal discs that expressed human p53 or dREX2 using this GMR-GAL4 driver, S-phase cells corresponding to the second mitotic wave were almost completely abolished (Yamaguchi et al. 1999; Otsuki et al. 2004). However, both in eye discs which express GAL4 alone and in eye discs which express FLAG-dG9a, BrdU incorporation was similarly observed anterior to the MF and in a stripe just posterior to the MF (Fig. 6C). These data indicate that the over-expression of dG9a does not affect entry into S-phase or progression through S-phase in eye imaginal discs.

Photoreceptor cells have been found to be generated in a stereotype order: R8 is generated first, with movement posterior from the MF, then cells are added pair wise (R2 and R5, R3 and R4, and R1 and R6), and R7 is the last photoreceptor to be added to each cluster. To determine if dG9a inhibits the differentiation of photoreceptors, we used several enhancer trap lines expressing a nucleus-located form of β-galactosidase depending on the specific enhancer-promoter located nearby the P-element to mark each photoreceptor cell. We used two enhancer trap lines, BB02 (Mlodzik et al. 1990a) and AE127 (Mlodzik et al. 1990b), to specifically mark photoreceptor cells of late R8 (Fig. 7A a,d) and R3/R4/R1/R6 (Fig. 7A b,e), respectively. Furthermore, we carried out immunostaining of eye imaginal discs with anti-Elav antibody for pan-neural marker. Elav normally expressed in the posterior portion of the eye imaginal disc. In eye imaginal discs of dG9a over-expressing flies, all eight photoreceptor cells appeared to differentiate normally (Fig. 7A) and the orientation of rhabdomeres as visualized with phalloidin staining appeared normal (Fig. 7B a,c).


Figure 7
View larger version (46K):
[in this window]
[in a new window]

  		Figure 7  Ectopic expression of dG9a in eye imaginal discs interferes with differentiation of cone cells. (A) Immunostaining of eye imaginal discs with {alpha}-β-galactosidase and {alpha}-Elav. GMR-GAL4 (a–c) or GMR-GAL4; UAS-FLAG-dG9a (c–d) line were crossed with photoreceptor-cell specific enhancer trap lines, BB02 (a, d), AE127 (b, e), and expressed β-galactosidase in R8 and R3/4/1/6, respectively. (c, f) Immunostaining of eye imaginal discs with {alpha}-Elav. (c) GMR-GAL4/+; +. (f) GMR-GAL4/+; UAS-FLAG-dG9a/+. (B) Immunostaining of eye imaginal discs and pupal retinae. (a, d) Immunostaining of eye imaginal discs with phalloidin. (b, e) Immunostaining of the pupal retinae with {alpha}-Cut. Arrowheads show irregular cone cells. (c, f) Immunostaining of the pupal retinae with {alpha}-Dlg. (a–c) GMR-GAL4/+; +, (d–f) GMR-GAL4/+; UAS-FLAG-dG9a/+. (b, c, e, f) Scale bars, 10.0 µM.

 
We therefore next examined the pattern formation in pupal ommatidia of dG9a over-expressing flies (Fig. 7B b,d). Differentiation of photoreceptors, cones, and 1°, 2°, and 3° pigment cells was completed by about 42 h after pupal formation (APF) at 28 °C (Fig. 7B c). However, in pupal retinae of dG9a over-expressing flies, the staining with an antibody to Cut which marks cone cells shows that dG9a over-expressing cone cells are significantly larger than wild-type cells and that the position of cone cells appears to be irregular (Fig. 7B b,e). Furthermore, monitoring apical cell junctions as visualized anti-Discs large (Dlg) staining shows that ommatidia were not hexagonal in shape and 1° pigment cells were smaller than wild-type (Fig. 7B c,f). In contrast, the ommatidia expressing FLAG-dG9a{Delta}1532–1538 showed same differentiation pattern as wild-type (data not shown). These observations suggest that 2° and 3° pigment cells are also abnormal. These data indicated that dG9a interferes with differentiation of pupal ommatidial cell types, especially cone cells probably by repressing expression of genes involved in the differentiation process.

Pc-G mutant alleles enhance the dG9a-induced rough eye phenotype

To test whether chromatin component proteins interact with the function of dG9a, we examined genetic interactions between dG9a and other chromatin modifier genes. A Pc mutant, Pc3, strongly enhanced the rough eye phenotype induced by over-expression of dG9a (Fig. 8C,D). Other alleles of Pc, Pc1, Pc6 or Pc7, also enhanced the rough eye phenotype (Table 4). PcG and trxG proteins are renowned for their essential role in maintaining for homeotic (Hox) gene expression during development (Schuettengruber et al. 2007). To silence or activate gene expression, PcG and trxG proteins bind to specific regions of DNA and direct the histone modification respectively. PcG proteins mainly form two different classes of complexes, PRC1 and PRC2. PRC2 contains four core components: E(Z), ESC, SU(Z)12 and NURF-55. E(Z) trimethylates H3-K27 and this methylation marker is specifically recognized by the PC, a component of the PRC1 complex. It is noted that PRC1 represses transcription of particular target genes. (Schwartz et al. 2007).


Figure 8
View larger version (55K):
[in this window]
[in a new window]

  		Figure 8 The rough eye phenotype induced by dG9a is enhanced by reducing the gene dose of PcG. (A–L) Scanning electron micrographs of adult compound eyes. (A, B) GMR-GAL4/+; UAS-FLAG-dG9a/+, (C, D) GMR-GAL4/+; UAS-FLAG-dG9a/Pc3, (E, F) GMR-GAL4/ph504; UAS-FLAG-dG9a/+, (G, H) GMR-GAL4/+; E(Pc)1/+; UAS-FLAG-dG9a/+, (I) GMR-GAL4/+; +, (J) GMR-GAL4/+; Pc3//+, (K) GMR-GAL4/ph504; +, (L) GMR-GAL4/+; E(Pc)1/+. (A–D and I–L) Scale bars, 50.0 µM. (E–H) Scale bars, 14.2 µM. (M–N) Immunostaining of eye imaginal discs with {alpha}-FLAG. (M) GMR-GAL4/+; UAS-FLAG-dG9a/+, (N) GMR-GAL4/+; UAS-FLAG-dG9a/Pc3. (M, N) Scale bar, 50.0 µM.

