Evidence for 3' Untranslated Region-Dependent Autoregulation of the Drosophila Gene Encoding the Neuronal Nuclear RNA-Binding Protein ELAV

Corresponding author: Marie-Laure Samson, Department of Biochemistry and Molecular Biology, University of Nebraska Medical Center, 984525 Nebraska Medical Center, Omaha, NE 68198-4525., msamson{at}molbio.unmc.edu (E-mail).

Communicating editor: L. L. SEARLES

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The Drosophila locus embryonic lethal abnormal visual system (elav) encodes a nuclear RNA-binding protein essential for normal neuronal differentiation and maintenance of neurons. ELAV is thought to play its role by binding to RNAs produced by other genes necessary for neuronal differentiation and consequently to affect their metabolism by an as yet unknown mechanism. ELAV structural homologues have been identified in a wide range of organisms, including humans, indicating an important conserved role for the protein. Analysis of elav germline transformants presented here shows that one copy of elav minigenes lacking a complete 3' untranslated region (3' UTR) rescues null mutations at elav, but that two copies are lethal. Additional in vivo experiments demonstrate that elav expression is regulated through the 3' UTR of the gene and indicate that this level of regulation is dependent upon ELAV itself. Because ELAV is an RNA-binding protein, the simplest model to account for these findings is that ELAV binds to the 3' UTR of its own RNA to autoregulate its expression. I discuss the implications of these results for normal elav function.

THE Drosophila gene embryonic lethal abnormal visual system (elav) is the first identified member of a conserved multigene family with homologues in many species, including humans (ROBINOW et al. 1988 ; SZABO et al. 1991 ; KIM and BAKER 1993 ; KING 1994 ; KING et al. 1994 ; MANLEY 1994 ; SAKAI et al. 1994 ; GOOD 1995 ; PERRON et al. 1995 ; ABE et al. 1996A , ABE et al. 1996B ; MA et al. 1996 ; STELLER et al. 1996 ; U. ATASOY, J. WATSON and J. D. KEENE, personal communication). The two Drosophila family members so far identified, elav and rbp9, are strictly expressed in neuronal nuclei (CAMPOS et al. 1987 ; KIM and BAKER 1993 ). Null and severe mutations at the elav locus cause altered neurite organization and are embryonic lethal (CAMPOS et al. 1985 ; JIMENEZ and CAMPOS-ORTEGA 1987 ). The lack of elav function in an embryo does not prevent the generation of neurons, but rather leads to their abnormal differentiation. Flies carrying temperature-sensitive elav mutations develop normally at permissive temperature, but initiate neuronal degeneration when shifted to nonpermissive temperature, showing that elav function is also necessary for neuronal maintenance (CAMPOS et al. 1985 ; HOMYK et al. 1985 ). No mutations of the rbp9 gene have been reported.

The products of elav and of the other members of this gene family contain three RNA recognition motifs, referred to as RRMs, that are present in a large number of proteins involved in diverse processes of RNA metabolism and translation (for reviews, see KENAN et al. 1991 ; MATTAJ 1993 ; BURD and DREYFUSS 1994 ). Because ELAV is a nuclear RNA-binding protein, it seems likely that it promotes normal neuronal differentiation by modulating the metabolism of RNA produced from genes whose expression is required in neurons. The mechanism of this proposed modulation remains unknown.

Several genes homologous to elav have been described in vertebrates, including humans. On the basis of sequence similarity, the vertebrate ELAV sequences can be split into four groups, referred to here as ELAV-A, -B, -C, and -D, following the nomenclature introduced by GOOD 1995 . Human, mouse, and Xenopus have one each of these four ELAV subtypes. Additionally rat ELAV-B and -D as well as zebrafish and chicken ELAV-A, -C, and -D have been identified (SZABO et al. 1991 ; KING 1994 ; KING et al. 1994 ; MANLEY 1994 ; SAKAI et al. 1994 ; GOOD 1995 ; PERRON et al. 1995 ; ABE et al. 1996A , ABE et al. 1996B ; MA et al. 1996 ; STELLER et al. 1996 ; HIROTAKA and DARNELL 1997 ; WAKAMATSU and WESTON 1997 ; U. ATASOY, J. WATSON and J. D. KEENE, personal communication). Amino acid sequence identities range from 85–99% among the members of each group. Interestingly, each structural group has distinct tissue-specific and subcellular localizations. The human, Xenopus, and mouse ELAV-A members are ubiquitously transcribed (GOOD 1995 ; MA et al. 1996 ; U. ATASOY, J. WATSON and J. D. KEENE, personal communication). In contrast, the ELAV-B, -C, and -D types are neuronal (SZABO et al. 1991 ; LEVINE et al. 1993 ; KING et al. 1994 ; SAKAI et al. 1994 ; GOOD 1995 ; PERRON et al. 1995 , PERRON et al. 1997 ; ABE et al. 1996A , ABE et al. 1996B ; STELLER et al. 1996 ). Low levels of Xenopus, mouse, and rat ELAV-B and mouse ELAV-D transcripts are also detected in testis/ovaries (KING et al. 1994 ; GOOD 1995 ; ABE et al. 1996A , ABE et al. 1996B ). Human, mouse, and Xenopus ELAV-B are found both in the cytoplasm and to various extents in the nucleus, possibly reflecting shuttling between those compartments (ABE et al. 1996A , ABE et al. 1996B ; GAO and KEENE 1996 ; PERRON et al. 1997 ). Human ELAV-C and -D are mostly nuclear (SZABO et al. 1991 ; SAKAI et al. 1994 ). Many of these proteins can bind U/AU-rich sequences in vitro (LEVINE et al. 1993 ; GAO et al. 1994 ; LIU et al. 1995 ; ABE et al. 1996A , ABE et al. 1996B ; CHAGNOVICH and COHN 1996 ; CHUNG et al. 1996 , CHUNG et al. 1997 ; MA et al. 1996 , MA et al. 1997 ; JAIN et al. 1997 ; MYER et al. 1997 ), but this binding depends upon different RRMs in the proteins, respectively the third RRM domain of human ELAV-B (Hel-N1), the first RRM of mouse ELAV-C (mHuC), and the first two RRMs of human ELAV-D (HuD). The third RRM of ELAV-A, -C, and -D (human HuR, mouse HuC, and human HuD) shows different specificity, and binds poly(A) in vitro (ABE et al. 1996A ; MA et al. 1997 ).

