| Nucleic Acids Research | Pages |
Synthesis of polyacrylamides N-substituted with PNA-like oligonucleotide mimics for molecular diagnostic applications
Introduction
Materials And Methods
Synthesis of ODNs, mimic oligomers and their conjugates with acrylamide
Preparation of PAA-oligomer co-polymers and their attachment to a solid surface
Hybridisation procedures
Results And Discussion
Synthesis of acrylamide-modified oligomers
Co-polymerisation of acrylamide-mimic conjugates with acrylamide and matrix preparation
Hybridisation properties of PAA-oligomer conjugates
Sandwich hybridisation assays
Conclusion
Acknowledgements
References
Synthesis of polyacrylamides N-substituted with PNA-like oligonucleotide mimics for molecular diagnostic applications
Received July 30, 1999; Revised and Accepted September 28, 1999
ABSTRACT Two types of oligonucleotide mimics relative to peptide nucleic acids (PNAs) were tested as probes in nucleic acid hybridisation assays based on polyacrylamide technology. One type of mimic oligomers represented a chimera constructed of PNA and phosphono-PNA (pPNA) monomers, and the other one contained pPNA residues alternating with PNA-like monomers on the base of trans-4-hydroxy-L-proline (HypNA). A chemistry providing efficient and specific covalent attachment of these DNA mimics to acrylamide polymers using a convenient approach based on the co-polymerisation of acrylamide and some reactive acrylic acid derivatives with oligomers bearing 5[prime]- or 3[prime]-terminal acrylamide groups has been developed. A comparative study of polyacrylamide conjugates with oligonucleotides and mimic oligomers demonstrated the suitability and high potential of PNA-pPNA and HypNA-pPNA chimeras as sequence-specific probes in capture and detection of target nucleic acid fragments to serve current forms of DNA arrays.
INTRODUCTION
In recent years, synthetic oligonucleotides have assumed a central role in a multitude of molecular biological techniques as linkers, primers, gene probes and segments for the construction of artificial DNA fragments. More recently, their potential as therapeutic agents based on the oligonucleotide hybridisation technique has attracted increasing interest (1,2). Hybridisation is widely used for various purposes to determine the presence in nucleic acids of a sequence that is complementary to the oligonucleotide probe. Particularly, it is a very attractive technique for detecting target DNAs, or RNAs, in diagnostics and for sequencing by hybridisation to oligonucleotide microchips, which can provide a simple, fast and inexpensive alternative to conventional direct sequencing methods (3-5). Hybridisation on solid supports involves the immobilisation of one of the interacting nucleic acids on a surface, while the other is free in solution. It offers convenient formats for the detection of nucleic acid complexes since unreacted molecules can be easily washed away after hybridisation, leaving only those which are specifically bound. Despite the general utility of this procedure, there are several problems: low rates of hybridisation, steric constraints, low capture and hybridisation efficiencies, and molecular and biological instability of probes. Numerous groups are developing novel approaches to DNA target and matrix preparation, probe labelling and read-out to make array design more flexible (for a review see 6). Among various supports proposed for solid phase hybridisation (polystyrene, nitrocellulose, nylon, activated dextran, diazotised cellulose, glass, etc.), polyacrylamide (PAA) matrices have been shown to have some advantages and convenience, including easy accessibility, and they have been successfully applied in DNA analysis and microchip technology (7-12).
Although natural oligodeoxyribonucleotides (ODNs) have been successfully used as sensor molecules for rapid analysis of nucleic acids in many cases, some problems still remain. Thus, ODNs have the inherent disadvantages of insufficient molecular stability in biological fluids and a relatively low stability of their duplexes. Over the past few years, a huge number of DNA analogues with modifications of the sugar-phosphate backbone and heterocyclic bases have been reported to improve the affinity and specificity of the target-probe interaction (for a short review see 13). Peptide nucleic acids (PNAs) are one of the most interesting DNA mimics due to their excellent binding properties and stability to the action of peptidases and nucleases (14,15). Classical PNAs represent achiral molecules with the sugar-phosphate backbone replaced by N-(2-aminoethyl)glycine units. It was shown that the use of PNAs as sensor molecules in diagnostically relevant assay formats leads to a simultaneous increase in both specificity and sensitivity, because of their higher affinity and ability to discriminate single base pair differences better than DNA probes (16-20). By virtue of their neutral backbone, PNAs can hybridise to complementary regions of nucleic acids in low salt concentrations that prevent the target nucleic acid from intra-strand folding and increases the accessibility of its target sequences. Nevertheless, the use of PNAs has some disadvantages connected with their poor water solubility and tendency to self-aggregate and sediment in low salt concentrations (20-22).
In an attempt to improve the potency of PNAs as diagnostic probes and therapeutics, a novel class of ODN mimics representing phosphono-PNA analogues (pPNAs), in which monomer units are connected with phosphonester bonds, has recently been synthesised (23,24). Investigation of their properties revealed that they are fully stable to the action of nucleases, and the introduction of negative charges into the PNA backbone led to excellent solubility characteristics. However, the stability of pPNA complexes with complementary nucleic acid fragments is lower than the stability of the equivalent PNA-DNA(RNA) complexes (25,26). It was also found that chimeric oligomers composed of PNA and pPNA residues form more stable complexes than do the equivalent pPNAs, and PNA-pPNA chimeras containing a few pPNA residues embedded into a PNA chain form complexes with complementary single-stranded nucleic acids with melting temperatures comparable to those of complexes formed by pure PNAs (25,26). More recently, we reported PNA-like monomers on the base of trans-4-hydroxy-L-proline (HypNA) representing conformationally constrained chiral PNA analogues (27,28). HypNA-pPNA hetero-oligomers constructed of alternating pPNA and HypNA residues at a 1:1 ratio demonstrated very strong binding to complementary DNA (RNA) strands, and these complexes exhibited stabilities very close to that of PNA complexes (28). In general, PNA-pPNA and HypNA-pPNA chimeras combine the high hybridisation characteristics of PNAs with the good water solubility of pPNAs, which makes them very promising for further evaluation as potential antisense and antigene therapeutic agents as well as for diagnostic purposes.
