A Drosophila melanogaster homologue of the human DEAD-box gene DDX1

Gene, 171 (1996) 225-229 0 1996 Elsevier Science B.V. All rights reserved.

225

0378-l 119/96/$15.00

GENE 09632

A Drosophila melanogaster gene DDXl

homologue of the human DEAD-box

(RNA-binding proteins; cell transformation;

Rafti”3b, Dimitrios

Fiorella

gene mapping; helicases; hnRNP U

Scarvelisb and Paul F. Laskob

“Istituto Internazionale di Genetica e Biqfisica (IIGB), Consiglio Nazionale delle Ricerche, Via Marconi 10, 80125 Napoli, Italy; and bDepartment of Biology, McGill University, Montkal, Qudbec, Canada, H3A 1 Bl Received by J.A. Engler: 30 August

1995; Revised/Accepted:

13 November

1995; Received at publishers:

8 January

1996

SUMMARY

DEAD-box genes are found throughout evolution and encode RNA-binding proteins. Such proteins include eukaryotic initiation factor-4A, which is essential for protein translation, Vasa, which is essential for germ line development, and a number of nuclear and mitochondrial RNA splicing factors. Transcription of a human DEAD-box gene, DDXl, is elevated in two retinoblastoma cell lines as a result of amplification of the immediate chromosomal region surrounding it, suggesting an important role for this gene in control of cell growth and division. We have isolated a Drosophila melanogaster (Dm) homologue (Ddxl) of DDXl which is strikingly similar to the human gene. The similarity (58.3% amino acid (aa) identity over 720 aa) extends beyond regions conserved in all DEAD-box proteins and covers the entire lengths of the proteins. The 2.7-kb Dm Ddxl RNA is expressed throughout development, but its levels are elevated in early embryos. Ddxl maps to polytene chromosome band 79D4 on the left arm of Dm chromosome 3.

INTRODUCTION

The DEAD-box family of putative RNA unwinding proteins includes a large number of similar proteins from bacteria, yeast, plants, and animals (Linder et al., 1989; Koonin, 1992). The best studied DEAD-box protein is eIF-4A, which is part of the cap-binding protein complex eIF-4F, and as part of the complex acts as an ATP-depenCorrespondence University, Canada.

to: Dr. P. Lasko,

1205 Avenue Docteur Tel. (1-514) 398-6721;

e-mail: Paul_ Abbreviations:

Department

of Biology,

Penfield, Montreal,

McGill

Quebec H3A 181,

Fax (1-514) 398-8051;

[email protected] aa, amino

tary DNA; DDXI, human

acid(s); bp, base pair(s); cDNA, gene encoding

putative

complemen-

RNA helicase; Ddxl,

Dm gene homologous to DDXI; DEAD-box, conserved protein domain of sequence Asp-Glu-Ala-Asp; Dm, Drosophila melanogaster; Dv, Drosophila vi&s; eIF, eukaryotic initiation factor; Hs, Homo sapiens; hnRNP U, heterogeneous nuclear ribonucleoprotein U; kb, kilobase or 1000 bp; Myr, million years; nt, nucleotide(s); PCR, polymerase chain reaction; RFLP, restriction fragment length sodium dodecyl sulfate, SSC, 0.15 M NaCl/O.OlS PII SO378-1119(96)00034-O

polymorphism; M Na,citrate

SDS, pH 7.6.

dent RNA helicase (Rozen et al., 1990). eIF-4A ranges from 390 to 414 aa in various species; all other known DEAD family proteins are larger than eIF-4A (ranging from 421 to 849 aa), and share 30-50% amino acid (aa) identity with eIF-4A and with each other throughout a region of approximately 350 aa. This region includes eight very highly conserved motifs (Schmid and Linder, 1992). This 350-aa conserved region is flanked by relatively unique aa sequences of variable lengths on the Nand C-terminal ends of the larger proteins. DEAD-box family proteins has been shown to be involved in translation initiation, RNA splicing, and assembly of spliceosomes and ribosomes (Wassarman and Steitz, 1991; Schmid and Linder, 1992). The DEADbox gene vasa is required for germ cell formation in Drosophila (Hay et al., 1988; Lasko and Ashburner, 1988), indicating a role for at least one of these proteins in a specific developmental event. Recently, homologues of vasa with germ-line specific expression have been identified in C. elegans, Xenopus, and mouse (Roussell and

