Skip Navigation

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (377K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (19)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Bailey, C. P.
Right arrow Articles by Weeks, D. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bailey, C. P.
Right arrow Articles by Weeks, D. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Nucleic Acids Research Pages 4860-4867  

Cationic oligonucleotides can mediate specific inhibition of gene expression in Xenopus oocytes
Introduction
Materials And Methods
   Oligonucleotides
   In vitro triplex assay
   Plasmids used for injection studies
   Chloramphenicol acetyltransferase (CAT) assays
   [beta]-Galactosidase assays
   Oocyte injections
Results
   Triplex forming oligonucleotides and target sequences
   TFO inhibition of CAT activity
   In vitro specificity of DEED-modified TFO
   In vivo specificity of triplex formation
   Polarity of the triplex target site does not alter triplex-mediated inhibition of gene expression
   Order of addition affects inhibition of gene expression
Discussion
Acknowledgements
References

Cationic oligonucleotides can mediate specific inhibition of gene expression in Xenopus oocytes

Cationic oligonucleotides can mediate specific inhibition of gene expression in Xenopus oocytes

Cheryl P. Bailey, John M. Dagle1, Daniel L. Weeks*

Department of Biochemistry and 1Department of Pediatrics, University of Iowa, Iowa City, IA 52242, USA

Received July 20, 1998; Revised and Accepted September 10, 1998

ABSTRACT

Base-specific hydrogen bonding between an oligonucleotide and the purines in the major groove of a DNA duplex provide an approach to selective inhibition of gene expression. Oligonucleotide-mediated triplex formation in vivo may be enhanced by a number of different chemical modifications. We have previously described an in vitro analysis of triplex formation using oligonucleotides containing internucleoside phosphate linkages modified with the cation N,N-diethyl-ethylenediamine (DEED). When compared with unmodified oligonucleotides of identical base composition, DEED-modified oligonucleotides were better able to form DNA triplexes under conditions that approximate the pH, magnesium and potassium levels found in vivo. Here we report the ability of DEED-modified oligonucleotides to inhibit the expression of plasmid DNA injected into Xenopus oocytes. Inhibition is specific to plasmids containing a triplex formation target and sensitive to sequence alteration in the triplex forming target site. Inhibition of gene expression was nearly complete when oligonucleotide and plasmid were mixed together prior to injection. Inhibition was partial when oligonucleotide was injected first and not evident when plasmid was injected and allowed to form chromatin prior to oligonucleotide injection. Thus, access to DNA is a determining factor in effective triplex inhibition of gene expression.

INTRODUCTION

The potential use of oligonucleotides as regulators of gene expression centers on base-specific hydrogen bond formation. When oligonucleotides are used for antisense inhibition of mRNA utilization, nucleic acid duplexes are formed following Watson-Crick base pairing rules.

Alternatively, oligonucleotides can bind to duplex DNA, forming a nucleic acid triplex (recently reviewed in 1). Formation of triplex DNA using oligonucleotides utilizes hydrogen bond opportunities in the major groove of the duplex and typically requires that the composition of one of the strands in the DNA duplex be composed predominantly of purine bases. The oligonucleotides participating in forming the triplex use either Hoogsteen or reverse Hoogsteen base pairing rules to bind the purine-rich strand of the duplex. Hoogsteen base pairing (parallel motif) occurs when the oligonucleotide is composed of pyrimidines. T binds to the A:T duplex base pairs and C binds to the G:C duplex base pairs parallel to the purine-rich strand of the duplex. Using this approach cytosine bases in the oligonucleotide must be protonated to participate in hydrogen bond formation, leading to a requirement of low pH. Reverse Hoogsteen base pairing (anti-parallel motif) is used when A (or T) binds to A:T duplex base pairs and G binds to G:C base pairs. Purine-rich oligonucleotides can bind at physiological pH, but are inefficient at triplex formation at intracellular levels of K+ (130 mM) or Mg2+. At least part of this inefficiency is due to the self-association of the purine motif oligonucleotides, making them unavailable for triplex formation. Recent use of triplex forming ODNs to inhibit transcription both in vitro (2-7) and in vivo (8,9) has been encouraging, even though the in vivo effects seen are relatively modest. Triplex forming oligonucleotides (TFO) that compete for transcription factor binding sites, those that bind between the transcription start and RNA polymerase binding site and those that are beyond the transcription start site have been reported to reduce transcription (2-8,10-17).

There are several different strategies that have been used to enhance the use of oligonucleotide-mediated triplex formation in vivo. The low pH required for parallel motif oligonucleotides can be altered by methylation of C residues to allow protonation at neutral pH (reviewed in 1,13,18). Chemical alteration of G residues (using 2[prime]-deoxy-6-thioguanine) has been shown to relieve some of the potassium inhibition normally seen with G-rich anti-parallel rule oligonucleotides (1,13,18). In vivo stability of oligonucleotides can be increased by internucleoside linkage modifications. Although some of these modifications may decrease the ability of oligonucleotides to form triplexes (19), others, like the modifications that use peptide-like backbones (PNA), have been reported to enhance the association of oligonucleotide with duplex DNA (20,21).

