(Received for publication, November 19, 1996, and in revised form, February 24, 1997)
From the Institut für Organische Chemie, Two hammerhead ribozymes targeted to point
mutations in codon 13 of the N-ras oncogene were
synthesized and their catalytic activity and substrate specificity
evaluated in vitro and ex vivo. In
vitro studies showed that these ribozymes were specific for the
oncogenic form of N-ras, since cleavage was observed only in a 849-nucleotide-long transcript containing mutant but not wild-type
N-ras sequences. For the ex vivo studies, the
ribozymes were 2 Catalytic RNAs include Group I and Group II introns and ribozymes
of the hammerhead, hairpin, and hepatitis delta virus type and the
subunit of RNase P (1-3). The presence of divalent metal ions,
e.g. Mg2+ or Mn2+, is essential for
their activity (4, 5). Hammerhead ribozymes are the smallest catalytic
RNA found to date. They consist of three stems connected by a conserved
core region. They recognize substrates containing an NUH base triplet
(N can be any base; H can be A, C, or U) and cleave the phosphodiester
bond on the 3 Inhibition of gene expression by trans-acting hammerhead ribozymes has
been reported in, e.g., plant cells (8), mammalian cells
(9-14), and Xenopus oocytes (15, 16). Ribozymes have also
been successfully used against oncogenes such as bcr-abl (17), c-fos, c-Ha-ras (12, 18, 19), or viral RNAs
(20). Thus, like other strategies, e.g. antisense and triple
helix formation, ribozymes appear to be promising agents to inhibit
expression of specific oncogenes.
In general, the most difficult step toward an ex vivo
application of ribozymes is the delivery of the ribozyme into the
cells. Ribozymes can be applied directly to cells (exogenous delivery), or plasmid or viral vectors can be used to express the ribozyme within
living cells (endogenous delivery).
For the exogenous application, ribozymes are delivered into cells by
microinjection (10) or transfection (21). In the latter case,
CaCl2 or cationic liposomes have been mostly used as
transfection reagents. Additional exogenous delivery methods are based
on the conjugation of the oligonucleotides to polylysine compounds (22)
or to lipophilic groups, like cholesterol (23, 24).
A problem that arises from exogenous delivery is the low stability of
unmodified ribozymes in cell culture supernatant containing fetal calf
serum. Since the 2 The growth and differentiation of cells depends on a variety of
parameters and signal transduction pathways. The proteins coded by the
three ras genes (Ha-ras, N-ras, and
Ki-ras) are involved in cell signal transduction and are
members of the supergene family of small GTP/GDP-binding proteins (32,
33). ras mutations have been detected in a wide variety of
tumors such as pancreatic carcinomas (34, 35) and tumors of the stomach
and breast (36, 37). N-ras mutations have also been found in
neuroblastoma, melanoma, acute myeloblastic leukemia, chronic
myelogenous leukemia, and multiple myeloma (38). Studies of
ras oncogenes in tumors have revealed several point
mutations in codons 12, 13, 59, or 61, which cause structural changes
in the GTP binding site and reduce GTPase activity (39). Binding of GTP
leads to an active state of the Ras proteins, whereas hydrolysis of GTP
to GDP causes inactivation. Mutant Ras proteins, having a reduced
ability to hydrolyze GTP, remain in the active state and thus stimulate
cell growth or differentiation autonomously. The inhibition of the incorrect signal transduction by ribozymes may lead to an efficient anti-cancer therapy.
In the present study we report the synthesis and catalytic properties
of several hammerhead ribozymes targeted against mutant N-ras transcripts. These hammerhead ribozymes were tested
for efficiency and specificity in vitro and ex
vivo. In addition, the effect of 2 Materials
Ribonucleoside phosphoramidites and control pore glass columns
were obtained from PerSeptive Biosystems. The sulfurizing reagent C-6
Thiol was purchased from Chemgens. The 2 Chemical Synthesis of Ribozymes
Oligoribonucleotides were prepared on an Applied Biosystems
model 380B DNA Synthesizer on a 1-µmol scale. The
oligoribonucleotides were base deprotected by incubation of the glass
support with 3 ml of aequeous concentrated ammonia (33%)/ethanol (3:1
(v/v)) at 55 °C for 16 h. After complete removal of the solvent
by Speed-Vac evaporation, the 2 Mass Spectrometry of Oligoribonucleotides
All oligoribonucleotides (unmodified and modified ribozymes)
were characterized by MALDI-TOF1 mass
detection as shown in Fig. 1. The MALDI mass spectra
were recorded on a VG TofSpec. For a MALDI-TOF spectrum, the RNA
solution was mixed with the matrix compound.