226

a

Bennett, 1993; Komiya et al., 1994; Fujiwara et al., 1994), suggesting high evolutionary conservation of its developmental function. A human DEAD-box gene, DDXl, was isolated from a screen for cDNA clones of transcripts preferentially expressed in retinoblastoma cell lines (Godbout and Squire, 1993). Co-amplification of DDXl and the protooncogene MYCN has been demonstrated in both retinoblastoma and neuroblastoma cell lines (Godbout and Squire, 1993; Squire et al., 1995). From a PCR-based screen for novel Dm DEAD-box genes, we have isolated a homologue of DDXl. The very high level of conservation between the human and Dm genes supports an essential role for this protein.

EXPERIMENTAL

c

b

d

AND DISCUSSION

(a) Cloning of Dm Ddxl

Previously, we isolated novel DEAD-box genes from Dm by PCR amplification using two degenerate primers which encode two characteristic aa motifs, DEAD and Y(I/V)HRIG (Lavoie et al., 1993). In most DEAD-box proteins these two motifs are separated by 182-186 aa, so the expected size of the major amplification product is approximately 550 bp. In the human DDXl protein, however, two additional segments of 39 aa and 9 aa are present between the DEAD and Y(I/V)HRIG motifs which are not found in other DEAD-box proteins (Godbout and Squire, 1993). Therefore, the amplification product that would result from a DDXl homolog using these degenerate primers would be expected to be approximately 700 bp, rather than 550 bp. In this study we used the two degenerate primers we used previously to amplify DEAD-box genes from a related species of Drosophila, D. virilis (Dv), which is separated from Dm by about 60-65 Myr (Spicer, 1988). Comparison of Dv homologues with their Dm counterparts has been valuable in identifying functionally important regions of proteins (Blackman and Meselson, 1986; Curtis et al., 1995). From Dv DNA we obtained prominent amplification products of 550 and 700 bp (Fig. 1). The 700-bp band was isolated and subcloned, and nt sequence analysis showed it to be a homologue of DDXl. This partial Dv clone was used to screen phage libraries to isolate a Dv genomic clone, and the Dm cDNA clone discussed below. High-stringency Southern hybridizations for the Dv genomic library (Blackman and Meselson, 1986) were performed as described in Lavoie et al., 1993. The partial Dv clone was used to screen a O-2 h embryonic Dm cDNA library (Larochelle and Suter, 1995) under the following reduced-stringency conditions: hybridization, 58°C overnight in 3 x SSC/O.l%

Fig. 1. PCR amplification Template library and

constructed Meselson,

(length = 12,

1986).

(length=

was carried

out using

ling (55°C

Primers

degeneracy

Cetus thermal

phage

= 16)

sequences

DNA

in the hEMBL3

A(Y)(R)TA

extension

of DEAD-box

was CsCl-purified

vector

used and

R. Blackman

were

R=A/G,

(Perkin-Elmer)

cycler for 28 cycles of denaturation

2 min), and extension

(b) no template

Gibco

DNA;

700-bp and 550-bp products

1 pg Du template are marked

PCR

in a Perkin-Elmer (94°C

(72”C, 1 min), followed

BRL); lanes (b-d)

(cd)

R)TG(C/R)-

Y=C/T).

at 72°C and cooling to 4°C. Lane (a) molecular

(1 kb DNA ladder,

DNA

(Blackman

GA(Y)GA(R)GCNGAT

CC( R/T)ATNCG(

17, degeneracy=288; AmpliTaq

from D. uiriEis (Du).

from a Du genomic

1 min), anneaby a IO-min size standards

PCR amplification described

above.

with The

with arrows.

