| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Molecular Biology,
* Division of Neurosciences,
Division of Biology, Beckman Research Institute of the City of Hope, Duarte, California 91010, USA; and
Forschung Biochemie, Aventis Crop Science, 65926 Frankfurt, Germany
1Correspondence: Department of Molecular Biology and Graduate School of Biological Sciences, Beckman Research Institute of the City of Hope, 1450 E. Duarte Rd., Duarte, CA 91010, USA. E-mail: jrossi{at}coh.org
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: 3'UTR • RNA localization • hammerhead ribozyme • Drosophila bicoid and oskar mRNAs
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Ribozymes have been shown to function in trans and can be directed against foreign target sequences by flanking the catalytic core with sequences complementary to the target (2
, 3)
. The hammerhead is the smallest of the known ribozyme motifs and therefore readily amenable to experimental manipulation (for a review, see ref 4
). Hammerhead ribozymes, therefore, have broad potential as therapeutic agents for the selective control of gene expression (see refs 1
, 3
, 5
for a review). However, two of the most important problems confronting the use of hammerhead ribozymes as genetic tools or therapeutic agents are finding accessible binding cleavage sites and colocalization with target RNAs.
One method for ribozyme target colocalization takes advantage of the properties of some messenger RNAs (mRNAs) to localize within specific subcellular compartments. The first evidence for cytoplasmic mRNA localization came from the observation that actin transcripts are unevenly distributed in ascidian embryos (6)
. Subsequently, several maternal mRNAs that are localized during oogenesis were identified in Drosophila (7)
and Xenopus (8)
; many mRNAs are localized in neurons (9
10
11)
and oligodendrocytes (12)
. Localized mRNAs have also been discovered in somatic cells (13
, 14)
. In myoblasts, different mRNAs can occupy different cytoplasmic compartments. ß-Actin mRNA is localized to the leading lamellae and 
-actin mRNA is localized to the perinuclear area via 3' untranslated regions (UTRs) (14)
. Subcellular localization signals for several other mRNAs are also positioned in the 3' UTR. We have previously used the human - and ß-actin 3'UTRs as signals for colocalizing hammerhead ribozymes with a lacZ target mRNA (15)
. The percentage of colocalization using the matched - or ß-actin 3'UTR (– or ß–ß) was enhanced 
threefold relative to unmatched 3'UTRs. The increase in ribozyme-mediated inhibition of ß-galactosidase activity observed when matched 3'UTRs were used was consistent with the percentage of colocalization.
In Drosophila, the maternal bicoid and oskar mRNAs are localized to the anterior and posterior poles of the egg, respectively. The spatial organization of their encoded protein products is thought to be essential for establishing the basic body plan of the fly (16
17
18
19
20
21
22)
. In Xenopus, Vg1 mRNA localizes to the vegetal pole of the oocyte (8
, 23)
. In Saccharomyces cerevisiae, ASH1 mRNA is tightly localized within the budding daughter cell (24
, 25)
. Each mRNA has its localization signal in the 3'UTR. These localized messages are tightly confined to their subcellular sites during specific stages of development or during yeast budding, and thus may provide stringent tests for colocalizing ribozymes with target mRNAs. The present study provides the first direct test of using mRNA localization signals in an animal by taking advantage of the discrete localization of oskar and bicoid mRNAs in Drosophila oocytes. The strategies for ribozyme target site selection and colocalization might be applied to endogenous gene transcripts as a surrogate genetic approach for studying gene function.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
) were kindly provided by the Ephrussi lab. To obtain ribozyme expressing transgenic flies, P element-mediated germ line transformation of Drosophila melanogaster was performed using a standard protocol (27)
. Ribozyme (Rbz) constructs described below were dissolved in 0.1 mM potassium phosphate buffer (pH 7.8) containing 5 mM KCl at 0.8 mg/ml along with 0.1 mg/ml of pTurbo helper plasmid: pCaSperRosk5'-Rbz3-osk3'UTR and pCaSperRosk5'-Rbz3-bcd3'UTR (Fig. 1
). A mutant ribozyme was used as a negative control for ribozyme function. The recipients for ribozyme constructs were white- and transformants were selected by eye color. Linkage analyses of the inserted P elements were carried out by testing for segregation from second or third balancer chromosomes marked with Curly (Cy) and Ultrabithorax (Ubx), respectively.