Plasmid Constructions
The pcN1 plasmid (41) containing most of the
N-ras cDNA was kindly provided by A. Hall (Medical
Research Council, London, United Kingdom). The N-ras
sequence was PCR-amplified and cloned in the plasmid pBluescript II KS
(Stratagene) to generate pMS1-NRAS.
A plasmid containing the full-length N-ras cDNA sequence
was constructed according to the strategy described by Khorana (42). For this purpose four overlapping oligodeoxynucleotides (99-102 nucleotides each) containing the 5 Two point mutations in codon 13 of N-ras were introduced
into plasmid pMS5-NRAS by PCR-based technology. Plasmid pMS5A-NRAS contains a G A
N-ras/luciferase reporter gene plasmid containing the first
452 bp of N-ras fused in frame with the luciferase gene was constructed. For this the translation initiation codon of the luciferase gene was mutated to ATA by insertion of a 70-bp linker containing the altered sequence at the XbaI site of plasmid
pBHEluc (kindly provided by H. Hauser, Gesellschaft für
Biotechnologische Forschung, Braunschweig, Germany).
Subsequently plasmids pMS5-NRAS, pMS5A-NRAS, and pMS5B-NRAS were
digested with BamHI and PflMI, and the so
obtained 452-bp DNA fragments containing either wild-type or any of the
mutated N-ras sequences were fused in frame to the firefly
luciferase gene. The N-ras/luciferase fusion genes were
cloned in the expression plasmid pcDNA3 (InVitrogen), and the
resulting plasmids were named pcDNA3-LucFUW (wild-type
N-ras), pcDNA3-LucFUC (GGT In Vitro Transcription
For in vitro transcription pMS5-NRAS, pMS5A-NRAS, and
pMS5B-NRAS were linearized with EcoRI, phenol-extracted, and
ethanol-precipitated. In vitro transcription was carried out
in 100 µl of mixture containing 50 ng/µl linearized plasmid DNA, 10 mM DTT, 40 mM Tris-Cl, pH 7.5, 50 mM NaCl, 8 mM MgCl2, 2 mM spermidine, 500 µM rNTPs, 0.8 unit/µl
RNase inhibitor, 2 µCi/µl [ Kinetics with Synthetic Substrates
Kinetic constants Km and
kcat were determined from Eadie-Hofstee plots
carried out with 32P-labeled substrates. Ribozyme and
substrate were heated separately for 1 min at 75 °C in 50 mM Tris-Cl, pH 7.5, and for 5 min at 37 °C. Then 100 mM MgCl2 were added to a final concentration of 10 mM and the solutions were incubated for an additional 5 min at 37 °C. "Steady state" reactions were carried out in a
volume of 100 µl with substrate concentrations between 20 and 500 nM and ribozyme concentrations from 2 to 5 nM
in 50 mM Tris-Cl, pH 7.5, and 10 mM
MgCl2 at 37 °C. Reactions were initiated by addition of
ribozyme. The reaction was stopped by mixing the ribozymes with an
equal volume of stop solution (8 M urea, 25 mM
EDTA). The cleavage reactions were analyzed on 20% denaturing
polyacrylamide gels (8 M urea) and scanned on a Molecular
Dynamics PhosphorImager.
Kinetics with in Vitro Transcribed RNA
"Single-turnover" reactions were performed in a volume of 10 µl containing 20-1200 nM ribozyme, 50 mM
Tris-Cl, pH 7.5, and 10 mM MgCl2. The reaction
was initiated by addition of 2-10 nM 32P-labeled RNA substrate. Under these conditions the
concentration of ribozyme is at least in a 10-fold excess over
substrate. After incubation at 37 °C for 1 h, the reaction was
stopped by addition of 8 µl of stop mix to each reaction. The
reaction products were analyzed as described above. The single turnover
kreact/Km values were
determined as described (30).
Analysis of the Stability of Oligoribonucleotides
NIH 3T3 cells were maintained in Dulbecco's modified Eagle's
medium supplemented with 10% heat-inactivated FCS and 100 units/ml penicillin, 100 µg/ml streptomycin, and 1 mM
L-glutamine in 96-well plates. The oligoribonucleotide
solutions (32 µl) containing modified or unmodified ribozymes were
added to the cell culture supernatant (525 µl) to reach a final
ribozyme concentration of 5 µM. Aliquots (67 µl) were
taken at different time points and shock-frozen into liquid nitrogen to
stop nuclease activity. After Speed-Vac evaporation, the pellets were
resuspended in formamide and analyzed on a 20% polyacrylamide gel (8 M urea). Nucleic acid bands were visualized by silver
staining.