SDS/2 x Denhardt’s solution/0.5% Na.pyrophosphate/ 100 ug of salmon sperm DNA per ml; washes 2 x 30 min in 2 x SSC/O.l% SDS. A partial 1.7-kb clone that resulted from this screen was used to screen a O-4 h embryonic Dm cDNA library (Brown and Kafatos, 1988) at high stringency; the resulting 2423-bp clone is described below. (b) Dm Ddxl encodes a protein extremely similar to human DDXl

The nt sequence of Dm Ddxl and deduced aa sequence are presented in Fig. 2. Ddxl encodes a polypeptide of 725 aa (81 157 Da), as compared to the human protein of 734 aa (81805). The similarity between the two proteins extends virtually throughout their entire lengths. Overall, the proteins are 58.9% identical (424 identities out of 720 aa compared), and regions surrounding the eight motifs conserved in the DEAD-box family (Schmid and Linder, 1992) are even more highly conserved. For instance, there is a stretch of 3 l/33 identities surrounding motif Ia (aa 24-56 in the Dm protein) and a region of 32134 identities surrounding motif VII (aa 579-612 in the Dm protein).

227 120 20 CGGG'TGCCTTCTGCCTCCCCAT 240 60 LP I

1 AAATAAATCAACCGGCT'ITC~~~A~C~~AGA~~GTAT~TC~~AC~AT~G~GAG~~GT~~CC~A~~~~CACAGAC~TT MTAFEEFGYLPELGMBTDEL 1 121 GGACTGGACCCTGCCCACCGATGTCCAGGCTGAGGCCATT 2lgwTLETPVQBEAIE41LcGGoYLMBBETPSBSTGBEc

motif Ia 360 GGGGAAGGC'IC&GCAAGGGCGGTGCAATTGGAGZAGCTG'IGACTCCG'i'GACCATCXCCT"l?TTCGACCGAGGAAACGCGCT 241 TCTGCAAATCGTGTGC~~~~~A~ 100 DSYTMSFFPBENBL RDLEEEKAGQGRCNWRSC 61 LQIYWElCL 480 361 TGCCGTCACTCCTGATGGTCTGCGTTGCCAGTCGCGGGAG 140 101AVT PPGLRSQ$EEFKEWBG CRATTGVRGKc&FXFgATVTQA zr 600 481 CGAGGGACTGTGTCG'TGTGGXIGGTCTACACAGCA~AAACCTCGA TMGGGCACCTGTCGCATGGX 'MTtXA'ITCGGCGGCACCGGCAAGAAGTCCAA04ACCGCCAGT~GATGA 180 141EGLCRYGW S TQQBNLq&LsTC RMGEGESSTGBESNERQEPD = zz 720 'ITn;AAAAACCAGGAAGTGAGX'TCACCA_AGAATGGCCAGAACC~GTAGCCTTCCGCCTGCCGGACAA 601 CTACGGTGAGGCATTCGAGATGTCATCGGCTGCCTGCTCGA 220 181-f GEAF GKADV I rcL&n LKNQEvAS E TKNG QN&GV&FRLEDN = 840 GACTGACTTCAAGTA?GCCCCGGGTAACGGTTTCGT%GCGCTT'XCAGGC 721 CCTJXCA?.GGAGACCTmACCCG>GGTGCTG~GAACGCCG?UA%CAGlTTAACTTTkAAA 260 221LAKE TFY PAVVL ENBEMQFNFCK T DESYAEGNGEYGAC QA

=

G~~'YXAGCACCGCCAAACCGGCGCC-~%ACGCCCCTCAGGC SWS TAB PA_PNAPQA I

841 TGGACCCGAGCATAGCACAATCCGATAACCGGTCCCGACA 261G PEH S KAN P I TG PD 961 AACCTACAACCAGATCGAG~G~~TACCACCTGAG 301z~NQ~ E KF_KYH L