The sequences of the two Drosophila proteins, ELAV (ROBINOW et al. 1988 ) and RBP9 (KIM and BAKER 1993 ), are similarly related to each of the four classes of human ELAV sequences (52–54% amino acid identity in the case of ELAV and 56–61% amino acid identity for RBP9). On the basis of their strict neuronal nuclear localization (ROBINOW and WHITE 1991 ; KIM and BAKER 1993 ; YANNONI and WHITE 1997 ), it can be inferred that they are more related to the ELAV-C or -D types, although ELAV is singled out by its systematic and exclusive neuronal expression. Together, the different subcellular localizations and the diversity of in vitro binding specificities underline differences in the function of ELAV-A, -B, -C, and -D proteins, but the strong structural conservation indicates the possibility that all mediate the same general function in RNA metabolism, possibly triggering differentiation of specific cell types, both for the exclusively neuronal and the ubiquitous members of the family.

Analysis of the ELAV family suggests two possible functions for ELAV-related genes. First, these proteins have been proposed to be implicated in the modulation of mRNA turnover/translation, processes that are intimately interdependent (reviewed in JACOBSON and PELTZ 1996 ). This model was proposed because ELAV family members from all four ELAV subtypes bind AU-rich sequences in vitro. These AU-rich binding sites resemble and in some instances correspond to the loosely defined ARE sequences typically present in the 3' untranslated region (3' UTR) of short-lived mRNA (CHEN and SHYU 1995 ). Consistent with a role in mRNA turnover/translation, the human Hel-N1 protein is associated with polysomes in neuroblastoma and medulloblastoma cells (CHAGNOVICH et al. 1996 ; GAO and KEENE 1996 ), while overexpression of Hel-N1 in a preadipocyte cell line leads to upregulation of the rate of GLUT1 mRNA stability and translational inititation (JAIN et al. 1997 ). Second, it was proposed that ELAV functions as an alternative splicing factor. The basis for this model is the influence of the elav genotype (elav ectopic expression versus elav gene deletions) on the ratio between the neuronal and nonneuronal forms of neuroglian protein that derive from alternatively spliced mRNAs (KOUSHIKA et al. 1996 ). Although performed in vivo, these experiments do not exclude an indirect effect of ELAV, for example on the stability of a neuroglian-specific splicing factor.

These disparate views are not necessarily contradictory, but require better documentation. It is worth noting the parallel between elav and another Drosophila gene, Sex-lethal (Sxl). Both genes are required for initiation and maintenance of developmental pathways: neuronal differentiation in the case of elav, and sex determination in the case of Sxl. Both encode related proteins that contain RRMs and show high affinity for poly U in vitro. Sxl autoregulates, and the data presented here show that elav also autoregulates. The molecular mechanism of action of Sxl protein (SXL) is better understood than that of elav. SXL has multiple activities in the regulation of pre-mRNA splicing and in translation, presumably through interactions with different transcript regions (BASHAW and BAKER 1997 ; GRANADINO et al. 1997 ; KELLEY et al. 1997 ).

In this article, I analyze the regulation of elav expression using functional elav transgenes carrying an altered 3' UTR. Remarkably, these transgenes are associated with dosage-dependent lethality and behave as neomorphic elav mutations. I show that the elav 3' UTR confers gene dosage independence on elav expression and that elav expression is autoregulated. Given the properties of ELAV, autoregulation is most likely to occur through its direct binding to the 3' UTR of elav mRNA.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Transformation vectors:
A partial sequence of 9285 nucleotides has been published for the elav gene (ROBINOW et al. 1988 ). The gene extends about 8 kb 3' to this sequenced fragment (CAMPOS et al. 1987 ). The positions of nucleotides in the gene are referred to according to their positions in the published sequence.

The transformation vector pe350 that carries a 15.5-kb genomic fragment (see Figure 1) was built as follows: The BamHI-EcoRI (8402–9280) fragment of pe332 (SAMSON et al. 1995 ), a pUC18 derivative that contains a SmaI-XbaI insert composed of the elav SmaI-EcoRI (6639–9280) fragment fused to transcription termination and polyadenylation signals of the ß1 tubulin gene, was replaced by the 7.5-kb BamHI restriction fragment ending at position 8402, yielding pe348. A 10.1-kb SmaI-XbaI (composed of the 9.3-kb elav SmaI-BamHI beginning at nucleotide 6639 fused to transcription termination and polyadenylation signals of the ß1 tubulin gene) was purified from pe348. A three-way ligation between the 5.9-kb PstI-SmaI (757–6639) elav fragment, the 10.1-kb SmaI-XbaI fragment from pe348 and the 7.8-kb pCaSpeR cut with PstI and XbaI yielded pe350. A similar strategy was used to build pe353, after replacing the wild-type BstXI-BamHI fragment (8269–8402) from pe332 by the homologous fragment from the elav^FliJ2 allele.

View larger version (14K): In this window In a new window Download PPT slide   Figure 1. Structure of the elav locus. A restriction map is shown as a continuous line. The structure and splicing pattern of cDNA 1 (a partial embryonic cDNA, one of the two similar elav cDNAs characterized) and its translational AUG initiation codon are shown. Transcripts extend much farther downstream, spanning over 16 kb, as shown, but their detailed structures have not been determined (YAO et al. 1993 ). The structure and properties of the four classes of tested transformants (see Table 1 for details), respectively, correspond to the 8.5-kb PstI-EcoRI genomic fragment, mutant versions of this fragment, the 13.5-kb XbaI-BamHI genomic fragment, and the 15.5-kb PstI-BamHI genomic fragment. TfL, long-transformants (function and 3' UTR); TfS, short-transformant (function and truncated 3' UTR); TfSm, mutant-short-transformant (little or no function and truncated 3' UTR). B, BamHI; E, EcoRI; H, HindIII; P, PstI; S, SmaI; X, XbaI.

Germline transformations:
Embryos from the stock y w^67c23 were coinjected with 200 ng/µl transformation vectors and 50 ng/µl p25.7 (KARESS and RUBIN 1984 ; PIRROTTA 1988 ; THUMMEL et al. 1988 ). Standard procedures were used to identify the transformants and generate stocks.

