QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online December 12, 2006
doi: 10.1242/10.1242/dev.02687

This Article Summary Full Text Full Text (PDF) Supplementary Material Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Anne, J. Articles by Mechler, B. M. Search for Related Content PubMed PubMed Citation Articles by Anne, J. Articles by Mechler, B. M.

Arginine methyltransferase Capsuléen is essential for methylation of spliceosomal Sm proteins and germ cell formation in Drosophila

¹ Department of Developmental Genetics, Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany.
² Laboratoire de Biologie Moléculaire de la Drosophile, Département de Biologie Moléculaire, Institut Pasteur, Paris F-75015, France.
³ Developmental Biology Unit, European Molecular Biology Laboratory, Heidelberg D-69117, Germany.

View larger version (38K): [in this window] [in a new window]   Fig. 1. Embryonic phenotype of csulP. (A-D) The csul gene belongs to the `posterior-grandchildless' group. Immunohistochemical detection (horseradish peroxidase) of Vas protein (A,B) and cuticle preparations of first instar larvae (C,D) in wild-type (A,C) and csulP (B,D). Anterior is to left. Pole cell formation and abdominal patterning are defective in csulP. (E-H) csul activity is required for the localization of maternal determinants. Distribution of nos (E,F) and gcl (G,H) transcripts detected by whole-mount in situ hybridization in wild-type (E,G) and csulP (F,H) embryos. In csulP the accumulation of nos mRNA is reduced at the posterior pole (F), whereas gcl mRNA is evenly distributed (H). (I-L) csulP suppresses the bicaudal phenotype induced by osk-bcd3'UTR. Immunohistochemical staining of Vas protein (I,J) and cuticle preparations of first instar larvae (K,L) in osk-bcd3'UTR (I,K) and csulP; osk-bcd3'UTR (J,L). Although the cuticle has a normal polarity in csulP; osk-bcd3'UTR, the head is malformed, indicating that the embryo contains a residual amount of nos activity at the anterior pole. (M-P) Distribution of Osk protein in wild-type (M,O) and csulP (N,P) embryos. Immunofluorescent detection of Osk (red). DNA detected using Oli-Green (green). After nuclear cycle 6, the level of Osk is strongly reduced at the posterior pole in csulP embryos (P). In older csulP embryos (insert in P) trace amounts of Osk can be detected at the posterior pole. (Q-T) Distribution of Vas protein in wild-type (Q,S) and csulP (R,T) embryos. Immunofluorescent detection of Vas (red); DNA (green). Similar to Osk, Vas staining is strongly reduced in the pole plasm of csulP embryos (T) and is rarely detected at the posterior pole of syncytial embryos (insert in T). (U,V) Distribution of Tud protein in wild-type (U) and csulP (V) embryos. Immunofluorescent detection of Tud (red); DNA (green). No localized Tud staining is observed in csulP embryos.

View larger version (70K): [in this window] [in a new window]   Fig. 2. csul activity contributes to nuage and pole plasm assembly during oogenesis. Distribution of Osk (A,B), GFP-Vas (C,D,G,H), Tud (E,F,I,J), and Mael (K,L) proteins in stage 10 egg chambers (A-F) and previtellogenic egg chambers (G-L). Left and right columns show wild-type and csulP egg chambers, respectively. (A,B) Osk immunostaining (red) and DNA (green). Osk is correctly synthesized and positioned at the posterior pole of csulP oocytes. (C,D,G,H) GFP-Vas (green) and DNA staining (red). GFP-Vas is only detected in the nuage of previtellogenic csulP egg chambers and disappears in later stages of egg chamber development. In comparison with wild type, the amount of GFP-Vas is markedly reduced in the pole plasm of csulP stage 10 oocytes. (E,F,I,J) Tud immunostaining (red) and DNA (green). Although Tud is synthesized in csulP egg chambers, it is absent both from nuage and pole plasm. (K,L) Mael immunostaining. Mael localization in the nuage is abolished in csulP egg chambers. o, oocyte; n, nurse cell nucleus.

View larger version (75K): [in this window] [in a new window]   Fig. 3. Molecular analysis of the csul locus. (A) Restriction map of the genomic DNA covering the csul locus with the location of the P element insertion in csulP. Below the map are shown three transcription units in this locus and their orientation, as well as the three genomic fragments used for constructing transgenes. Transcription orientation of the fidipidine (), csul (ß), and Kinesin heavy chain (Khc) genes is indicated by arrows. The analysis of complementation of the csulP mutation by transgenes shown on the right demonstrates that the ß gene corresponds to csul. (B) Restriction map of the csul gene, with alignment of the csul transcript. Exons are drawn as boxes and the putative open reading frame is indicated in black. (B) BamHI; (P) PstI; (R) EcoRI; (Xb) XbaI; (Xh) XhoI. (C) Northern blot showing a single ovarian csul transcript of 2 kb. RNA marker sizes (in kb) are indicated on the left. (D) Western blot detection of Csul in wild-type and csulP ovarian protein extracts using anti-Csul antibodies. A band of 65 kDa (arrow) is detected in wild-type but not in csulP. Protein load in each well was normalized by using anti-actin antibodies (data not shown). The anti-Csul antibodies recognize another protein of 38 kDa unrelated to Csul. (E) Alignment of the amino acid sequences of Csul and homologous proteins performed using the Pileup program of the Wisconsin Package (Genetics Computer Group) reveals that csul encodes a protein-arginine methyltransferase similar to human PMRT5. Gaps in the amino acid sequence, indicated by dots, were introduced for optimal alignment. At each position of the alignment, residues identical in all sequences are background-shaded blue, and functionally conserved (i.e. more than half of the amino acids of a column) residues with strong or weak similarities are background-shaded red and orange, respectively. The asterisks over the positions 243 and 244 indicate the substitutions made in the csulG343A;R344L transgene. GenBank accession number sequences are as follows: CP4855_AG: EAA14767; PRMT5_HS: AF167572; zC14A17.2_DR: CAD60861; F6E21.40_AT: T10666; Skb1_SP: U59684; YLB5_CE: U10402; B208.200_NC: T49572; Hsl7p_SC: U65920.

