- Email: [email protected]
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko1*, V. Helin2, J.R. Bertrand2, M. Gottikh3, C. Malvy2 1 Bioalliance Pharma SA, 59, boulevard M.-Valin, 75015 Paris, France CNRS UMR 8121, Institut Gustave-Roussy PR2, 39, rue Camille-Desmoulins, 94805 Villejuif Cedex, France 3 Belozersky Institute of Physicochemical Biology, Lomonosov Moscow State University, Leniskie Gory, 119992 Moscow, Russia *Correspondence: [email protected] 2
The aim of the present investigation was to study the behavior of PAMAM dendrimer-nucleic acid complexes in vitro and living cells. We demonstrated the rapid and sensitive detection of mRNA in living cells using molecular beacon pair, one with a donor and the other with a quenching fluorophore that hybridises to adjacent regions on the same mRNA target, resulting in fluorescence resonance energy transfer (FRET). The molecular beacon was composed of a 13-nt loop structure containing the antisense sequence that can hybridise with the AUG translational start site of the Friend env gene. It was shown that SuperFect may stimulate the antisense ON-RNA hybridisation. The secondary structure of antisense oligonucleotide was optimized. An antisense sequence-specific inhibition of 75% was obtained for one reporter gene with a stem-loop ODN containing four phosphorothioate groups, two at each end. Key words : Antisense oligonucleotide – Molecular beacon – PAMAM dendrimer.
PAMAM dendrimers are cationic polymers that have been used for the delivery of genes and oligonucleotides to cells [1-4]. However, little is known about the behavior of dendrimer-nucleic acid complexes once they reach the cell interior. Factors limiting the use of antisense oligonucleotides (ONs) as therapeutic agents are inefficient cellular uptake and intracellular transport to RNA target. To overcome these obstacles, the molecular beacon (MB) oligonucleotide structure used in these studies was designed to study the antisense oligonucleotide interaction in the presence of the dendrimer with RNA target in vitro and inside of the cells [5]. The molecular beacon pair, one with a donor and the other with a quenching fluorophore that hybridises to the mRNA target, which results in fluorescence resonance energy transfer (FRET) [6]. The molecular beacon was composed of a 13-nt loop structure containing the antisense sequence that can hybridise with the AUG translational start site of the Friend env gene. We could therefore detect env mRNA in living cells by FRET. We clearly demonstrated the molecular beacon interaction with target inside of the cells using the dendrimer approach. In addition, in order to evaluate the optimal secondary structure of the antisense ON in enhancing the inhibition of the targeted genesʼ expression, we used a rapid screening system which measures the transient expression of two reporter genes, one used as target (GFP), the other as control (β-gal) [3]. The plasmids were co-transfected simultaneously into Hela cells using SuperFect. The ON complexed with the same transfecting reagent was added to cells 2 h before the plasmids. β-gal and GFP expression was measured in the cell lysate, 16 h after plasmid co-transfection. The ratio between activities (Rc) was determined using an ON that was not complementary to either plasmid to eliminate possible variations in gene expression due to a polyanionic effect caused by any short nucleic acid sequence. Despite great individual variations of GFP and β-gal
with the control ON, the GFP/β-gal ratio remains constant. This argues for a linear response of both GFP and β-gal. This system was validated through investigating the effect of the dendrimer vector on ON biological activity. Antisense sequence-specific inhibition of more than 75% of one reporter gene was obtained with a chimeric-structured ON containing four phosphorothioate groups, two at each end.
I. MATERIALS AND METHODS 1. Materials
Ethylenediamine dihydrochloride, β-mercaptoethanol, Hoechst 33342, 1-ethyl-3(3ʼ-dimethylaminopropyl)carbodii mide, dimethylacetamid and FITC isomer I were purchased from Sigma. 5ʼ-Phosphate-ON, 3ʼ-dabcyl CPG 500 were purchased from Glen Research. Propidium iodide and CPRG were purchased from Boehringer Mannheim. The pCMVβ-gal and pEGFP-N1 plasmids were purchased from Clontech. The degraded PAMAM dendrimer SuperFect (3 mg/ml) was purchased from Qiagen. SuperFect is degraded at high temperature. The dendrimeric structure presents 140 terminal NH2 groups on its surface (sixty of which are positively charged at pH 7) and has a molecular weight of 35 kDa after purification. Oligonucleotides purchased from Eurogentec (Searing, Belgium) were desalted on G25 Sephadex Columns (see Tables I and II for ODN sequences). The oligonucleotides AS6-AS8 and Ct2 contain two phosphorothioate bonds at the 5ʼ and 3ʼ ends. All oligonucleotides were quantified by their absorbance at 260 nm, their purity was confirmed by mass spectrometry. The oligonucleotides MB AS and MB Ct (Table I) with a phosphate flanking at the 5ʼ-end and dabcyl at 3ʼ-end were 75
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko, V. Helin, J.R. Bertrand, M. Gottikh, C. Malvy
Table I - The design of molecular beacons and target oligonucleotides. Name
Sequence (5’-3’) CA C GC
MB AS
a
5’- FITC - GCGCTTA T GAA 3’- Dabcyl - CGCGAATT CTT A
MB Ct
GC T G
Note Loop 13/Stem 11
C A T
Loop 13/stem 11
AG C T G T A C C CG
T
5’- FITC - CGCGAAT CTT 3’- Dabcyl - GCGCTTA T GAAC A
CCAGCAGAATCGACACATGGCGTGTTCAACGCT
Perfect match
DNA target (31DM)
CCAGCAGATTCGACACATGACGTGTTCAACGCT
DM: double base mismatch
RNA target (31R)
CCAGCAGAAUCGACACAUGGCGUGUUCAACGCU
Perfect match
DNA target (31D)b c
MB, molecular beacons. AS, antisense sequence. Ct, control sequence. The upper case bold letters correspond to residues participating in target binding. b The 21 base target sequence complementary to antisense fragment of MB AS is underlined. c The upper case bold letters correspond to mismatched bases in the target. a
HeLa (human epitheloid carcinoma cells; F. Clavel, Institut Pasteur, Paris, France) cell lines were grown in DMEM medium supplemented, respectively, with 10% of heat-inactivated FBS (Gibco BRL), streptomycin (100 µg/ml), and penicillin (100 U/ml). Cell lines were incubated at 37°C in 5% CO2. The culture media, trypsin, and PBS were pre-warmed to 37°C before adding on cells.