|
Ribozyme constructs
The hammerhead ribozyme motif used in this study contains 12 and 10 nucleotides flanking the catalytic core that are complementary to the lacZ message (Fig. 2
C). To make catalytically inactive ribozyme variants, we mutated G5 in the core to an A (28)
. BstEII-BamHI restriction fragments containing the ribozyme sequences were prepared from synthetic oligonucleotides (upper: 5'-gggtaacccCAGGCTGCGCCTGATGAGTCCGTGAGGAC; lower: 5'-cgggatcccgCTTCCCAACAGTTTCGTCCTCACGGAC), which share 12 bases of complementary sequence (underlined) at their 3' ends (29)
. The double-stranded ribozyme gene was then amplified using several rounds of PCR. The wild-type and mutant ribozyme genes were cloned into the BstEII-BamHI site of pSP72osk5'-lacZ-osk3'UTR after removing the lacZ gene using the BstEII and BamHI restriction enzymes, yielding pSP72osk5'-Rbz3-osk3'UTR (Fig. 1)
. To clone into pCaSpeR, the BstEII-NotI fragment containing the lacZ and osk3'UTR sequences was removed from pCaSpeRosk5'-lacZ-osk 3'UTR, and the BstEII-NotI fragment containing the ribozyme and osk3'UTR sequences from pSP72osk5'-Rbz3-osk3'UTR was subcloned into the sites of pCaSpeRosk5'-lacZ-osk 3'UTR (Fig. 1)
, yielding pCaSperRosk5'-Rbz3-osk3'UTR. For cloning the bcd3'UTR, a XbaI fragment containing bcd3'UTR from p2411 (gift of Paul Macdonald, University of Texas, Austin, TX) was rendered blunt-ended and ligated into the SmaI site of pBluescriptTM KS+ to yield pBS KS+-bcd3'UTR (figure not shown). The BamHI-EcoRV fragment containing the bcd3'UTR from the pBS KS+-bcd 3'UTR was cloned into a BamHI-NotI (blunt) site of pSP72osk5'-Rbz3-osk3'UTR after removing the osk3'UTR, yielding pSP72osk5'-Rbz3-bcd3'UTR. The BstEII-EcoRV fragment containing the ribozyme and bcd3'UTR was then subcloned into the BstEII-EcoRV site of pCaSpeRosk5'-Rbz3-osk 3'UTR after removing the ribozyme and osk3'UTR, yielding pCaSperRosk5'-Rbz3-bcd3'UTR (Fig. 1)
. The same method described above was used to clone the mutant ribozyme gene. Each construct was analyzed and the structures were confirmed by restriction enzyme analysis and DNA sequencing.
|
Crosses
Virgin females from each lacZ transgenic fly line (with osk or bcd3'UTRs) were mated to the ribozyme transgenic males: 1) wtRbz3-osk3'UTR, 2) mtRbz3-osk3'UTR, 3) wtRbz3-bcd3'UTR, 4) mtRbz3-bcd3'UTR, and 5) UAS-GFP (mock control) (Fig. 1)
. After 10 days, F1 female progeny from each cross were dissected to obtain egg chambers or ovaries for RNA localization and ß-galactosidase analyses
Accessibility of antisense oligodeoxyribonucleotides (ASOs) and ribozymes
Ovaries were removed by manual dissection from the lacZ transgenic females reared at 25°C on yeast for 2–3 days and placed in ice-cold PBS (137 mM NaCl, 2.6 mM KCl, 10 mM NaHPO4, 1.7 mM KH2PO4) supplemented with 0.1% Triton X-100 (PBT). The following experiments were performed on ice. Ovaries were transferred individually to microfuge tubes containing 100 µl of cold PBS with RNasin® [40 U; Promega, Madison, WI (PBR)] and homogenized using a pestle inserted directly into the microfuge tubes, followed by centrifugation at 15,000 rpm. The supernatants, which contain endogenous mRNAs, RNase H, and RNA binding proteins, were used in experiments to probe for accessibility of the lacZ mRNA for binding of ASOs and ribozymes. The RNase H-mediated mRNA cleavage experiments were performed in a total volume of 30 µl as described previously (30)
, with minor modifications. Ovary extract (27 µl) was incubated with 1 mM DTT, RNasin®, and 50 nM of each ASO for 5 min or ribozyme for 20 min at 37°C. To prevent further activity of the ribozyme, 200 nM of an oligonucleotide complementary to the ribozyme sequence was added and incubated for 5 min at 37°C. The mixtures were digested with 10 µl of DNase I (10 U/µl) for 1 h at 37°C, followed by phenol extraction and ethanol precipitation. The precipitates were resuspended in 8 µl of DEPC water. Reverse transcription using 4 µl of the resuspension was carried out in a total volume of 20 µl, using Moloney murine leukemia virus (Mo-MLV) reverse transcriptase (RT), according the manufacturer’s instructions (Life Technologies, Grand Island, NY). First-strand priming was performed with 50 ng of random hexamer primers and 1 µl of Mo-MLV RT (10 U/µl). Seventeen microliters of the RT reaction were used for further PCR amplification with a tetrachlorofluorescein (Tet, red; Glen Research, Sterling, VA) -labeled 5' primer and a 3' unlabeled primer for lacZ mRNA. As an internal control, 3 µl of the RT reaction was used for amplifying endogenous ß-tubulin with a 5' unlabeled primer and a 3' primer labeled with fluorescein (green: Glen Research, VA). The PCR cycling for lacZ was 1 min at 95°C, 1 min at 62°C, and 1 min at 72°C for 40 cycles. The same cycling conditions were used for ß-tubulin, except that 30 cycles of PCR were carried out. All of the reactions were performed in a volume of 50 µl. The RT-PCR products were first visualized after electrophoresis in a 2% agarose gel, then analyzed and quantified on an Applied Biosciences PrismTM 377 DNA Sequencer using Genescan analysis (version 2.1, ABI).