View larger version (64K): [in this window] [in a new window]   Fig. 4. Symmetrical methylation of SmB and SmD3 proteins is dependent on Csul and Vls. (A) Csul and Vls are required for sDMA synthesis. Protein extracts of wild-type (WT), csulRM, vls1, osk84/pXT103, vasQ7, vasO11 and aubHN2/aubN11 ovaries were separated by SDS-PAGE, blotted and probed with -SYM10 (left panel) and -ASYM24 antibodies (right panel). Anti-ribosomal p40 antibodies (-p40) were used as a loading control (left panel, lower blot). The intensity of four SYM10-reactive protein bands in the range of 15 to 26 kDa was strongly reduced in csulRM and vls1 protein extracts. No change of the intensity of the ASYM24-reactive bands was observed in any mutant background. [B] and [C] indicate the position of protein bands shown in B and C, respectively. (B) SmB is symmetrically dimethylated in vivo. Ovarian protein extracts of wild-type and SmBBG02775/CyO females were separated by SDS-PAGE, blotted and probed with -SYM10 and -Y12 antibodies. By comparison to wild type, the intensity of the SYM10/Y12-reactive protein band is significantly reduced in SmBBG02775/CyO. (C) SmD3 is symmetrically dimethylated in vivo. Protein extracts of wild-type and homozygous SmD3l(2)k131-07 larvae were separated by SDS-PAGE, blotted and probed with -SYM10 antibodies. Arrow indicates the position of SmD3. (D) SmB immunostaining using -Y12 antibodies. sDMA-SmB localizes in the nuclei of the nurse cells and somatic follicular cells of wild-type egg chambers but is undetected in csulRM egg chambers.

View larger version (23K): [in this window] [in a new window]   Fig. 5. SmB binds to Csul and Vls. Left panel: GST-Csul and GST-Vls proteins stained with Coomassie Blue. Middle panel: [35S]S*Tag-SmB fragments were separated by SDS-PAGE and detected by fluorography. Right panel: Following incubation with GST-Csul or GST-Vls, the bound [35S]S*Tag-SmB fragments were separated and detected as described above. No binding was detected with GST alone (data not shown). Below these panels the binding results of SmB and derivatives as well as those of SmD1 and SmD3 to Csul and Vls are summarized graphically. Size and structure of the SmB, SmD1 and SmD3 proteins are depicted with the two Sm domains and the RG dipeptides shown as orange boxes and circles, respectively. N- and C-terminal amino acid residues of the Sm proteins and fragments are indicated.

View larger version (20K): [in this window] [in a new window]   Fig. 6. SmB-binding domain in Csul. (Upper panel) Full-length GST-Csul or derivatives were purified from bacteria and incubated with [35S]S*Tag-SmB. Bound [35S]S*Tag-SmB proteins were separated by SDS-PAGE and detected by fluorography. Amino acid numbers are given across the top. `Input' was one tenth of the synthesized [35S]S*Tag-SmB proteins. The smallest fragment of Csul showing binding to SmB encompasses amino acids 1-200. (Lower panel) The amount of GST-Csul polypeptides was visualized by Coomassie Blue staining. (Right) Representation of the GST-Csul fragments used for the mapping and the summary of the results, with the identified domain required for SmB-binding in Csul shown in blue.

View larger version (56K): [in this window] [in a new window]   Fig. 7. A domain in the N-terminal region of Csul is essential for the methylation of Sm proteins and localization of Tud. (A-C) Identification of the Tud9A1-N-binding domain in Csul. (A) Fulllength GST-Csul or derivatives were purified from bacteria and incubated with S*Tag-Tud9A1-N. Bound S*Tag-Tud9A1-N was detected as described in Fig. S1 of the supplementary material (upper panel). The amount of GST-Csul proteins was visualized by Coomassie Blue staining (lower panel). Amino acid numbers are given across the top. `Input' was one tenth of the S*Tag-Tud9A1-N extract. The smallest fragment of Csul showing binding to Tud9A1-N encompasses the first 111 amino acids of Csul and may thus be distinct from the SmB binding site. (B) Delimitation of the Tud-binding domain using N-terminal truncation of Csul. Detection of S*Tag-Tud9A1-N bound to the N-terminal truncated Csul polypeptides revealed that the sequence following amino acid residue 60 is required for Tud9A1-N binding. (C) Representation of the GST-Csul fragments used for the mapping and the summary of the binding results are indicated on the right. The identified domain of Csul necessary for Tud-binding is shown in blue. (D,E) Use of interstitial deletions in the N terminus region of Csul to determine the domain necessary for Tud localization in the nuage and sDMA methylation of SmB. (D) The Tud9A1-N-binding domain of Csul is required for Tud localization in the nuage and pole cell formation. Egg chambers and embryos from homozygous csulRM females expressing the different transgenes were stained using anti-Tud (red; left column) and anti-Vas (red; right column) antibodies, respectively. DNA staining is in green. The deletions uncovering residues 21-40 and 41-60 can properly target Tud in the nuage and rescue the csul phenotype. (E) The N-terminal region of Csul is necessary for sDMA synthesis on SmB protein. Transgenes expressing full-length or modified Csul proteins were introduced into the csulRM background and ovarian extracts were prepared from the homozygous females. Western blot analysis using -SYM10 antibodies indicates that none of the deletion transgenes are able to rescue the methylation of the SmB protein (arrow).