Cell Lines and DNA Transfection
HeLa cells were cultured in Dulbecco's modified Eagle's medium
supplemented with 10% heat-inactivated FCS, 100 units/ml
penicillin/streptomycin (100 µg/ml), and 1 mM
L-glutamine. Stable DNA transfections were performed by the
calcium-phosphate precipitation method (43) with 20 µg of
CsCl-purified plasmid DNA containing either wild-type or mutant
N-ras/luciferase fusion gene and the neomycin-resistance gene. After transfection cells were split at a ratio 1:40 and incubated
for an additional 24 h before selection. Neomycin-resistant clones
were obtained after 2 weeks of selection in medium containing 800 µg/ml G418 (50% biological activity). Individual cell clones were
isolated, expanded, and tested for luciferase expression by the Triton
X-100 lysis method (44).
Ribozyme Experiments with Cells
HeLa cells were grown at a density of 3·104
cells/ml in 96-well plates in Opti-MEM® (Life Science) for
24 h. Different amounts of ribozymes, ranging from 6 µM to 13 µM, were mixed with 3 µl of
LipofectAMINETM (2 µg/µl) and 10 µl of Opti-MEM® and
incubated for 45 min at room temperature. After the addition of 60 µl
of Opti-MEM®, the mixture was given to the cells. Cells
were incubated with the LipofectAMINETM-oligoribonucleotide complex for
8 h at 37 °C. Transfected cells were cultured for additional
24-36 h before testing for luciferase activity and mRNA
analysis.
Luciferase Assay
Luciferase activity was determined by the Triton X-100 lysis
method (44). Protein concentrations determined with the Bradford protein assay were used to standardize the luciferase activity (45).
Total RNA Isolation and RT-PCR
Total RNA was isolated from HeLa cells using a commercially
available kit (GlassMAX RNA Microisolation Spin Cartridge System, Life
Technologies, Inc.). After isolation and before the DNase I digestion
step, the internal standard RNA (0.5 amol; see below) was added to the
cellular RNA and the mixture was digested with DNase I for 1 h at
37 °C, followed by phenol chloroform extraction and ethanol
precipitation. Reverse transcription was performed in 20 µl of
reaction volume in the presence of 10 mM DTT, 50 mM Tris-HCl, pH 8.3, 3 mM MgCl2, 75 mM KCl, 1 mM dNTPs, 0.5 unit of RNasin, 50 ng
of hexamer primer, and 10 units of Moloney murine leukemia virus
reverse transcriptase. The reaction mixture was incubated for 10 min at
room temperature, followed by 90 min at 42 °C. The reaction was
stopped by incubation at 94 °C for 7 min. An aliquot of the RT
reaction was amplified under standard Perkin-Elmer Cetus PCR conditions
(1 min at 94 °C, 1 min at 60 °C, and 2 min at 72 °C) in a
total volume of 50 µl. In parallel, a Construction of an Internal Control Plasmid for RT-PCR
To obtain an internal control for the RT-PCR, a 50-bp linker was
inserted into the N-ras gene. The linker was cloned into the
5 Two hammerhead ribozymes targeted against point mutations in codon
13 of the N-ras gene were synthesized. A GC transversion at
position 763 generates a GUC triplet, which was targeted by the
ribozyme MRE763C. An additional ribozyme, MRE764U, was targeted against
a second GT transversion at position 764, which generates a GUU triplet
(Fig. 2A). For comparison purposes, three
ribozymes recognizing either a GUC triplet at codon 89 (RE990) or a GUA triplet at codons 64 (RE917) and codon 103 (RE1035) of the wild-type N-ras mRNA were also investigated (Fig. 2B).
At first, short synthetic oligoribonucleotides 15 nucleotides in length
containing the cleavage site were used to test for cleavage activity
under steady-state conditions. The resulting Km and
kcat values show equal catalytic efficiency for
the ribozymes MRE763C, MRE764U, and RE917 (0.2·106
s
Table I.
Catalytic efficiencies of different ribozymes with synthetic substrates
To study the cleavage efficiency of the ribozymes on larger RNA
substrates, transcripts containing the full-length wild-type or mutant
N-ras sequence were synthesized in vitro and used under single-turnover conditions in the cleavage reaction (see
"Experimental Procedures" and Table II).
Table II.
Catalytic efficiencies of different ribozymes with in vitro transcribed
full-length N-ras transcripts
Georg-Speyer-Haus,
ABSTRACT
INTRODUCTION
EXPERIMENTAL PROCEDURES
RESULTS
DISCUSSION
FOOTNOTES
ACKNOWLEDGEMENTS
REFERENCES
-modified to protect them against degradation by
nucleases. 2
-Fluoro-2
-deoxyuridine/cytidine-substituted ribozymes
were nearly as active as their unmodified counterparts, but had a
prolonged stability in cell culture supernatant containing fetal calf
serum. The stability of the modified ribozymes increased by
introduction of terminal phosphorothioates groups without significant
influence in their catalytic efficiency. A sensitive assay based on the use of N-ras/luciferase fusion genes as a reporter system
was established to detect ribozyme-mediated cleavage in HeLa cells. A
reduction of nearly 60% in luciferase activity was observed in cells
expressing mutant but not wild-type N-ras/luciferase fusion
transcripts. Moreover, cleavage of N-ras transcripts in HeLa cells was directly confirmed by a semi-quantitative RT-PCR assay.
side of H. The GUC base triplet is cleaved most
efficiently (6, 7). Thus, any RNA containing this motif can be cleaved by a hammerhead ribozyme if the target sequence is accessible for
ribozyme binding.
