ABSTRACT
Peptide nucleic acids (PNA) mimic DNA and RNA by forming complementary duplex structures following Watson-Crick base pairing. A set of reporter compounds that bind to DNA by intercalation are known, but these compounds do not intercalate in PNA/DNA hybrid duplexes. Analysis of the hybrid PNA duplexes requires development of reporter compounds that probe their chemical and physical properties. We prepared a series of anthraquinone (AQ) derivatives that are linked to internal positions of a PNA oligomer. These are the first non-nucleobase functional groups that have been incorporated into a PNA. The resulting PNA(AQ) conjugates form stable hybrids with complementary DNA oligomers. We find that when the AQ groups are covalently bound to PNA that they stabilize the hybrid duplex and are, at least partially, intercalated.
Peptide nucleic acids (PNA) are remarkable DNA/RNA mimics in which the sugar-phosphate spine of the natural nucleic acid is replaced by a synthetic peptide backbone formed from N-(2-aminoethyl)-glycine units (1,2) (Scheme 1). Surprisingly, PNA oligonucleotides containing purine and pyrimidine nucleobases hybridize with complementary DNA and RNA strands to form right-handed, double-helical complexes according to the Watson-Crick rules of hydrogen bond mediated base pair formation (3,4). The hybrid duplexes formed by PNA with DNA generally have higher thermal stabilities than their duplex DNA counterparts and show unique ionic strength effects because the PNA strand does not bear negatively charged phosphate groups (5). Examination of the PNA/DNA hybrid structure by NMR spectroscopy reveals a unique double helical conformation that has features of both A- and B-form DNA (4). These properties have sparked interest in PNA for application in antisense and antigene drugs and as diagnostic reagents (6-8).
Scheme 1. DNA, PNA and modified PNA groups.
DNA oligomers were purchased from Midland Certified Reagent Company, purified by gel filtration and characterized by MALDI-TOF mass spectrometry. The extinction coefficients of DNA oligomers were calculated using the nearest-neighbor values (11): for DNA(1), [epsilon]260 = 238 200 M-1cm-1; DNA(2), [epsilon]260 = 120 200 M-1cm-1. Similarly, the PNA oligomer extinction coefficient, [epsilon]260 = 233 000 M-1cm-1, was determined using DNA values with AQ substituted for adenine. The samples for all experiments were prepared in a 10 mM sodium phosphate buffer at pH = 7.0. Spectra were recorded with the following instruments: Cary 1E (UV-Vis), SPEX 1681 FLUOROLOG (Phosphorescence).
3,6-Diaza(N3-Boc-aminoethyl)-4,7-dioxo-7-(2-anthraquinoyl)-heptanoic acid (AQ2) methyl 4-(2-anthraquinoyl)-4-oxo-3-aza-butanoate. Anthraquinone-2-carboxylic acid (3 g, 11.9 mmol), DIEA (3.1 ml, 24 mmol) and HBTU (4.5 g, 11.9 mmol) in CH2Cl2 (30 ml) was stirred for 10 min at room temperature. Methyl glycinate hydrochloride (1.5 g, 11.9 mmol) was added after which the product started to precipitate. The mixture was stirred for an additional 3 h. The precipitated material was collected by filtration. The volume of the mother liquor was reduced to 1/3 and an additional crop was collected. Yield 3.65 g (93%).
4-(2-Anthraquinoyl)-4-oxo-3-aza-butanoic acid. Methyl (2-anthraquinoyl)-3-aza-butanoate (3.65 g, 11.3 mmol) was stirred for 1 h at room temperature in 1 M LiOH (20 ml) and THF (3 ml) whereby the starting material dissolved. The pH of the solution was adjusted to 2.8 with 2 M NaHSO4 and the target molecule precipitated. The organic solvents were removed from the suspension under reduced pressure and the precipitate was collected by filtration and dried. Yield 3.5 g (100%). 1H-NMR DMSO-d6 d: 9.37 (t, 1 H, NH); 8.67 (d, 1 H, H-1aq); 8.36 (dd, 1 H, H-3aq); 8.30 and 8.28 (1 H, H-4aq); 8.23 (m, 2 H, H-5 and H-8aq); 7.95 (m, 2 H, H-6 and H-7aq); 4.00 (d, 2 H, CH2).