RNA detection
Ovaries were dissected from F1 females derived from crosses with the ribozyme and the target lacZ expressing transgenic flies. For the bcd3'UTR, ovary extracts were obtained as described above. For the osk3'UTR, ovaries were further dissected with sharp forceps to obtain stage 10 egg chambers identified by oocytes occupying more than 50% of the volume. The egg chambers were transferred to 48-well plates maintained on ice. The chambers were broken using a needle under a dissecting microscope. Subsequently, 100 µl of cold PBR was added to each well, followed by mixing, and the solutions were transferred to microfuge tubes. The ovary homogenates or the egg chamber solutions were centrifuged for 10 min at 14,000 rpm at 4°C in a microcentrifuge. The supernatants were collected in new tubes to detect RNA using RT-PCR. The extracts were digested with 10 µl of DNase I (10 U/µl) for 1 h at 37°C, followed by phenol extraction and ethanol precipitation. The precipitates were resuspended in 8 µl of DEPC water. RT-PCR was performed as described above with minor modification. From 20 µl of the RT reaction, 5.5 µl for lacZ mRNA, 12.5 µl for ribozyme, and 2 µl for endogenous ß-tubulin (as an internal control) were used for further PCR amplification. Each primer for PCR is shown in Fig. 1
. The primer sequences are: for lacZ, 5' TCACCAGCCGGCAGAGCAG and 3' AGCGCCCGTTGCACCACAGA labeled by Tet (582 nt); for ribozyme, 5' ATGAGTCCGTGAGGACGAAACTG labeled by fluorescein and 3' GCATTTGATTATTTTACACAGCT for osk3'UTR (395 nt) and GGGTAATCATTGTATGAGATTACG for bcd3'UTR (435 nt); and for ß-tubulin, 5' CCTTGGCGGCGGCACTGG and 3' AGATGGCGGCGACGGTAAGGTAGC labeled by Tet (536 nt). The PCR cycling for lacZ and ribozyme was 1 min at 95°C, 1 min at 62°C, and 1 min at 72°C for 35 cycles. The same cycling conditions were used for ß-tubulin, except the number of cycles was 30.
Analysis of lacZ expression in egg chambers from the lacZ transgenic females
Ovaries were removed from females reared at 25°C on yeast for 2–3 days by manual dissection in PBT. ß-Galactosidase activity was determined as described previously, with minor modifications (31)
. The ovaries from each female were further dissected with sharp forceps to obtain the egg chambers, which were fixed with 4% formaldehyde in PBS for 20 min at room temperature on a shaking platform at 350 rpm. The egg chambers were washed once with PBT and incubated in PBT for 5 min at room temperature. The PBT was removed and 1 ml X-gal staining solution (10 mM sodium phosphate, pH 7.2, 150 mM NaCl, 1 mM MgCl2, 3 mM K4[Fe(CN)6], 3 mM K3[Fe(CN)6], 0.3% Triton®X-100, 20 µl of 10% X-gal) were added, followed by incubation at 37°C for 4 h or at 4°C for overnight. The staining solution was removed and samples were rinsed with PBT. The X-gal-stained egg chambers were mounted on slides in 70% glycerol/PBS. Specimens were analyzed and photographed with an Olympus BX50 microscope and a DEI-750 video camera (Optronics, Goleta, CA) or an Olympus AX70 microscope and a Real14 CCD camera.