-hydroxyl group plays an important role in the
degradation mechanism by nucleases, ribozymes have been protected
against degradation by modification of the 2
-hydroxyl position (25,
26). For this 2
-deoxyribonucleotide (27, 28), 2
-O-methyl
groups (29), 2
-fluoro- and/or 2
-amino groups (30), or
2
-deoxyribonucleotides together with phosphorothioate linkages (31)
have been used.
-modifications on the cleavage
efficiency and stability of the ribozymes in cell culture media
containing fetal calf serum was analyzed. Furthermore, a reporter gene
system based on luciferase gene expression was established to evaluate
the catalytic properties of 2
-modified ribozymes ex vivo. A
semi-quantitative RT-PCR was used to assess mRNA cleavage by the
ribozymes. Our study demonstrates that ribozymes targeted against
mutant N-ras sequences are highly specific and efficient
in vitro and ex vivo.
-fluoro-modified ribonucleosides were a kind gift of Dr. T. Wittmann and Isis
Pharmaceuticals. The 2
-amino-2
-deoxyuridine was a kind gift from K. Jahn-Hoffmann (Universität Frankfurt). The
2
-O-methylribonucleosides and 2
-deoxyribonucleosides were
obtained from PerSeptive Biosystems. Nucleoside triphosphates were
purchased from Boehringer Mannheim. Radiolabeled nucleoside triphosphates [
-32P]ATP and [
-32P]ATP
with the specific activity 3000 Ci/mmol were obtained from Amersham.
-silyl group was removed by overnight
incubation at room temperature in 1 M tetrabutylammonium
fluoride in tetrahydrofuran. After addition of 0.5 ml of 3 M sodium acetate solution (pH 5.2), the tetrahydrofuran was
removed on a Speed-Vac concentrator and the aqueous phase was extracted
twice with 1 ml of ethyl acetate. The oligoribonucleotide was
precipitated by addition of 2.5 volumes of absolute ethanol followed by
centrifugation at 13,000 rpm. The pellet was dissolved in 1 ml of
water. The RNA solution was purified on polyacrylamide gels (12% or
20%) containing 8 M urea. The RNA was visualized under UV
light, excised from the gel, and eluated in 0.05 M ammonium
acetate solution (pH 7.0) overnight. Thereafter, the RNA solution was
loaded onto a Sephadex G-25 column, and fractions of 1 ml were
collected and stored frozen at
20 °C. The homogeneity of the
ribozyme RNA and substrate RNA was checked by mass spectrometry and
analytical PAGE. RNA concentration was determined by assuming an
extinction coefficient at 260 nm of 6.6 × 103
M
1 cm
1 (40).
Fig. 1.
MALDI mass spectrum of the 32-mer
oligoribonucleotide ribozyme MRE763C. All pyrimidine nucleosides
were replaced by the corresponding 2-fluoro-2
-deoxyanalogues. The
spectrum was recorded on a VG TofSpec with
2,4,6-trihydroxyacetophenone/NH4-citrate = 2:1 as
matrix as described under "Experimental Procedures."
[View Larger Version of this Image (18K GIF file)]
-termini sequences of
N-ras were synthesized and cloned into pMS1-NRAS to generate
pMS5-NRAS. Thus, pMS5-NRAS contains N-ras sequence from the
transcription initiation site to the transcription termination site.
The sequence was confirmed by DNA sequence analysis employing standard
procedures. In vitro transcription of pMS5-NRAS gave the
expected 849-nucleotide-long RNA product.
T transversion at position 764, whereas plasmid pMS5B-NRAS contains a G
C transversion at position 763. All sequences were confirmed by DNA sequence analysis.
CGT mutation), and
pcDNA3-LucFUT (GGT
GTT mutation). The sequence of the fusion genes were confirmed by DNA sequence analysis.
-32P]ATP, and 2.5 units/µl T7 RNA polymerase. After 1 h of incubation at 37 °C,
DNase I (25 units) was added and the mixture incubated for another 10 min at 37 °C. After subsequent phenol extraction, the aqueous phase
was transferred into a Centricon-100 tube and centrifuged at 3,400 rpm
for 30 min. The RNA solution was checked for homogeneity by UV
absorption and in a 6% analytical PAGE (8 M urea). The RNA
was stored frozen at
20 °C.