Table II - Inhibition of GFP expression by vectorized ONs. ODN
Possible secondary structure
Tm
a
(°C)
AS1
3 ’-CGGCAG T T CTCGAC-5 ’ T T GAGCTGCACGCTGCCGTC
AS2
3 ’-CGGCAG 5 ’-GAGCTGCACGCTGCCGTC
AS3
5 ’-GCGA GCTGCA C G CGT 3 ’-CGCT GC C
AS4 5 ’-GA 3 ’-CT
AS5
Ct 1
AS6
AS7
GCT C GCA G CGT C GC
5 ’-GCGTA GA 3 ’-CGCAT CT
GCT C GCA G CGT GC C
GTC GCC A 5 ’-GCGTAGA C 3 ’-CGCATCT GC G TGC
5 ’-G*A* 3 ’-C*T*
GCT C GCA G CGT GC C
GCTGCA C 5 ’-G*C*GA G 3 ’-C*G*CT GC CGT C
AS8 C
Ct 2
T T
3 ’- G*G*CAG C T CGCTG C C GTC C CGA* G*-5 ’ G T A
5 ’-T*G*AACACGCCATGTCGATTC*T* -3 ’
% of the pEGFPN1 inhibition
60
20
56
20
52
20
48
10
56
15
50
0
48
75
55
35
32
20
2. Molecular Beacon synthesis
The oligonucleotides MB AS and MB Ct (Table I) with a phosphate flanking at the 5ʼ-end and Dabcyl at 3ʼ-end were synthesized using the routine β-cyanoethyl phosphoramide method, 3ʼ-dabcyl CPG 500 polymer and 5ʼ-phosphate-ON as the phosphorylating agent. A solution containing the 3ʼ-dabcyl-5ʼ-phophorylated oligonucleotide (0.1-0.2 µmol) in 700 µl of water was supplemented with 240 mg of ethylenediamine dihydrochloride (1.8 mmol) and 120 mg of 1-ethyl-3(3ʼdimethylaminopropyl) carbodiimide (0.6 mmol). The mixture was vortexed and incubated at room temperature for 3 h. The oligonucleotide ethylenediamine derivative was then separated from excess reagents by size exclusion on a NAP-10 column (Pharmacia LKB Biotechnology), followed by precipitation by adding a 10-fold excess of 2% LiClO4 in acetone. This derivative was subsequently dissolved in 60 µl of water and the solution was supplemented with 20 µl of 1 M carbonate buffer, pH 11, 80 µl of dimethylacetamide, then 1 mg of FITC (2.5 µmol) was added. The mixture was vortexed and incubated at room temperature in the dark for 2-4 h. The dabcyl-3ʼ-ODN-5ʼ-FITC conjugate was separated from salts and free FITC by size exclusion on a NAP-10 column and purified by 20% PAGE in the presence of 7 M urea. The purity of MB AS and Ct was confirmed by mass spectrometry.
3. Molecular Beacon assay -
Ten nM of MB was incubated at 37°C in the presence of nucleic acid target in 10 mM phosphate buffer containing 50 mM NaCl. The reaction was performed in a quartz cuvette (final volume 0.4 ml) and fluorescence was measured using a fluorometer SFM 25 Kontron and real-time computed with the attached software WIND25 1.50. Excitation was at 488 nm and emission at 515 nm. Fluorescence was expressed in rf units.
0
synthesized in the laboratory of Nucleic Acids Chemistry of Lomonosov Moscow State University by Dr E. Romanova and Dr. E. Volkov. 76
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko, V. Helin, J.R. Bertrand, M. Gottikh, C. Malvy
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
4. Preparation of plasmid and oligonucleotide complexed to the transfection agent
reaction buffer (4.6 mg/ml CPRG, 80 mM phosphate buffer pH 7.4, 0.7% β-mercaptoethanol and 9 mM MgCl2) in a 96-well plate and incubated at 37°C for 20 min. β-Galactosidase activity was then detected by reading the absorbance at 570 nm using an automatic reader spectrophotometer (Dynatech Laboratories). The amount of green fluorescent protein was directly measured on the remaining supernatant with a spectrofluorimeter (Kontron SFM 23/B) using a 488 nm excitation wavelength and reading the emission at 507 nm.
For the fluorescence analysis, 1.2 µg of env plasmid and 5 µg of MB were correspondingly mixed with 6 µl (18 µg) and 5 µl (15 µg) of SuperFect in a final volume of 150 µl of 10 mM Hepes (pH 7.0) containing 150 mM NaCl for 10 min at room temperature. For the biological assay, 1.2 µg plasmid (0.8 µg of pEGFPN1 and 0.4 µg of pCMVβ-gal) and 5 µg of ON were correspondingly mixed with 6 µl (18 µg) and 5 µl (15 µg) of SuperFect in a final volume of 150 µl of 10 mM Hepes (pH 7.0) containing 150 mM NaCl for 10 min at room temperature.