Quantitation of ribozyme-mediated reduction in ß-galactosidase activity
Ovaries were dissected from F1 females derived from crosses with the ribozyme and the target lacZ expressing transgenic flies. UAS-GFP transgenic flies were used as a negative control for ribozyme. Ovary extracts for the bcd3'UTR and stage 10 egg chamber solutions for the osk3'UTR were obtained as described above for RNA detection. The fluorogenic substrate 3-carboxyumbeliferyl ß-D-galactopyranoside (CUG; Diagnostic Chemicals Ltd., Oxford, CT) was used for assaying ß-galactosidase activity in the ovary or egg chamber extracts. Fifty microliters of the extract was mixed with 100 µl of reaction buffer (100 mM sodium phosphate buffer, pH 7.0, 1 mM MgCl2, 10 mM ß-mercaptoethanol, and 0.1% Triton X-100) and incubated for 10 min at room temperature. Fifty microliters of 1 mM CUG was added to each mixture. After incubation for 1 h at 37°C, 100 µl of stop buffers (500 mM glycine, 10 mM EDTA, pH 12) was added to each tube. After incubation for 10 min at room temperature, the fluorescence was determined with a Fluorometer (Bio-Rad VersaFluorTM Quick) at 460 nm using a 390 nm excitation filter.
RNA localization
Whole mount in situ hybridization of ovaries was performed as described previously (32
, 33)
with some modifications. Ovaries were dissected in PBT from well-fed 2- or 3-day-old lacZ or ribozyme transgenic females or F1 progeny females obtained after appropriate crossing. Prehybridization, hybridization, and washing were performed at 55°C. RNA probes were prepared by in vitro transcription using digoxygenin- or biotin-labeled UTP from the DNA templates containing the lacZ (pBSKS+-HHlacZ) or ribozymes (pSP72osk5'-Rbz3-no3'UTR) (Fig. 1)
using the Genius4 RNA labeling kit (Boehringer Mannheim, Mannheim, Germany). After ethanol precipitation, the probe was resuspended in 100 µl RNase-free water. For the lacZ RNA probe (630 nt), alkaline hydrolysis was performed to reduce the size of the RNA probes to less than 250 nt.
After color detection of alkaline phosphatase (AP) for 5-bromo-4-chloro-3-indolyl phosphate and 4-nitro blue tetrazolium chloride, the stained ovaries were washed several times in PBT for a total of 45 min to stop the AP reaction and equilibrated overnight at 4°C in 60% glycerol in PBT. They were then mounted in 60% glycerol in PBT and observed as described above.
For localization of mtRbz-3'bcd, ovaries were incubated for 30 min in PBT/1% bovine serum albumin at room temperature after hybridization and washing. The ovaries were washed 3x for 10 min each in PBT. An Extravidin (1/2000 of 1 mg/ml, Sigma) in PBT was added and incubated for 2 h at 25°C. After washing 4x with PBT for 30 min, an anti-avidin-biotin conjugate (1/2000, Sigma) in PBT was added and incubated for 1.5 h at 25°C. After washing 4x with PBT for 30 min, an Extravidin-Cy3 conjugate (1/2000, Sigma) in PBT was added and incubated for 1.5 h at 25°C. The ovaries were washed in PBT for 1–2 h, mounted in 70% glycerol + 2% DABCO in PBT, and visualized microscopically with a rhodamine filter under an Olympus 10x objective.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). RNase H selectively degrades RNA in RNA/DNA hybrids and the amount of cleavage is directly proportional to accessibility of the ASOs to base pairing (30)
. The reduction in substrate was evaluated by RT-PCR using primers that flank the regions chosen for site selection (Fig. 2A
). Quantitation of the target RNA reduction was performed by comparative, quantitative Genescan analysis of the amplified target RNA vs. amplification of a nontargeted internal standard, endogenous ß-tubulin mRNA. Results from these analyses are summarized in Table 1
. The greatest reduction (44%) was directed by ASO3. In contrast to ASO3, ASO1 gave very poor reduction of the RT-PCR lacZ product (8%). When the sites screened by antisense selection were used for ribozyme screening, ribozyme 3 (Rbz3) resulted in a 45 ± 1% reduction of the endogenous lacZ product in the fly ovary extracts. This result is consistent with the percent reduction obtained with ASO3 and with our previous results comparing antisense and ribozyme accessibility (34
, 35)
. The target site for ASO3, and hence Rbz3, was selected for application in the in vivo colocalization studies.