Transformed lines:
The 8.5-kb genomic fragment included between nucleotides 757 to 9285 (fused to termination of transcription and polyadenylation signals) is sufficient for transformation rescue (DmORF series; YAO and WHITE 1991 ). Transformed Drosophila stocks of the 336, 339, FliJ1, FliJ2, and TS1 series carry the 8.5-kb fragment, modified by the insertion of an oligonucleotide or by point mutations (SAMSON et al. 1995 ). The DvORF lines also contain the Drosophila melanogaster 8.5-kb genomic fragment, where most of the D. melanogaster coding sequence is replaced by its D. virilis counterpart, generating a larger ELAV protein (YAO and WHITE 1991 ). Transformed lines 17 and 20 carry the 8.5-kb genomic DNA, where the most 5' intron is deleted, and transformed lines 7 and 22 carry the 8.5-kb genomic DNA depleted of its two introns (a gift from K.-M. Yao and K. White).

The lines P15.2 and P62.2 (a gift from K. White), carry a 13.5-kb-long transforming genomic fragment whose 5' end corresponds to nucleotide 2528 of the elav gene, and are similar to those described by CAMPOS et al. 1987 . The lines 350 and 353 carry a 15.5-kb-long transforming genomic fragment whose 5' end corresponds to nucleotide 757 of elav. Lines of the 350 series carry a wild-type sequence, while line 353 carries the mutation elav^FliJ2, as does the FliJ2 series (SAMSON et al. 1995 ).

Genetic analysis:
All crosses were maintained on standard cornmeal medium at 25°, or in specified instances at 18°. Fertility and viability of specific phenotypes were measured as detailed in the table legends. The hypothesis that the flies carrying two transgenes have normal viability was tested using a chi-square test with 1 d.f., comparing expected and observed values of the number of progeny carrying two transgenes and the number of progeny carrying no or one transgene.

Protein preparation and immunoblot analysis:
Drosophila head protein extracts were prepared and resolved as previously described (SAMSON et al. 1995 ). All flies were raised at 25°. Immunodetection was performed using chemiluminescence (Amersham ECL, Arlington Heights, IL). A mouse anti-ELAV monoclonal antibody (ascites fluid obtained after injection of hybridoma cell line 9F8A9 provided by G. Rubin) was used for primary incubation (1:5000 dilution). The secondary antibody was peroxidase-conjugated anti-mouse IgG (Boehringer 1317377, 1:5000 dilution; Boehringer Mannheim, Mannheim, Germany). An internal loading control was performed by detecting ß-tubulin immunoreactivity via a 1:10,000 mouse primary antibody (Sigma, St. Louis, MO) and 1:10,000 peroxidase-conjugated anti-mouse IgG secondary antibody. Quantification was done with a computing laser densitometer (ImageQuant software; Molecular Dynamics, Sunnyvale, CA).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Functional elav transgenes missing the 3' UTR are homozygous lethal, while transgenes containing the 3' UTR are homozygous viable:
To analyze the structure of elav and its function, elav constructs were reintroduced into the Drosophila germline. An 8.5-kb genomic fragment rescues elav mutations, although it does not include the entire elav 3' UTR (Figure 1). Nine independent lines carrying such transgenes [transformant-short (TfS)], were analyzed. In all cases, attempts to establish homozygous stocks carrying two copies of a given TfS failed, whether in a wild-type or a null elav background. In balanced heterozygous stocks carrying these constructs, flies with two copies of the transgene represent 0–10% of the total progeny, as opposed to the expected 33%, and they are invariably sterile (Table 1). In contrast flies carrying one TfS copy are obtained in the expected number in the progeny of the crosses (Table 1, see rescue of elav^e5), indicating that a single copy of TfS has no deleterious effect. Transgenes carrying mutated elav forms [mutant transformant-short (TfS^m)] that fail to provide elav⁺ function were also tested. Most (15 out of 17) are homozygous viable (Table 1). The two exceptions, FliJ1-5 and TS1-271, probably correspond to recessive lethal mutations generated by the insertion of the transgene.

To test whether the lethality/sterility observed in theTfS homozygotes can be accounted for by the truncation of the elav 3' UTR, the effect of constructs carrying a larger elav 3' UTR was analyzed. Constructs including a 6-kb fragment of the 3' UTR [transformant-long (TfL)] rescue elav^null function and, in contrast to the TfS, can be recovered and maintained as homozygotes. Two lines of a 13.5-kb construct (CAMPOS et al. 1987 ) and two lines of a 15.5-kb construct were tested (Figure 1 and Table 1).

Combinations of different functional elav minigenes lacking the normal 3' UTR induce high levels of lethality or sterility:
To exclude further the possibility that the homozygous lethality of the TfS was due to the generation of deleterious mutations, males and females carrying different TfS insertions were mated and their progeny examined. A representative sample of the results of 116 crosses is shown in Table 2.

Effects of varying severity were found in progeny carrying two different TfS insertions. In the majority of these crosses, a significant deficit of flies carrying two transgenes was observed (7 of 10 cases, crosses 3, 4, 8, 10, 13, 19, and 20, Table 2). When they could be recovered as adults, flies carrying two TfS had reduced fertility. In less extreme cases, flies carrying two TfS did not show significantly decreased viability, but still showed decreased fertility (crosses 5, 15, and 18, Table 2). Overall, viability was reduced in males more than in females, and fertility was decreased in females more than in males. Deleterious effects were seen both in normal and elav^null backgrounds.

The sterility of the flies carrying two TfS transgenes was evidenced in multiple ways (data not shown): some adults died within 24 hr of eclosion leaving no progeny, some females did not lay eggs, while the embryos of other females failed to hatch. In some instances, only female fertility was affected, while males were normal (for instance, see crosses 5 and 13, Table 2).

As a control, functional TfS transgenes were combined with functional TfL or nonfunctional TfS^m transgenes. Expected numbers of flies carrying either one or two transgenes were obtained in all of these crosses (Table 2). The fertility of both male and female progeny was normal, with the exception of females carrying two transgenes arising from cross 12 (Table 2). The lowered fertility of these females is unrelated to the transgene, because a similar cross involving the same transgene in a different genetic background yielded fertile females (cross 7, Table 2). The consequence of X chromosome-dependent genetic background, or perhaps a maternal effect, on the outcome of these crosses becomes apparent by comparing reciprocal crosses, as for instance crosses 15 and 20 (Table 2). These lead either to a significant deficit of flies carrying two transgenes (cross 20), or to normal viability of the sterile females carrying two transgenes (cross 15).

In summary, the effect of combinations of two different insertions of TfS is similar to the effect of homozygosity for a given transgene insertion. In each case, the presence of two TfS elements leads to abnormal phenotypes ranging from sterility to lethality. The milder effects observed when combining different TfS presumably reflect some aspect of the genetic background.