We recently examined the relative ability of unmodified, neutral (methoxyethylamine) and positively charged (N,N-diethyl-ethylenediamine; DEED) ODNs to bind to a target duplex (22). The choice of target duplex was based on an enhancer sequence for the GS17 gene of Xenopus laevis that mediates embryonic activation. The GS17 target was 17 bp long with two pyrimidines interrupting the purine-rich strand of the duplex. In our initial studies we compared the binding of unmodified and neutral anti-parallel ODNs under conditions typically used to measure triplex formation that contain no K+ and 10 mM Mg2+. We found that neutral ODNs were several fold better at forming triplexes than unmodified ODNs. Based on the assumption that the enhanced binding was due to decreased electrostatic repulsion, we next examined if cationic ODNs would similarly improve triplex formation, first using standard in vitro conditions that were devoid of K+ and then under conditions that more closely approximate physiological, with 130 mM K+. We estimated that the Kd of unmodified ODN is at least 104 times greater than a fully modified ODN and was not inhibited by 130 mM K+. We found strong oligonucleotide-mediated triplex formation even though the duplex target was relatively small (17 bp) and contained two pyrimidine interruptions in the purine predominant strand. Importantly, we found that non-specific interaction with DNA was much lower than sequence-directed interaction. The finding that cationic internucleoside linkages can stabilize triplex formation has also been reported by others (23,24).

We present here an examination of the ability of cationic modified oligonucleotides to form triplex DNA in vitro when the sequence of the DNA duplex it targets is progressively altered. We have extended these and our previous studies by in vivo analysis of the effect of cationic TFOs on the expression of reporter plasmids in Xenopus oocytes. We show that the cationic modified TFOs inhibit gene expression when their target site is after the transcriptional start site of a gene and that the same levels of unmodified TFOs have no detectable affect on gene activity. We show that mismatches between the TFO and the target can greatly reduce the effective inhibition. We also found no difference in inhibition when the purine-rich strand is on the template (3[prime]->5[prime]) or non-template (5[prime]->3[prime]) strand of the DNA duplex.

The order of exposure of plasmid to oligonucleotide and in vivo DNA binding proteins was examined. Preforming triplex prior to injection leads to near complete inhibition of expression, injection of plasmid into nuclei pre-injected with oligonucleotide leads to a partial inhibition of gene expression, while injection of plasmid followed by injection of oligonucleotide did not alter gene expression.

MATERIALS AND METHODS

Oligonucleotides

Unmodified oligonucleotides were purchased from Gibco BRL (Life Technologies Inc., Gaithersberg, MD). DEED (Fig. 1A)-modified oligonucleotides were synthesized on an ABI PCR-mate DNA synthesizer (Perkin-Elmer Corp., Foster City, CA) using protocols for H-phosphonate (25) synthesis provided by Perkin-Elmer. DEED was purchased from Aldrich Chemical Co. (Milwaukee, WI) and all oligonucleotide synthesis reagents were purchased from Glen Research Co. (Stering, VA). Protocols for oxidative amidation using DEED and purification of modified oligonucleotides have been previously described (22,26-29). DNA duplexes were formed and 5[prime]-end-labeled as previously described (22) using T4 polynucleotide kinase (Promega, Madison WI).


Figure 1. Schematic of target site, variants and modification. (A) The structure of the phosphate modification used in this study. N,N-Diethyl-ethylenediamine phosphoramidate (DEED) modification results in a positively charged TFO at neutral pH. (B) The triplex target sequence used in this paper (30mer) contains a TFO binding site (bold). The sequence of the TFO (17mer) binds anti-parallel to the purine strand of the target sequence. (C) Target sequences containing varying degrees of mismatches (30mer) are shown with the mismatches indicated in lower case and underlined. M1-M4 have mismatches in the middle of the target sequence, while E1-E4 have mismatches near the end of the target sequence. R4 has mismatches placed randomly through the sequence.


In vitro triplex assay

Triplex assays were performed according to Dagle and Weeks (22). Buffer conditions for triplex formation were 20 mM Tris (pH 7.5), 130 mM KCl, 1 mM MgCl2 and 0.1 µg/µl tRNA. The formation of triplex DNA was analyzed by non-denaturing gel electrophoresis (15% polyacrylamide, acrylamide:bisacrylamide 100:1). The amount of radioactivity present in the duplex and triplex forms was determined by electronic autoradiography (InstantImager; Packard Instument Co., Meriden, CT). The fraction of target duplex bound by a TFO ([thetas]) was calculated using the equation:

[thetas] = Striplex/(Striplex + Sduplex)

where Sduplex and Striplex represent the electronic autoradiographic signal for the duplex and triplex bands respectively. The Kd for an oligonucleotide in triplex formation was determined from the concentration of the compound that caused 50% of the target duplex to shift to the triplex form.