|
Localization of lacZ target mRNA and ß-galactosidase protein in oocytes
The 3' UTRs from the bicoid or oskar mRNAs (bcd3'UTR or osk3'UTR) are necessary and sufficient for correct localization of the transcripts and capable of localizing heterologous transcripts to the anterior or posterior regions (respectively) of oocytes during oogenesis (18
, 36
37)
. pCaSpeRosk5'-lacZ-osk3'UTR contains 414 nt of the 5'osk derepressor element for translation upstream of the lacZ coding sequence, allowing expression of the lacZ gene at the posterior part of the oocyte (26)
. Otherwise, expression of the lacZ gene could not be detected at the oocyte posterior in absence of the 5'osk derepressor element. To confirm the localization of the lacZ target RNAs using the bcd3'UTR or the osk3'UTR and analyze their expression patterns, we performed X-gal staining and in situ hybridization of the mRNA in the lacZ transgenic fly egg chambers. We focused on oogenesis stages 9 through 14 for characterization of the subcellular distribution of LacZ protein and lacZ mRNA in egg chambers, since the anterior and posterior regions are easily distinguished in oocytes during these stages and normal oskar and bicoid mRNAs are discretely localized at these stages (18
, 36
37
38)
. The ovaries dissected from pCaSpeRosk5'-lacZ-osk3'UTR transgenic females were separated into ovarioles and then into independent egg chambers. The separated egg chambers were stained using X-gal (Fig. 3
). The osk3'UTR directed lacZ mRNA and hence ß-galactosidase expression to the posterior region of the oocytes in stage 9 through 14 egg chambers (Fig. 3A
). ß-Galactosidase localization using the osk3'UTR was more discrete in stage 10 egg chambers (Fig. 3A
), where the oocyte occupies half of the egg chamber. At stage 14, X-gal staining diffused throughout the egg chambers, with a slightly stronger localization in the posterior region. To confirm posterior localization of lacZ mRNA using the osk3'UTR, an antisense lacZ RNA probe was used for in situ hybridization in the separated egg chambers. The osk3'UTR directed the lacZ message to the posterior region of the oocytes in stage 9 through 14 egg chambers (data not shown). Posterior localization of lacZ mRNA was similar to X-gal staining in that it was slightly more discrete in stage 10 egg chambers than in stage 14 (Fig. 4
A). Stage 10 egg chambers were therefore selected for additional studies examining colocalization of target and ribozyme transcripts using the osk3'UTR
|
|
In contrast to the osk3'UTR, the bcd3'UTR resulted in LacZ localization in the anterior region of stage 9 through 14 egg chambers during oogenesis (Fig. 3B
). At stage 9, X-gal staining was still present in nurse cells, subsequently moving to the anterior of the oocyte at stage 10. A little more discrete staining and hence ß-galactosidase localization at the anterior of the oocyte was observed in stage 14 egg chambers, even though diffuse X-gal staining was observed throughout the egg chambers. To determine the localization of lacZ mRNA, we used antisense lacZ RNA probes for in situ hybridization on the separated egg chambers (Fig. 4B
). The bcd3'UTR directed the lacZ mRNAs to the anterior region of the oocytes in stage 9 through 14 egg chambers (data not shown). As observed for the protein, localization of lacZ mRNA in later stages (stage 14) was a little more discrete than at stage 10 (Fig. 4B
). For quantitation of the ß-galactosidase activity and RNA expression with the bcd3'UTR, we used ovaries rather than individual egg chambers since they are primarily occupied with later stage oocytes.
Localization of ribozymes in egg chambers from transgenic flies
Using the site accessibility assay described in Materials and Methods, the sequence complementary to ASO3 and Rbz3 was determined to be the most accessible. Functional and mutant ribozymes for this target site were constructed (wtRbz3 and mtRbz3) and fused to the osk3'UTR or bcd3'UTR. These constructs were expressed from the same promoter system used to express the lacZ target mRNA (posk) (Fig. 1)
and include wtRbz3-osk3'UTR, mtRbz3-osk3'UTR, wtRbz3-bcd3'UTR, and mtRbz3-bcd3'UTR. Transformation recipients for the ribozyme constructs were screened by eye color (red) and the presence of the ribozyme gene was confirmed by PCR using genomic DNA as the template (data not shown). Linkage analyses of the transgenes indicated that wtRbz3-osk3'UTR, mtRbz3-osk3'UTR, and mtRbz3-bcd3'UTR were integrated into chromosome 2 and wtRbz3-bcd3'UTR was integrated into chromosome 3.
In situ hybridization was carried out to investigate the localization of ribozyme transcripts in egg chambers obtained from the ribozyme containing transgenic flies (Fig. 5
). Discrete localization of the ribozymes was seen in the posterior regions of oocytes at stage 10 egg chambers using the osk3'UTR (Fig. 5A
). The osk3'UTR directed both wt and mt ribozyme transcripts to the posterior regions of oocytes at stage 9 through 14 egg chambers (data not shown). As anticipated, wt and mt ribozymes harboring the bcd3'UTR were confined to the anterior section of oocytes at stage 9 through 14 egg chambers. The localization signal from the bcd3'UTR facilitated more discrete localization of mutant ribozyme constructs to the anterior parts of oocytes in later stage egg chambers (Fig. 5B
) even though the differences were not clear between stages 10 and 14. Stage 10 egg chambers were therefore selected for further studies incorporating the osk3'UTR, and ovaries were selected for studies using the bcd3'UTR.