Based on these data, we extended our investigations to developing procedures for the construction of arrays for nucleic acid hybridisation analysis using these PNA-like mimics as sensor molecules. Here, we report the results of some preliminary studies on the evaluation of the properties of PNA-pPNA and HypNA-pPNA chimeras as capture and detection probes in comparison with natural ODNs and PNAs. As a test system, we chose assay formats utilising PAA-based technology in view of its easy accessibility and efficiency. We have developed a convenient approach to the synthesis of 5[prime]- and 3[prime]-acrylamide conjugates of ODNs and mimic oligomers via a universal intermediate. Also, we describe here soluble comb-type co-polymers of ODNs and mimics with PAA, their immobilisation on a solid surface and evaluation of these branched co-polymers in DNA array hybridisation as capture and detection/amplification probes.
MATERIALS AND METHODS
Solvents were purchased from commercial suppliers and were used without further purification. Acrylamide, bis-acrylamide and sodium cyanoborohydride were purchased from Sigma. The introduction of radioactive 5[prime]-labels to ODNs was performed with the use of [[gamma]-32P]ATP (or [[gamma]-33P]ATP) and polynucleotide kinase following the standard protocols. Fluorescein-labelled non-modified ODNs were synthesised using fluorescein phosphoramidite (Glen Research). Radioactivity was measured using a Beckman LS6000 counter, and the relative fluorescence was determined in a Perkin Elmer 50B spectrofluorometer. The preparation of 10% polyacrylamide gel (PAAG) containing covalently attached oligomers on glass slides and in the wells of polystyrene microtiter plates as well as the estimation of oligomer immobilisation efficiency by electrophoresis were performed as described (12). The direct immobilisation of 3[prime]-amino-modified ODNs to a glass surface derivatised with phenylisothiocyanate groups was performed as described (29). Chemical reactions and hybridisation procedures on the surface of a glass microscope slide were carried out under a coverslip in chambers of a 0.5 mm thick frame mounted onto a surface and containing evenly spaced holes ([empty] 5 mm, or 10 × 10 mm).
Synthesis of ODNs, mimic oligomers and their conjugates with acrylamide
The solid phase synthesis of ODNs and mimic oligomers was accomplished on the 10 µmol scale using an Applied Bio-systems DNA Synthesiser. ODNs were synthesised using the standard protocol for the phosphoramidite method. The introduction of a 5[prime]-amino group to ODNs was performed with the use of N-MMT-aminolinker phosphoramidite (Cruachem) as a 5[prime]-terminal unit. Chain elongation in the solid phase synthesis of oligomer mimics was carried out using the properly protected monomer and dimer units according to previously developed protocols (25,26). A support (1) derivatised with an abasic ribo unit was prepared by acylation of LCAA CPG amino groups with 3[prime]-succinate of 2[prime]-O-benzoyl-5[prime]-O-dimethoxytrityl-1-deoxy-D-ribofuranose (Scheme 1) (30). The introduction of an acrylic residue to the 3[prime]-termini of oligomers was performed after the completion of chain elongation, removal from the support and deprotection. The oligomer of type 3 containing a 3[prime]-ribo unit (0.1 µmol) was dissolved in 0.5 ml of water, and its 3[prime]-terminal cis-glycol group was oxidised with 0.1 M NaIO4 (100 µl) for 15 min at room temperature to generate dialdehyde unit 4 (31). The excess NaIO4 was reduced with 100 µl of 0.2 M sodium hypophosphite for 20 min. Sodium acetate (0.5 M, pH 4.0) was added to 50 mM concentration, followed by 300 µl of 20 mM N-(2-aminoethyl)acrylamide hydrochloride. The mixture was incubated with 150 µl of 0.2 M NaCNBH3 in acetonitrile for 30 min at 20°C. Then, the reaction mixture was diluted with water to a volume of 1.5 ml and acrylamide-modified oligomer 5 was isolated by gel filtration on a Pharmacia NAP-25 column. Similarly, the synthesis of 3[prime]-amino-modified ODNs and mimics was carried out using hexamethylenediamine (31). The attachment of an acrylic acid residue to support-bound 5[prime]-amino-modified oligomer 6 was performed by the action of a 0.3 M solution of acrylic acid anhydride in the presence of triethylamine (0.3 M) in pyridine/acetonitrile (1:3 v/v) for 15 min at room temperature, followed by an acetonitrile wash and drying of the column material.
Scheme 1.