Plasmids used for injection studies

pCAT-control, pSV[beta]-galactosidase control and the enzymes used to manipulate these plasmids were purchased from Promega Biotech (Madison, WI). The insertion of triplex target sites was accomplished by StuI digestion of pCAT-control to provide an insertion site for the triplex target site. Duplexes of 30 bp containing the triplex binding site were formed from two complementary oligonucleotides and inserted into the StuI site. A ligation reaction with the linear pCAT-control and the triplex target site was transformed into Escherichia coli (DH5[alpha]) cells and successful insertion of the triplex target site was identified by colony hybridization. The insertion of the triplex target site and the orientation of the site was confirmed by DNA sequence analysis. Four different triplex target-containing vectors were examined. pCAT-target and pCAT-target-rev both contain the target sequence shown in Figure 1B; pCAT-target contains the purine-rich strand of target on the template strand of the gene while pCAT-target-rev contains the pyrimidine-rich strand on the opposite strand of the gene. pCAT-M1, contains the 30 bp M1 sequence and pCAT-E4 the 30 bp E4 sequence shown in Figure 1C.

Chloramphenicol acetyltransferase (CAT) assays

CAT assays were based on a combination of protocols provided by Promega and reported in Jones et al. (30). Reagents were purchased from Promega (Madison, WI) except where noted. Briefly, 10-12 oocytes were homogenized by sonication in 200 µl 1× reporter lysis buffer with 1 mM PMSF. The extracts were cleared by centrifugation for 3 min at 4°C in a microcentrifuge(12 000 r.p.m.). Part of the extract (70 µl) was set aside for the [beta]-galactosidase assay. The CAT assay was started by adding 5 µl n-butyryl CoA (5 mg/ml) and 3 µl [14C]chloramphenicol (0.05 mCi/ml; Amersham, Arlington Heights, IL). These assays were incubated for 1-3 h at 37°C. After incubation, 500 µl ethyl acetate (EM Science, Gibbstown, NJ) was added, the samples were vortexed for 1 min and the phases separated by centrifugation for 3 min. The upper phase was evaporated under vacuum, resuspended in 20 µl ethyl acetate and spotted onto a silica gel TLC plate (Whatman, Clifton, NJ). The silica plate chromatography was developed using chloroform:methanol (97:3) as the mobile phase, for ~1 h in a closed chamber, removed and allowed to dry. Detection and quantitation of [14C]chloramphenicol was carried out using an Instant Imager (Packard Instrument Co., Meriden, CT) with

% conversion = converted counts/total counts.

CAT activity in pCAT-control injected oocytes was determined to be linear with time over a range of 10-75% conversion of chloramphenicol to acetylated chloramphenicol. To allow comparison of assays from different experiments CAT activity was normalized using values from pCAT-target injected oocytes (without TFO) as 100% activity.

[beta]-Galactosidase assays

An independent measure of gene expression was provided bythe inclusion of pSV[beta]-galactosidase control in all injections.[beta]-Galactosidase assays were performed by a modified Promega protocol. As indicated above, extracts made from injected oocytes were split into two aliquots to allow simultaneous evaluation of CAT activity and [beta]-galactosidase activity. Prior to assaying for [beta]-galactosidase activity, the extract was extracted with an equal volume of chloroform:isoamyl alcohol (24:1) to remove lipid and other particulate matter that interfered with the absorbance readings. After 5 min centrifugation (12 000 r.p.m.), 50 µl of the aqueous phase was transferred to a 96-well titer plate (Costar, Cambridge, MA). To each well, 50 µl of 2× assay buffer (200 mM sodium phosphate buffer, pH 7.3, 2 mM MgCl2, 100 mM [beta]-mercaptoethanol, 1.33 mg/ml ONPG) was added and the samples were mixed. The plate was covered and incubated for 15-30 min at 37°C until a faint yellow color appeared and the reaction was stopped with 150 µl of 1 M sodium carbonate. The absorbance of the samples at 414 nm was measured using a Titertek Microskan TCC/340 plate reader. Each titer plate included a serial dilution of [beta]-galactosidase enzyme that had been extracted with chloroform:isoamyl alcohol (24:1) prior to the assay, in order to calculate a standard curve for the evaluation of [beta]-galactosidase activity in injected oocytes. Values obtained for [beta]-galactosidase activity routinely varied by <10% and were used to normalize CAT activity.


Figure 2. Schematic of constructs. Linear representations of plasmids are not drawn to scale. (A) pCAT-control was modified by insertion of triplex target sequences in a StuI site to form pCAT-target, pCAT-M1 and pCAT-E4. pSV-[beta]-galactosidase was co-injected as a control. (B) Insertion site of triplex target into the pCAT plasmids is 30 bp downstream from the start site of transcription, but still upstream from the start of translation.