SBZEVB

ShL

960 300

IMEESEEhZiEQ

motif Ib '~XGAAGAGCAAAAGGCGCAGCTGATGCAGGGCACTCA 1080 340 E EQKAQLMQGTH

LIGGYRL motif II

1081 CATAGTGGTGGGCACGCC GGGGCGGCTGGAGGAGATGATCAAG4GTGGACTGGTCCTG ~~~IVYGTE&_RI!E EM I N SGLVLL T HCGF FYLPEBPALLKQ&XT motif Iv motif III 1201 CGAGCn;A?TGACCGTCTACACAAACAAATTCCCAAGATCCC~GATCACATC~A~~~CG~~AGA~~G~~~~CACG~~A~~A~~~~~~G 381E LIDBLHKQJEK ITSPGRBLQMVYCSBTLHAEEYBqMAER motif V GGGG04GGAn;C?Y;TCCCTGAGRCGG?TCACCACGTCGT 1321 CCTGA'IGCACTTCCCCACCTGGGG'IGATCTGAA 421LMHE9TWGPLBGEeAYE ETY_HHYYC LYDEQMQTTWQ SLRQ 1441 ACCCATTGGCACCGACGGCGTCCACGACCGTGACAACGTT 461PIGTpGYH D R_DNVHr_GNHS KETLSQBVgLLBGEXCYHP~D 1561 TAAACACAATATGGACCGCATTATC'ITCn;CCGCACCG 501KHNMpR8IIr6E~qQp~gNZERF L RQRDKHYSGYCLBGBB 1681 AAAGCCGCAGGACXGTAAAGAAAAC CTGWLRATG~~~GGCAACAAGTC~G~~A~~AC~ACG~~TCGA~T~A~~~AC~~CGT~A~AT~ 54lqPQEBBENLEMEBRQ QYKEhLclPYBBBE LPlTP L? FMI_N motif VI 1801 'lG'IGACGC?GCCCGA?GACAAGACCAATTACG'ITCATCXXATCGGACG~ TGGGCAGGGCGGAACGCA~~CA~AG~~~CAG~C~A~~~TAC~ ~~~~~LEPDYTNYYBE~G~Y~~~~~~~~~~~~~~~V PEKYllJXH motif VII 1921 CGGCGAGTGGTGCAAGTCGC~~CGTA~~~C~CAC~~~CCG~GTCC~~~~A~~TAC~C~CC~CCTAC~CG~G~A~ACCACT~~ 621G EWCKSRPR SCNNTNLTEVRG~_?CIWXIJE PNLLAEVEDBLN 2041 CATCACCATCCA~A~~ACCATGGATGTGCCA~A~CAG~~~AC~C~C~C~~G~TAC~TC~G~T~~G~C~~~TAC~GACCACG~GA~A 6611 TIQ QY D K T MDYEYN DEQG PVVXGQSNL R Tg S GX E DBY E 2161 GCTGGAGCCAACTGTACGCAAACTGACGGAGCTAGAGTM; 701L EETYRKLT ELELQ SQ S LEL K RL KV. 2281

A’lYXGGTAAGACCAACTCACGC~TAC~TGCGGAAATA

2401

AAA&&&&CATTTAATTAAACT(A),

Fig. 2. Nucleotide templates McGill accession proteins

site (AATAAA; nt 2404-2409)

subcloned University, number

in the BlueScript Montreal,

for the partial

are underlined;

(open triangle).

Sequences

and the Sequenase

DC clone (sequence

aa identical

of 13 aa which was introduced

is underlined.

SK vector (Stratagene)

Canada)

were aligned

Sequencing

DDXl

sequences

polypeptide

was carried

and human to optimize

using the GeneJockey

hnRNP

primers

The GenBank

is listed under the nt sequence.

underlined.

with hnRNP

package

(Biosoft,

by dideoxy

(purchased

accession

sequencing

from the Sheldon

number

Those aa which are identical U are doubly

alignment

II software

sequence

out on both strands

using oligodeoxynucleotide system (Amersham).

not shown) is U34779.

in the two DDXl

in the DDXl

1320 '420 1440 460 1560 500 1680 540 1800 580 1920 620

Q

CYM-TTWGGCACAAACACACAAmTATATGTAGGAA

of the Dm Ddxl cDNA clone. The predicted

sequence

polyadenylation

1200 380

between

U; similarly, Cambridge,

A consensus

of double-stranded

Biotechnology

for the Dm sequence the predicted

The solid triangle

Centre,

is U34773, the

Dm and Hs DDXl

after aa 140 indicates

a gap of 2 aa was introduced UK) and minor

2040 660 2160 700 2280 725 2400

adjustments

a gap

after aa 202 were made

by hand.