Deblocking and isolation procedures for ODNs and mimics were essentially as described (25). The deprotection of oligomers containing pPNA units was started by the removal of catalytic P-protective groups by the action of piperidine as described previously (23). The cleavage of oligomers from the support and parallel removal of N-protecting groups from heterocycles were achieved by treatment with concentrated aqueous NH3 for 24 h at room temperature. Oligomers were purified by anion exchange chromatography using a Pharmacia Mono-Q column/FPLC system and a linear gradient of NaCl (0-1.2 M) in 0.02 M NaOH (pH 12) or reversed phase FPLC using a Pharmacia ProRPC column and a linear gradient of acetonitrile (0-30%) in 0.1 M triethylammonium acetate (pH 7), and desalted. The identity of oligomers was confirmed by mass spectrometry (MALDI-TOF)
Preparation of PAA-oligomer co-polymers and their attachment to a solid surface
The co-polymerisation of acrylamide (100 mM) and 5[prime]- or 3[prime]-acrylamide-oligomer conjugates (0.5-1 mM) was performed in the presence of ammonium persulfate (0.1% w/v) and TEMED (0.1% v/v) for 16 h at room temperature in a nitrogen atmosphere with stirring. The polymer was precipitated with ethanol (5 vol), dissolved in water and applied to a Bio-Gel A 0.5 m (BioRad) column. Fractions containing a soluble PAA-oligomer conjugateof type 9 (Scheme 2A) were collected (UV monitoring) and lyophilised.
Scheme 2.
Scheme 3.
Soluble PAA-oligomer co-polymers containing reactive aminohexyl 10a or bromoacetamide 10b groups (Scheme 2B) were obtained by the co-polymerisation of acrylamide (95 mM), N-(6-aminohexyl)acrylamide hydrochloride obtained as described (8) or its N-bromoacetylated derivative (5 mM) and 5[prime](or 3[prime])-acrylamide-oligomer conjugate (0.5-1 mM) in 50% aqueous DMF in the presence of ammonium persulfate (0.1% w/v) and TEMED (0.05% v/v) for 16 h at 25°C in a nitrogen atmosphere with stirring. After precipitation with ethanol, the polymer was dissolved in 10% aqueous acetonitrile and purified by gel filtration as described above. Molecular weights of acrylamide co-polymers were estimated by FPLC using a Superose-6 column (Pharmacia) and molecular weight markers (32). The content of aminohexyl groups in the co-polymers was determined by treatment with an excess of fluorescein isothiocyanate (Sigma), as was described for fluorescent labelling of 3[prime]-alkylamine-modified RNA (31), followed by fluorescence quantification ([lambda]ex = 498 nm, [lambda]em = 513 nm) against a standard curve of dye in water (32). To estimate the content of bromoacetamide groups, the corresponding polymer was treated with an excess of 2-mercaptoethylamine hydrochloride (Sigma) at pH 7.5 for 5 h under conditions (33) to create amino functions, which were fluorescently labelled and quantified as described above. The oligomer content in a co-polymer was quantified by absorption at 260 nm. When a PAA co-polymer contained oligomers with two or three different sequences, their relative content was estimated by hybridisation of the co-polymer (10 µg/ml) with a set of the corresponding complementary 5[prime]-labelled ODNs (5-fold excess over the polymer) under the conditions described below. Each of these ODNs contained a different type of label (32P, 33P or fluorescein). After hybridisation, the excess of labelled ODNs was removed by gel filtration on a Bio-Gel A 0.5 m column. The amount of specific radioactive label in the sample was determined by differential counting, whereas the relative fluorescence was measured using a standard curve prepared with a dilution series of fluorescently labelled ODN.
For the attachment of soluble PAA co-polymers, the surface of non-derivatised microscope slides was treated with either monoethoxydimethylsilylbutanal, 3-mercaptopropyltrimethoxysilane or 3-aminopropyltrimethoxysilane (Aldrich) according to published protocols (29,34,35). To attach co-polymer 10a containing HCl·H2N(CH2)6 functions to aldehyde-functionalised glass (Scheme 3), a mixture of the co-polymer solution (1 µl, 2 mg/ml) in 50 mM sodium acetate (pH 4) and NaCNBH3 (1 µl, 0.1 M) was applied to the surface. After immobilisation for 2 h, 0.1 M ethanolamine (1 µl) was added, and the reaction mixture was allowed to stand for an additional 1 h. Then, the matrix 11 was washed with water. To attach PAA co-polymer 10b carrying bromoacetamide groups to the sulfhydryl-modified glass surface, 2 µl of the polymer solution (1 mg/ml) in 0.1 M triethylammonium phosphate (pH 9) was arrayed onto mercaptosilane-coated glass. After reaction for 5-6 h under nitrogen at room temperature in a humid chamber, an excess of [beta]-mercaptoethanol was added to cap unreacted bromacetamide groups, and the matrix 13 was washed with 0.1 M sodium phosphate (pH 7) and water. Attachment of the amino-modified co-polymer 10a (1 mg/ml) to a glass surface activated with phenylisothiocyanate groups, which was obtained by the treatment of aminated glass with 1,4-phenylene diisocyanate (29), was performed in 2 µl of 100 mM sodium carbonate/bicarbonate buffer (pH 9) (36) for 12 h at 37°C in a humid chamber, followed by washing with water and methanol. The unreacted surface groups were deactivated by a 2 h treatment with ethanolamine (100 mM) in DMF, and the matrix 12 was washed with DMF, methanol and water.