Oocyte injections

Prior to use in these assays female Xenopus laevis frogs (Xenopus I, Ann Arbor, MI) were screened to identify those with oocytes that were able to actively transcribe pCAT-control. Variation in pCAT-control expression between individual frogs was as large as 50- to 100-fold, although oocytes from an individual frog gave consistent results in multiple trials.

Oocytes were obtained as described (31) and incubated with 0.2% collagenase (Sigma Chemical Co., St Louis, MO) in OR2 (82.50 mM NaCl, 2.5 mM KCl2, 1 mM CaCl2, 1 mM MgCl2, 1 mM Na2HPO4, 5 mM HEPES, pH 7.8) for 2 h. After collagenase treatment, the oocytes were incubated at 18°C in OR2 overnight. The oocytes were then manually defolliculated prior to direct nuclear injection of 10 nl of solution using a Singer MK.1 micromanipulator (Singer Instruments, Somerset, UK) and a Inject+matic microinjector (Inject+matic, Geneva, Switzerland). One source of variation included occasional injections that missed the oocyte nucleus, although 80-90% of injections were estimated to be nuclear based on trial injections using a 0.2% solution of Trypan blue (31). CAT-containing plasmid DNA at a final concentration of 1 µg/µl either by itself or mixed with TFO (final concentration 0.5 µg/µl) was placed in solution with pSV[beta]-galactosidase control plasmid (final concentration 1 µg/µl) in 100 mM HEPES, pH 7.5, 130 mM KCl and 1 mM MgCl2 at 25°C ~1 h prior to injection. For order of addition experiments sequential injections of plasmid or TFO were separated by 15 min. After injection, the oocytes were incubated for 24 h in OR2 with penicillin (0.5 U)/streptomycin (0.5 µg) at 18°C. Oocytes were frozen on dry ice and stored at -80°C prior to use in enzymatic assays.

RESULTS

Triplex forming oligonucleotides and target sequences

Figure 1A shows the structure of the phosphate modification used in this study (22). The DEED is shown protonated and charged, as would be expected at cellular pH. The purine-rich modified oligonucleotides [triplex forming ODN (TFO) in Fig. 1B] follow the anti-parallel triplex motif and were previously shown to form triplex DNA in vitro (22).

The target sequence used in this study is shown in Figure 1B. The target duplex is 30 bp in length and contains a 17 bp polypurine:polypyrimidine region with two C:G inversions. In designing the TFO, thymidine was chosen to interact with the two C:G inversions based on findings that demonstrated significant T:C:G binding (32). Figure 1C shows the target sequences used for the investigation of the specificity of the TFO-duplex interaction. The same TFO sequence was used in all assays, but target sequences were altered to contain varying degrees of mismatches. Changes in the target were made by either replacing G or A residues with C residues in the middle (M series), end(E series) or randomly (R4) in the duplex sequence.

The CAT and [beta]-galactosidase vectors used for this study are shown in Figure 2A. The pCAT control vector contains an SV40 promoter and the CAT gene followed by an SV40 3[prime]-end and an SV40 enhancer. The pSV[beta]-galactosidase control vector contains an SV40 promoter and enhancer in addition to the [beta]-galactosidase gene. Figure 2B shows the insertion site of the triplex target sites into the pCAT plasmids 30 bp downstream from the start site of transcription, but still upstream from the start of translation.

TFO inhibition of CAT activity

DEED-modified TFOs were tested in vivo, to determine whether they could inhibit transcription. Figure 3A shows representative CAT assays of oocytes injected with the pCAT control or pCAT-target either with or without co-injection of TFO. CAT activity from companion groups of oocytes injected with the same plasmid were within 10% of each other.


Figure 3. TFO inhibition of CAT activity. (A) CAT activity present in oocytes injected with pSV-[beta]-galactosidase and pCAT-control or pCAT-target in the presence or absence of TFO with a DEED-modified backbone. (B) CAT activity present in oocytes injected with pSV-[beta]-galactosidase and pCAT-target with either modified (DEED) or unmodified TFO was compared. The percent CAT activity was normalized for [beta]-galactosidase expression and presented as a mean ± SE of three independent trials. pCAT-target was arbitrarily set to 100% to allow for interassay comparison.

The insertion of the triplex target sequence, in the absence of TFO, did not alter the expression of the pCAT reporters. CAT activity present in oocytes injected with pCAT-control was not changed beyond the variation inherent in the assay by the inclusion of TFO. pCAT-target injected oocytes had similar CAT activity to oocytes injected with pCAT-control when TFO was not co-injected. However, extracts made using oocytes co-injected with TFO had much lower CAT activity (2-10% of pCAT-target alone) and were generally difficult to distinguish from background levels in non-injected oocytes.