-actin PCR was performed to
control for the RNA input in the RT-PCR. The following primers were
used: 5
-actin, 5
-GTGGGGCGCCCCAGGCACCA-3
; 3
-actin,
5
-CTCCTTAATGTCACGCACGCTTTC-3
; RT1-N-ras,
5
-GGGAGACCCAAGCTTGGTACC-3
; RT2-N-ras,
5
-ACTCGCTTAATCTGCTCCCTGTAGAG-3
. Reaction products were separated on a
4% NuSieve agarose gel in 1 × TAE and stained with ethidium
bromide.
-nontranslated region (BamHI restriction site) of the
N-ras gene in the expression plasmid pcDNA3-LucFUC. The
construct was named pcDNA3-LucFUCL. To generate the internal
standard RNA transcript, plasmid pcDNA3-LucFUCL was linearized with
BstEII, in vitro transcribed as described above,
and quantified by UV spectrophotometry.
1·M
1), whereas ribozyme
RE990 has a kcat/Km value 2 orders of magnitude lower (Table I and Fig.
3). RE1035 was found to have an intermediate
kcat/Km value.
Fig. 2.
Design of the chemically synthesized
hammerhead ribozymes. A, ribozymes targeted against mutant
N-ras mRNA. B, ribozymes targeted against the
wild-type N-ras mRNA. The cleavage site is indicated by
an arrow, and the catalytic core sequences are
boxed. Stems are numbered according to Hertel et
al. (46).
[View Larger Version of this Image (24K GIF file)]
Ribozyme
kcat
Km
kcat/Km
min
1
nM
106·s
1·M
1
RE917
0.78 ± 0.20
78
± 10
0.2
RE990
0.24 ± 0.08
404 ± 45
0.009
RE1035
0.78 ± 0.18
277 ± 25
0.05
MRE763C
0.90 ± 0.19
82 ± 19
0.2
MRE764U
0.72 ± 0.10
65 ± 5
0.2
Fig. 3.
Kinetics of ribozyme cleavage with short
synthetic substrate. The cleavage kinetic of ribozymes MRE763C
(A) and RE990 (B) is shown. Short synthetic
oligoribonucleotides served as substrates. The ribozyme concentrations
used were 2 nM, while the substrate concentration was 100 nM. Aliquots were taken at the indicated time points (2, 5, 10, 20, 40, and 60 min; lanes 2-7, respectively) and loaded
onto a polyacrylamide gel. Lane 1 shows the untreated oligoribonucleotide used as substrate.
[View Larger Version of this Image (26K GIF file)]
Ribozyme
kreact
Km
kreact/Km
10
6·s
1
nM
s
1·M
1
RE917
59 ± 8.2
63 ± 20
938
RE990
72 ± 7.3
234 ± 28
307
RE1035
44
± 11
425 ± 41
103
MRE763C
266 ± 25
71
± 13
3752
MRE764U
137 ± 12.2
113
± 21
1212
Ribozyme MRE763C showed the best catalytic activity. A computer-assisted folding analysis of the RNA substrate revealed a number of mismatches (bulges) at the binding region of ribozyme MRE763C. The same applies for the binding region of ribozyme MRE764U. In contrast the cleavage site of ribozyme RE1035 is embedded within a region that folds in a stable hairpin structure. Thus, the differences in the catalytic efficiency of the enzyme and the large RNA substrate correlate well with the predicted secondary structure of the 849-nucleotide-long transcript. According to the kreact/Km values, the most effective ribozymes (MRE763C, MRE764U, and RE917) were chosen for further investigations.
The specificity of one of the ribozymes (MRE763C) to cleave only
N-ras sequences containing a point mutation at position 763 was examined next. Incubation of MRE763C with a 849-nucleotide-long mutant N-ras transcript resulted in the expected cleavage
products of 534 and 315 bases (Fig. 4A). The
same result was obtained for ribozyme MRE764U (data not shown). In
contrast, incubation of MRE763C with transcripts containing the
wild-type N-ras sequences did not result in any detectable
substrate cleavage, demonstrating the absolute specificity of the
ribozyme for its substrate in vitro (Fig.
4B).
Since the final goal of our experiments was to demonstrate the
efficiency of these ribozymes in cell culture, pyrimidine
ribonucleotides containing 2-OH substitutions were used for the
synthesis of ribozymes to increase the stability of the
oligoribonucleotides against degradation by RNases. Several 2
-modified
ribonucleotides such as
2
-O-methyl-2
-deoxyuridine/cytidine,
2
-deoxyuridine/cytidine, or 2
-fluoro-2
-deoxyuridine/cytidine were
combined with terminal phosphorothioate linkages (see Table
III). The modified ribozymes were incubated in cell
culture media at 37 °C for up to 120 h. At specific time
points, aliquots were taken from the supernatant, frozen, and dried.