The 3' UTR is a regulator of ELAV protein levels:
Because two copies of the TfS minigenes lead to abnormalities, but two copies of TfL minigenes containing an elav 3' UTR are normal, it seemed possible that the elav 3' UTR was modulating elav expression. I therefore looked at the dosage dependence of elav expression respectively from a TfL minigene (350-83-1), from a TfS minigene [Tf(2) DmORF3], and from the endogenous locus, all of which provide full elav function (Table 1 and Table 3). Surprisingly, immunoblot analysis shows that the same amount of ELAV protein is produced from one or two doses of the endogenous elav locus, and from one or two doses of the TfL transgene 350-83-1 (Figure 2, lanes 1–4 and 9–11). In contrast, expression from the TfS transgene, missing the 3' UTR, is directly proportional to its copy number (Figure 2, lanes 5–8). Thus the 3' UTR confers gene copy number independence of ELAV expression.

View larger version (60K): In this window In a new window Download PPT slide   Figure 2. Production of ELAV in flies carrying one or two copies of elav minigene or gene: immunoblot analysis with the mouse anti-ELAV monoclonal antibody of protein extracts from the heads of males (10 µl of protein extracts at about 1 µg/µl in odd-numbered lanes and 5 µl in even-numbered lanes). Tubulin is used as an internal loading control. (1 and 2) elave5/Y; 350-83-1/350-83-1, (3 and 4) elave5/Y; 350-83-1/+, (5 and 6) elave5/Y; Tf(2)DmORF3/Tf(2)DmORF3, (7 and 8) elave5/Y; Tf(2)DmORF3/+, (9 and 10) y w/y+ sc Y, (11) y w/Y. The Y chromosome y+ sc Y carries a translocation of the X chromosome including elav+. The amount of ELAV protein in each band, as determined by laser scanner densitometry, is indicated below the gel in arbitrary units. Quantification of one other gel gives consistent results.

Not all TfS produce less protein than the endogenous locus. Two copies of the fully functional TfS minigene Tf(2) DmORF3 yield lower amounts (about 50%) of protein than wild-type flies (Figure 2). However, the same transgene inserted in a different location [Tf(2) DmORF2] yields normal amounts of ELAV (when present as two copies) but is also expressed in a gene dosage-dependent fashion (not shown).

There is an inverse correlation between expression of the endogenous elav locus and expression of the D. virilis minigene missing the 3' UTR:
Immunoblot analysis was used to examine the expression of ELAV protein in flies carrying different doses of the endogenous elav gene and of an elav minigene. A D. virilis minigene, which encodes a 55-kD ELAV containing an extended N terminus compared to its 50-kD D. melanogaster counterpart, was used (YAO and WHITE 1991 ). The D. virilis transgene differs from the D. melanogaster gene by the absence of the 3' UTR, as well as the replacement of D. melanogaster sequences with those of D. virilis in the 460 nucleotides most 3' of the second intron and through most of the open reading frame (ORF; Figure 1; YAO and WHITE 1991 ). Similar to the TfS 8.5-kb D. melanogaster transformants that provide elav⁺ function, the D. virilis transformants that provide elav⁺ function are homozygous lethal or sterile (Table 1).

When combining the two types of elav loci, I found that each locus affects the expression of the other. First, the truncated D. virilis elav minigene alone produces ELAV in direct relationship to its copy number (Figure 3, lanes 1, 2, and 7), as does the D. melanogaster minigene (Figure 2, lanes 5–8 ). However, expression of the D. virilis minigene is lowered in genetic combinations where the ratio between the copy numbers of the D. virilis and D. melanogaster loci is 1 (Figure 3, lanes 3, 4, and 6); the lower the ratio, the greater the reduction. Second, as demonstrated, the endogenous elav locus alone produces ELAV in a dose-independent fashion (Figure 2, lanes 9–11, and Figure 3, lanes 8 and 9). However, protein expression from the endogenous elav locus is reduced in genetic combinations where the ratio between the copy numbers of the D. virilis and D. melanogaster loci is 1 (Figure 3, lanes 3, 4, and 5); the higher the ratio, the greater the reduction.

View larger version (45K): In this window In a new window Download PPT slide   Figure 3. Production of ELAV in flies carrying zero, one, or two copies of the D. virilis minigene (encoding a 55-kD protein) and zero, one, or two copies of the D. melanogaster gene (encoding a 50-kD protein): immunoblot analysis with the mouse anti-ELAV monoclonal antibody of protein extracts from the heads of females (10 µl of protein extracts at about 1 µg/µl in lane 1 and 5 µl in other lanes). Tubulin is used as an internal loading control. (1 and 2) elave5/elave5; elavDvORF/CyO, (3) +/+; elavDvORF/elavDvORF, (4) elave5/+; elavDvORF/CyO, (5) elave5/+; elavDvORF/elavDvORF, (6) +/+; elavDvORF/CyO, (7) elave5/elave5; elavDvORF/elavDvORF, (8) +/+, (9) elave5/+. The amount of ELAV protein in each band, as determined by laser scanner densitometry, is indicated below the gel in arbitrary units. Quantification of two other gels gives consistent results. Note that the amount of D. virilis protein produced from the flies carrying one copy of D. virilis transgene and one copy of D. melanogaster minigene is close to, but slightly lower than the expected value.

Interestingly, the D. melanogaster gene and the D. virilis transgene do not respond coordinately to regulation. When the ratio between the copy numbers of the D. virilis and D. melanogaster loci is at its lowest (flies carrying one copy of the D. virilis minigene and two copies of the D. melanogaster gene), the D. virilis transgene produces about 60% (0.4/0.6 or 0.7) of normal D. virilis ELAV levels, but the D. melanogaster gene is normally expressed. In contrast, when the ratio reaches its highest (flies carrying two copies of the D. virilis minigene and one copy of the D. melanogaster gene), the D. melanogaster gene produces only about 30% (0.3/1) of normal D. melanogaster ELAV levels, but the D. virilis minigene is normally expressed. Indeed, there is an inverse correlation between the levels of the D. virilis ELAV and D. melanogaster ELAV (Figure 3, lanes 3–6). This suggests the existence of different, but correlated, modes of regulation of the D. virilis transgene and the endogenous D. melanogaster locus.

elav minigenes have differential effects, but are similarly expressed in males and females:
As noted previously, viability is more affected in males and fertility is more affected in females carrying two TfS transgenes (Table 2). Since the genomic elav locus is located on the X chromosome, it seemed plausible that these differences might reflect dosage compensation of elav transgenes when integrated on autosomes. When dosage compensated, a given X-linked gene is expressed at levels roughly twice as high in males as in females (KELLEY and KURODA 1995 ).