An independent measure of the changes in gene expression was gained by analysis of the levels of [beta]-galactosidase activity produced by the pSV[beta]-galactosidase control vector, which was always co-injected. There was no indication of TFO-mediated reduction in [beta]-galactosidase activity. Thus, the only expression altered by the presence of the TFO was from plasmid containing the triplex target site.

In Figure 3B, CAT assays of oocytes injected with pCAT-target in the presence of either modified or unmodified TFO were compared. The two oligonucleotides were incubated with the pCAT-target prior to injection under slightly different conditions. To allow optimal binding of the unmodified TFO, the 130 mM KCl was omitted in the incubation, while all other components were the same. Under these conditions, the CAT activity of pCAT-target alone did not differ from oocytes with pCAT-target and unmodified oligonucleotide. As seen before, modified (DEED) oligonucleotide dramatically lowers CAT activity. These results suggest that unmodified oligonucleotide did not remain bound to the target CAT vector or, if bound, did not inhibit transcription.

In vitro specificity of DEED-modified TFO

The length of consecutive purine-rich base pairs that allow triplex structures to form and inhibit gene expression greatly affects the probability of finding an appropriate sequence in a gene and the probability of finding another such sequence in a gene not being targeted. To determine the specificity of the DEED-modified TFO, we first compared binding of TFO with different target sequences. The triplex targets shown in Figure 1C were used in in vitro triplex assays shown in Figure 4. The triplex formation of the different duplexes was compared in a range of TFO concentration (2 nM-2 µM). Interestingly, M1-M4 and R4 showed no detectable triplex formation, as compared with duplex without TFO present. M1-M4 contained either 10 or nine consecutive matching base pairs to the TFO, however, this was not sufficient for triplex formation. The E1-E4 duplexes did allow triplex formation, with 13 consecutive matching base pairs to the TFO. The presence of several shifted species may reflect different triplex conformations or oligonucleotide dissociation during electrophoresis. The apparent Kd (1 × 10-7 M) for E1-E4 was similar to the apparent Kd for parental triplex target duplex.


Figure 4. In vitro specificity of triplex formation. End-labeled triplex target sequences (0.2 nM) shown in Figure 1 were used in in vitro triplex assays with increasing concentrations of TFO (0-2 µM) (A) Triplex target sequences with 0 (DUP), 1 (E1), 2 (E2), 3 (E3) and 4 (E4) mismatches near the end of the sequence. E1-E4 contain 13 consecutive matches to the TFO. (B) Triplex target sequences with 1 (M1), 2 (M2), 3 (M3) and 4 (M4) mismatches in the middle of the triplex target sequence. M1-M4 contain 9-10 consecutive matches to the TFO. R4 has four mismatches throughout the triplex target. These autoradiograms are representative of five independent trials.


In vivo specificity of triplex formation

Duplex variants M1 and E4 were tested in vivo by insertion into pCAT-control at the same site as we had previously inserted the triplex target site in pCAT-target. The expectation from the in vitro trials was that M1 would not serve as a useful target in vivo, while E4 would. pCAT M1 or pCAT E4, along with pSV-[beta]-galactosidase, were mixed with TFO prior to injection into Xenopus oocyte nuclei and CAT and [beta]-galactosidase activity were assayed after 24 h. Figure 5 represents a typical assay. When the triplex target site was the M1 sequence, the TFO did not alter CAT activity. The E4 target, unlike what was predicted from the in vitro assays, did not lead to complete loss of CAT activity, but rather reduced activity to ~25% of that seen without TFO present. Since the triplex was preformed prior to the injection of the plasmid into the nucleus, this finding is consistent with the suggestion that the E4 target is less stable in vivo than a target that fully complements the TFO.


Figure 5. In vivo specificity of triplex formation. CAT activity present in oocytes injected with pSV-[beta]-galactosidase and one of three vectors with different target sequences. TFO has a DEED-modified backbone. pCAT-target contains no mismatches to the TFO, while pCAT-E4 contains four mismatches (13 consecutive matches to the TFO) and M1 contains one mismatch (10 consecutive matches to the TFO). The percent CAT activity was adjusted for [beta]-galactosidase expression and is given as the mean ± SE of four independent trials. pCAT-target was arbitrarily set to 100% to allow for interassay comparison.


Polarity of the triplex target site does not alter triplex-mediated inhibition of gene expression

The hydrogen bonds that form between the duplex DNA and the TFO are through interaction with the purine-rich strand of the target, thus one strand of the duplex DNA provides the bulk of the interaction with the TFO. As transcription occurs, one strand, the template strand, serves to direct the insertion of the appropriate nucleotides to form the RNA transcript while the other does not. By placing the triplex target sequence in both possible orientations (pCAT-target and pCAT-target-rev) we tested whether the polarity of the triplex target had an impact on the ability to inhibit gene expression in the assay used in this study. Figure 6 shows representative data of such an experiment. The similar level of CAT activity leads to the conclusion that orientation of the target site has little effect on TFO-mediated gene inhibition.