 

View this table:
[in this window]
[in a new window]

  		Table 4  Subset of PcG genes genetically interact with dG9a
 
Next, we crossed other PcG mutants with the dG9a over-expression fly line. The polyhomeotic (ph) and Enhancer of Polycomb (E(Pc)) mutants enhanced the dG9a-induced phenotype (Fig. 8E–H). No difference in the level of FLAG-dG9a was observed between control dG9a over-expressing flies and those crossed with Pc mutant flies as shown by immunostaining with anti-FLAG antibody (Fig. 8M–N). The Pc and ph genes belong to those encoding PRC1 components. In contrast, mutants of genes encoding PRC2 components such as E (Z), Drosophila HDAC, RPD3 exerted no effect on the rough eye phenotype (Table 4). Flies heterozygous for all of these mutations exhibited an apparently normal eye morphology (Fig. 8I–L). Pc, ph, E(Pc) mutants demonstrated no effect on the expression of FLAG-dG9a (Fig. 8M,N).

To determine whether PcG over-expression suppresses the rough eye phenotype induced by over-expression of dG9a, we crossed dG9a and PcG over-expressing fly lines (Fig. S1 in Supplementary Material). Transgenic flies over-expressing PC (Dietzel et al. 1999) or PH (Martunez et al. 2006) have been successfully established. Over-expression of PC alone in eye imaginal discs exerted no apparent effect on the adult eye morphology (Supplementary Fig. S1A e,f). Co-expression of PC weakly but significantly suppressed the rough eye phenotype induced by over-expression of dG9a (compare Supplementary Fig. S1A g,h and c,d). The results are consistent with the observations with Pc mutants. We also tried to examine effect of over-expression of PH on the rough eye phenotype induced by over-expression of dG9a. However, since over-expression PH alone resulted in severe rough eye phenotype (Supplementary Fig. S1B), it was difficult to assess effect on the rough eye phenotype induced by over-expression of dG9a. In any events, some PRC1 components and E (PC) appear to play a role in negative regulation of the function of dG9a, whereas PRC2 components do not.

lid, encodes histone demethylase, interacts with dG9a genetically

We also crossed trxG mutants with dG9a over-expressing flies. Although most trxG mutants did not influence the rough eye phenotype induced by over-expression of dG9a, a lid mutant, lidk06801 effectively suppressed the rough eye phenotype (Fig. 9A–D, Table 5). Another allele of lid, lid10424 moderately suppressed the rough eye phenotype (Fig. 9E–F). No difference in the level of FLAG-dG9a in the eye imaginal discs was observed between control dG9a over-expressing flies and those crossed with lid mutant flies (Fig. 9G–H), confirming that the suppression of the rough eye phenotype is not due to the reduced expression of dG9a. Furthermore co-expression of Lid resulted in enhancement of the rough eye phenotype induced by dG9a (Fig. 9A–B,K–L), although expression of Lid alone exerted no apparent effect on eye morphology (Fig. 9I–J). These results, taken together indicate that Lid plays a role in positive regulation of the dG9a function and this regulation is independent of other trxG genes.


Figure 9
View larger version (59K):
[in this window]
[in a new window]

  		Figure 9 Half dose reduction of lid gene suppresses the rough eye phenotype induced by ectopic expression of dG9a. (A–D, E–P) Scanning electron micrographs of adult compound eyes. (A, B) GMR-GAL4/+; UAS-FLAG-dG9a/+, (C, D) GMR-GAL4/+; lid k06801/+; UAS-FLAG-dG9a/+, (E, F) GMR-GAL4/+; lid10424/+; UAS-FLAG-dG9a/+, (I, J) GMR-GAL4/+; UAS-lid/+, (K, L) GMR-GAL4/+; UAS-lid/+; UAS-FLAG-dG9a/+, (M, N) GMR-GAL4/+; UAS-lid-JmjC*/+, (O, P) GMR-GAL4/+; UAS-FLAG-dG9a/UAS-lid-JmjC*. (A, C, E, I, K, M, O) Scale bars, 50.0 µM. (B, D, F, J, L, N, P) Scale bars, 14.2 µM. (G–H) Immunostaining of eye imaginal discs with {alpha}-FLAG. (G) GMR-GAL4/+; UAS-FLAG-dG9a/+, (H) GMR-GAL4/+; lidk06801/+; UAS-FLAG-dG9a/+. (G, H) Scale bars, 50.0 µM.

 

View this table:
[in this window]
[in a new window]

  		Table 5  Subset of TrxG genes genetically interact with dG9a
 
To further examine the enhancement of dG9a-induced rough eye phenotype by co-expression of Lid is dependent on its H3-K4 demethylase activity, we crossed the catalytically inactive Lid JmjC* (Secombe et al. 2007) to dG9a over-expressing flies. The adult flies expressing Lid JmjC* showed that the bristles in the compound eye were slightly reduced in posterior region (Fig. 9M–N). Although one copy of Lid JmjC* enhanced the dG9a-induced rough eye phenotype moderately, the extent of enhancement appears to be less than that observed with one copy of wild-type Lid (compare Fig. 9L,P). Considering that two copies of Lid showed similar expression levels as that of one copy of Lid JmjC* in eye imaginal discs (Secombe et al. 2007), the demethylase activity of Lid is more required for dG9a function during eye development.

To further examine physical interaction between Lid and dG9a, we carried out the immunoprecipitation of dG9a and HA-Lid from extracts of Drosophila Kc cells over-expressing HA-Lid. However, we detected no signal indicating physical interaction between dG9a and Lid (Supplementary Fig. S2). Therefore, positive regulation of the dG9a function by Lid might be indirect as discussed below.


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Experimental procedures
 References
 
Here we demonstrated that dG9a is a euchromatic and telomere regions specific H3-K9 HMTase and that dG9a is important for the localization of HP 1 in euchromatic regions. Furthermore, we identified the genetic interactants with dG9a.