Hybridisation procedures
The melting temperatures (Tm) of polymer solutions were measured at 260 nm using a heating/cooling rate of 0.5°C/min in 0.1 M NaCl, 10 mM sodium phosphate (pH 7), 5 mM EDTA, 10 mM MgCl2. The dissociation temperatures (Td) of duplexes formed by solid phase attached oligomers with the complementary ODNs were determined by subsequent 5 min washes in the same hybridisation buffer at gradually (5°C steps) increased temperatures. Direct capture of labelled ODNs on oligomer-derivatised solid supports was performed for 1-2 h using 200 nM solutions (20-50 µl) of labelled ODNs in 0.1 M NaCl, 10 mM sodium phosphate (pH 7), 5 mM EDTA, 0.1% w/v SDS, followed by washing with the same buffer. The hybridisation temperature was determined by the type of molecule attached to the surface (usually at 10-20°C below Td). The hybridisation-based sandwich assays were performed in 150 mM NaCl, 60 mM Na citrate (pH 7), 1 mg/ml sonicated salmon sperm DNA, 5 mM EDTA, 0.1% SDS as hybridisation buffer for 2 h. Prior to hybridisation, the double-stranded target DNA was denatured in 0.05 M NaOH for 5 min at room temperature, chilled on ice and neutralised with 0.1 M acetic acid. After washing with 150 mM NaCl, 60 mM Na citrate (pH 7), 0.1% SDS for 30 min, the surface was treated in the same buffer with the detection probe, a comb-type PAA-oligomer conjugate containing a 15mer detection probe complementary to the target and multiple 15mer amplifier probes using a 10:1 molar ratio of detection probe to target DNA. After washing, incubation with a 32P-labelled ODN probe complementary to the amplifier sequence (~5000 µCi/nmol) followed by washing and detection of signal were performed. The following oligomer sequences were used in the sandwich hybridisation system: for the single-stranded 40mer target the capture probe was CTGCAAAGGACACCATGA and the detection probe was TCACTCAACACTCAC; for double-stranded long target the capture probes were CTCGAGGAAGATCTG (-21 to -7), ATGGAACCGAAATCT (1-15) and AAAACTCACACCT-GC (22-36) and the detection probes were TCCGTTATGC-ACGAA (636-650), AACCACTACACCCAG (660-674) and GGGAAATAAGGATCC (696-710). The sequence of the amplifier probe was ACTACTACTACTACT and the labelled ODN probe was d([32P]AGTAGTAGTAGTAGT).
RESULTS AND DISCUSSION
The attractive features of PAA supports are their hydrophilicity, relatively high chemical and thermal stability, low non-specific adsorption of biological macromolecules as well as ease of derivatisation with reactive functional groups and manipulation of probe density (8). For immobilisation of biomolecules to the preliminary prepared PAA supports, several methods, including activation of gels, probes or both with reactive groups, have been proposed (7-9). An alternative approach is the co-polymerisation of acrylamide-modified ODN probes in acrylamide-based co-polymers using standard polymerisation techniques (10-12). In this study, we used a co-polymerisation approach to obtain PAA polymers containing ODN or mimic ligands in view of its feasibility and specificity.
Synthesis of acrylamide-modified oligomers
The synthesis of oligomer chains was performed by the solid phase technique using a CPG support derivatised with properly protected 1-deoxy-D-ribofuranose 1 (30) (Scheme 1). Mimic oligomers were constructed from the appropriate monomers as well as PNA-pPNA and HypNA-pPNA dimers. Chain elongation was carried out automatically under the conditions developed by us previously for the formation of phosphonester and amide bonds (24,25), whereas elongation of ODN chains was performed by the standard phosphoramidite method. The same procedures were used for the introduction of acrylamide residues into ODNs or mimic oligomers. A common type of oligomer intermediate 2 was used to introduce acrylamide residues to the 5[prime]- or 3[prime]-terminus of oligomers (pseudo 5[prime]- and 3[prime]-ends for mimics). To attach the acrylamide residue to the 3[prime]-end of an oligomer, intermediate 2 was removed from the support and deprotected to obtain 3 containing a 3[prime]-terminal cis-glycol group. The latter was oxidised with sodium periodate to obtain oligomer 4 with a 3[prime]-dialdehyde function, which was coupled to aminoalkylacrylamide, followed by stabilisation of the aldimine bond by reduction with sodium cyanoborohydride to obtain acrylamide oligomer 5 at a yield of 80-90%. Attachment of an acrylamide residue to the 5[prime]-terminus of an oligomer was carried out on a CPG support similarly to a procedure described earlier (11). In the first step, intermediate 2 was elongated by a spacer containing a terminal amino group, followed by 5[prime]-acylation of support-bound oligomer 6 with acrylic acid anhydride to obtain derivative 7 at almost quantitative yield. Acrylamide-containing ODNs, PNAs, PNA-pPNAs and HypNA-pPNAs of types 5 and 8, whose general structures are depicted in Figure 1, were isolated by gel filtration and purified by anion exchange or reversed phase chromatography, and their purity and identity were confirmed by mass spectrometry.
Figure 1. General chemical structures of acrylamide conjugates with ODNs, PNAs and PNA-related mimic oligomers.