Figure 6. Polarity of triplex target does not alter triplex-mediated inhibition of gene expression. CAT activity present in oocytes injected with pSV-[beta]-galactosidase and pCAT-target or pCAT-target-rev. pCAT-target contains the triplex target sequence on the template strand, while pCAT-target-rev contains the triplex target sequence on the non-template strand. TFO is DEED modified. The percent CAT activity was adjusted for [beta]-galactosidase expression and is given as the mean ± SE of two independent trials. pCAT-target was arbitrarily set to 100% to allow for interassay comparison.


Order of addition affects inhibition of gene expression

In the experiments detailed above, plasmid and oligonucleotides were co-injected. We have also examined CAT activity when plasmid and oligonucleotide are injected sequentially. In vitro chromatin assembly assays have previously shown that nucleosome position is affected by pre-existing triplex, but that once nucleosomes are established TFOs no longer bind to duplex DNA (33). The oocyte injection assay used here examined TFO-mediated gene regulation under conditions where DNA could be transcribed, but was not being replicated. In addition, it assayed gene expression under conditions where there was a large pool of the proteins associated with chromatin assembly in preparation for the rapid cell divisions associated with early development. When TFOs were injected first, we routinely saw a sequence-specific reduction in CAT activity from half to one third of the level seen in controls (Fig. 7A). When plasmid was injected first, followed by oligonucleotide, CAT activity was unaffected (Fig. 7B). Thus, at least for this particular target sequence, although the TFO can compete for target site binding with histones, they are unable to displace nucleosomes already formed that block the TFO binding site.


Figure 7. Order of addition of pCAT-target and TFO affects ability to inhibit gene expression. CAT activity is compared for both pCAT-control and pCAT-target either (A) injected with TFO first or (B) injected with vector first. TFO has a DEED-modified backbone. The percent CAT activity was adjusted for [beta]-galactosidase expression and is given as a mean ± SE of at least three independent trials. pCAT-target was arbitrarily set to 100% to allow for interassay comparison.


DISCUSSION

The cationic (DEED-modified) oligonucleotides used in this study showed a dramatic ability to inhibit gene expression from the reporter plasmid containing the triplex binding site. Equally important was that the inhibition was sequence specific. This conclusion is based on the lack of oligonucleotide inhibition of co-injected plasmid expressing [beta]-galactosidase, on pCAT-control plasmid or on pCAT-M1, which has a single base mutation in the middle of the target site.

The difference in in vivo effectiveness when the cationic oligonucleotide was compared with an unmodified oligonucleotide of identical base sequence was clear, as the unmodified oligonucleotide had no effect. Others have reported inhibition in vivo with unmodified oligonucleotides (8,9,14) and our findings may differ for several reasons that include length of the target site, sequence of the target site and nuclease activity of cell type. Typically, reports of in vivo inhibition of gene expression using oligonucleotides have used target sites that range from 23 to 38 bp (7,14). The ability to show inhibition with a 17mer increases the potential targets available. The triplex target sites used in this study were suboptimal, as there were two pyrimidines within the purine-rich strand of the targeted duplex. Again, the ability to identify triplex target sites is increased if pyrimidine interruptions in the target can be accommodated. Finally, we and others have reported on the robust single-stranded nuclease activity found in Xenopus oocytes and embryos (29,34). Uncomplexed, unmodified oligonucleotides may have a half-life of only minutes. Thus, even a temporary release of the unmodified TFO from the duplex may lead to its degradation, where the modified TFOs would be nuclease resistant. Cells with less active single-stranded nuclease activity may be more tolerant of transient disassociation of unmodified TFOs.

Specificity of triplex formation using cationic modified oligonucleotides gives initial information on the sequence requirements for an effective triplex forming site in vivo. As a first approximation, we suggest that at least 13 consecutive appropriate contacts are required for effective triplex formation. Notably, with only 13 consecutive base pair contacts the in vivo efficacy of inhibition of gene expression was not as complete as predicted by our in vitro study. If mismatches are less well tolerated in vivo, then functional specificity may be better than indicated by in vitro binding assays alone. We view this finding as a guideline for potential target sites for triplex formation. We do not know whether 11 or 12 consecutive matching base pairs to the TFO would also be marginally sufficient for triplex formation, however, it is likely that the absolute number of consecutive matches will be influenced by the sequence that is being targeted.

The assay used in this study placed the triplex target site 30 bp after the transcription initiation site. The design of the oligonucleotide rules out inhibition due to Watson-Crick duplex formation with transcription products. The sequence of the oligonucleotide is not complementary to transcript and, even if it were, the full modification of the internucleoside linkages would not allow RNase H-mediated mRNA degradation (29). The mechanism of inhibition is under investigation, but may be due to the premature termination of transcription or the inhibition of binding of the initiation complex. Other studies have also reported the use of triplex inhibition where triplex binding sites are downstream of the transcription start site (5,10-12,15,17). Whatever the mechanism, the polarity of the triplex target site made little difference to the ability of the TFO to inhibit gene expression.