SU(VAR)3–9 is a major H3-K9 specific HMTase in Drosophila. In Su(var)3–9 null mutants, H3-K9 dimethylation and trimethylation thus disappear from the chromocentre of the salivary gland chromosome but not from the heterochromatic fourth chromosome (Schotta et al. 2002; Ebert et al. 2004). Although it has been reported that dG9a localizes in euchromatic region but not in heterochromatic centromere on polytene chromosome (Stabell et al. 2006), the localization of dG9a on the heterochromatic fourth chromosome has not been reported. As revealed by immunostaining of polytene chromosomes with anti-dG9a antibody in the present study, dG9a is exclusively associated with chromosomes on euchromatic region and no signals on the heterochromatic fourth chromosome were detected. In addition, dG9a signals were not found to overlap with Pol II signals on polytene chromosomes. Furthermore, Pol II signals were dramatically increased on polytene chromosome in dG9a RNAi fly line. These results indicate that dG9a is a HMTase playing a role in the euchromatic region as a key component of the mechanism that negatively regulates gene expression.

In addition to euchromatic regions of the polytene chromosomes, dG9a was also found to be localized on the telomere regions in the polytene chromosomes. Though HP1 binds to methylated H3-K9 in telomeres in mammals (García-Cao et al. 2004), in Drosophila, HP1 binds directly to telomeric DNA and subsequently trimethylation of H3-K9 occurs in telomere regions (Perrini et al. 2004). SU(VAR)3–9 is associated with H3-K9 methylation in the centromere heterochromatin and therefore is not responsible for H3-K9 methylation in euchromatic and telomere regions (Schotta et al. 2002). The localization of Enhancer of zeste (E(Z)), the E(z) complex, does not correlate with the few euchromatic bands of H3-K9 dimethylation, but corresponds to H3-K9 trimethylation in the euchromatic region (Czermin et al. 2002). The absence of E(Z) seems to affect only the H3-K9 trimethylation in the euchromatic localization (Perrini et al. 2004). Therefore, it appears that dG9a is involved in H3-K9 dimethylation in both euchromatic and telomere regions. In addition, some unknown HMTases other than E(Z) may contribute to H3-K9 trimethylation in the telomere regions.

HP1 binds to methylated H3-K9 by its chromodomain and establishes a silent state of chromatin (Bannister et al. 2001; Lachner et al. 2001). In Drosophila, HP1 mainly localizes in chromocenters, but is also localized in some distinct euchromatic regions (Perrini et al. 2004). In SU(VAR)3–9 deficient salivary gland nuclei, the association of HP1 to chromocentre heterochromatin was found to be strongly reduced (Schotta et al. 2002). In salivary gland nuclei of dG9a knockdown flies, HP1 localized on the euchromatic region but not on the chromocentre was strongly reduced. These results indicate that distribution of HP1 is dependent on dG9a in euchromatic regions. Since enzymatically active dG9a can induce histone H3-K9 dimethylation in euchromatic regions in vivo, H3-K9 methylation by dG9a may serve as a mark for the binding of HP1 to establish silent state of chromatin in the euchromatic regions.

To investigate the function of dG9a in vivo, we here analyzed the rough eye phenotype induced by over-expression of dG9a in eye imaginal discs. G1-S transition and the progression of S phase and the differentiation of photoreceptor cells were not affected in larval eye imaginal discs over-expressing dG9a. However, in pupae, the differentiation of ommatidial cell types was disturbed in flies over-expressing dG9a. In addition, ectopic expression of dG9a{Delta}1532–1538 carrying a mutation in the catalytic motif did not induce the rough eye phenotype. Therefore over-expression of dG9a may regulate the expression of genes involved in development of the sensory organ precursor or interommatidial bristle through ectopic H3-K9 methylation.

PcG proteins mainly form two different classes of complexes, PRC1 and PRC2. PRC2 contains E(Z) that trimethylates H3-K27 and this methylation marker is specifically recognized by the Pc, a component of the PRC1 complex. In mammals, G9a and HPC2 are components of the same CtBP complex (Shi et al. 2003). However Drosophila CtBP mutants did not demonstrate any effect on the rough eye phenotype induced by over-expression of dG9a (data not shown), suggesting that the enhancement of the dG9a-induced rough eye phenotype by Pc mutation is not mediated by the CtBP complex. Enhancement of the dG9a-induced rough eye phenotype was however observed with Pc, ph, E(Pc) mutants. Most of these gene products are components of the PRC1. We therefore predict that some PRC1 components, including PC, act as negative regulators of dG9a function. Because E(z) mutant did not affect on the rough eye phenotype, this negative regulation seems to be independent of the H3-K27 methylation. Therefore, dG9a may interact with PRC1 that has not been recruited to the H3-K27 locus and this interaction may inhibit enzyme activity of dG9a.

The trxG protein Lid is a JmjC domain-containing trimethyl H3-K4 demethylase (Eissenberg et al. 2007; Lee et al. 2007; Secombe et al. 2007). Its mammalian orthologue, SMCX/JARID1C is a component of the transcription repressor REST complex including G9a (Tahiliani et al. 2007). Based on our finding of suppression of the dG9a-induced rough eye phenotype by lid mutation and enhancement of the rough eye phenotype by over-expression of Lid, we predict that Lid–G9a function may conserved between mammals and Drosophila. To examine interaction between Lid and dG9a, we tried immunoprecipitation of dG9a and HA-Lid from Drosophila Kc cell extracts. However, we could not obtain any data indicating physical interaction between dG9a and Lid (Supplementary Fig. S2). Therefore in Drosophila dG9a may not physically interact with Lid. It has been reported that SU(VAR)3–3, human LSD1 amine oxidase orthologue, demethylates di- and trimethyl H3-K4 (Di Stefano et al. 2007; Rudolph et al. 2007). Interestingly, loss of Su(var)3–3 prevented extension of dimethyl H3-K9 at pericentric heterochromatin (Rudolph et al. 2007). Therefore, elevated di and trimethyl H3-K4 by the reduction of Lid may prevent the expansion of dimethyl H3-K9 catalyzed by dG9a and consequently suppressed the dG9a-induced rough eye phenotype.