The pellets were subsequently resuspended in formamide and analyzed on
a polyacrylamide gel. While the unmodified ribozyme was degraded within
30 s, the introduction of three phosphorothioate linkages at the
3
-end of the ribozyme and one at the 5
-end increased the half-life of
ribozymes to 2-3 min. Other modifications (e.g. 2
-fluoro-2
-deoxyuridine) led to an additional increase in stability (Fig. 5A). Finally, complete substitution of
all pyrimidine nucleotides (e.g. 2
-fluoro-2
-deoxycytidine)
prevented ribozyme degradation for up to 80 h (Fig.
5B). Most of the modified ribozymes were still capable of
cleaving the N-ras transcripts. The introduction of the
three 3
- and one 5
-terminal phosphorothioate groups led only to a
minor decrease in catalytic potency (Table IV). However, the catalytic efficiency of ribozymes containing
2
-fluoro-2
-deoxyuridine nucleotides and phosphorothioate groups was
very low compared with the unmodified ones
(kreact/Km = 150 s
1·M
1 versus 938 s
1·M
1 for the modified and
unmodified ribozymes, respectively). The catalytic potency of ribozymes
containing phosphorothioate groups and 2
-deoxyuridine or
2
-deoxycytidine decreased 50-60-fold compared with the unmodified
ones. Substitution of the 2
-hydroxy group by
2
-O-methyl-2
-deoxyuridine/cytidine led to a complete loss of activity. In contrast, additional introduction of
2
-fluoro-2
-deoxycytidine into ribozymes containing
2
-fluoro-2
-deoxyuridine led to a 2-fold increase in catalytic
activity independently of the presence or absence of terminal
phosphorothioate groups (compare RE917 (S, FU) and RE917 (FU, FC) in
Table IV). According to these results, modified ribozymes containing
2
-fluoro-2
-deoxycytidine and 2
-fluoro-2
-deoxyuridine with or
without phosphorothioate groups were used for the studies in cell
culture.
|
|
To examine the cleavage properties of the modified ribozymes ex
vivo, a N-ras/luciferase fusion minigene was
constructed as shown in Fig. 6. A 452-bp
N-ras DNA fragment, containing 50-bp 5-untranslated
sequences, the N-ras translation initiation codon, and
sequences coding for the first 134 amino acids of wild-type or mutant
N-ras, was fused in frame with the firefly luciferase gene.
In addition the AUG translation initiation codon of the luciferase gene
was mutated to ATA. In this construct the expression of the luciferase
gene depends on an intact AUG on the N-ras/luciferase fusion
mRNA. Thus, cleavage of N-ras sequences by ribozymes can be assessed by the reduction in luciferase activity.
HeLa cells were transfected with plasmids containing either wild-type
or mutant N-ras/luciferase fusion gene under the
transcriptional control of a cytomegalovirus promoter/enhancer element.
Neomycin-resistant clones were isolated and tested for luciferase
activity. The HeLa cell clones C#3 (GGT CGT mutation at position
763), T#4 (GGT
GTT transversion at position 764), and W#2
(wild-type N-ras sequence) showed the highest luciferase activity and
were chosen for all subsequent experiments.
These clones were transiently transfected with the ribozymes MRE763C and MRE764U using LipofectAMINETM as described under "Experimental Procedures." As a control for unspecific cleavage, a ribozyme containing an active catalytic site but no homology to the target N-ras sequence was used (nonsense ribozyme). Similarly, catalytically inactive ribozymes containing an adenosine residue instead of guanosine at position 5 (iMRE763C and iMRE764U) were used to estimate the reduction in luciferase activity caused solely by the hybridization of the ribozymes to the target sequences (antisense effect).
Treatment of HeLa cells with LipofectAMINETM alone caused a significant reduction in luciferase activity. This effect was attributed to the toxicity of cationic liposomes on HeLa cells since a large number of cells died shortly after the addition of the compound. This toxicity was, however, counteracted by the presence of nucleic acid in the transfection mixture. For this reason we decided to use HeLa cells treated with the nonsense ribozyme as a control for our experiments (100% luciferase activity).
The ribozyme MRE763C, targeted against the mutation at position 763 of
the N-ras/luciferase mRNA, was most effective in
inhibiting luciferase gene expression. An inhibition of up to 60% was
observed at a ribozyme concentration of 10 µM (Fig.