To examine this possibility, the level of elav function provided by transgenes was compared in males vs. females. Two situations were found regarding the efficiency of rescue by elav transgenes. First, transgenes carrying normal D. melanogaster elav sequences providing 25–100% of the normal ELAV level (Table 3) fully rescue elav^null function similarly in males and females (Table 3). Second, transgenes that carry mutated elav minigenes encoding an impaired ELAV protein and providing partial function rescue significantly better (at least five-fold) in males than in females (Table 3). I also examined levels of ELAV expression in males and females carrying either a TfL minigene that fully rescues (350-83-1, Table 3) or a mutant transgene TfL^m that partially rescues (353-66-2, Table 3). While, as seen by YAO et al. 1993 , there is some fluctutation in the level of ELAV protein between males and females, there is clearly much less than a twofold difference, relative to the control (Figure 4). Thus males and females express similar levels of ELAV protein in heads (Figure 4). It is a possibility that differential expression between males and females occurs earlier in development or in a subset of neurons, leading to the differential activity of elav transgenes in males and females. The sexual dimorphism of defects due to elav misexpression is reminiscent, for instance, of the sexual dimorphism seen with some alleles of mushroom body miniature that leads to central brain lesions (DE BELLE and HEISENBERG 1995 ) or of the homeotic gene Transabdominal (CELNIKER and LEWIS 1993 ). For these genes as for elav, the sexual dimorphism is unexpected because the genes are not, at least directly, implicated in sex determination, and the observation remains unexplained to date.

View larger version (70K): In this window In a new window Download PPT slide   Figure 4. Production of ELAV in males vs. females: immunoblot analysis with the mouse anti-ELAV monoclonal antibody of protein extracts from the heads of males (10 µl of protein extracts at about 1 µg/µl per lane, except 5 µl in lane 3). Tubulin is used as an internal loading control. (1) females elav+/elav+, (2 and 3) males elav+/Y, (4) females elave5/elave5; 350-83-1/350-83-1, (5) males elave5/Y; 350-83-1/350-83-1, (6) females elave5/elave5; 353-66-2/353-66-2, (7) males elave5/Y; 353-66-2/353-66-2. The amount of ELAV protein in each band, as determined by laser scanner densitometry, is indicated below the gel in arbitrary units.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The long 3' UTR is an important functional part of elav:
Detection of elav RNA and protein in situ shows that the expression of the gene is restricted to neurons (ROBINOW et al. 1988 ; ROBINOW and WHITE 1991 ). Multiple developmentally regulated mRNA transcripts have been detected (CAMPOS et al. 1987 ; YAO et al. 1993 ) that span approximatively 16 kb. Alternative forms differing in their 3' UTRs exist (Figure 1; CAMPOS et al. 1987 ; ROBINOW et al. 1988 ; YAO et al. 1993 ; M.-L. SAMSON, unpublished data). The two characterized elav cDNAs are each about 2.5 kb long. Both have a full length 5' end, contain the same ORF, and are truncated at their 3' end. These cDNAs each terminate within five nucleotides of the sequence 5'-A₆GA₅-3' (noncoding strand), suggesting that reverse transcription initiated in this region of the RNA rather than at its true 3' polyadenylated end (ROBINOW 1989 ). The size of elav transcripts (4.7 kb to 9.5 kb) is consistent with an extended 3' UTR. The complexity of RNA transcript patterns, their temporal regulation, and the differential expression of elav RNA in different parts of the nervous system are in contrast to the ubiquitous and continuous expression of the 50-kD ELAV in neurons (ROBINOW and WHITE 1991 ; SAMSON et al. 1995 ). In this article, the functional importance of the elav 3' UTR in the regulation of elav expression of the gene is directly demonstrated by analyzing the properties of elav minigenes.

Minigenes containing 8.5 kb of the elav gene including the ORF but lacking the normal 3' UTR provide apparently normal function when present as a single copy, although flies carrying two of these short minigenes have reduced viability and/or fertility. The dosage properties of the TfS minigenes differ from those of the endogenous locus and from those of the TfL transgenes carrying an additional 7 kb of the elav 3' UTR, because multiple copies of the latter are viable. In addition, TfS^m 8.5-kb minigenes with alterations (point mutations or oligonucleotide insertions) that eliminate elav function are homozygous viable, indicating that the 8.5-kb DNA fragment per se is not responsible for the observed defects. The defects resulting from the presence of two copies of TfS thus directly reflect the absence of the normal elav 3' UTR.

The 3' UTR is responsible for gene dosage independence of elav expression:
Insight into the function of the elav 3' UTR was gained by monitoring ELAV protein expression levels produced from the endogenous gene and from minigenes. Surprisingly, ELAV expression was found to be independent of elav gene dosage. Males carrying one (y w/Y) or two (y w/y⁺ sc Y) copies of the gene express similar levels of ELAV, as do females carrying one vs. two doses of the elav gene (unpublished results). Similar to the endogenous elav locus, transgenes carrying the 3' UTR show dosage independence of elav expression. In contrast, TfS minigenes produce ELAV in a dosage-dependent fashion. Thus, the 3' UTR is responsible for normalizing ELAV amount produced, regardless of the gene copy number.

It is interesting to note that, before generation of an ELAV antibody, BIER et al. 1988 proposed that the monoclonal antibody Mab44C11 identified elav protein, on the basis of the pattern of embryonic staining, the size of the corresponding antigen, and the absence of this antigen in elav mutants. However, reservations were expressed because the quantity of ELAV protein did not depend upon the copy number of the elav gene. The data presented here explain this discrepancy, because they provide direct evidence for gene dosage independence of elav expression. Mab44C11 thus is a bona fide ELAV antibody. Furthermore, the gene dosage independence of elav expression mediated via the 3' UTR suggests the existence of a feedback regulation mechanism by which control of ELAV synthesis occurs.

The elav gene autoregulates:
The possible autoregulation of elav was investigated by producing increasing amounts of a 55-kD D. virilis ELAV from a minigene (without 3' UTR) and examining levels of ELAV produced from the endogenous elav gene (with 3' UTR) encoding the 50-kD D. melanogaster ELAV.