   Experimental procedures
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
References
 
Plasmid construction

A full-length dG9a cDNA in the vector PGVB-FLAG (Toyo ink.) was obtained by RT-PCR and PCR from EST cone (ResGen, clone ID: GM19374). The BglII/NotI-digested 250 bp cDNA fragment encoding amino acids (aa) 2–87 which had been amplified from PGVB-FLAG-dG9a2–1637 using primers 5'-AGAT CTACGGACTTTGTTGAGCTGATGA and 5'-TTACGAG GTACCTTGACAGCGGCCGCAT was ligated into BglII/NotI sites of pUAST-FLAG to create pUAS-FLAG-dG9a2–87. PGVB-FLAG-dG9a2–1637 was digested with NotI and SpeI, and the isolated fragment was inserted into NotI/XbaI sites of pUAS-FLAG-dG9a2–87 to create pUAS-FLAG-dG9a.

The cDNA fragment encoding aa1334–1637 which had been amplified from PGVB-FLAG-dG9a2–1637 using primers 5'-CATTCCGAATGAGGAGTCTGAG and 5'-AGATCTTC GACAACATCTGCCAAAAGTG was subcloned into the pT7 Blue-2 vector (Novagen) to create pT7-dG9a1334–1637. Site-directed mutagenesis of pT7-dG9a1334–1637 to delete aa1532–1538 was carried out using the Quik Change Site-Directed Mutagenesis Kit according to the manufacturer's instructions (Stratagene). pT7-dG9a1334–1637 was digested with AseI/SpeI and cloned into AseI/SpeI sites of PGVB-FLAG-dG9a to create PGVB-FLAG-dG9a{Delta}1532–1538. This was then digested with NotI and SpeI then the isolated fragment was inserted into NotI/XbaI sites of pUAS-FLAG-dG9a2–87 to create pUAS-FLAG-dG9a{Delta}1532–1538.

A 0.6-kb region from the N-terminus of dG9a containing the non-coding region was PCR-amplified from the EST clone GM19374 using primers 5'-GCAGTCGACTTCGCTGCCGC and 5'-AGATCTTCGACAACATCTGCCAAAAGTG, and the amplified fragment was cloned into the pT7 Blue-2 vector to create pT7–5'IR. XhoI/KpnI-digested cDNA fragments from pT7–5'IR were ligated into XhoI/KpnI sites of pUAST to create pUAS-5'IR. Then EcoRI/BglII-digested cDNA fragments from pT7–5'IR were ligated into EcoRI/BglII sites of pUAS-5'IR to create pUAS-5'IR-dG9a (tail to tail). A 0.5-kb cDNA fragment containing the C-terminal region of dG9a was amplified from pT7-dG9a1334–1637 using primers 5'-AGATCTTGCAGAATGG TACACGGACAC and 5'-TCGTAAACTAGTCTACGCGT GTCCAATTTTCTCC. Then pUAS-3'IR-dG9a (head to head) was created in the same way.

pGST-dG9a1–465 was generated by subcloning an EcoRI/XhoI fragment of dG9a encoding aa1–436 into pGEX-4T-1 (GE Healthcare). pGST-dG9a1314–1647 was generated by subcloning an SaII/NotI fragment of dG9a encoding aa1314–1647 into pGEX-4T-1.

pAct-HA-Lid was generated by cloning an ApaI/SpeI fragment of lid encoding aa2–1838 into pActSV-HA.

Expression of GST-dG9a1–465 fusion protein in Escherichia coli and production of anti-dG9a antibodies

The GST-dG9a1–465 fusion protein was expressed in E. coli BL21 (DE3) and lysates of cells were prepared by sonication in PBS containing 0.1% Triton X-100, 2 µg/mL pepstatin A, 10 µg/mL leupeptin, 10 µg/mL aprotinin and 100 mM PMSF and applied to Glutathione Sepharose 4B (GE Healthcare) to purify the GST-dG9a1–465 fusion protein as described previously (Yamaguchi et al. 1995). The purified GST-dG9a1–465 fusion protein and the purified GST protein were used to elicit polyclonal antibody production in rabbits. Polyclonal antibodies reacting with dG9a were affinity-purified with GST-dG9a-conjugated CNBr-activated Sepharose 4B (GE Healthcare) after passing through GST-conjugated Sepharose.

Western immunoblot analysis

To perform Western immunoblot analysis, third instar larvae were homogenized in SDS sample buffer and heated at 95 °C for 10 min and insolubles were pelletted by centrifugation for 10 min at 4 °C. Drosophila S2 cells were grown in M3 medium (Sigma) containing 10% fetal calf serum at 25 °C in 5% CO2 atmosphere. 2 x 106 cells were transfected with pAct5 c-GAL4 and pUAS-FLAG-dG9a using Cell-fectin reagent (Invitrogen). At 48 h thereafter, cell lysates were prepared with TNE buffer (10 mM Tris–HCl [pH7.5], 150 mM NaCl, 1 mM EDTA, 1% NP-40, 2 µg/mL pepstatin A, 10 µg/mL leupeptin, 10 µg/mL aprotinin and 100 mM PMSF). Protein samples were fractionated by SDS-PAGE and transferred to PVDF membranes (Bio-Rad). The membranes were blocked with 5% dry milk, and incubated with control rabbit IgG (0.691 µg/mL) (Sigma), {alpha}-dG9a antibody (0.691 µg/mL), {alpha}-FLAG M2 (2 µg/mL) (Sigma), {alpha}-HP1 C1A9 (1/500) (a gift from S. Elgin) and {alpha}-{alpha}-tubulin (1/5000) (Sigma) at 4 °C for 16 h. HRP-conjugated anti-rabbit and anti-mouse IgGs (GE Healthcare) were used as secondary antibodies. Visualization was carried out using ECL or ECL-plus (GE Healthcare).