Co-polymerisation of acrylamide-mimic conjugates with acrylamide and matrix preparation
The basic procedures for the introduction of acrylamide-oligomer conjugates into PAA chains and their attachment to a solid surface are depicted in Schemes 2 and 3. To obtain comb-type soluble polymers of type 9, acrylamide-modified oligomers were co-polymerised with acrylamide at a 1:100 molar ratio (Scheme 2A). Addition to the polymerisation mixture of bis-acrylamide along with other components enabled us to obtain gels containing immobilised mimic probes. Characterisation of gel-type co-polymers for oligomer content using a gel electrophoresis method (12) revealed that the average efficiency of immobilisation was ~90% for both the 5[prime]- and 3[prime]-acrylamide-modified ODNs and mimic oligomers. Control of immobilisation specificity with the addition of unmodified oligomers into the co-polymerisation reaction revealed that non-specific binding was <2% (data not shown). The co-polymers functionalised with reactive groups were obtained by co-polymerisation of the acrylamide derivatives bearing corresponding functional groups (alkylamino, bromoacetamide, etc.) with acrylamide at a 1:20 molar ratio (Scheme 2B). Such groups can be used for the introduction of fluorescent labels or side chains of other types to obtain multiple functionalised polymers, for the attachment of PAA co-polymers to solid surfaces as well as for modulation of the polymer properties with respect to hydrophobicity and charge. Apparent molecular weights for water soluble co-polymers obtained by this approach were estimated to be 250-350 kDa, and by gel filtration with UV monitoring it was determined that up to 85% of acrylamide oligomer was incorporated into the polymer chains (~15-20 oligomer molecules per chain). Estimation of the co-polymerisation efficiency, using fluorescent labelling of alkylamino and bromoacetyl groups on the polymer, demonstrated that ~65% of alkylamino- and ~90% of bromoacetyl-modified acrylamide was incorporated into the co-polymer.
Derivatisation of solid supports was performed using several approaches. The first approach, schematically shown in Figure 2A, was based on a previously described technique for the preparation of gel-bound ODN arrays (12). Acrylamide/bis-acrylamide mixtures containing acrylamide-oligomer probes were co-polymerised on the surface of a glass microscope slide coated with covalently attached acrylamide residues to introduce surface co-polymerisation groups. The amount of probe attached to PAAG was linearly related to the amount of oligomer introduced into the co-polymerisation reaction (10-50 µM). The second approach was similar to common methods for the attachment of ODN probes to a support by the direct chemical substitution of reactive groups on the surface (7,29,36). In contrast to these methods, we attached, instead of the oligomer, a previously prepared comb-type PAA precursor containing the tethered oligomer in multiple copies (Fig. 2B). In this study, we tested three well-known attachment chemistries depicted in Scheme 3, which meet several criteria including simplicity (one or two steps) and efficiency. Moreover, they provide connection with the surface through stable C-N and C-S bonds, do not produce non-specific bonding to the support and prevent any cross-reaction between PAA chains prior to immobilisation. In one of the attachment variants, a solution of co-polymer 10a containing reactive aminohexyl groups was reacted with aldehyde groups on the glass surface at pH 4.5 in the presence of sodium cyanoborohydride to obtain the matrix of type 11, followed by treatment of the surface with ethanolamine to block all remaining reactive aldehyde functions. The second chemistry included the interaction of amino groups of co-polymer 10a with phenylisothiocyanate groups on the glass surface at pH 9 to obtain the matrix 12. Unreacted phenylisothiocyanate groups were deactivated with ethanolamine to prevent non-specific DNA binding at later stages, causing a high background (29). We also tested the interaction of SH groups on the glass surface with PAA-oligomer conjugates modified with bromoacetyl groups on side chains to obtain the matrix of type 13. Capping of unreacted bromoacetyl groups was by the action of [beta]-mercaptoethanol. For comparison, we obtained matrix 14 by direct attachment of 3[prime]-amino-modified ODNs and mimics onto a phenylisothiocyanate-activated glass surface by a flexible spacer similar to a technique described for ODN immobilisation previously (29).
Figure 2. Concepts for the attachment of oligomer capture probes to a solid support via conjugation with PAAG (A) or PAA chains (B).
Common methods for evaluation of probe surface density use direct attachment of labelled oligomers or hybridisation assays with complementary labelled ODNs (8,12,29). We found that a convenient method to estimate probe capacity on a surface is spectroscopic analysis of the amount of dimethoxytrityl function liberated under the action of 60% HClO4 from the 5[prime]-end of an oligomer tethered to PAA via its 3[prime]-terminus. For this, 3[prime]-acrylamide oligomers containing 5[prime]-terminal dimethoxytrityl groups were introduced into the co-polymerisation reaction. Thus, the average fixation of oligomer probes using comb-type PAA co-polymers was estimated to be 350-500 fmol/mm2, and optimum attachment efficiency of PAA chains was observed in the case of thioether PAA-surface linkage (Table 1). The matrices of type 14 gave up to 250 fmol/mm2. At the same time, the PAAG loading was essentially higher. The test on the thermal stability of the attachment revealed that no systematic probe loss can be detected after 1 h heating in hybridisation buffer at 95°C (data not shown).
Table 1. Probe attachment density and hybridisation efficiencies on derivatised glass supportsa
| Probe attachment mode | Probe attachment densityb (fmol/mm2) | Hybridisation capacityc (fmol/mm2) | ||||
| ODN | PNA-pPNA | HypNA-pPNA | ODN | PNA-pPNA | HypNA-pPNA | |
| PAAG | 12 500 | 11 050 | 12 200 | 180 | 190 | 220 |
| -SH/BrCH2-PAA | 500 | 450 | 490 | 395 | 380 | 410 |
| -NCS/NH2-CH2-PAA | 420 | 440 | 450 | 340 | 360 | 385 |
| -CHO/NH2-CH2-PAA | 350 | 390 | 380 | 275 | 320 | 330 |
| -NCS/NH2-CH2-oligo | 210 | 250 | 230 | 45 | 60 | 50 |
bThe attachment density was estimated by the treatment of a dimethoxytrityl-containing probe on a surface (1 cm2) with 60% aqueous perchloric acid (50 µl) followed by measurment of the solution optical density at 495 nm ([epsiv] = 71.7/µmol) using a microvolume spectrophotometer cell.
cThe capacity of the probe-modified surface was determined by hybridisation with a 32P-labelled complementary ODN as described in Materials and Methods. After washing the radioactivity of each sample was counted.