The experiment reveals that the amounts of D. melanogaster ELAV and D. virilis ELAV are inversely correlated, in a fashion that works to maintain ELAV levels within a set range, indicating that both the D. melanogaster gene and the D. virilis minigene autoregulate. Autoregulation is consistent for a gene whose function is required not only for the initiation of differentiation, but also for its maintenance (for a review, see YAO et al. 1993 ). However, since elav gene and transgene expression are inversely correlated, the data suggest distinct mechanisms of autoregulation.

The specific autoregulation of the elav endogenous gene must depend upon sequences that are specific to the D. melanogaster gene. Candidates are the 3' UTR, and possibly the region of the D. melanogaster gene corresponding to 460 nucleotides most 3' of the second intron and most of the ORF, because they are replaced in the D. virilis transgene by the homologous D. virilis sequences. While the data do not allow exclusion of the region containing the second intron and the ORF of elav, there is no evidence for a regulatory role either. I favor the model in which autoregulation of the D. melanogaster gene depends upon the 3' UTR, shown in this article to confer gene dosage independence.

I propose that a second mechanism is responsible for the modulation of the level of the D. virilis minigene. It is likely that the D. melanogaster gene is also sensitive to this second level of autoregulation. The ratio of D. virilis and D. melanogaster ELAV levels is therefore the result of the balance between these regulatory mechanisms that work together to maintain ELAV levels within a set range.

ELAV protein contains three RNA recognition motifs diagnostic of a family of RNA-binding protein implicated in RNA metabolism and translation (BURD and DREYFUSS 1994 ). Indeed, ELAV binds homoribopolymers, as well as RNA corresponding to the 3' UTR of its own gene in vitro (C. D. BORGESON and M.-L. SAMSON, unpublished results). One of the mouse ELAV homologues, Mel-N1, also binds its own 3' UTR in vitro (ABE et al. 1996B ). On the basis of these properties, the possibility can be considered that 3' UTR-dependent autoregulation occurs through the direct binding of ELAV to the 3' UTR of elav RNA. Thus, it may be that ELAV promotes normal neuronal differentiation by binding to the 3' UTR of specific RNAs, influencing some aspect of their metabolism. Cis-regulatory elements present in the 3' UTR of mRNAs are known to influence the initiation of translation, intracellular localization, and multiple pathways of mRNA decay (reviewed in SACHS 1993 ; CURTIS et al. 1995 ; DREYFUSS et al. 1996 ). Such functions are consistent with the properties of RRM-containing proteins and could account for the role of ELAV.

The TfS behave as dosage-dependent neomorphic elav mutations:
The distinctive and unsual genetic properties of the transgenes shed some light on the abnormalities associated with the presence of two TfS copies. First, two copies of the TfS remarkably lead to abnormal fertility/viability, while one copy fully rescues elav^null function without conferring any deleterious effects. Second, the defects associated with the presence of two transgenes are seen independently of the elav⁺ or elav^null genetic backgrounds. Third, although the lowered viability of the individuals carrying two transgenes could be explained by altered elav expression, their increased sterility and the sex specificity of the defects are not easily explained by altered elav function, which is thought to be limited to development and maintenance of the nervous system. This third point suggests that a process other than that affected by elav might be altered in the flies carrying two TfS transgenes.

Therefore, it seems that the TfS produce or induce a molecule that is toxic to the flies past a given threshold. As shown, transgene dosage itself (DNA) does not explain the deleterious effects of two copies of TfS minigenes. The data presented suggest that the nuclear ELAV protein normally binds its 3' UTR, leading to autoregulation, by modulating a currently undetermined process in the metabolism of its RNA. The absence of a complete 3' UTR in the TfS alters this regulation, and it is thus possible that the observed deleterious effects seen in flies carrying two TfS minigenes result from the elevation of ELAV levels. Although I did not detect increased levels of ELAV in the proteins of head extracts from flies carrying two copies of transgenes, it is possible that elevated ELAV expression could happen either earlier in development and/or in a subset of neurons. Alternatively, it is possible that the production of an abnormal elav RNA, and not of the final gene product, triggers the deleterious phenotypes associated with two TfS copies. Going past a threshold of TfS RNA reached when two copies of TfS are present could lead to the observed deleterious effects, presumably by titrating a factor that binds the TfS RNA and forms an unprocessable complex. Clearly, additional experiments are required to test these possibilities.

Elucidating how the interaction of ELAV with the elav 3' UTR regulates the expression of the gene will contribute substantially to an understanding of ELAV function in promoting neuronal differentiation, assuming that the gene modulates the metabolism of RNA "targets" involved in neuronal differentiation in the same way it influences its own expression. Moreover, such understanding will also contribute more generally to the understanding of 3' UTR function. The minigenes provide a means to monitor the function of the elav 3' UTR in vivo, via assessing their viability in various genetic combinations. This tool, that has to my knowledge no equivalent in other systems, should greatly facilitate the analysis of elav function.

	ACKNOWLEDGMENTS

I thank B. Chase, M.-P. Furrer, L. Rabinow, and M. Wegnez for constructive comments on the manuscript, and H. Furneaux and J. Keene for communication of unpublished results. This work was supported by a Basic Research grant no. FY96-0995/FY97-0605 from the March of Dimes Birth Defect Foundation, and a University of Nebraska Medical Center seed grant no. 22-271-81201.

Manuscript received January 21, 1998; Accepted for publication June 24, 1998.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

ABE, R., E. SAKASHITA, K. YAMAMOTO, and H. SAKAMOTO, 1996a  Two different RNA binding activities for the AU-rich element and the poly(A) sequence of the mouse neuronal protein mHuC. Nucleic Acids Res. 24:4895-4901[Abstract/Free Full Text].

ABE, R., K. YAMAMOTO, and H. SAKAMOTO, 1996b  Target specificity of neuronal RNA-binding protein, Mel-N1: direct binding to the 3' untranslated region of its own mRNA. Nucleic Acids Res. 24:2011-2016[Abstract/Free Full Text].

BASHAW, G. J. and B. S. BAKER, 1997  The regulation of the Drosophila msl-2 gene reveals a function for Sex-lethal in translational control. Cell 89:789-798[Medline].

BIER, E., L. ACKERMAN, S. BARBEL, L. JAN, and Y. N. JAN, 1988  Identification and characterization of a neuron-specific nuclear antigene in Drosophila. Science 240:913-916[Abstract/Free Full Text].

BURD, C. G. and G. DREYFUSS, 1994  Conserved structures and diversity of functions of RNA-binding proteins. Science 265:615-621[Abstract/Free Full Text].