Immunoprecipitation

Drosophila Kc cells were transfected with a mixture of plasmids. Cells were lysed with TNE buffer containing a mixture of protease inhibitors (Nacalai tesque). Proteins in cell extracts equivalent to 1.5 mg were added to Protein A Sepharose beads (GE healthcare Life Science) and 2 µg of {alpha}-dG9a IgG or 10 µL culture supernatant of hybridoma cells-producing mouse {alpha}-HA (supplied by M. Inagaki). Normal rabbit IgG (Sigma) was used as a negative control. Samples were rotated for 16 h at 4 °C. Sepharose beads were washed three times with the wash buffer (10 mM Tris [pH7.5], 100 mM NaCl, 0.1% Nonidet P-40) and analyzed by Western blot using {alpha}-dG9a (1/1000), and culture supernatant of hybridoma cells-producing mouse {alpha}-HA (1/500). Visualization was carried out using ECL (GE healthcare Life Science).

Immunostaining of whole mount salivary glands and polytene chromosomes

Preparation of polytene chromosomes was performed as described previously (Schotta et al. 2002). The remaining steps tool place as described previously (Kato et al. 2007). The primary antibodies used in this study were {alpha}-HP1 C1A9 (1/100) (a gift from S. Elgin), {alpha}-dG9a (1/200) and {alpha}-RNA pol II H14 (1/200) (Covance). The secondary antibodies were Alexa Fluor 488 anti-mouse IgM (Molecular Probes), Alexa Fluor 488 anti-mouse IgG (Molecular Probes) or Alexa Fluor 594 anti-rabbit IgG (Molecular Probes). Preparations were mounted with FluoroGuard Antifade Reagent (Bio-Rad) and examined using a fluorescence microscope (Olympus BX-50) equipped with a cooled CCD camera (Hamamatsu Photo ORCA-ER) and the Aquacosmos image analysis software (Hamamatsu Photo ORCA-ER).

For immunostaining of whole mount salivary glands, these were prepared as described previously (Schotta et al. 2002) with the following modifications: third instar larvae were dissected in PBS containing 0.5% NP-40 and the salivary glands were fixed for 30 min in 4% paraformaldehyde/1% Triton X-100 at 25 °C. After permeabilizing with 1% Triton X-100 in PBS, the samples were blocked for 20 min with 1% goat serum/ 0.1% Triton X-100 in PBS and then incubated for 16 h at 4 °C with {alpha}-dG9a (1/400), {alpha}-H3-K9me2 (1/250) (Upstate Biotechnology 07–212), {alpha}-FLAG M2 (5 µg/mL) (Sigma) or {alpha}-HP1 C1A9 (1/100) (a gift from S. Elgin). The samples were then incubated for 2 h at 25 °C with Alexa Fluor 488 anti-mouse IgG, Alexa Fluor 488 anti-rabbit IgG (Molecular Probes) or Alexa Fluor 594 anti-rabbit IgG. DNA was stained with 4', 6-diamidino-2-phenylindole (DAPI) or propidium iodide (PI). Preparations were examined using a fluorescence microscope or a confocal laser scanning microscopy (Carl Zeiss LSM510).

BrdU labeling

To detect cells in S phase, a BrdU labeling method was applied as described previously with minor modifications (Otsuki et al. 2004) with the following modifications: the dissected imaginal discs were incubated in Grace's insect medium (Sigma), and in the presence of 75 µg/mL BrdU (Roche) for 1 h at 25 °C. Incorporated BrdU was detected using {alpha}-BrdU (Roche) and Alexa Fluor 488 anti-mouse IgG and examined using a fluorescence microscopy.

Immunostaining of eye imaginal discs and pupal retinae

Third instar larval imaginal discs were dissected in 0.7% NaCl and fixed for 20 min in 4% paraformaldehyde. For immunofluorescence analysis with {alpha}-Caspase-3, dissected imaginal discs were fixed for 10 min in 2% paraformaldehyde. Pupae were dissected in PBS at 42 h APF and fixed for 20 min in 4% paraformaldehyde. The eye imaginal discs and pupal retinae were permeabilized in PBS containing 0.3% Tritone X-100 (PBST). The remaining steps took place as described above. The following primary antibodies were used: {alpha}-FLAG M2 (5 µg/mL) (Sigma), {alpha}-cleaved Caspase-3 (Asp 175) (1:200) (Cell Signaling Technology), {alpha}-Dlg (1/500) (Developmental Studies Hybridoma Bank [DSHB] 4F3), {alpha}-Elav (1/500) (DSHB 7E8A10), {alpha}-Cut (1/500) (DSHB 2B10), {alpha}–β-Gal (1/500) (Promega), and Alexa Fluor 488 conjugated phalloidin (1/100) (Molecular Probes). Secondary antibodies were Alexa Fluor 488 anti-mouse IgG, Alexa Fluor 488 anti-rat IgG, Alexa Fluor 546 anti-mouse IgG, and Alexa Fluor 594 anti-rabbit IgG. Preparations were examined using a confocal laser scanning microscopy and a fluorescence microscopy.

Fly stocks

Fly stocks were cultured at 25 °C on standard food. Canton S was used as the wild-type strain. The engrailed-GAL4 (en-GAL4) fly stock was kindly provided by N. Dyson and the Pc-G mutant flies by K. Ohno and V. Pirrotta. UAS-lid and UAS-lid JmjC* flies were kindly provided by J. Secombe and R. Eisenman. UAS-Pc-GFP fly and UAS-ph transgenic lines were kindly provided by R. Paro and F. Maschat, respectively. Flies bearing GMR-DIAP1 and GMR-DIAP2 transgenes were provided by G. Rubin. Enhancer trap lines carrying the lacZ markers BB02 and AE127 were obtained from Y. Hiromi. The UAS-Lid was obtained from Establishment of lines carrying GMR-GAL4 was as described earlier (Robertson et al. 1988; Takahashi et al. 1999). All other stocks used in this study were obtained from the Bloomington Drosophila stock center and the Drosophila Genetic Resource Center in Kyoto.