These studies examined the simultaneous injection of reporter plasmid with TFO into the nucleus as well as the sequential injection of plasmid and TFO. Xenopus oocytes used in the assays are in the last stage (stage VI) of development. The chromosomes are condensed in a meiotic G2 arrest, allowing very little endogenous transcription. Plasmids injected into oocytes can be actively transcribed, but they do not replicate. In addition, proteins involved in chromatin assembly are plentiful, enough to fulfil the requirements of several thousand somatic cell nuclei (21). Our findings suggest that under these conditions, when DNA is allowed to assemble into chromatin prior to exposure to TFO, TFOs did not inhibit transcription. On the other hand, when TFOs were present and competing for target site binding they significantly reduced reporter gene activity. Thus, our data suggest that if the triplex target site occupancy is a competition between the TFO and histones, that priority matters, and that TFO displacement of established chromatin structure did not occur. We assayed the inhibitory activity of a triplex formed within the transcribed region of a gene, where the TFO is presumably competing with histones and other general DNA binding proteins for site occupancy. If the TFO were competing with a transcription factor for target site binding the result could be different, as the target site may be more exposed. Since DNA is not replicating in oocytes, further studies on dividing cells where, at least transiently, chromatin structure is disrupted and reformed, will be needed to judge the possible use of cationic TFOs in cells where DNA is being amplified.

Some of the variables that remain to be tested include those that will resolve the ability of cationic TFOs to find their way into the nucleus when direct injection is not the mode of delivery and the ability of the TFOs to interact with endogenous genes. However, the initial findings presented here indicate cationic TFOs have promise as sequence-specific inhibitors of transcription.

ACKNOWLEDGEMENTS

The authors acknowledge helpful discussion of this work with other members of the Weeks laboratory and funding by the National Institutes of Health. Part of this work was done during the tenure of an established investigatorship (DLW) of the American Heart Association.

REFERENCES

1. Vasquez,K.M. and Wilson,J.H. (1998) Trends Biochem. Sci., 23, 4-9. MEDLINE Abstract

2. Cooney,M., Czernuszewicz,G., Postel,E.H., Flint,S.J. and Hogan,M. (1988) Science, 241, 456-459. MEDLINE Abstract

3. Durland,R.H., Kessler,D.J., Gunnell,S., Duvic,M., Pettitt,B.M. and Hogan,M.E. (1991) Biochemistry, 30, 9246-9255. MEDLINE Abstract

4. Maher,L.J. (1992) Biochemistry, 31, 7587-7594. MEDLINE Abstract

5. Duval-Valentin,G., Thuong,N.T. and Helene,C. (1992) Proc. Natl Acad. Sci. USA, 89, 504-508. MEDLINE Abstract

6. Noonberg,S.B., Scott,G.K., Hunt,C.A., Hogan,M.E. and Benz,C.C. (1994) Gene, 149, 123-126. MEDLINE Abstract

7. Kim,H.G. and Miller,D.M. (1995) Biochemistry, 34, 8165-8171. MEDLINE Abstract

8. Ing,N.H., Beekman,J.M., Kessler,D.J., Murphy,M., Jayaraman,K., Zendegui,J.G., Hogan,M.E., O'Malley,B.W. and Tsai,M.J. (1993)Nucleic Acids Res., 21, 789-2796.

9. Kovacs,A., Kandala,J.C., Weber,K.T. and Guntanka,R.V. (1996)J. Biol. Chem., 271, 1805-1812. MEDLINE Abstract

10. Giovannangeli,C., Perrouault,L., Escude,C., Nguyen,T. and Helene,C. (1996) Biochemistry, 35, 10539-10548. MEDLINE Abstract

11. Giovannangeli,C., Perrouault,L., Escude,C., Gryaznov,S. and Helene,C. (1996) J. Mol. Biol., 261, 386-398.

12. Hacia,J.G., Dervan,P.B. and Wold,B. (1994) Biochemistry, 33, 61192-6200.

13. Maher,L.J., Dervan,P.B. and Wold,B. (1991) In Wickstrom,E. (ed.), Prospects for Antisense Nucleic Acid Therapy of Cancer and AIDS. Wiley Liss, New York, NY, pp. 227-242.

14. McShan,W.M., Rossen,R.D., Laughter,A.H., Trial,J., Kessler,D.J., Zendegui,J.G., Hogan,M.E. and Orson,F.M. (1992) J. Biol. Chem., 267, 5712-5721. MEDLINE Abstract

15. Rando,R.F., DePaolis,L., Durland,R.H., Jayaraman,K., Kessler,D.J. and Hogan,M.E. (1994) Nucleic Acids Res., 22, 678-685. MEDLINE Abstract

16. Samadashwily,G.M. and Mircin,S.M. (1994) Gene, 149, 127-136. MEDLINE Abstract

17. Young,S.L., Krawczyk,S.H., Matteucci,M.D. and Toole,J.J. (1991)Proc. Natl Acad. Sci. USA, 18, 10023-10026.

18. Durland,R.H., Kessler,D.J. and Hogan,M.H. (1991) In Wickstrom,E. (ed.), Prospects for Antisense Nucleic Acid Therapy of Cancer and AIDS.Wiley Liss, New York, NY, pp. 219-226.