|
Enhancement of in vivo ribozyme function using colocalization with synonymous 3'UTRs
Virgin transgenic females containing lacZ with either the osk or bcd 3'UTR (lacZ-osk3'UTR or lacZ-bcd3'UTR) were crossed with the various ribozyme transgenic flies to establish the desired combinations of ribozyme and target 3'UTRs. The wtRbz3-bcd3'UTR males were crossed to the lacZ-osk3'UTR females for use as noncolocalization controls. A nonribozyme-containing UAS-GFP construct was used as a mock control. After 10 days, 8–10 F1 females from each cross were dissected to obtain stage 10 egg chambers from different animals. The chambers were stained for ß-galactosidase activity with X-gal and the results are shown in Fig. 6
. Colocalization of the ribozyme with the lacZ target using localization signals from synonymous osk3'UTRs (osk-osk) resulted in enhanced inhibition of ß-gal activity relative to the noncolocalized (osk-bcd) or negative control (osk-GFP) (Fig. 6)
. When target and ribozyme both contained the same osk3'UTR (osk-osk), eight of ten egg chambers had no detectable ß-gal activity at the posterior ends, whereas two egg chambers showed some staining (Fig. 6A
). In contrast, 50% of the egg chambers expressing the noncolocalized osk-bcd transcripts had readily detectable ß-gal activity at the posterior ends (Fig. 6B
) whereas all of the nonribozyme-expressing eggs (osk-GFP) stained at the posterior ends (Fig. 6C
). In the case of the bcd3'UTR, 10 F1 females from each cross were dissected to obtain stage 14 egg chambers from different animals. Eight of 10 egg chambers expressing the colocalized bcd-bcd transcripts had barely detectable ß-gal activity at the anterior ends (80% inhibition) whereas 60% of the egg chambers expressing the noncolocalized bcd-osk transcripts had readily detectable ß-gal activity at the anterior ends (40% inhibition); all of the nonribozyme-expressing eggs (bcd-GFP) stained strongly at the anterior ends (0% inhibition) (data not shown).
|
To quantitate the reduction of ß-gal activity, ß-galactosidase assays were performed on stage 10 egg chamber extracts for osk3'UTR or ovary extracts for bcd3'UTR constructs (Fig. 7
) using the fluorogenic ß-galactosidase substrate CUG. When the synonymous osk3'UTR was used for the substrate and ribozyme (osk-osk), ß-gal activity was reduced by 57% (P<0.0002) relative to the negative control (osk-Gfp) and 52% (P<0.007) relative to the noncolocalized combinations (osk-bcd) (Fig. 7A
). In the case of bcd3'UTR, the mtRbz3-bcd3'UTR males were crossed to the lacZ-bcd3'UTR females to establish controls for ribozyme function. The ß-gal activity also was reduced by 44% (P<0.02) relative to the negative control (bcd-Gfp), 59% (P<0.001) relative to the noncolocalized combinations (bcd-osk), and 31% (P<0.04) relative to the mutant ribozyme (bcd-bcd*) (Fig. 7B
). These results verify that colocalization of the substrate and the ribozyme improves ribozyme function. The inhibitory activity of the colocalized mutant ribozyme (although less than the functional ribozyme) further supports the enhancing function of colocalization by increasing the opportunities for antisense activity.
|
Reduction of lacZ mRNA transcripts by colocalization of target and ribozyme RNAs
To evaluate the reduction in target lacZ mRNAs via ribozyme cleavage, RT-PCR assays were performed on stage 10 egg chamber RNAs for osk3'UTR experiments or ovary extracts for bcd3'UTR experiments. Virgin females from lacZ-osk3'UTR or lacZ-bcd3'UTR transgenic flies were mated with the appropriate ribozyme transgenic male flies. After 10 days, F1 females from each cross were dissected to obtain egg chamber or ovary extracts. The egg chamber and ovary extracts were used for RT-PCR. The RT-PCR products for the target lacZ were detected after agarose gel electrophoresis and ethidium bromide staining. The expected product from the primer set flanking the ribozyme cleavage site is 582 nt, and this was the size of the experimentally observed product (Fig. 8
). Colocalization of the ribozyme (wt-osk) to the target using the osk3'UTR localization signal (lacZ-osk 3' UTR) resulted in a decrease of the lacZ amplified product of 58% relative to the noncolocalized (wt-bcd), a 97% decrease relative to the mutant (mt-osk), and a 94% decrease relative to the negative control (No) (Fig. 8A
). (The RT-PCR data were normalized by taking into account the relative amounts of ribozyme and ß-tubulin transcript levels.) The expression level of the wt ribozyme with osk3'UTR(wt-osk) was 1.5-fold less than that of the mt ribozyme with osk3'UTR (mt-osk) and 1.5-fold higher than that of the wt ribozyme with bcd3'UTR (wt-bcd).