II. RESULTS AND DISCUSSION
A 35-nt stem-loop structure (MB AS) composed of 11-nt complementary flanking sequences with one mismatch and a 13-nt intervening loop sequence was synthesized (Table I). The molecular beacon contained the 21-member sequence 5ʼTGAACACGCCATGTCGATTCT-3ʼ complementary to the translation initiation region of env RNA of the Friend murine leukaemia retrovirus. FITC, the fluorescent donor, and DABCYL, the acceptor, were conjugated to the 5ʼ and 3ʼ ends of the molecule, respectively, through spacer arms. In the stem-loop configuration, the FITC and DABCYL moieties are in close enough proximity so that FRET occurs and no fluorescent signal is observed. In the presence of a complementary target sequence, a bimolecular helix is formed causing the stem-loop structure to open and fluorescent signal is detected. The molecular beacon with reverse antisense sequence (MB Ct) was used as control. To test the specificity of the structures used, MBs with different loop sequences were mixed in solution with 31-nt DNA (31D) and RNA (31R) oligonucleotide targets and examined by a spectrofluorimeter (SFM 25 Kontron) using a 488 nm excitation wavelength and reading the emission at 510 nm. As expected, the fluorescence signal was detected only when MB AS was mixed with 31D and 31R targets to which it could hybridise (Figure 1A). No fluorescent signal was observed in the presence of oligonucleotide targets with mismatches (data not shown). The optimal ratio between ODN targets and MB AS was measured (Figure 1B). It corresponds to 3 for 31D and 5 for 31R oligonucleotides. This last ratio was used to study the MBs-RNA interactions. We examined the ability of MB AS to hybridise with denaturated and non-denaturated forms of RNA targets with different lengths (Figure 2). The 380-nt (380R) and 2600-nt (2600R) fragments of the env RNA were synthesized. 2600R corresponds to the complete mRNA of env gene. The MB AS hybridisation with 31R and denaturated 380R was completed within 30-50 min. The MB AS hybridisation with non-denatured 380R is very slow. The same tendency was observed for MB AS hybridisation with denaturated and non- denaturated 2600R, where the reaction was complete after 12 h of incubation. The hybridisation efficacy was 100% for the small target and 4050% for 380-nt and 2600-nt RNA target. The secondary RNA structure is more thermodynamically stable than the MB/ RNA double stranded structure. These results confirmed the influence of the secondary RNA structure on MB hybridisation [5, 6] and can be explained if no fluorescent signal was observed in the presence of MB Ct (data not shown). We also examined the efficacy of MB hybridisation in the
5. Cell transfection protocol
The FBS used for cell transfection was always heat-inactivated for 30 min at 56°C. For the biological assay, cells were seeded on 6-well plates to obtain 60-80% confluency (4 x 105 HeLa cells). The next day, cells were washed with PBS and treated with 150 µl of the SuperFect-ON preparation diluted with 700 µl of 10% FBS DMEM (with antibiotics), 2 h before the addition of 150 µl of SuperFect- pBK-57Env plasmid or mixture of pEGFP-N1 and pCMVβ-gal plasmids.
6. Flow cytometry analysis
For the flow cytometry analysis, cells were seeded on 24well plates (8 x 104 HeLa cells). The following day, 5 µg of MB complexed to SuperFect was incubated with cells for different times. Adherent cells were trypsinized and then all cells were washed with PBS. Cells were centrifuged for 10 min at 400 g and resuspended in 400 µl PBS containing 10 µg/ml propidium iodide. Mean cellular fluorescence intensities for 5,000 or 10,000 viable cells were determined on a Coulter EPICS Elite dual-laser flow cytometer. Dead cells were excluded by two means: through forward and side scatter gatings and by using propidium iodide to stain the dead cells. For cell cycle measurements, cells were resuspended in PBS containing 10 µg/ml propidium iodide and 20 µg/ml Hoechst 33342, incubated in the dark for 30 min at 37°C [7] and analyzed by dual-laser flow cytometer.
7. MB stability inside cells
MB (5 µg) delivered by SuperFect was incubated for 24 and 48 h with HeLa cells in a 12-well plate. Cells were washed three times with PBS, trypsinized, centrifuged, and washed again three times with PBS. The pellet was resuspended in 1 ml of water, vortexed, and kept for 30 min at - 20°C. MB was then isolated by phenol extraction, precipitated by adding a 10-fold excess of 2% LiClO4 in acetone, and analyzed by 20% PAGE containing 7 M urea. The gel was read using a fluoroimager (Storm 840, Molecular Dynamics).
8. Biological assay system
β-Galactosidase and green fluorescent protein [8] expression was detected 16 h after plasmid co-transfection. Cells were washed twice with PBS, and 400 µl of reporter lysis buffer (Promega) was added for 15 min. Cells were then scraped and centrifuged at 10,000 g for 15 min at 4°C. For the β-galactosidase assay, 5 µl of supernatant was mixed with 100 µl of a CPRG 77
180 160 140 120 100 80 60 40 20 0
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko, V. Helin, J.R. Bertrand, M. Gottikh, C. Malvy
A
% of MB hybridisation
Fluorescence Intesivity, f.u.
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
31D+MB AS 31R+MB AS 31D+MB Ct 31R+MB Ct
A
100 75 50 25 0
4
0
8
12
16
20
Time (h)
0
10
20
30
40
380RD
380RN
380RSU
31R
Time (min)
% of MB hybridisation
100
% of MB hybridisation
60
B
75
B
45 30 15 0 4
0
50
8
12
16
20
Time (h)
31D
25
2600RD
31R 1
2
3
4
2600RSU
Figure 2 - The influence of the target structure and its length on MB AS hybridisation. We used fragments of the env mRNA with sizes of 380 (A) and 2600 (B) bases. 0.1 nmol (1 µg) of the MB in the presence or absence of SuperFect (3 µg) was mixed with 0.5 nmol (4 mg for 380mer RNA and 20 mg for 2600-mer RNA) of the env mRNA fragment in 500 µl of 10 mM phosphate buffer containing 50 mM NaCl. 31R: 31-mer RNA target, RD: hybridisation with denatured RNA at 90°C for 5 min, RN: hybridisation with native RNA, RSU: hybridisation in the presence of SuperFect. The hybridisation efficacy was calculated as Q = [(F - Fmin)/(Fmax - Fmin)] x 100%, where Fmax = maximum fluorescence, Fmin = minimum fluorescence, F = detected fluorescence.