The similarities, however, are not confined to common regions of DEAD-box proteins. Both the human and Dm proteins have an aberrantly long distance between motifs Ia and Ib; aa 84-280 in the human protein (aa 91-286 in the Dm protein) are unique among DEAD-box proteins to DDXl (Godbout et al., 1994). This region shares some sequence similarity with human hnRNP U, which may be involved in RNA processing (Kiledjian and Dreyfuss, 1992). Within this unique region the two DDXl proteins are 53.8% identical, including a stretch of 28/36 identical residues. The similarity includes the region of DDXl which is shared by hnRNP U; in a region of 130 aa, 42 (32.3%) are identical in all three proteins and an additional 17 (13.1%) differ by only conservative substitutions.

With regard to the two additional segments between the DEAD and Y (I/V)HRIG motifs (motifs IV and VII), the 39-aa segment found in human DDXl is well conserved, with 22 identities in a corresponding 38-aa segment in the Dm protein (aa 462-499). However, the shorter 9-aa segment is mostly missing in the Dm protein. The partial Du DDXl peptide sequence we obtained from cDNA and genomic clones shares 300/347 identities (86.5%) with the Dm sequence. It differs by the inclusion of a 18-aa segment between residues corresponding to aa 577 and 578 in the Dm sequence, and by the lack of a lo-aa segment corresponding to aa 633-642 in the Dm sequence. An additional 33 aa are present at the C terminus of the Dv polypeptide.

228 (c) Ddxl is expressed throughout Dm development, most

(d) Mapping the Ddxl gene

abundantly in early embryos

In situ hybridization of the Ddxl cDNA clone to Dm larval salivary gland polytene chromosomes allowed mapping of the gene to band 79D4 on the left arm of chromosome 3 (data not shown). We have not found any RFLPs in genomic DNA prepared from any of the transposon-insertion or aberration mutations we have obtained in this chromosomal region, so we do not as yet have any candidate Ddxl mutations. We plan to obtain information regarding the phenotypic consequences of alterations in Ddxl expression by creating transgenic flies which overexpress the gene, or which express an antisense construct. As there are two P-element insertions in chromosomal interval 79Dl-4 (1(3)neo30 and 1(3)01544; Cooley et al., 1988; Karpen and Spradling, 1992) the local hopping strategy (Golic, 1994) will be a feasible one for generating Ddxl mutations.

To investigate the temporal control of Ddxl expression, we probed a Northern blot containing poly(A)+ RNA from various developmental stages with the partial Ddxl clone (Fig. 3). We found that Ddxl is expressed as a major 2.7-kb transcript in all developmental stages, with particularly high levels in O-2 h embryos and reduced levels in late embryos. These results are consistent with a high level of Ddxl expression during oogenesis, and with storage of the Ddxl RNA in oocytes. Tissue in situ hybridizations on ovaries indicated high levels of Ddxl RNA in nurse cell cytoplasm from stage 1 of oogenesis onward, but no asymmetric accumulation of this RNA within the oocyte or early embryo (data not shown). In early (O-4 h) embryos we reproducibly observed low levels of a smaller, 1.3-kb mRNA despite our hybridizations being carried out under high stringency conditions. We have not investigated whether this may result from alternative splicing of the Ddxl primary transcript; such a small transcript would be expected to encode a severely truncated DDX 1 protein.