|
A similar pattern in target product reduction was observed when the bcd3'UTR was used for colocalization (Fig. 8B
). Colocalization of the ribozyme (wt-bcd) and target using the bcd3'UTR (lacZ-bcd 3'UTR) resulted in a 75% decrease of the lacZ product relative to the noncolocalized (wt-osk), 94% relative to the mutant (mt-bcd), and 96% relative to the negative control (No) (Fig. 8B
). [The expression level of the wt ribozyme with the bcd3'UTR (wt-bcd) was 9.4- and 4.2-fold lower, respectively, than the mt ribozyme with bcd3'UTR (mt-bcd) and the wt ribozyme with the osk3'UTR (wt-osk).]
Thus, at the RNA level, colocalization of the ribozyme with the lacZ target mRNA using synonymous 3'UTRs consistently resulted in reduction of the target message compared with the noncolocalized, mutant, or negative control ribozyme transcripts.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
- and ß-actin 3'UTRs in chicken embryo fibroblast cells for colocalizing hammerhead ribozymes with a lacZ target mRNA (15)
. The ribozyme-mediated inhibition of ß-galactosidase activity using the matched 3'UTRs was enhanced 30% relative to unmatched 3'UTRs in this system. In this study, the RNA localization signals present in the Drosophila bcd and osk 3'UTRs were used to test the effect of discrete ribozyme target colocalization on efficacious ribozyme activity in vivo. Colocalization of the ribozymes with the respective lacZ targets using synonymous localization signals from the osk3'UTR or bcd3'UTR resulted in 50% enhancement of ribozyme-mediated inhibition of ß-galactosidase activity in Drosophila oocytes when compared with the noncolocalized controls (Fig. 7)
. When target and ribozyme both contained the same osk3'UTR (osk-osk), X-gal staining was undetectable at the posterior ends of 80% (n=10) of stage 10 egg chambers obtained from F1 females (Fig. 6)
. In contrast, 50% of stage 10 egg chambers containing the noncolocalized osk-bcd transcripts showed ß-gal activity at the posterior ends of oocytes, and all of the egg chambers expressing the osk-GFP control construct showed X-gal staining at the posterior ends. These results demonstrate that discrete mRNA localization signals in the osk3'UTR can be used to colocalize ribozymes and target mRNAs to the posterior of the oocyte. The result of this colocalization was enhanced ribozyme function as measured by reduced ß-galactosidase activity and mRNA levels (Fig. 8A
). Similar results were obtained when the bcd 3' UTR was used on both ribozyme and target transcripts (Fig. 8B
).
oskar mRNA appears to be strictly localized to the posterior pole from stages 9 through 14, and bicoid mRNA is abundant throughout the anterior of the oocyte after stage 9 (18
, 36
37
38)
. Our data demonstrate both inhibition of ß-galactosidase activity and reduction of lacZ mRNAs by ribozymes colocalized with the target mRNAs via the osk3'UTR or bcd3'UTR (50%) (Figs. 7
and 8)
. These results suggest that the quantity of mRNA localization using the osk3'UTR is similar to that obtained from the bcd3'UTR. Cellular factors other than colocalization may play some role in limiting ribozyme target interactions. For instance, the transgenic 3'UTR fusion constructs must compete with the endogenous transcripts for factors that bind the UTR elements and direct RNA trafficking. It is not apparent at this time how important a role factor titration might play in the strategy tested in this study. We are now studying inhibition of endogenous oskar transcripts using an anti-oskar ribozyme colocalized with the osk 3'UTR.
The osk and bcd mRNAs localize to opposite poles of the Drosophila oocyte. The pathways for osk and bcd mRNA movement from nurse cells to oocytes may therefore involve different sets of 3' UTR interacting factors. The bcd mRNA is localized to the anterior of the oocyte (22
, 33)
and the Bcd protein initiates a series of concentration-dependent transcriptional programs that establish the anterior pattern of the embryo (39)
. In contrast, osk mRNA is transported from the nurse cells to the anterior of the oocyte, but is ultimately localized to the posterior of the oocyte where it becomes stably anchored (20
, 21)
. The Osk protein synthesized at this location recruits additional components that are required for the formation of the abdomen and germ cells. Various studies have shown that localized messages are organized into particles (12
, 40
41)
, suggesting that a large protein complex may be involved in recognizing, transporting, and anchoring localized messages. Transport of localized messages from the nucleus to their final destinations occurs along either actin filaments or microtubule tracks, and the transcripts are anchored at their sites of localization through attachments to cytoskeletal elements. Based on the inefficiency of ribozyme activity using heterologous 3' UTRs on the target and ribozyme, these trafficking mRNAs may not have the opportunity to interact with one another during trafficking as RNA–protein particles, even though a ribonucleoprotein complex involved in mRNA localization in Drosophila oocytes contains bcd and osk mRNAs simultaneously (33)
. The rates and directionality of the movement both in and from the nurse cells to oocytes in Drosophila may preclude significant interaction of the ribozyme and target mRNAs during this process.