0 0
2600RN
5
[ON target]/[MB AS] ratio Figure 1 - (A) Monitoring of the MBs hybridisation with 31D and 31R targets. (B) Hybridisation efficacy of MB AS with 31D and 31R targets as a function of the [target]/[MB AS] ratio. The hybridisation efficacy was calculated as Q = [(F - Fmin)/(Fmax - Fmin)] x 100%, where Fmax = maximum fluorescence, Fmin = minimum fluorescence, F = detected fluorescence.
presence of SuperFect. No fluorescent signal is observed in the presence of 31R or 31D targets (data not shown). It was possible that small oligonucleotides interacted with non-bounded amino groups of PAMAM-dendrimer and could not hybridise with MB. The MB interaction with long RNA fragments (380R and 2600R) in the presence of SuperFect was quicker than without dendrimer (Figure 2). One hypothesis to support this result would be that SuperFect interacts with RNA by electrostatic forces and destroys the RNA secondary structure. Owing to this change of RNA structure the MB would be more efficient. We therefore propose that SuperFect stimulates the MB/RNA hybridisation. We then studied the kinetics of MB hybridisation with env RNA in living cells. To obtain the env RNA expression in HeLa cells the plasmid pBK-57Env was transfected by SuperFect 16 h before MBs transfection. At various times after MB AS or MB Ct transfection, the cell fluorescence was measured by flow cytometry analysis (Figure 3). The mean fluorescence was measured taking into account living cells only. Higher levels of cellular fluorescence were observed in cells treated with MB AS vectorized by dendrimer after 20 h of oligonucleotide incubation when compared to cells treated
cells
% of fluorescent cells
60
MB AS
50
MB Ct
40 30 20 10 0 0
8
16
24
32
40
48
Incubation time (h) Figure 3 - Percentage of fluorescent cells after HeLa cells incubation in the presence and absence of the MBs. The plasmid pBK-57Env was transfected with SuperFect 16 h before MBs transfection. The fluorescence was measured by flow cytometry analysis according to time.
with the MB Ct. No MB degradation was shown after 24 hrs of oligonucleotide incubation (data not shown). Non- treated cells displayed no discernible fluorescence. When compared to the background an increase of fluorescence was seen with 78
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko, V. Helin, J.R. Bertrand, M. Gottikh, C. Malvy
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
MB-treated cells. Therefore with the Ct MB the fluorescence observed at 24 h might originate in non specific interactions. After 48 h of MB Ct incubation the increasing of fluorescence may be explained by partial oligonucleotide degradation. The increase of fluorescent cells number after MB AS transfection is consistent with the long time required for ON hybridisation with the RNA target (Figure 3). Based on these results, one important conclusion may be drawn. SuperFect may stimulate the antisense ON-RNA hybridisation. To determine the effective secondary structure of antisense oligonucleotide we used a rapid screening system, which measures the transient expression of two reporter genes, one used as a target, the other as a control [3]. With this system we have investigated the effect of the dendrimer vector on ODN biological activity. For this aim we constructed the antisense oligonucleotide with different secondary structures (Table II). An antisense sequence-specific inhibition of 75% was obtained for one reporter gene with a stem-loop ODN containing four phosphorothioate groups, two at each end. Its sequence corresponds exactly to the target sequence. At this time we do not know the exact role of dendrimers since they are used for the transfection of oligonucleotides in the cells. We do not know whether, once inside the cell, the dendrimer oligonucleotide dissociation takes place. However, we were able to observe intracellularly for 2 stem loop antisense oligonucleotides in the presence of dendrimers, mRNA binding in the cells for the first one and an antisense effect for the second one. Real-time visualization of specific endogenous mRNA expression in vivo has the potential to revolutionize medical diagnosis, drug discovery, developmental and molecular biology. The dendrimer molecular beacons approach promises to open new and exciting opportunities in sensitive gene detection and quantification in vivo.
REFERENCES 1. 2. 3.
4.
5. 6. 7.
8.
YOO H., SAZANI P., JULIANO R.L. - PAMAM dendrimers as delivery agents for antisense oligonucleotides. - Pharm. Res., 16, 1799-1804, 1999. BIELINSKA A.U., CHEN C., JOHNSON J., BAKER J.R., JR. - DNA complexing with polyamidoamine dendrimers: implications for transfection. - Bioconjug. Chem., 10, 843-850, 1999. HELIN V., GOTTIKH M., MISHAL Z., SUBRA F., MALVY C., LAVIGNON M. - Cell cycle-dependent distribution and specific inhibitory effect of vectorized antisense oligonucleotides in cell culture. - Biochem. Pharmacol., 58, 95-107, 1999. DELONG R., STEPHENSON K., LOFTUS T., FISHER M., ALAHARI S., NOLTING A., JULIANO R.L. - Characterization of complexes of oligonucleotides with polyamidoamine starburst dendrimers and effects on intracellular delivery. - J. Pharm. Sci., 86, 762-764, 1997. SOKOL D.L., ZHANG X., LU P., GEWIRTZ A.M. - Real time detection of DNA-RNA hybridization in living cells. - Proc. Natl. Acad. Sci. USA, 95, 11538-11543, 1998. TSOURKAS A., BEHLKE M.A., BAO G. - Structure-function relationships of shared-stem and conventional molecular beacons. - Nucleic Acids Res., 30, 4208-4215, 2002. GREGOIRE M., HERANDEZ-VERDUN D., BOUTEILLE M. - Visualization of chromatin distribution in living PTO cells by Hoechst 33342 fluorescent staining. - Exp. Cell Res., 152, 3846, 1984. MISTELI T., SPECTOR D.L. - Applications of the green fluorescent protein in cell biology and biotechnology. - Nat. Biotechnol., 15, 961-964, 1997.
ACKNOWLEDGEMENTS We thank E. Volkov and E. Romanova for molecular beacon synthesis. This work was supported by the Association de Recherches sur le Cancer (Grant 4310) and the Russian Foundation for Basic Research.
MANUSCRIPT Received 12 July 2004, accepted for publication 26 November 2004.