CAMPOS, A. R., D. GROSSMAN, and K. WHITE, 1985  Mutant alleles at the locus elav of Drosophila melanogaster lead to nervous system defects: a developmental-genetic analysis. J. Neurogenet. 2:197-218[Medline].

CAMPOS, A. R., D. R. ROSEN, S. N. ROBINOW, and K. WHITE, 1987  Molecular analysis of the locus elav in Drosophila melanogaster: a gene whose embryonic expression is neural specific. EMBO J. 6:425-431[Medline].

CELNIKER, S. E. and E. B. LEWIS, 1993  Molecular basis of transabdominal, a sexually dimorphic mutant of the bithorax complex of Drosophila. Proc. Natl. Acad. Sci. USA 90:1566-1570[Abstract/Free Full Text].

CHAGNOVICH, D. and S. L. COHN, 1996  Binding of a 40-kDa protein to the N-myc 3' untranslated region correlates with enhanced N-myc expression in human neuroblastoma. J. Biol. Chem. 271:33580-33586[Abstract/Free Full Text].

CHAGNOVICH, D., B. E. FAYOS, and S. L. COHN, 1996  Differential activity of ELAV-like RNA-binding proteins in human neuroblastoma. J. Biol. Chem. 271:33587-33591[Abstract/Free Full Text].

CHEN, C. Y. and A. B. SHYU, 1995  AU-rich elements: characterization and importance in mRNA degradation. Trends Biochem. Sci. 20:465-470[Medline].

CHUNG, S., L. JIANG, S. CHENG, and H. FURNEAUX, 1996  Purification and properties of HuD, a neuronal RNA-binding protein. J. Biol. Chem. 271:11518-11524[Abstract/Free Full Text].

CHUNG, S., M. ECKRICH, N. PERRONE-BIZZOZERO, D. KOHN, and H. FURNEAUX, 1997  The Elav-like proteins bind to a conserved regulatory element in the 3'-untranslated region of GAP-43 mRNA. J. Biol. Chem. 272:6593-6598[Abstract/Free Full Text].

CURTIS, D., R. LEHMANN, and P. D. ZAMORE, 1995  Translational regulation in development. Cell 81:171-178[Medline].

DE BELLE, J. S. and M. HEISENBERG, 1995  Genetic, neuroanatomical and behavioural analyses of the mushroom-body-miniature gene in Drosophila melanogaster. J. Neurogenet. 10:24.

DREYFUSS, G., M. HENTZE, and A. I. LAMOND, 1996  From transcript to protein. Cell 85:963-972[Medline].

GAO, F. B. and J. D. KEENE, 1996  Hel-N1/Hel-N2 proteins are bound to poly(A)+ mRNA in granular RNP structures and are implicated in neuronal differentiation. J. Cell Sci. 109:579-589[Abstract/Free Full Text].

GAO, F. B., C. C. CARSON, T. LEVINE, and J. D. KEENE, 1994  Selection of a subset of mRNAs from combinatorial 3' untranslated region libraries using neuronal RNA-binding protein Hel-N1. Proc. Natl. Acad. Sci. USA 91:11207-11211[Abstract/Free Full Text].

GOOD, P. J., 1995  A conserved family of elav-like genes in vertebrates. Proc. Natl. Acad. Sci. USA 92:4557-4561[Abstract/Free Full Text].

GRANADINO, B., L. O. F. PENALVA, M. R. GREEN, J. VALCÁRCEL, and L. SÁNCHEZ, 1997  Distinct mechanisms of splicing regulation in vivo by the Drosophila protein Sex-lethal. Proc. Natl. Acad. Sci. USA 94:7343-7348[Abstract/Free Full Text].

HIROTAKA, J. and R. DARNELL, 1997  A hierarchy of Hu RNA binding proteins in developing and adult neurons. J. Neurosci. 17:3024-3037[Abstract/Free Full Text].

HOMYK, T. J., K. ISONO, and W. L. PAK, 1985  Developmental and physiological analysis of a conditional mutation affecting photoreceptor and lobe development in Drosophila melanogaster. J. Neurogenet. 2:309-324[Medline].

JACOBSON, A. and S. PELTZ, 1996  Interrelationships of the pathways of mRNA decay and translation in eukaryotic cells. Annu. Rev. Biochem. 65:693-739[Medline].

JAIN, R. G., L. G. ANDREWS, K. M. MCGOWAN, P. H. PEKALA, and J. D. KEENE, 1997  Ectopic expression of Hel-N1, an RNA-binding protein, increases glucose transporter (GLUT1) expression in 3T3-L1 adipocytes. Mol. Cell. Biol. 17:954-962[Abstract].

JIMENEZ, F. and J. A. CAMPOS-ORTEGA, 1987  Genes in the subdivision 1B of the Drosophila melanogaster X chromosome and their influence on neuronal development. J. Neurogenet. 4:179-200[Medline].

KARESS, R. and G. M. RUBIN, 1984  Analysis of P-transposable element functions in Drosophila. Cell 38:135-146[Medline].

KELLEY, R. L. and M. I. KURODA, 1995  Equality for X chromosomes. Science 270:1607-1610[Abstract/Free Full Text].

KELLEY, R. L., J. WANG, L. BELL, and M. I. KURODA, 1997  Sex lethal controls dosage compensation in Drosophila by a non-splicing mechanism. Nature 387:195-199[Medline].

KENAN, D. J., C. C. QUERY, and J. D. KEENE, 1991  RNA recognition: towards identifying determinants of specificity. Trends Biochem. Sci. 16:214-220[Medline].

KIM, Y. J. and B. S. BAKER, 1993  The Drosophila gene rbp9 encodes a protein that is a member of a conserved group of putative RNA binding proteins that are nervous system-specific in both flies and humans. J. Neurosci. 13:1045-1056[Abstract].

KING, P. H., 1994  Hel-N2: a novel isoform of Hel-N1 which is conserved in rat neural tissue and produced in early embryogenesis. Gene 151:261-265[Medline].

KING, P. H., T. D. LEVINE, R. T. FREMEAU, JR., and J. D. KEENE, 1994  Mammalian homologs of Drosophila ELAV localized to a neuronal subset can bind in vitro to the 3' UTR of mRNA encoding the Id transcriptional repressor. J. Neurosci. 14:1943-1952[Abstract].

KOUSHIKA, S. P., M. J. LISBIN, and K. WHITE, 1996  ELAV, a Drosophila neuron-specific protein, mediates the generation of an alternatively spliced neural protein isoform. Curr. Biol. 6:1634-1641[Medline].