To enhance the efficacy of ribozyme-mediated target mRNA cleavage, the target site chosen for ribozyme cleavage must also be accessible for ribozyme base pairing in the context of a particular cell. We identified a cleavage site that was accessible to base pairing using a native RNA-ASO screening approach in which endogenously expressed lacZ mRNA was targeted in ovary extracts. The most accessible ribozyme target site among six randomly chosen sites was identified and used for the in vivo studies. ASO3 and concomitantly Rbz3 directed the greatest reduction in an in vivo lacZ transcript under the conditions used. Both reagents gave an approximate 45% reduction in the target transcript relative to controls (Table 1)
. We chose to target the region near the translation initiation codon for the lacZ mRNA, since cleavage in this part of the message would produce a truncated protein even if subsequent destruction of the cleaved message did not occur (42)
. Consequently, we may have overlooked more accessible target sites. Since the entire lacZ mRNA is >3000 nucleotides long, a random or semi-random target site selection approach may facilitate identifying even more accessible cleavage sites than those in this study. It would be of interest to identify other potential targets using the defined oligo screening approach taken in this study.
A locally restricted interaction between transacting factors and a 5' translational derepressor element may be the link between mRNA localization and translational activation (26)
. When oskar mRNA reaches the posterior pole of the oocyte, its translation is derepressed by a 5' derepressor element. The derepressor element interacts with a repressor element in the oskar 3'UTR and activates oskar mRNA translation at the posterior pole. The accessibility of a ribozyme to its target may also depend on elements mediating translational regulation. It is possible that an mRNA, whose translation is strictly regulated, is accessible to base pairing for a much shorter period after translation activation than an mRNA that is constitutively translated. A comparison of the nonrepressed and nonderepressed transgenic flies with wild-type may lead to further insights into accessibility.
Colocalization was accompanied by a two- to threefold enhancement of ribozyme function when compared with combinations with nonsynonymous 3'UTRs. These studies thus provide the first demonstration of functional ribozyme target colocalization mediated by the 3'UTR localization signals in an animal model and lay the foundation for effective ribozyme utilization in future studies of gene expression during development. Since ribozymes can down-regulate transcripts to varying levels, they can be effective tools for studying the effects of incomplete perturbations of gene expression. The use of localizing elements further enhances the probability that ribozyme targeting will be efficacious at some level. We are exploring the use of a ribozyme modulation of the endogenous oskar mRNA using synonymous as well as heterologous UTR elements.
The alternative surrogate genetic approach to ribozymes and antisense is RNAi, which has been demonstrated as a potent mechanism for inhibition of gene expression in Drosophila cell culture and whole flies (43
44
45)
. The development of a hairpin RNAi expression system (43)
allows the temporally controlled expression of RNAi in Drosophila. The results of lacZ down-regulation in embryos using the hairpin RNAi (43)
were comparable to what we report here for the ribozyme approach. A ribozyme approach may be more useful than RNAi when it is important to control the amount of down-regulation. Thus, by increasing or decreasing the expression level of the ribozyme or using ribozymes with suboptimal targeting accessibility, it should be possible to achieve a controlled level of mRNA targeted down-regulation that is not possible with RNAi. On the other hand, RNAi may be more useful as a near complete ‘knockout’ approach, since it signals cellular enzymatic machinery that amplifies degradation of the targeted mRNA; the inhibitory effect remains for several generations of cell division, even upon loss of the original RNAi.
|
ACKNOWLEDGMENTS |
|---|
Received for publication February 7, 2001.
Revision received July 5, 2001.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  K. D. Viles and B. A. Sullenger Proximity-dependent and proximity-independent trans-splicing in mammalian cells RNA, June 1, 2008; 14(6): 1081 - 1094. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||
) indicate the position of the PCR primers and the asterisks indicate the labeled primers; tetrachlorofluorescein (TET, red) for the target lacZ and FAM (green) for the ribozymes. The vertical arrow (
) depicts the cleavage site for Rbz3. The lines with numbers depict the RT-PCR products. B: BamHI; Bt: BstEII; Ev: EcoRV; H: HpaI; N: NotI; P: PstI; Sm: SmaI.