7). Lower ribozyme concentrations (5 µM)
led to a lower inhibitory effect. A minor decrease in luciferase
activity (20%) was seen with the inactive ribozymes iMRE764U and
iMRE763C, which can be explained by an antisense effect. For this study
eight independent experiments were evaluated, whereas the deviation of
the mean was calculated for each value from six parallel reactions.
The reduction in luciferase activity caused by MRE763C was specific. Treatment of the HeLa cell clone W#2, which expresses wild-type N-ras sequences, with MRE763C led to a 20% reduction in luciferase activity (mean value of 4 experiments, Fig. 7). This value is within the range of the luciferase activity obtained from HeLa cells treated with the inactive form of the ribozyme (iMRE763C) and thus is probably due to an antisense effect of ribozyme MRE763C on the wild-type N-ras/luciferase mRNA.
Although the reduction in luciferase activity suggests that the ribozymes MRE764U and MRE763C cleave the fusion mRNA efficiently, a quantitative assessment of RNA molecules cleaved by the ribozymes is not possible by this assay. To investigate the reduction in the amount of N-ras/luciferase mRNA, a semi-quantitative RT-PCR reaction was established. For this total RNA was isolated from clone C#3 by the guanidine-isothiocyanate method. Since cleavage of target RNA may occur during the RNA isolation procedure (47, 48), an internal control for the RT-PCR was generated to estimate the degree of N-ras/luciferase mRNA cleavage during the extraction protocol. For this, a 50-bp oligonucleotide was cloned within the ras sequences in the expression vector pcDNA3-LUCFUC. From this construct an in vitro transcribed RNA, 50 nucleotides longer than the N-ras/luciferase mRNA, was generated and served as internal control in the RT-PCR. The amplified region of the internal standard thus can be easily distinguished from the RT-PCR product originated from the N-ras/luciferase mRNA on agarose gels.
Total RNA obtained from the HeLa cell clone C#3 was mixed with the internal standard RNA at a molar ratio of approximately 1. The samples were treated with DNase I to destroy any residual DNA remaining in the RNA preparation. At this stage any ribozyme-mediated cleavage of RNA will affect equally the internal standard transcript and the N-ras/luciferase target sequences. Upon reverse transcription and PCR, the internal standard control should generate a 450-bp-long PCR product, while amplification of a segment of the N-ras/luciferase fusion mRNA should generate a 394-bp DNA fragment.
The RT-PCR done on RNA obtained from clone C#3 treated with the
nonsense ribozyme gave the expected product of 394 bp (Fig. 8, lane 5). When the ethidium bromide
intensity of this band was compared with that obtained from untreated
cells (Fig. 8, lane 4), no significant reduction was
observed, suggesting that treatment of the cells with a nonsense
ribozyme does not reduce the amount of N-ras/luciferase
transcripts. In addition, cleavage of the internal control was not
observed (compare the ethidium bromide intensity of the 450-bp PCR
product in Fig. 8, lanes 2 and 5). In contrast,
treatment of cells with the active ribozyme MRE763C caused a
significant reduction in N-ras transcripts. A densitometric scanning analysis of the PCR products observed in the agarose gel
revealed at least a 10-fold reduction in the amounts of
N-ras/luciferase mRNA. A reduction of the internal
control was also observed, confirming the observations of others that
during RNA extraction a cleavage of target RNA occurs (47, 48).
However, only a 3-fold reduction in the amounts of control RNA were
observed, suggesting that most of the N-ras/luciferase
mRNA cleavage occurred inside the cells and not during the
extraction procedure.
In the present study we examined the catalytic efficiencies,
specificities, and intracellular activities of hammerhead ribozymes targeted against N-ras transcripts containing wild-type
sequences or point mutations at codon 13. For short synthetic
substrates, the Km values of MRE763C and MRE764U
ranged between 65 and 82 nM and the
kcat values between 0.7 min 1 and
0.9 min
1, respectively, resulting in catalytic
efficiencies (kcat/Km) of
0.2·106 s
1·M
1.
The catalytic data obtained with the larger RNA substrate correlates with predictions of the FOLD program, which suggests that the N-ras RNA folds into a secondary structure with a number of
hairpins, double-stranded regions, bulges, and mismatches, which may
hinder the binding of ribozymes to the target. The ribozymes RE917,
MRE764U, and MRE763C, which show best catalytic properties under
single-turnover conditions, bind to a region of the N-ras
transcript composed of double-stranded structures interrupted by
unpaired ribonucleotides. These mismatches produce particularly
single-stranded regions where ribozyme binding is facilitated. RE990
and RE1035, in contrast, show low cleavage efficiency, probably because
their target sites are embedded within the stem of stable hairpin
structures.