Hybridisation properties of PAA-oligomer conjugates
The hybridisation properties of polymeric molecules and the control with respect to binding selectivity to DNA/RNA fragments were examined for both 3[prime]- and 5[prime]-conjugated oligomers starting from UV melting experiments in solution. As model sequences, we used homo-Thy oligomers and an 18mer sequence (CTGCAAAGGACACCATGA) corresponding to a region of the TNF-[alpha] gene (positions 138-156 with respect to the transcription start site). In accord with the data published for PAA-bound ODNs (11), we found that the viscosity of diluted polymer solutions had almost no influence on the hybridisation characteristics of polymeric probes. The mode of oligomer immobilisation (5[prime] or 3[prime]) also had no influence on the Tm of its complex with complementary DNA/RNA single-stranded targets (data not shown). Control experiments with oligomers having non-complementary sequences, or mismatches, confirmed that hybridisation of all PAA-mimic conjugates occurs in a sequence-specific manner. Thus, we did not detect formation of a complex between PAA-mimic probes and non-complementary DNA/RNA targets. The introduction of a mismatch in the centre of the mimic sequence gave a significant drop in the Tm value of its complex with the target ([Delta]Tm from -17 to -23°C) (Table 2). Under the same conditions, mismatched ODN duplexes gave a drop of only -12 to -14°C. Oligomers with a mixed purine·pyrimidine sequence containing two mismatches situated at positions 6 and 10 were unable to form stable complexes with the targets. In general, PNA-pPNA and HypNA-pPNA probes showed better mismatch discrimination properties than ODNs, and their characteristics were similar to those for PNAs. Comparison of the association rates for four types of fully complementary 18mer duplexes formed by soluble PAA-oligomer co-polymers with complementary ODN targets revealed that they were close for all types of mimics (Fig. 3A). It required ~10 min for the PNA, PNA-pPNA and HypNA-pPNA probes and <3 min for ODN probes to reach equilibrium at 20°C. On a solid phase, the dissociation temperatures of the fully matched duplexes were determined to range between 68 and 74°C for mimic/DNA duplexes and the Td of the DNA/DNA duplex was 55°C (Table 2). The stability of duplexes formed by a DNA target and a PNA-pPNA or HypNA-pPNA mimic oligomer attached to a solid surface can be described as being in general 0.72-0.89°C higher per base pair than that of the corresponding DNA/DNA duplex, whereas for PNAs this value was ~1.05°C.
Figure 3. Association kinetics of the 18mer (CTGCAAAGGACACCATGA) conjugated with PAA and the complementary ODN target d(TCATGGTGTCCTTTGCAG). (A) Rate of duplex formation between soluble PAA-oligomer conjugates and the target in 5 µM solutions (UV absorbance monitoring). (B). Kinetic plot of the hybridisation of a 32P-labelled ODN target (0.2 µM) with the complementary probe attached to a glass surface via PAAG (1-3) or PAA chains (4-6) (Fig. 2) using thiol/bromacetamide chemistry. The hybridisation was performed at 40°C for the ODN probe (1 and 4) or at 50°C for the PNA-pPNA (2 and 5) and HypNA-pPNA (3 and 6) mimic probes. The average result from the three experiments is shown.
Table 2. The effect of a mismatch on stability of complexes formed by d(A15) and d(TCATGGTGTCCTTTGCAG) targets with complementary and mismatched PAA-conjugated oligomer probes conjugated to soluble PAA polymer or PAAG Acrylamide-containing mimics were also tested on solid-phase hybridisation in a gel situated in wells of a standard polystyrene microtitre plate. After co-polymerisation, non-immobilised capture probes were removed by washing, and the gel was hybridised with the complementary fluorescein-labelled ODN at temperatures near the optimum for hybridisation kinetics (usually 10-20°C below the Tm of the corresponding duplex). After washing, the scanning wells show that the tethered mimic probes work well in the same way but give increased signal intensities compared to ODNs, probably due to their higher binding affinity (Fig. 4). Moreover, mimic probes showed better mismatch discrimination properties (discrimination factor 7.3:1 for PNA-pPNA probes, 7.2:1 for HypNA-pPNA probes and 7.4:1 for PNA probes) than ODN probes (discrimination factor 4.9:1).