79
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko1*, V. Helin2, J.R. Bertrand2, M. Gottikh3, C. Malvy2 1 Bioalliance Pharma SA, 59, boulevard M.-Valin, 75015 Paris, France CNRS UMR 8121, Institut Gustave-Roussy PR2, 39, rue Camille-Desmoulins, 94805 Villejuif Cedex, France 3 Belozersky Institute of Physicochemical Biology, Lomonosov Moscow State University, Leniskie Gory, 119992 Moscow, Russia *Correspondence: [email protected] 2
The aim of the present investigation was to study the behavior of PAMAM dendrimer-nucleic acid complexes in vitro and living cells. We demonstrated the rapid and sensitive detection of mRNA in living cells using molecular beacon pair, one with a donor and the other with a quenching fluorophore that hybridises to adjacent regions on the same mRNA target, resulting in fluorescence resonance energy transfer (FRET). The molecular beacon was composed of a 13-nt loop structure containing the antisense sequence that can hybridise with the AUG translational start site of the Friend env gene. It was shown that SuperFect may stimulate the antisense ON-RNA hybridisation. The secondary structure of antisense oligonucleotide was optimized. An antisense sequence-specific inhibition of 75% was obtained for one reporter gene with a stem-loop ODN containing four phosphorothioate groups, two at each end. Key words : Antisense oligonucleotide – Molecular beacon – PAMAM dendrimer.
PAMAM dendrimers are cationic polymers that have been used for the delivery of genes and oligonucleotides to cells [1-4]. However, little is known about the behavior of dendrimer-nucleic acid complexes once they reach the cell interior. Factors limiting the use of antisense oligonucleotides (ONs) as therapeutic agents are inefficient cellular uptake and intracellular transport to RNA target. To overcome these obstacles, the molecular beacon (MB) oligonucleotide structure used in these studies was designed to study the antisense oligonucleotide interaction in the presence of the dendrimer with RNA target in vitro and inside of the cells [5]. The molecular beacon pair, one with a donor and the other with a quenching fluorophore that hybridises to the mRNA target, which results in fluorescence resonance energy transfer (FRET) [6]. The molecular beacon was composed of a 13-nt loop structure containing the antisense sequence that can hybridise with the AUG translational start site of the Friend env gene. We could therefore detect env mRNA in living cells by FRET. We clearly demonstrated the molecular beacon interaction with target inside of the cells using the dendrimer approach. In addition, in order to evaluate the optimal secondary structure of the antisense ON in enhancing the inhibition of the targeted genesʼ expression, we used a rapid screening system which measures the transient expression of two reporter genes, one used as target (GFP), the other as control (β-gal) [3]. The plasmids were co-transfected simultaneously into Hela cells using SuperFect. The ON complexed with the same transfecting reagent was added to cells 2 h before the plasmids. β-gal and GFP expression was measured in the cell lysate, 16 h after plasmid co-transfection. The ratio between activities (Rc) was determined using an ON that was not complementary to either plasmid to eliminate possible variations in gene expression due to a polyanionic effect caused by any short nucleic acid sequence. Despite great individual variations of GFP and β-gal
with the control ON, the GFP/β-gal ratio remains constant. This argues for a linear response of both GFP and β-gal. This system was validated through investigating the effect of the dendrimer vector on ON biological activity. Antisense sequence-specific inhibition of more than 75% of one reporter gene was obtained with a chimeric-structured ON containing four phosphorothioate groups, two at each end.
I. MATERIALS AND METHODS 1. Materials
Ethylenediamine dihydrochloride, β-mercaptoethanol, Hoechst 33342, 1-ethyl-3(3ʼ-dimethylaminopropyl)carbodii mide, dimethylacetamid and FITC isomer I were purchased from Sigma. 5ʼ-Phosphate-ON, 3ʼ-dabcyl CPG 500 were purchased from Glen Research. Propidium iodide and CPRG were purchased from Boehringer Mannheim. The pCMVβ-gal and pEGFP-N1 plasmids were purchased from Clontech. The degraded PAMAM dendrimer SuperFect (3 mg/ml) was purchased from Qiagen. SuperFect is degraded at high temperature. The dendrimeric structure presents 140 terminal NH2 groups on its surface (sixty of which are positively charged at pH 7) and has a molecular weight of 35 kDa after purification. Oligonucleotides purchased from Eurogentec (Searing, Belgium) were desalted on G25 Sephadex Columns (see Tables I and II for ODN sequences). The oligonucleotides AS6-AS8 and Ct2 contain two phosphorothioate bonds at the 5ʼ and 3ʼ ends. All oligonucleotides were quantified by their absorbance at 260 nm, their purity was confirmed by mass spectrometry. The oligonucleotides MB AS and MB Ct (Table I) with a phosphate flanking at the 5ʼ-end and dabcyl at 3ʼ-end were 75
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko, V. Helin, J.R. Bertrand, M. Gottikh, C. Malvy
Table I - The design of molecular beacons and target oligonucleotides. Name
Sequence (5’-3’) CA C GC
MB AS
a
5’- FITC - GCGCTTA T GAA 3’- Dabcyl - CGCGAATT CTT A
MB Ct
GC T G
Note Loop 13/Stem 11
C A T
Loop 13/stem 11
AG C T G T A C C CG
T
5’- FITC - CGCGAAT CTT 3’- Dabcyl - GCGCTTA T GAAC A
CCAGCAGAATCGACACATGGCGTGTTCAACGCT
Perfect match
DNA target (31DM)
CCAGCAGATTCGACACATGACGTGTTCAACGCT
DM: double base mismatch
RNA target (31R)
CCAGCAGAAUCGACACAUGGCGUGUUCAACGCU
Perfect match
DNA target (31D)b c
MB, molecular beacons. AS, antisense sequence. Ct, control sequence. The upper case bold letters correspond to residues participating in target binding. b The 21 base target sequence complementary to antisense fragment of MB AS is underlined. c The upper case bold letters correspond to mismatched bases in the target. a
HeLa (human epitheloid carcinoma cells; F. Clavel, Institut Pasteur, Paris, France) cell lines were grown in DMEM medium supplemented, respectively, with 10% of heat-inactivated FBS (Gibco BRL), streptomycin (100 µg/ml), and penicillin (100 U/ml). Cell lines were incubated at 37°C in 5% CO2. The culture media, trypsin, and PBS were pre-warmed to 37°C before adding on cells.