3,6-Diaza(N3-Boc-aminoethyl)-4,7-dioxo-7-(2-anthraquinoyl)-heptanoic acid (AQ2). 4-(2-Anthraquinoyl)-4-oxo-3-aza-butanoic acid (3.5 g 11.5 mmol), DCC (2.6 g, 12.65 mmol), HOBT (1.7 g, 12.65 mmol) and methyl [N-(2-Boc-aminoethyl)] glycinate (3.2 g, 13.8 mmol) were stirred in DMF (50 ml) at room temperature for 48 h. The reaction mixture was filtered and the residue was washed with CH2Cl2 (2 × 30 ml). The organic phase was extracted with dilute NaHCO3 (3 × 30 ml), 2 M NaHSO4 (2 × 30 ml), brine, dried with MgSO4 and evaporated to dryness. Crystallization from ethyl acetate gave 4.0 g (67%) of methyl 3,6-diaza-(N3-Boc-aminoethyl)-4,7-dioxo-7-(2-anthraquinoyl)-heptanoate. This compound was then hydrolyzed at room temperature in 1 M LiOH and 10% THF. The pH of the solution was adjusted to 2.8 with 2 M NaHSO4 and the target molecule separated as an oil. The oil was extracted with ethyl acetate and the organic phase was dried with MgSO4. The volume was reduced to 2 ml and, while stirring, hexane (250 ml) was added, whereby 3,6-diaza(N3-Boc-aminoethyl)-4,7-dioxo-7-(2-anthraquinoyl)-heptanoic acid (AQ2) precipitated. Yield 3.7 g (92%). 1H-NMR DMSO-d6 d: 12.64 (s, 1 H, COOH); 9.14 (m, 1 H, NH); 8.69 (d, 1 H, H-1aq); 8.67 and 8.35 (dd, 1 H, H-3aq); 8.32 and 8.30 (dd, 1 H, H-4aq); 8.25 (m, 2 H, H-5 and H-8aq); 7.96 (m, 2 H, H-6 and H-7aq); 6.90 and 6.75 (t, 1 H, BocNH); 4.24 (d, 2 H, CH2CO); 4.24 and 4.12 (dd, 2 H, CH2CO); 3.90 (s, 2 H, CH2CO); 3.44, 3.17 and 3.06 (m, 2 H, CH2); 1.40 and 1.38 (s, 9 H, Boc).
The monomers described above were incorporated by the solid phase peptide synthesis procedure previously described (12). The oligomers were cleaved from the resin with trifluoromethanesulphonic acid and purified by HPLC. The identity of each oligomer was confirmed by mass spectroscopy. PNA(AQ1): calcd. m/e = 5429; found m/e = 5430. PNA(AQ2): calcd. m/e = 5496; found m/e = 5496. PNA(Ac): calcd. m/e = 5237; found m/e = 5239.
Samples were prepared consisting of equimolar concentrations of PNA and DNA oligomers (1.0 or 2.0 µM each) in 1.0 ml of 10 mM sodium phosphate buffer (pH = 7.0). (The PNA strand will often precipitate upon addition of the phosphate, presumably due to complexation of the lysine units by the phosphate. Mixing and addition of the DNA strand results in dissolution of the PNA.) Samples were placed in cuvettes (1.5 ml capacity, 1.0 cm path length) and sealed with tape to prevent evaporation of water during heating/cooling cycles. The absorbance of the samples at 260 nm was monitored as a function of temperature for three consecutive runs: heating at 1.0°C/min, then cooling followed by reheating at 0.5°C/min. For experiments in which the absorbance was monitored at 330 nm, 20 µM of each oligomer was used due to the low extinction coefficient of the AQ chromophore.
The absorbance was plotted versus temperature for each sample. Melting temperatures (Tm) were determined as the maxima of plots of the first derivative of absorbance with respect to temperature, assuming a first order phase transition. Tm values given in the text and tables have error values of ±0.5°C.