Figure 4. Analysis of a mismatch effect in solid phase hybridisation of the fluorescently labelled ODN d(TCATGGTGTCCTTTGCAG) to immobilised oligomer probes with complementary (CTGCAAAGGACACCATGA), mismatched at position 9 (CTGCAAAGCACACCATGA) and non-complementary (CTTTCTTTTCTCTTCTCC) sequences. The immobilisation was performed by co-polymerisation of acrylamide-oligomer probes with PAAG in wells of a microtitre plate. An average integrated fluorescence intensity from each set of three wells is shown. Comparison of the hybridisation process on glass surfaces coated with PAAG or PAA chains conjugated with oligomers revealed that the mode of derivatisation had some influence on the hybridisation rate and surface capacity. Hybridisation with complementary 32P-labelled ODNs was at equilibrium within ~30 min in the case of glass slides modified with comb-type PAA co-polymers (matrices 11-13). At the same target concentration in solution, for a surface covered with PAAG the annealing period should be extended to ~1 h to reach equilibrium (Fig. 3B). On a solid support, hybridisation was complete in almost the same period of time for the mimic probes as for the native ODN probe. The apparent capture probe densities for gel-type matrices was estimated as ~200 fmol/mm2, which represents only a small part of the capture probe amount attached to the gel (Table 1). The attachment of comb-type co-polymers increased the effective surface capacity to >400 fmol/mm2 of hybridisable probe (~80% of the surface-bound oligomer). For comparison, the capacity of matrices 14 obtained by direct immobilisation of oligomers to an activated glass surface was estimated to be 45-60 fmol/mm2, which corresponds to only ~20% of the surface-bound capture probe, probably the result of steric constraints (for comparison see also 29,35,37). So, it can be concluded that an approach to the construction of PAA-type arrays based on surface immobilisation of presynthesised PAA chains containing multiple tethered probes has advantages over traditional PAAG-coated glass plates. The gel permits high probe loading, but only short targets can diffuse into it due to steric hindrances, and, apparently, the target-probe interaction occurs mainly on the gel surface. The gel-bound ligand is not so free to move as it would be in the more flexible chain-bound variant, which must reduce the rate of duplex formation. A surface modified with PAA chains, which play the role of a flexible spacer for oligomer probes, is closer to the solution state and more likely allows probes to react freely with dissolved target molecules and, probably, decreases steric hindrance. In addition, we tested soluble PAA-oligomer co-polymers as branched signal amplification probes for nucleic acid detection. The traditional strategy of signal amplification employs branched multimers containing a primary sequence complementary to the target sequence and a set of secondary sequences for signal amplification. These dendrimeric ODNs are usually synthesised by solid phase chemical methods, which is a difficult and tedious procedure (38,39). An alternative strategy is to prepare a polymer of non-nucleotide nature containing ODNs covalently coupled to this polymer at multiple sites. A successful application of conjugates between ODNs and linear co-polymers of N-vinylpyrrolidone with N-acryloxysuccinimide for signal amplification was published by Erout et al. (32). Similar signal amplification comb-type branched probes were constructed by us on the base of PAA with tethered ODN, PNA, PNA-pPNA or HypNA-pPNA ligands. In this case, a co-polymerisation procedure was used to incorporate two different oligomers into the same PAA molecule in the desired ratio. Each of them has its own function: detection of a target or signal amplification. Since the oligomers are incorporated in high yields, their concentration and composition in a co-polymer is easily adjustable. Moreover, the amount of each oligomer can be estimated by hybridisation with a specifically labelled complementary probe. The functional properties of the PAA-based detection probes were examined in a sandwich hybridisation assay format schematically shown in Figure 5. As targets, we chose a model 40mer synthetic ODN d(TCATGGTGTCCTTTGCAGTTTT-TTTGTGAGTGTTGAGTGA) and a double-stranded ~720 bp XhoI-BamHI DNA fragment representing a cloned artificial gene for the Fc domain of human IgG1 (40). The system consisted of a comb-type capture probe immobilised on a glass surface and a comb-type PAA-based detection/amplification probe. The latter was designed to bind the target at a site separate from that recognised by the capture probe and contained multiple amplifier sequences, which were capable of binding to the complementary labelled ODN probes. We used a 1:10 molar ratio of detection/amplifier probes in a comb-type co-polymer. For a short target, this format made use of one sequence in the capture and one sequence in the detection reaction. The format for a long DNA target made use of PAA-based capture and detection/amplifier probes with three sequences for capture and three sequences for detection of multiple sites on the target DNA. For signal detection, we used 32P-labelled ODNs complementary to the secondary amplifier sequences. The direct comparison of ODNs and mimics in the same formats showed that the use of mimic oligomers in target capture and detection probes as well as in signal amplification significantly increased the sensitivity of detection without increasing the background (Table 3). We found that the assay system with PNA, PNA-pPNA and HypNA-pPNA oligomers was capable of detecting 5 amol of a target with the use of radioactively labelled probes complementary to the amplifier sequences. This level of sensitivity is comparable to published methods (7,32,41) and is suitable for a variety of biological testing applications. It is obvious that in practice this assay system can be made considerably more sensitive by the introduction of preamplifier and capture extender probes, as was proposed by Collins et al. (41), or/and by the application of more sensitive chemiluminiscent labels for target detection.
Figure 5. The format for target detection by sandwich hybridisation using the PAA-based capture and branched amplifier detection probes. Table 3. Comparison of ODN and mimic probes in the sandwich hybridisation DNA assay using a glass surface coated with PAA-conjugated capture probes by the thiol/bromoacetamide method and PAA-based signal amplification probes as shown in Figure 5
The data presented here demonstrate the rather high potential of PNA-like DNA mimics bearing negative charges, particularly PNA-pPNA and HypNA-pPNA chimeras, as probes in PAA-based diagnostics. The evaluation of hybridisation characteristics of these mimics in direct capture and detection hybridisation reactions has shown that the interaction between the immobilised recognition mimic oligomer and the complementary target is based on a sequence-specific hybridisation, and mimic probes can effectively discriminate between single base mismatches in the target sequence. The use of PNA-related probes increases the detection sensitivity as compared to that of natural ODNs. Because the stability of mimic/DNA duplexes is higher than that of DNA/DNA, binding is more specific, and single-base mismatches are more detectable. It should be noted that soluble comb-type PAA-oligomer conjugates might be useful as capture and detection probes for signal amplification in nucleic acid quantification assays. The main advantages of PAA-based branched probes are ease of preparation and possibility of alteration of their chemical properties as well as oligomer content and type of label. Further evaluation of PNA-related mimics to extend the range of their applications in biological assays is now in progress. Nevertheless, it can now be concluded that in view of their hybridisation properties, stability under conditions, which natural ODNs cannot withstand, and good water solubility, PNA-pPNA and HypNA-pPNA mimics can provide the necessary robustness and automatability for routine clinical analysis. The authors gratefully acknowledge Prof. A. D. Mirzabekov, Dr N.V. Bovin and Dr K. Jayaraman for helpful discussions on this work.