Table II - Inhibition of GFP expression by vectorized ONs. ODN
Possible secondary structure
Tm
a
(°C)
AS1
3 ’-CGGCAG T T CTCGAC-5 ’ T T GAGCTGCACGCTGCCGTC
AS2
3 ’-CGGCAG 5 ’-GAGCTGCACGCTGCCGTC
AS3
5 ’-GCGA GCTGCA C G CGT 3 ’-CGCT GC C
AS4 5 ’-GA 3 ’-CT
AS5
Ct 1
AS6
AS7
GCT C GCA G CGT C GC
5 ’-GCGTA GA 3 ’-CGCAT CT
GCT C GCA G CGT GC C
GTC GCC A 5 ’-GCGTAGA C 3 ’-CGCATCT GC G TGC
5 ’-G*A* 3 ’-C*T*
GCT C GCA G CGT GC C
GCTGCA C 5 ’-G*C*GA G 3 ’-C*G*CT GC CGT C
AS8 C
Ct 2
T T
3 ’- G*G*CAG C T CGCTG C C GTC C CGA* G*-5 ’ G T A
5 ’-T*G*AACACGCCATGTCGATTC*T* -3 ’
% of the pEGFPN1 inhibition
60
20
56
20
52
20
48
10
56
15
50
0
48
75
55
35
32
20
2. Molecular Beacon synthesis
The oligonucleotides MB AS and MB Ct (Table I) with a phosphate flanking at the 5ʼ-end and Dabcyl at 3ʼ-end were synthesized using the routine β-cyanoethyl phosphoramide method, 3ʼ-dabcyl CPG 500 polymer and 5ʼ-phosphate-ON as the phosphorylating agent. A solution containing the 3ʼ-dabcyl-5ʼ-phophorylated oligonucleotide (0.1-0.2 µmol) in 700 µl of water was supplemented with 240 mg of ethylenediamine dihydrochloride (1.8 mmol) and 120 mg of 1-ethyl-3(3ʼdimethylaminopropyl) carbodiimide (0.6 mmol). The mixture was vortexed and incubated at room temperature for 3 h. The oligonucleotide ethylenediamine derivative was then separated from excess reagents by size exclusion on a NAP-10 column (Pharmacia LKB Biotechnology), followed by precipitation by adding a 10-fold excess of 2% LiClO4 in acetone. This derivative was subsequently dissolved in 60 µl of water and the solution was supplemented with 20 µl of 1 M carbonate buffer, pH 11, 80 µl of dimethylacetamide, then 1 mg of FITC (2.5 µmol) was added. The mixture was vortexed and incubated at room temperature in the dark for 2-4 h. The dabcyl-3ʼ-ODN-5ʼ-FITC conjugate was separated from salts and free FITC by size exclusion on a NAP-10 column and purified by 20% PAGE in the presence of 7 M urea. The purity of MB AS and Ct was confirmed by mass spectrometry.
3. Molecular Beacon assay -
Ten nM of MB was incubated at 37°C in the presence of nucleic acid target in 10 mM phosphate buffer containing 50 mM NaCl. The reaction was performed in a quartz cuvette (final volume 0.4 ml) and fluorescence was measured using a fluorometer SFM 25 Kontron and real-time computed with the attached software WIND25 1.50. Excitation was at 488 nm and emission at 515 nm. Fluorescence was expressed in rf units.
0
synthesized in the laboratory of Nucleic Acids Chemistry of Lomonosov Moscow State University by Dr E. Romanova and Dr. E. Volkov. 76
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko, V. Helin, J.R. Bertrand, M. Gottikh, C. Malvy
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
4. Preparation of plasmid and oligonucleotide complexed to the transfection agent
reaction buffer (4.6 mg/ml CPRG, 80 mM phosphate buffer pH 7.4, 0.7% β-mercaptoethanol and 9 mM MgCl2) in a 96-well plate and incubated at 37°C for 20 min. β-Galactosidase activity was then detected by reading the absorbance at 570 nm using an automatic reader spectrophotometer (Dynatech Laboratories). The amount of green fluorescent protein was directly measured on the remaining supernatant with a spectrofluorimeter (Kontron SFM 23/B) using a 488 nm excitation wavelength and reading the emission at 507 nm.
For the fluorescence analysis, 1.2 µg of env plasmid and 5 µg of MB were correspondingly mixed with 6 µl (18 µg) and 5 µl (15 µg) of SuperFect in a final volume of 150 µl of 10 mM Hepes (pH 7.0) containing 150 mM NaCl for 10 min at room temperature. For the biological assay, 1.2 µg plasmid (0.8 µg of pEGFPN1 and 0.4 µg of pCMVβ-gal) and 5 µg of ON were correspondingly mixed with 6 µl (18 µg) and 5 µl (15 µg) of SuperFect in a final volume of 150 µl of 10 mM Hepes (pH 7.0) containing 150 mM NaCl for 10 min at room temperature.