Establishment of transgenic flies

P-element-mediated germ line transformation was carried out as described previously (Spradling et al. 1986). F1 transformants were selected on the basis of white-eye color rescue (Robertson et al. 1988) and 9 lines were established for the pUAS-FLAG-dG9a. Established transgenic strains carrying UAS-FLAG-dG9a constructs and their chromosomal linkages are listed in Table 1. We used line 47 carrying pUAS-FLAG-dG9a on chromosome III in the detailed studies. Ten lines were established for the pUAS-FLAG-dG9a as also listed in Table 1. We used line 100 carrying pUAS-FLAG-dG9a{Delta}1532–1538 on chromosome III in the detailed studies. Fifteen lines were established for pUAS-5'IR-dG9a and four for the pUAS-3'IR-dG9a, as listed in Table 2. We used line 61 carrying UAS-5'IR-dG9a on chromosome II in the detailed studies.

Scanning electron microscopy

Adult flies were anesthetized, mounted and observed under a scanning electron microscope (Keyenece VE-7800).

Total RNA preparation and semi-quantitative RT-PCR

0–2 h embryos and third instar larvae were collected and total RNA was isolated with TRIzol (Invitrogen) and treated with DNase I. For RT-PCR, mRNA was purified using an Oligotex-dT30 <Super> mRNA Purification kit (Takara). First cDNA was synthesized using oligo(dT) and Bca PLUS RTase (TaKaRa). The following primer sequences were used to amplify the genes: for dG9a 5'-CATTCCGAATGAGGAGTCT-3' and 5'-CTACGC GTGTCCAATTTTCTCC-3', for rp49 5'-ATGGCAACAAGC GCCAAACTG-3' and 5'-TCAGAAGCCCTTCTCCTTGAG-3'.


   Acknowledgements
 
We thank M. Moore and S. Cotterill for comments on the English language in the manuscript. We are grateful to J. Secombe and R. Eisenman for a full-length Lid cDNA, S. Elgin for anti-HP1 monoclonal antibody (C1A9), M. Inagaki for culture supernatant of hybridoma cells-producing mouse anti-HA monoclonal antibody and R. Rubin, Y. Hiromi, N. Dyson, J. Secombe and R. Eisenman, K. Ohno and V. Pirrotta, F. Maschat, R. Paro for providing fly stocks. We acknowledge all the members of the Yamaguchi laboratory for helpful discussion and advice. This study was partially supported by grants from the Ministry of Education, Culture, Sports, Science and Technology of Japan.


   Footnotes
 
Communicated by: Fumio Hanaoka

aPresent address: Developmental Biology, Stanford School of Medicine, 279 Campus Dr B371, Stanford, CA 94305, USA. Back

* Correspondence: Email: myamaguc{at}kit.ac.jp


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Experimental procedures
 References
 
Bannister, A.J., Zegerman, P., Partridge, J.F., Miska, E.A., Thomas, J.O., Allshire, R.C. & Kouzarides, T. (2001) Selective recognition of methylated lysine 9 on histone H3 by the HP1 chromo domain. Nature 410, 120–124.[CrossRef][Medline]

Clough, E., Moon, W., Wang, S., Smith, K. & Hazelrigg, T. (2006) Histone methylation is required for oogenesis in Drosophila. Development 134, 157–165.[CrossRef]

Crook, N.E, Clem, R.J. & Miller L.K. (1995) An apoptosis-inhibiting baculovirus gene with a zinc finger-like motif. J. Virol. 67, 2168–2174.

Czermin, B., Shotta, G., Hülsmann, B.B., Brehm, A., Becker, P.B., Reuter, G. & Imhof, A. (2001) Physical and functional association of SU(VAR)3–9 and HDAC1 in Drosophila. EMBO Rep. 2, 915–919.[CrossRef][Medline]

Czermin, B., Melfi, R., McCabe, D., Seitz, V., Imhof, A. & Pirrotta, V. (2002) Drosophila Enhancer of Zeste/ESC complexes have a histone H3 methyltransferase activity that marks chromosomal polycomb sites. Cell 111, 185–196.[CrossRef][Medline]

Di Stefano, L., Ji, J.Y., Moon, N.S., Herr, A. & Dyson, N. (2007) Mutation of Drosophila Lsd1 disrupts H3-K4 methylation, resulting in tissue-specific defects during development. Curr. Biol. 17, 808–812.[CrossRef][Medline]

Dietzel, A., Niemann, H., Brückner, B., Maurange, C. & Paro, R. (1999) The nuclear distribution of Polycomb during Drosophila melanogaster development shown with a GFP fusion protein. Chromosoma 108, 83–94.[CrossRef][Medline]

Ebert, A., Schotta, G., Lein, S., Kubicek, S., Krauss, V., Jenuwein, T. & Reuter, G. (2004) Su(var) genes regulate the balance between euchromatin and heterochromatin in Drosophila. Genes Dev. 18, 2973–2983.[Abstract/Free Full Text]

Eissenberg, J.C., Lee, M.G., Schneider, J., Ilvarsonn, A., Shiekhattar, R. & Shilatifard, A. (2007) The trithorax-group gene in Drosophila little imaginal discs encodes a trimethylated histone H3 Lys4 demethylase. Nat. Struct. Mol. Biol. 14, 344–346.[CrossRef][Medline]

Fischle, W., Wang, Y. & Allis, C.D. (2003) Extending the histone code: modification cassettes and switches. Nature 425, 475–479.[CrossRef][Medline]

García-Cao, M., O'Sullivan, R., Peters, A.H., Jenuwein, T. & Blasco, M.A. (2004) Epigenetic regulation of telomere length in mammalian cells by the Suv39h1 and Suv39h2 histone methyltransferases. Nat. Genet. 36, 94–99.[CrossRef][Medline]

Gyory, I., Wu, J., Fejér, G., Seto, E. & Wright, K.L. (2004) PRDI-BF1 recruits the histone H3 methyltransferase G9a in transcriptional silencing. Nat. Immunol. 5, 299–308.[CrossRef][Medline]