We thank Michel Harvey for technical assistance, Stephane Larochelle and Beat Suter for the developmental Northern and for the hZAP cDNA library, and Akira Nakamura for helpful discussions. This work was supported by a research grant to P.F.L. from the National Cancer Institute of Canada with funds from the Canadian Cancer Society, and by a team grant from the FCAR. F.R. was the recipient of an MRC (Canada)-CNR (Italy) Exchange Fellowship, and is a Research Scientist of the Consiglio Nazionale delle Ricerche of Italy. D.S. was supported by an NSERC University Undergraduate Student Research Award. P.F.L. is a Research Scientist of the National Cancer Institute of Canada.

abcdefahiiklmnoD

A

2.7+

-

9.5

-

7.5

-

2.4

ACKNOWLEDGEMENTS

1.3+

REFERENCES Blackman,

R.K. and Meselson,

comparisons Fig. 3. Expression

pattern

of the Ddxf gene. A: Polyadenylated

RNA

from (a) O-2 h embryos; (b) 2-4 h embryos; (c) 4-8 h embryos; (d) 8-12 h embryos; (e) 12-20 h embryos; (f) first-instar larvae; (g) secondinstar larvae; (h) early third-instar early pupae;

(k) late pupae;

larvae; (i) late third-instar

(1) adult appendages;

(m) adult

larvae; (j) heads; (n)

adult abdomens; (0) adult males; (p) adult females was separated by electrophoresis and probed with radiolabelled Ddxl cDNA. RNA isolation techniques

and hybridization

conditions

were as described

in Diirig

et al. (1988) and Suter et al. (1989). Autoradiographic exposure was 2 days. The major 2.7-kb and minor 1.3-kb transcripts are indicated by arrows. The sizes (in kb) listed on the right margin show the migration distances of RNA molecular size standards (0.24-9.5 kb RNA ladder, Gibco-BRL). B: The same filter as in A probed with RpSlSa cDNA (encoding small ribosomal subunit protein autoradiographic exposure was 16 h.

in Dm; Lavoie et al., 1994);

the Drosophila Brown,

N.H.

M.: Interspecific

used to identify and

regulatory

nucleotide

and structural

sequence features

of

hsp82 gene. J. Mol. Biol. 188 (1986) 499-515. Kafatos,

F.C.:

Functional

cDNA

libraries

from

Drosophila embryos. J. Mol. Biol. 203 (1988) 425-437. Cooley, L., Kelley, R. and Spradling, A.C.: Insertional mutagenesis the Drosophila 1121l1128. Curtis, D., Apfeld, conserved

genome

with single P-elements.

J. and Lehmann,

organizer

R.: Nanos

of anterior-posterior

of

Science 239 (1988) is an evolutionarily

polarity.

Development

121

(1995) 189991910. Dorig, R.E., Suter, amber nonsense formation

B., Gray, mutation

of an amber

M. and Ktibli, E.: Identification in the rosy516 gene by germline

suppressor

tRNA gene. EMBO

of an trans-

J. 7 (1988)

2579-2584. Fujiwara, Y., Komiya, T., Kawabata, H., Sato, M., Fujimoto, H., Furusawa, M. and Note, T.: Isolation of a DEAD-family protein

229 gene that

encodes

specific expression

a murine

of Drosophila

homolog

unsa and its

in germ cell lineage. Proc. Natl. Acad. Sci. USA

R. and Squire, J.: Amplification

in retinoblastoma

of a DEAD box protein

gene

cell lines. Proc. Natl. Acad. Sci. USA 90 (1993)

7578-7582. Godbout,

R., Hale, M. and Bisgrove, homology

D.: A human

to heterogeneous

DEAD box protein

nuclear

ribonucleoprotein

ter and recombination specific recombinase.

between Genetics

duplicated

polar granules

is encoded

in

P-element Kiledjian,

the

U protein:

a site-

A.C.: Analysis

of subtelomeric

minichromosome

sim-

by

single

132 (1992) 737-753.

structure

and binding

RNA through

characterization

K. and Furusawa,

of a novel gene of the DEAD

which is specifically

expressed

activity

box protein

and family

in germ cells of Xenopus laeuis. Dev.