II. RESULTS AND DISCUSSION
A 35-nt stem-loop structure (MB AS) composed of 11-nt complementary flanking sequences with one mismatch and a 13-nt intervening loop sequence was synthesized (Table I). The molecular beacon contained the 21-member sequence 5ʼTGAACACGCCATGTCGATTCT-3ʼ complementary to the translation initiation region of env RNA of the Friend murine leukaemia retrovirus. FITC, the fluorescent donor, and DABCYL, the acceptor, were conjugated to the 5ʼ and 3ʼ ends of the molecule, respectively, through spacer arms. In the stem-loop configuration, the FITC and DABCYL moieties are in close enough proximity so that FRET occurs and no fluorescent signal is observed. In the presence of a complementary target sequence, a bimolecular helix is formed causing the stem-loop structure to open and fluorescent signal is detected. The molecular beacon with reverse antisense sequence (MB Ct) was used as control. To test the specificity of the structures used, MBs with different loop sequences were mixed in solution with 31-nt DNA (31D) and RNA (31R) oligonucleotide targets and examined by a spectrofluorimeter (SFM 25 Kontron) using a 488 nm excitation wavelength and reading the emission at 510 nm. As expected, the fluorescence signal was detected only when MB AS was mixed with 31D and 31R targets to which it could hybridise (Figure 1A). No fluorescent signal was observed in the presence of oligonucleotide targets with mismatches (data not shown). The optimal ratio between ODN targets and MB AS was measured (Figure 1B). It corresponds to 3 for 31D and 5 for 31R oligonucleotides. This last ratio was used to study the MBs-RNA interactions. We examined the ability of MB AS to hybridise with denaturated and non-denaturated forms of RNA targets with different lengths (Figure 2). The 380-nt (380R) and 2600-nt (2600R) fragments of the env RNA were synthesized. 2600R corresponds to the complete mRNA of env gene. The MB AS hybridisation with 31R and denaturated 380R was completed within 30-50 min. The MB AS hybridisation with non-denatured 380R is very slow. The same tendency was observed for MB AS hybridisation with denaturated and non- denaturated 2600R, where the reaction was complete after 12 h of incubation. The hybridisation efficacy was 100% for the small target and 4050% for 380-nt and 2600-nt RNA target. The secondary RNA structure is more thermodynamically stable than the MB/ RNA double stranded structure. These results confirmed the influence of the secondary RNA structure on MB hybridisation [5, 6] and can be explained if no fluorescent signal was observed in the presence of MB Ct (data not shown). We also examined the efficacy of MB hybridisation in the
5. Cell transfection protocol
The FBS used for cell transfection was always heat-inactivated for 30 min at 56°C. For the biological assay, cells were seeded on 6-well plates to obtain 60-80% confluency (4 x 105 HeLa cells). The next day, cells were washed with PBS and treated with 150 µl of the SuperFect-ON preparation diluted with 700 µl of 10% FBS DMEM (with antibiotics), 2 h before the addition of 150 µl of SuperFect- pBK-57Env plasmid or mixture of pEGFP-N1 and pCMVβ-gal plasmids.
6. Flow cytometry analysis
For the flow cytometry analysis, cells were seeded on 24well plates (8 x 104 HeLa cells). The following day, 5 µg of MB complexed to SuperFect was incubated with cells for different times. Adherent cells were trypsinized and then all cells were washed with PBS. Cells were centrifuged for 10 min at 400 g and resuspended in 400 µl PBS containing 10 µg/ml propidium iodide. Mean cellular fluorescence intensities for 5,000 or 10,000 viable cells were determined on a Coulter EPICS Elite dual-laser flow cytometer. Dead cells were excluded by two means: through forward and side scatter gatings and by using propidium iodide to stain the dead cells. For cell cycle measurements, cells were resuspended in PBS containing 10 µg/ml propidium iodide and 20 µg/ml Hoechst 33342, incubated in the dark for 30 min at 37°C [7] and analyzed by dual-laser flow cytometer.
7. MB stability inside cells
MB (5 µg) delivered by SuperFect was incubated for 24 and 48 h with HeLa cells in a 12-well plate. Cells were washed three times with PBS, trypsinized, centrifuged, and washed again three times with PBS. The pellet was resuspended in 1 ml of water, vortexed, and kept for 30 min at - 20°C. MB was then isolated by phenol extraction, precipitated by adding a 10-fold excess of 2% LiClO4 in acetone, and analyzed by 20% PAGE containing 7 M urea. The gel was read using a fluoroimager (Storm 840, Molecular Dynamics).
8. Biological assay system
β-Galactosidase and green fluorescent protein [8] expression was detected 16 h after plasmid co-transfection. Cells were washed twice with PBS, and 400 µl of reporter lysis buffer (Promega) was added for 15 min. Cells were then scraped and centrifuged at 10,000 g for 15 min at 4°C. For the β-galactosidase assay, 5 µl of supernatant was mixed with 100 µl of a CPRG 77
180 160 140 120 100 80 60 40 20 0
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko, V. Helin, J.R. Bertrand, M. Gottikh, C. Malvy
A
% of MB hybridisation
Fluorescence Intesivity, f.u.
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
31D+MB AS 31R+MB AS 31D+MB Ct 31R+MB Ct
A
100 75 50 25 0
4
0
8
12
16
20
Time (h)
0
10
20
30
40
380RD
380RN
380RSU
31R
Time (min)
% of MB hybridisation
100
% of MB hybridisation
60
B
75
B
45 30 15 0 4
0
50
8
12
16
20
Time (h)
31D
25
2600RD
31R 1
2
3
4
2600RSU
Figure 2 - The influence of the target structure and its length on MB AS hybridisation. We used fragments of the env mRNA with sizes of 380 (A) and 2600 (B) bases. 0.1 nmol (1 µg) of the MB in the presence or absence of SuperFect (3 µg) was mixed with 0.5 nmol (4 mg for 380mer RNA and 20 mg for 2600-mer RNA) of the env mRNA fragment in 500 µl of 10 mM phosphate buffer containing 50 mM NaCl. 31R: 31-mer RNA target, RD: hybridisation with denatured RNA at 90°C for 5 min, RN: hybridisation with native RNA, RSU: hybridisation in the presence of SuperFect. The hybridisation efficacy was calculated as Q = [(F - Fmin)/(Fmax - Fmin)] x 100%, where Fmax = maximum fluorescence, Fmin = minimum fluorescence, F = detected fluorescence.