Hay, B.A., Wassarman, D. & Rubin, G.M. (1995) Drosophila homologs of baculovirus inhibitor of apoptosis proteins function to block cell death. Cell 83, 1253–1262.[CrossRef][Medline]

James, T.C., Eissenberg, J.C., Craig, C., Dietrich, V., Hobson, A. & Elgin, S.C. (1989) Distribution patterns of HP1, a heterochromatin-associated non histone chromosomal protein of Drosophila. Eur. J. Cell Biol. 50, 170–180.[Medline]

Kato, M., Kato, Y., Nishida, M., Hayakawa, T., Haraguchi, T., Hiraoka, Y., Inoue H.Y. & Yamaguchi, M. (2007) Functional domain analysis of human HP1 isoforms in Drosophila. Cell Struct. Funct. 32, 57–67.[CrossRef][Medline]

Komarnitsky, P., Cho, E.J. & Buratowski, S. (2000) Different phosphorylated forms of RNA polymerase II and associated mRNA processing factors during transcription. Genes Dev. 14, 2452–2460.[Abstract/Free Full Text]

Lachner, M., O’Carroll, C.D., Rea, S., Mechtler, K. & Jenuwein, T. (2001) Methylation of histone H3 lysine 9 creates a binding site for HP1 proteins. Nature 410, 116–120.[CrossRef][Medline]

Lachner, M., O'Sullivan, R.J. & Jenuwein, T. (2003) An epigenetic road map for histone lysine methylation. J. Cell Sci. 16, 2117–2124.

Lee, D.Y., Northrop, J.P., Kuo, M.H. & Stallcup, M.R. (2006) Histone H3 lysine 9 methyltransferase G9a is a transcriptional coactivator for nuclear receptors. J. Biol. Chem. 281, 8476–8485.[Abstract/Free Full Text]

Lee, N., Zhang, J., Klose, R.J., Erdjument-Bromage, H., Tempst, P., Jones, R.S. & Zhang, Y. (2007) The trithorax-group protein Lid is a histone H3 trimethyl-Lys4 demethylase. Nat. Struct. Mol. Biol. 14, 341–343.[CrossRef][Medline]

van Leeuwen, F., Gafken, P.R. & Gottschling, D.E. (2002) Dot1p modulates in yeast by methylation of the nucleosome core. Cell 109, 745–756.[CrossRef][Medline]

Martunez, A.M., Colomb, S., Déjardin, J., Bantignies, F. & Cavalli, G. (2006) Polycomb group-dependent Cycline A repression in Drosophila. Genes Dev. 20, 501–513.[Abstract/Free Full Text]

Mis, J., Ner, S.S. & Grigliatti, T.A. (2006) Identification of three histone methyltransferases in Drosophila: dG9a is a suppressor of PEV and is required for gene silencing. Mol. Genet. Genomics 275, 513–526.[CrossRef][Medline]

Mlodzik, M., Baker, N.E. & Rubin, G.M. (1990a) Isolation and expression of scabrous, a gene regulating neurogenesis in Drosophila. Genes Dev. 4, 1848–1861.[Abstract/Free Full Text]

Mlodzik, M., Hiromi, Y., Weber, U., Goodman, C.S. & Rubin, G.M. (1990b) The Drosophila seven-up gene, a member of the steroid receptor gene superfamily, controls photoreceptor cell fates. Cell 60, 211–224.[CrossRef][Medline]

Moses, K. & Rubin, G.M. (1991) Glass encodes a site-specific DNA binding protein that is regulated in response to positional signals in the developing Drosophila eye. Genes Dev. 5, 583–593.[Abstract/Free Full Text]

Nishida, M., Kato, M., Kato, Y., Sasai, N., Ueda, J., Tachibana, M., Shinkai, Y. & Yamaguchi, M. (2007) Identification of ZNF200 as a novel binding partner of histone H3 methyltransferase G9a. Genes Cells 12, 877–888.[Abstract/Free Full Text]

Nishio, H. & Walsh, M.J. (2004) CCAAT displacement protein/cut homolog recruits G9a histone lysine methyltransferase to repress transcription. Proc. Natl. Acad. Sci. USA 101, 11257–11262.[Abstract/Free Full Text]

Ogawa, H., Ishiguro, K., Gaubatz, S., Livingston, D.M. & Nakatani, Y. (2002) A complex with chromatin modifiers that occupies E2F- and Myc-responsive genes in G0 cells. Science 296, 1132–1136.[Abstract/Free Full Text]

Otsuki, K., Hayashi, Y., Kato, M., Yoshida, H. & Yamaguchi, M. (2004) Characterization of dRFX2, a novel FRX family protein in Drosophila. Nucleic Acids Res. 32, 5636–5648.[Abstract/Free Full Text]

Perrini, B., Piacentini, L., Fanti, L., Altieri, F., Chichiarelli, S., Berloco, M., Turano, C., Ferraro, A. & Pimpinelli, S. (2004) HP1 controls telomere capping, telomere elongation, and telomere silencing by two different mechanisms in Drosophila. Mol. Cell 15, 467–476.[CrossRef][Medline]

Rea, S., Eisenhaber, F., O’Carroll, D., Strahl, B.D., Sun, Z.W., Schmid, M., Opravil S, Mechtler, K., Ponting, C.P., Allis, C.D. & Jenuwein, T. (2000) Regulation of chromatin structure by site-specific histone H3 methyltransferases. Nature 406, 593–599.[CrossRef][Medline]

Robertson, H.M., Preston, C.R., Phillis, R.W., Johnson-Schlitz, D.M., Benz, W.K. & Engels, W.R. (1988) A stable genomic source of P-element transpose in Drosophila melanogaster. Genetics 118, 461–470.[Abstract/Free Full Text]

Roopra, A., Qazi, R., Schoenike, B., Daley, T.J. & Morrison, J.F. (2004) Localized domains of G9a-mediated histone methylation are required for silencing of neuronal genes. Mol. Cell 14, 727–738.[CrossRef][Medline]