*To whom correspondence should be addressed. Tel: +7 095 336 5911; Fax: +7 095 336 5911; Email: eva{at}ibch.siobc.ras.ru
This article has been cited by other articles:
bThe duplexes are formed by solid phase attached oligomers (10 × 10 mm matrix of type 13) with complementary 32P-labelled ODNs in 0.5 ml of hybridisation buffer. Aliquots were withdrawn and the amount of labelled ODN in solution was measured.
Sandwich hybridisation assays
aSS, single-stranded short 40mer target; DS, denatured 720 bp double-stranded target.
Target DNA
Probe signal (c.p.m.) (signal/noiseb)
Conc. (amol)
Typea
ODN
PNA-pPNA
HypNA-pPNA
PNA
0
SS
10
12
10
10
DS
13
15
11
12
5
SS
31 (1.4)
59 (3.2)
62 (3.4)
60 (3.2)
DS
18 (1.0)
44 (2.9)
50 (3.3)
49 (3.3)
10
SS
84 (3.4)
165 (7.5)
182 (8.2)
180 (5.0)
DS
65 (2.3)
124 (6.3)
131 (5.1)
134 (5.8)
20
SS
205 (9.5)
312 (18.1)
326 (19.8)
320 (18.6)
DS
232 (9.2)
340 (19.6)
354 (17.6)
315 (18.4)
100
SS
1140 (22.7)
1802 (45.0)
1923 (47.2)
1960 (44.5)
DS
961 (19.4)
1380 (37.4)
1405 (40.5)
1410 (40.3)
500
SS
8443 (49.4)
10 522 (83.2)
10 850 (95.6)
DS
6180 (40.5)
8395 (94.5)
9421 (92.5)
1000
SS
25 230 (116.3)
27 900 (128.5)
28 120 (177.4)
DS
17 847 (89.8)
19 324 (92.8)
19 537 (100.9)
bNon-complementary DNA target was used as a control to determine the level of non-specific binding in the assay. Signal/noise was expressed as the ratio of 32P counts obtained with complementary and non-complementary targets. The data presented here are the average of four determinations.CONCLUSION
ACKNOWLEDGEMENTS
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: jnl.info{at}oup.co.uk
Last modification:
Copyright© Oxford University Press, 1999.
CiteULike 
Connotea 
Del.icio.us What's this?
D. Yang, Q. Chen, S. B. Su, P. Zhang, K. Kurosaka, R. R. Caspi, S. M. Michalek, H. F. Rosenberg, N. Zhang, and J. J. Oppenheim
Eosinophil-derived neurotoxin acts as an alarmin to activate the TLR2-MyD88 signal pathway in dendritic cells and enhances Th2 immune responses
J. Exp. Med.,
January 21, 2008;
205(1):
79 - 90.
[Abstract]
[Full Text]
[PDF]
K. Okubo, F. Sakai, E. L. Lau, G. Yoshizaki, Y. Takeuchi, K. Naruse, K. Aida, and Y. Nagahama
Forebrain Gonadotropin-Releasing Hormone Neuronal Development: Insights from Transgenic Medaka and the Relevance to X-Linked Kallmann Syndrome
Endocrinology,
March 1, 2006;
147(3):
1076 - 1084.
[Abstract]
[Full Text]
[PDF]
V. A. Efimov, K. R. Birikh, D. B. Staroverov, S. A. Lukyanov, M. B. Tereshina, A. G. Zaraisky, and O. G. Chakhmakhcheva
Hydroxyproline-based DNA mimics provide an efficient gene silencing in vitro and in vivo.
Nucleic Acids Res.,
January 1, 2006;
34(8):
2247 - 2257.
[Abstract]
[Full Text]
[PDF]
V. A. Efimov, O. G. Chakhmakhcheva, J. Archdeacon, J. M. Fernandez, O. N. Fedorkin, Yu. L. Dorokhov, and J. G. Atabekov
Detection of the 5'-cap structure of messenger RNAs with the use of the cap-jumping approach
Nucleic Acids Res.,
November 15, 2001;
29(22):
4751 - 4759.
[Abstract]
[Full Text]
[PDF]
This Article 
Abstract
Print PDF (918K)
Alert me when this article is cited
Alert me if a correction is posted
Services
Email this article to a friend
Similar articles in this journal
Similar articles in ISI Web of Science
Similar articles in PubMed
Alert me to new issues of the journal
Add to My Personal Archive
Download to citation manager
Search for citing articles in:
ISI Web of Science (23)
Request Permissions
Commercial Re-use Guidelines
for Open Access NAR Content
Google Scholar
Articles by Efimov, V. A.
Articles by Chakhmakhcheva, O. G.
Search for Related Content
PubMed
PubMed Citation
Articles by Efimov, V. A.
Articles by Chakhmakhcheva, O. G.
Related Collections
DNA characterisation
Genomics
Social Bookmarking
What's this?