0 0
2600RN
5
[ON target]/[MB AS] ratio Figure 1 - (A) Monitoring of the MBs hybridisation with 31D and 31R targets. (B) Hybridisation efficacy of MB AS with 31D and 31R targets as a function of the [target]/[MB AS] ratio. The hybridisation efficacy was calculated as Q = [(F - Fmin)/(Fmax - Fmin)] x 100%, where Fmax = maximum fluorescence, Fmin = minimum fluorescence, F = detected fluorescence.
presence of SuperFect. No fluorescent signal is observed in the presence of 31R or 31D targets (data not shown). It was possible that small oligonucleotides interacted with non-bounded amino groups of PAMAM-dendrimer and could not hybridise with MB. The MB interaction with long RNA fragments (380R and 2600R) in the presence of SuperFect was quicker than without dendrimer (Figure 2). One hypothesis to support this result would be that SuperFect interacts with RNA by electrostatic forces and destroys the RNA secondary structure. Owing to this change of RNA structure the MB would be more efficient. We therefore propose that SuperFect stimulates the MB/RNA hybridisation. We then studied the kinetics of MB hybridisation with env RNA in living cells. To obtain the env RNA expression in HeLa cells the plasmid pBK-57Env was transfected by SuperFect 16 h before MBs transfection. At various times after MB AS or MB Ct transfection, the cell fluorescence was measured by flow cytometry analysis (Figure 3). The mean fluorescence was measured taking into account living cells only. Higher levels of cellular fluorescence were observed in cells treated with MB AS vectorized by dendrimer after 20 h of oligonucleotide incubation when compared to cells treated
cells
% of fluorescent cells
60
MB AS
50
MB Ct
40 30 20 10 0 0
8
16
24
32
40
48
Incubation time (h) Figure 3 - Percentage of fluorescent cells after HeLa cells incubation in the presence and absence of the MBs. The plasmid pBK-57Env was transfected with SuperFect 16 h before MBs transfection. The fluorescence was measured by flow cytometry analysis according to time.
with the MB Ct. No MB degradation was shown after 24 hrs of oligonucleotide incubation (data not shown). Non- treated cells displayed no discernible fluorescence. When compared to the background an increase of fluorescence was seen with 78
Real-time detection and efficacy of antisense oligonucleotides delivered by PAMAM dendrimers in living cells A. Maksimenko, V. Helin, J.R. Bertrand, M. Gottikh, C. Malvy
J. DRUG DEL. SCI. TECH., 15 (1) 75-79 2005
MB-treated cells. Therefore with the Ct MB the fluorescence observed at 24 h might originate in non specific interactions. After 48 h of MB Ct incubation the increasing of fluorescence may be explained by partial oligonucleotide degradation. The increase of fluorescent cells number after MB AS transfection is consistent with the long time required for ON hybridisation with the RNA target (Figure 3). Based on these results, one important conclusion may be drawn. SuperFect may stimulate the antisense ON-RNA hybridisation. To determine the effective secondary structure of antisense oligonucleotide we used a rapid screening system, which measures the transient expression of two reporter genes, one used as a target, the other as a control [3]. With this system we have investigated the effect of the dendrimer vector on ODN biological activity. For this aim we constructed the antisense oligonucleotide with different secondary structures (Table II). An antisense sequence-specific inhibition of 75% was obtained for one reporter gene with a stem-loop ODN containing four phosphorothioate groups, two at each end. Its sequence corresponds exactly to the target sequence. At this time we do not know the exact role of dendrimers since they are used for the transfection of oligonucleotides in the cells. We do not know whether, once inside the cell, the dendrimer oligonucleotide dissociation takes place. However, we were able to observe intracellularly for 2 stem loop antisense oligonucleotides in the presence of dendrimers, mRNA binding in the cells for the first one and an antisense effect for the second one. Real-time visualization of specific endogenous mRNA expression in vivo has the potential to revolutionize medical diagnosis, drug discovery, developmental and molecular biology. The dendrimer molecular beacons approach promises to open new and exciting opportunities in sensitive gene detection and quantification in vivo.
REFERENCES 1. 2. 3.
4.
5. 6. 7.
8.
YOO H., SAZANI P., JULIANO R.L. - PAMAM dendrimers as delivery agents for antisense oligonucleotides. - Pharm. Res., 16, 1799-1804, 1999. BIELINSKA A.U., CHEN C., JOHNSON J., BAKER J.R., JR. - DNA complexing with polyamidoamine dendrimers: implications for transfection. - Bioconjug. Chem., 10, 843-850, 1999. HELIN V., GOTTIKH M., MISHAL Z., SUBRA F., MALVY C., LAVIGNON M. - Cell cycle-dependent distribution and specific inhibitory effect of vectorized antisense oligonucleotides in cell culture. - Biochem. Pharmacol., 58, 95-107, 1999. DELONG R., STEPHENSON K., LOFTUS T., FISHER M., ALAHARI S., NOLTING A., JULIANO R.L. - Characterization of complexes of oligonucleotides with polyamidoamine starburst dendrimers and effects on intracellular delivery. - J. Pharm. Sci., 86, 762-764, 1997. SOKOL D.L., ZHANG X., LU P., GEWIRTZ A.M. - Real time detection of DNA-RNA hybridization in living cells. - Proc. Natl. Acad. Sci. USA, 95, 11538-11543, 1998. TSOURKAS A., BEHLKE M.A., BAO G. - Structure-function relationships of shared-stem and conventional molecular beacons. - Nucleic Acids Res., 30, 4208-4215, 2002. GREGOIRE M., HERANDEZ-VERDUN D., BOUTEILLE M. - Visualization of chromatin distribution in living PTO cells by Hoechst 33342 fluorescent staining. - Exp. Cell Res., 152, 3846, 1984. MISTELI T., SPECTOR D.L. - Applications of the green fluorescent protein in cell biology and biotechnology. - Nat. Biotechnol., 15, 961-964, 1997.
ACKNOWLEDGEMENTS We thank E. Volkov and E. Romanova for molecular beacon synthesis. This work was supported by the Association de Recherches sur le Cancer (Grant 4310) and the Russian Foundation for Basic Research.
MANUSCRIPT Received 12 July 2004, accepted for publication 26 November 2004.
79