Rudolph, T., Yonezawa, M., Lein, S, Heidrich, K., Kubicek, S., Schäfer, C., Phalke, S., Walther, M., Schmidt, A., Jenuwein, T. & Reuter G. (2007) Heterochromatin formation in Drosophila is initiated through active removal of H3K4 methylation by the LSD1 homolog SU(VAR)3–3. Mol. Cell 26, 103–115.[CrossRef][Medline]

Schotta, G, Ebert, A., Krauss, V., Fischer, A., Hoffmann, J., Rea, S., Jenuwein, T., Dorn, R. & Reuter, G. (2002) Central role of Drosophila SU(VAR)3–9 in histone H3-K9 methylation and heterochromatic gene silencing. EMBO J. 21, 1121–1131.[CrossRef][Medline]

Schuettengruber, B., Chourrout, D., Vervoort, M., Leblanc, B. & Cavalli, G. (2007) Genome regulation by polycomb and trithorax proteins. Cell 128, 735–745.[CrossRef][Medline]

Schwartz, Y.B. & Pirrotta, V. Polycomb silencing mechanisms and the management of genomic programmes. Nat. Rev. Genet. 8, 9–22.

Secombe, J., Li, L., Carlos, L. & Eisenman, R.N. (2007) The trithorax group protein Lid is a trimethyl histone H3K4 demethylase required for dMyc-induced cell growth. Genes Dev. 21, 537–551.[Abstract/Free Full Text]

Seum, C., Reo, E., Peng, H., Rauscher, F.J. 3rd, Spierer, P. & Bontron, S. (2007) Drosophila SETDB1 is required for chromosome 4 silencing. PLoS Genet. 3, E76.[CrossRef][Medline]

Shi, Y., Sawada, J., Sui, G., Affar, el B., Whetstine, J.R., Lan, F., Ogawa, H., Luke, M.P., Nakatani, Y. & Shi, Y. (2003) Coordinated histone modifications mediated by CtBP co-repressor complex. Nature 422, 735–738.[CrossRef][Medline]

Spradling, A.C. (1986) P-element-mediated transformation. In: Drosophila: A Practical Approach (ed. D.B. Roberts), pp. 175–197, Oxford: IRL Press.

Stabell, M., Eskeland, R., Bjørkmo, M., Larsson, J., Aalen, R.B., Imhof, A. & Lambertsson, A. (2006) The Drosophila G9a gene encodes a multi-catalytic histone methyltransferase required for normal development. Nucleic Acids Res. 34, 4609–4621.[Abstract/Free Full Text]

Tachibana, M., Sugimoto, K., Fukushima, T. & Shinkai, Y. (2001) SET Domain-containing protein, G9a, is a novel lysine-preferring specific selectivity to lysines 9 and 27 of histone H3. J. Biol. Chem. 276, 25309–25317.[Abstract/Free Full Text]

Tachibana, M., Sugimoto, K., Nozaki, M., Ueda, J., Ohta, T., Ohki, M., Fukuda, M., Takeda, N., Niida, H., Kato, H. & Shinkai, Y. (2002) G9a histone methyltransferase plays a dominant role in euchromatic histone H3 lysine 9 methylation and is essential for early embryogenesis. Genes Dev. 16, 1779–1791.[Abstract/Free Full Text]

Tahiliani, M., Mei, P., Fang, R., Leonor, T., Rutenberg, M., Shimizu, F., Li, J., Rao, A. & Shi, Y. (2007) The histone H3K4 demethylase SMCX links REST target genes to X-linked mental retardation. Nature 447, 601–605.[CrossRef][Medline]

Takahashi, Y., Hirose, F., Matsukage, A. & Yamaguchi, M. (1999) Identification of three conserved regions in the DREF transcription factors from Drosophila melanogaster and Drosophila virilis. Nucleic Acids Res. 27, 510–516.[Abstract/Free Full Text]

Tzeng, T.Y., Lee, C.H., Chan, L.W. & Shen, C.K. (2007) Epigenetic regulation of the Drosophila chromosome 4 by the histone H3K9 methyltransferase dSETDB1. Proc. Natl. Acad. Sci. USA 104, 12691–12696.[Abstract/Free Full Text]

Ueda, J., Tachibana, M., Ikura, T. & Shinkai, Y. (2006) Zinc finger protein Wiz links G9a/GLP histone methyltransferases to the co-repressor molecule CtBP. J. Biol. Chem. 281, 20120–20128.[Abstract/Free Full Text]

Yamaguchi, M, Hirose, F., Nishimoto, Y., Naruge, T., Ikeda, M., Hachiya, T., Tamai, K., Kuroda, K. & Matsukage, A. (1995) Expression patterns of DNA replication enzymes and the regulatory factor DREF during Drosophila development analyzed with specific antibodies. Biol. Cell 85, 147–155.[CrossRef][Medline]

Yamaguchi, M., Hirose, F., Inoue, Y.H., Shiraki, M., Hayashi, Y., Nishi, N. & Matsukage, A. (1999) Ectopic expression of human p53 inhibits entry into S phase and induces apoptosis in the Drosophila eye imaginal discs. Oncogene 18, 6767–6775.[CrossRef][Medline]

Yoshida, H., Kwon, E., Hirose, F., Otsuki, K., Yamada, M. & Yamaguchi, M. (2004) DREF is required for EGFR signalling during Drosophila wing vein development. Genes Cells 9, 935–944.[Abstract/Free Full Text]

Received: 15 October 2007
Accepted: 2 April 2008




This article has been cited by other articles:


Home page
 GeneticsHome page
B. Brower-Toland, N. C. Riddle, H. Jiang, K. L. Huisinga, and S. C. R. Elgin
Multiple SET Methyltransferases Are Required to Maintain Normal Heterochromatin Domains in the Genome of Drosophila melanogaster
Genetics, April 1, 2009; 181(4): 1303 - 1319.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kato, Y.
Right arrow Articles by Yamaguchi, M.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Kato, Y.
Right arrow Articles by Yamaguchi, M.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE ADVANCED SEARCH TABLE OF CONTENTS