LEVINE, T. D., F. GAO, P. H. KING, L. G. ANDREWS, and J. D. KEENE, 1993  Hel-N1: an autoimmune RNA-binding protein with specificity for 3' uridylate-rich untranslated regions of growth factor mRNAs. Mol. Cell. Biol. 13:3494-3504[Abstract/Free Full Text].

LIU, J., J. DALMAU, A. SZABO, M. ROSENFELD, J. HUBER, and H. FURNEAUX, 1995  Paraneoplastic encephalomyelitis antigens bind to the AU-rich elements of mRNA. Neurology 45:544-550[Abstract/Free Full Text].

MA, W. J., S. CHENG, C. CAMPBELL, A. WRIGHT, and H. FURNEAUX, 1996  Cloning and characterization of HuR, a ubiquitously expressed Elav-like protein. J. Biol. Chem. 271:8144-8151[Abstract/Free Full Text].

MA, W.-J., S. CHUNG, and H. FURNEAUX, 1997  The Elav-like proteins bind to AU-rich elements and to the poly(A) tail of mRNA. Nucleic Acids Res. 25:3564-3569[Abstract/Free Full Text].

MANLEY, G., 1994 Identification and analysis of anti-Hu paraneoplastic antigens. Ph.D. Thesis, Cornell University Medical College, New York.

MATTAJ, I. W., 1993  RNA recognition: a family matter? Cell 73:837-840[Medline].

MYER, V. E., X. C. FAN, and J. A. STEITZ, 1997  Identification of HuR as a protein implicated in AUUUA-mediated mRNA decay. EMBO J. 16:2130-2139[Medline].

PERRON, M., L. THEODORE, and M. WEGNEZ, 1995  Isolation and embryonic expression of Xel-1, a nervous system-specific Xenopus gene related to the elav gene family. Mech. Dev. 51:235-249[Medline].

PERRON, M., P. BOURLITIO, M. WEGNEZ, and L. THEODORE, 1997  Subcellular distribution of Xenopus XEL-1 protein, a member of the neuron-specific ELAV/Hu family, revealed by epitope tagging. DNA Cell Biol. 16:579-587[Medline].

PIRROTTA, V., 1988 Vectors for P-mediated transformation in Drosophila, pp. 437–456 in Vectors: A Survey of Molecular Cloning Vectors and Their Uses, edited by R. L. RODRIGUEZ and D. T. DENHARDT. Butterworth, Boston.

ROBINOW, S. and K. WHITE, 1991  Characterization and spatial distribution of the ELAV protein during Drosophila melanogaster development. J. Neurobiol. 22:443-461[Medline].

ROBINOW, S., A. R. CAMPOS, K. M. YAO, and K. WHITE, 1988  The elav gene product of Drosophila, required in neurons, has three RNP consensus motifs. Science 242:1570-1572[Abstract/Free Full Text].

ROBINOW, S. N., 1989 The elav gene of Drosophila melanogaster encodes a neuron-specific RNA binding protein which is required for the development and maintenance of the nervous system. Ph.D. Thesis, Brandeis University, Waltham, MA.

SACHS, A. B., 1993  Messenger RNA degradation in eukaryotes. Cell 74:413-421[Medline].

SAKAI, K., M. GOFUKU, Y. KITAGAWA, T. OGASAWARA, and G. HIROSE et al., 1994  A hippocampal protein associated with paraneoplastic neurologic syndrome and small cell lung carcinoma. Biochem. Biophys. Res. Commun. 199:1200-1208[Medline].

SAMSON, M.-L., M. J. LISBIN, and K. WHITE, 1995  Two distinct temperature-sensitive alleles at the elav locus of Drosophila melanogaster are suppressed nonsense mutations of the same tryptophan codon. Genetics 141:1101-1111[Abstract].

STELLER, U., S. KOHLS, B. MULLER, R. SOLLER, and R. MULLER et al., 1996  The RNA binding protein HuD: rat cDNA and analysis of the alternative spliced mRNA in neuronal differentiating cell lines P19 and PC12. Brain Res. Mol. Brain Res. 35:285-296[Medline].

SZABO, A., J. DALMAU, G. MANLEY, M. ROSENFELD, and E. WONG et al., 1991  HuD, a paraneoplastic encephalomyelitis antigen, contains RNA-binding domains and is homologous to Elav and Sex-lethal. Cell 67:325-333[Medline].

THUMMEL, C. S., A. M. BOULET, and H. D. LIPSHITZ, 1988  Vectors for Drosophila P-element mediated transformation and tissue-culture transfection. Gene 74:445-456[Medline].

WAKAMATSU, Y. and J. WESTON, 1997  Sequential expression and role of Hu RNA-binding proteins during neurogenesis. Development 124:3449-3460[Abstract].

YANNONI, Y. M. and K. WHITE, 1997  Association of the neuron-specific RNA binding domain-containing protein ELAV with the coiled body in Drosophila neurons. Chromosoma 105:332-341[Medline].

YAO, K.-M. and K. WHITE, 1991  Organizational analysis of elav gene and functional analysis of ELAV protein of Drosophila melanogaster and Drosophila virilis. Mol. Cell. Biol. 11:2994-3000[Abstract/Free Full Text].

YAO, K. M., M. L. SAMSON, R. REEVES, and K. WHITE, 1993  Gene elav of Drosophila melanogaster: a prototype for neuronal-specific RNA binding protein gene family that is conserved in flies and humans. J. Neurobiol. 24:723-739[Medline].

This article has been cited by other articles:

				G. Toba and K. White The third RNA recognition motif of Drosophila ELAV protein has a role in multimerization Nucleic Acids Res., March 27, 2008; 36(4): 1390 - 1399. [Abstract] [Full Text] [PDF]

				A. Ratti, C. Fallini, C. Colombrita, A. Pascale, U. Laforenza, A. Quattrone, and V. Silani Post-transcriptional Regulation of Neuro-oncological Ventral Antigen 1 by the Neuronal RNA-binding Proteins ELAV J. Biol. Chem., March 21, 2008; 283(12): 7531 - 7541. [Abstract] [Full Text] [PDF]

				A. Ratti, C. Fallini, L. Cova, R. Fantozzi, C. Calzarossa, E. Zennaro, A. Pascale, A. Quattrone, and V. Silani A role for the ELAV RNA-binding proteins in neural stem cells: stabilization of Msi1 mRNA. J. Cell Sci., April 1, 2006; 119(Pt 7): 1442 - 1452. [Abstract] [Full Text] [PDF]