19. Musso,M. and Van Dyke,M.W. (1995) Nucleic Acids Res., 23, 2320-2327. MEDLINE Abstract

20. Betts,L., Josey,J.A., Veal,J.M. and Jordan,S.R. (1995) Science, 270, 1838-1841. MEDLINE Abstract

21. Faruqi,A.F., Egholm,M. and Glazer,P.M. (1998) Proc. Natl Acad. Sci. USA, 95, 1398-1403.

22. Dagle,J.M. and Weeks,D.L. (1996) Nucleic Acids Res., 24, 2143-2149. MEDLINE Abstract

23. Mayfield,C. and Miller,D. (1994) Nucleic Acids Res., 22, 1909-1916. MEDLINE Abstract

24. Chaturvedi,S., Horn,T. and Letsinger,R.L. (1996) Nucleic Acids Res., 24, 2318-2323. MEDLINE Abstract

25. Froehler,B.C., Ng,P.G. and Matteucci,M.D. (1986) Nucleic Acids Res., 14, 5399-5407. MEDLINE Abstract

26. Weeks,D.L., Walder,J.A. and Dagle,J.N. (1991) Development, 111, 1173-1178. MEDLINE Abstract

27. Dagle,J.M., Walder,J. and Weeks,D.L. (1990) Nucleic Acids Res., 18, 4751-4757.

28. Dagle,J.M., Andracki,M.E., DeVine,R.J. and Walder,J.A. (1991)Nucleic Acids Res., 19, 1805-1810. MEDLINE Abstract

29. Dagle,J.M., Weeks,D.L. and Walder,J. (1991) Antisense Res. Dev., 1, 11-20.

30. Jones,N.C., Richter,J.D., Weeks,D.L. and Smith,L.D. (1983)Mol. Cell. Biol., 3, 2131-2142. MEDLINE Abstract

31. Colman,A. (1984) In Hames,D. and Higgins,S. (eds), Transcription and Translation-A Practical Approach. IRL Press, Oxford, UK, pp. 49-69.

32. Durland,R.H., Rao,T.S., Revankar,G.R., Tinsley,J.H., Myrick,M.A., Seth,D.M., Rayford,J., Singh,P. and Jayaraman,K. (1994)Nucleic Acids Res., 22, 3233-3240. MEDLINE Abstract

33. Westin,L., Blomquist,P., Milligan,J.F. and Wrange,O. (1995)Nucleic Acids Res., 23, 184-2191.

34. Prives,C. and Foukal,D. (1991) In Kay,B. and Peng,B. (eds), Methods in Cell Biology. Academic Press, San Diego, CA, Vol. 36, pp. 185-213. MEDLINE Abstract

*To whom correspondence should be addressed. Tel: +1 319 335 7918; Fax: +1 319 335 9570; Email: daniel-weeks@uiowa.edu

This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 17 Oct 1998
Copyright©Oxford University Press, 1998.
Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 Nucleic Acids ResHome page
T. Michel, C. Martinand-Mari, F. Debart, B. Lebleu, I. Robbins, and J.-J. Vasseur
Cationic phosphoramidate {alpha}-oligonucleotides efficiently target single-stranded DNA and RNA and inhibit hepatitis C virus IRES-mediated translation
Nucleic Acids Res., September 15, 2003; 31(18): 5282 - 5290.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
K. M. Vasquez, J. M. Dagle, D. L. Weeks, and P. M. Glazer
Chromosome Targeting at Short Polypurine Sites by Cationic Triplex-forming Oligonucleotides
J. Biol. Chem., October 12, 2001; 276(42): 38536 - 38541.
[Abstract] [Full Text] [PDF]

Home page
 Hum Mol GenetHome page
M. P. Knauert and P. M. Glazer
Triplex forming oligonucleotides: sequence-specific tools for gene targeting
Hum. Mol. Genet., October 1, 2001; 10(20): 2243 - 2251.
[Abstract] [Full Text] [PDF]

Home page
 Nucleic Acids ResHome page
C. Bailey and D. L. Weeks
Understanding oligonucleotide-mediated inhibition of gene expression in Xenopus laevis oocytes
Nucleic Acids Res., March 1, 2000; 28(5): 1154 - 1161.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Print PDF (377K) Freely available
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrow Search for citing articles in:
ISI Web of Science (19)
Right arrowRequest Permissions
Right arrow Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Right arrow Articles by Bailey, C. P.
Right arrow Articles by Weeks, D. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bailey, C. P.
Right arrow Articles by Weeks, D. L.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?