Ribozyme MRE763C was found to be extremely specific in
vitro. A single ribonucleotide change at position 763 (C G
transversion) abolished completely cleavage by the ribozyme. Thus,
ribozyme MRE763C could be a very useful in the context of a clinical
application, since it will cleave exclusively mutated N-ras
transcripts but will dispense the wild-type counterparts.
For a clinical application of ribozymes, high catalytic efficiency,
stability, and good availability has to be achieved. Rapid degradation
of oligoribonucleotides in living cells diminishes availability of the
ribozyme with a concomitant decrease in efficiency. Modifications such
as terminal phosphorothioate groups and 2-modifications lead to high
stability against digestion by nucleases but are often accompanied by a
decrease in catalytic efficiency. For example, replacement of
U4 within the catalytic core by its 2
-O-methyl analogue leads to a significant reduction of the ribozyme catalytic efficiency (28). This is in agreement with results presented in this
study, as 2
-O-methyl substitution causes complete catalytic inactivation of ribozyme RE917 (Table IV). In contrast, replacement of
2
-hydroxy groups by amino or fluorine groups reduce the catalytic efficiency by only 2-3-fold. Similarly, in a previous study
Heidenreich et al. (30) found that a ribozyme containing
2
-amino groups at positions 4 and 7 of the catalytic core was almost
as active as its unmodified counterpart. This was explained by
proton-donor and acceptor properties of the amino group, which are
similar to those of the hydroxyl group. Thus, these observations reveal that the 2
-position of U4 plays an essential role in the
ribozyme catalysis.
For the ex vivo studies, modified ribozymes containing
2-fluoro-2
-deoxyuridine/cytidine groups alone or in combination with terminal phosphorothioate linkages were used. The modified ribozymes were active ex vivo. The N-ras/luciferase
reporter system used in our studies provides a very sensitive assay to
detect ribozyme activity, since the expression of the luciferase
reporter gene depends on an intact N-ras sequence. This cell
experiments revealed a reduction in luciferase activity of up to 55%
in HeLa cells treated with MRE763C. A low but detectable antisense
effect was also observed since incubation of HeLa cells expressing
mutant N-ras/luciferase fusion transcripts with the inactive
forms of MRE763C or MRE764U resulted in a 20% reduction in luciferase
activity. The cleavage activity of ribozyme MRE763C was restricted to
mutated N-ras sequences, as the luciferase activity in HeLa
cells expressing a wild-type N-ras/luciferase transcript was
not reduced above the levels expected from an antisense effect.
Estimates of ribozyme activity ex vivo based on reporter assays have to take into account the half-life of the protein being measured, in our case a Ras/luciferase fusion protein. For example, a reduction in enzyme activity will not reflect faithfully the decrease in the amounts of the corresponding mRNA if the half-life of the protein is high. Since the half-life of a Ras/luciferase protein was not known, we estimated the cleavage efficiency at the RNA level by analyzing directly the amount of mRNA molecules cleaved by the ribozyme MRE763C. For this type of analysis an internal RNA standard is required, since cleavage of the substrate may also occur during the RNA extraction protocol (47, 48). The amount of RT-PCR product obtained from the standard RNA can be estimated by densitometric evaluation of the ethidium bromide bands in the agarose gel and thus can be correlated to the amounts of input RNA in the reaction. A systematic analysis of the amounts of ras/luciferase mRNA present in the HeLa clone C#3 was conducted by mixing different amounts of total cellular RNA with a constant amount of standard RNA. At a standard concentration of 0.5 amol (3·105 RNA molecules), equivalent RT-PCR signals from total cellular RNA and standard RNA were observed, suggesting that the amount of ras/luciferase mRNA expressed in this cell clone was roughly 3·105 molecules. After treatment of the cells with ribozyme MRE763C, no visible RT-PCR product was observed (Fig. 8, lane 6). Since the sensitivity of the RT-PCR reaction was 3·104 RNA molecules, the amount of ras/luciferase transcripts in the ribozyme-treated cells was reduced from 3·105 to less than 30,000 molecules. Not all of this reduction could be attributed to specific cleavage of the RNA template, since a decrease in the amount of internal standard RNA was also observed. Nevertheless, our observation suggests that a significant portion of the ras/luciferase transcripts was cleaved inside the cells and not during the RNA extraction procedure.
The usefulness of ribozyme MRE763C for clinical purposes, for example within purging strategies for leukemic cells in the context of autologous bone marrow transplantation, still remains to be demonstrated. The experiments presented here provide a starting point on which clinical application of MRE763C could be based.
We thank O. Heidenreich, S. Klein, P. Marschall, and M. Schnee for critical reading of the manuscript and helpful discussions and H. Brill for the measurement of the mass spectra. The Georg-Speyer-Haus is supported by the Bundesministerium für Gesundheit and the Hessisches Ministerium für Wissenschaft und Kunst.