Biol. 162 (1994) 354-363.

Larochelle,

S. and Suter, B.: Molecular

logue of the rat ribosomal 1261 (1995) 1477150. Lasko,

P.F. and Ashburner,

uasa is very similar (1988) 611-617.

protein

RNA helicases. Trends Biochem. cloning of the Drosophila homoLll gene. Biochim.

M.: The product

to eukaryotic

initiation

Biophys.

Acta

of the Drosophila

gene

factor

4A. Nature

Biochim.

M., Lee, H., Sonenberg,

plasmid

a

yeast

Biophys.

Acta

N. and Lasko,

cdc33 mutation

a Drosophila

expressing

P., Lasko,

Nature

P.F., Ashburner,

J. and

M., Leroy,

Slonimski,

P.:

of yeast by

ribosomal

protein.

P.P.:

Birth

P., Nielsen,

P.J., Nishi,

of the D-E-A-D

box.

337 (1989) 121-122. D.L. and Bennett,

K.L.: glh-1, a germ-line

case from Caenorhabditis,

has four zinc fingers.

putative Proc.

RNA hehNatl. Acad.

Sci. USA 90 (1993) 9300-9304. F., Edery,

I., Meerovitch,

Sonenberg,

N.: Bidirectional

translation

initiation

K., Dever, T.E., Merrick, RNA helicase

factors

Schmid,

S.R. and Linder,

helicases.

activity

W.C., and

of eukaryotic

4A and 4F. Mol. Cell. Biol. 10 (1990)

G.S.:

groups

P.: D-E-A-D

Mol. Microbial. Molecular

as indicated

335

protein

family of putative

RNA

6 (1992) 283-292.

evolution

among

by two-dimensional

some

Drosophila

electrophoresis.

species J. Mol.

Evol. 27 (1988) 250-260. Squire, J.A., Thorner,

P.S., Weitzman,

J., Hale, M. and Godbout

Koonin, E.V.: A new group of putative 17(1992)495-497.

assembly.

encodes

to a putative

1134-1144.

RGG box. EMBO M.: Isolation

Dbp45A

P.F.:

with similarity

of a temperature-sensitive

K., Schnier,

Spicer,

T., Itoh, K., Ikeneshi,

Lasko,

in ribosome

C., Tam, R., Clark,

a multicopy

Rozen,

heterochro-

DpZ187

Genetics

G.: Primary binding

sequence

J. 11 (1992) 265552664. Komiya,

Lavoie,

Roussell,

Cell 55 (1988) 577-587.

mutagenesis.

M. and Dreyfuss,

of the hnRNP

helicases.

Drosophila

insertional

using

of Drosophila

component

by vasa and has extensive

ilarity to ATP-dependent G.H. and Spradling,

elements

137 (1994) 551-563.

Hay, B., Jan, L.Y. and Jan, Y.N.: A protein

matin

involved

and

1216 (1993) 140-144.

Linder, of P elements in Drosophila melanogas-

Local transposition

M.

box protein

J. Biol. Chem. 269 (1994) 1462514630.

U. Gene 138 (1994) 243-245.

Karpen,

Harvey, DEAD

Suppression

with partial Golic, KG.:

C.A.,

Drosophila helicase

91 (1994) 1225812262. Godbout,

Lavoie,

S., Maggi, J.D., Dirks, P., Doyle,

R.: Co-amplification

DEAD box gene (DDXZ) in primary (1995) 1417-1422. Suter, B., Romberg,

L.M. and Steward,

of MYCN

and a

neuroblastoma.

Oncogene

10

R.: Bicaudal-D,

a Drosophila

gene involved in developmental asymmetry: localized transcript accumulation in ovaries and sequence similarity to myosin heavy chain tail domain. Wassarman, proteins.

Genes Dev. 3 (1989) 1957- 1968.

D.A. and Steitz, J.A.: RNA splicing. Nature

349 (1991) 463-464.

Alive with DEAD