Samples were prepared containing 5.0 µM of AQC(2), 5.0 µM of PNA(AQ) oligomer, or 5.0 µM each of PNA(AQ) and DNA-1X dissolved in 30% ethylene glycol/sodium phosphate (10 mM phosphate, pH = 7.0) buffer. Approximately 500 µl of sample was placed into a standard NMR tube which was then submerged into a liquid nitrogen Dewar flask with quartz windows situated in the sample compartment of a fluorescence spectrometer. Phosphorescence emission spectra from the frozen glass samples were recorded over the range 400-600 nm using the front-face detection mode with excitation at 330 nm. A blank sample containing only the ethylene glycol/phosphate buffer was used to record a baseline; subtraction of this spectrum from the phosphorescent sample spectra eliminates contributions from light scattering to the spectra at short wavelengths.
Computer models of PNA(AQ)/DNA hybrids were generated using SYBYL 6.0 (Tripos Associates, St. Louis) and the coordinates for a PNA/DNA duplex structure determined by Erikkson and Nielsen (4). One of the central base pairs was removed from that structure by excision of the nucleobase from both the PNA strand and the DNA strand, leaving a free secondary amine on the PNA backbone and an abasic deoxyribose on the DNA backbone. AQ units with the appropriate linker were docked to the PNA secondary amine while a hydrogen atom was placed on C-1' of the DNA. The AQ units were manipulated in order to assess their abilities to intercalate without disrupting the duplex structure. The structures were not subjected to energy minimization.
PNA oligomers hybridize with their complementary DNA oligomers in a 1:1 stoichiometry yielding right-handed, double-helical complexes (3). The thermal stability of these hybrid duplexes is strongly dependent on the orientation of the two strands and is sensitive to single-base mismatches. A series of thermal denaturation experiments was undertaken to investigate the sequence specificity and orientation preference of the PNA(AQ) conjugates.
Mixing PNA(AQ1) or PNA(AQ2) with DNA-1X yields stable hybrid complexes which exhibit cooperative, monophasic melting transitions (Fig. 1). The melting temperatures (Tm) for these hybrids, determined from the maxima of first-derivative plots, are 61.4°C and 68.8°C, respectively. No hysteresis is observed for the transitions recorded on sequential heating and cooling experiments, demonstrating rapid hybridization kinetics for these PNA-DNA duplexes. Significantly, the PNA-DNA hybrid formed with PNA(Ac), where an acetyl group replaces the AQ, gives a duplex with a lower Tm (56.6°C) than the AQ-substituted PNA. Clearly, the quinone is contributing to the stability of the hybrid duplex.
The data illustrated in Figure 1A were obtained by monitoring the absorbance at 260 nm as a function of temperature. Thermal denaturation experiments monitored at 330 nm, where only the AQ absorbs light, also show significant hyperchromicity upon melting (Fig. 1B). This is consistent with a structure where the [pi]-electrons of the quinone interact strongly with adjacent base pairs in the duplex.
Stable PNA(AQ)/DNA duplexes also form when a normal base is at the position directly opposite the quinone. DNA-1X contains an abasic residue (actually, the base at C-1 of the deoxyribose unit has been replaced by a hydrogen atom) designed to provide the most space for accommodation of the AQ into the helix. Critically, stable PNA(AQ)/DNA duplexes form with any of the four naturally occurring bases opposite the AQ and have melting temperatures that are only ~1-2°C below that formed with DNA-1X (Table 1). This finding assures that PNA(AQ) oligomers will routinely form duplexes with natural, complementary DNA.
AQ derivatives which have n[pi]* excited state electronic configurations (such as the AQ carboxamides studied here) intersystem cross to their triplet state in less than 10 ps, precluding fluorescence emission (13). Phosphorescence from the quinone triplet states is readily detected at 77 K (14). We have demonstrated that the phosphorescence of these anthraquinones is nearly completely quenched by electron transfer from a base when they are intercalated into DNA duplexes (15). In contrast, when the associated quinone is not intercalated, overlap with the bases is diminished and little or no quenching of the phosphorescence is observed (16).
The phosphorescence spectrum of AQC(2), a cationic, water-soluble anthraquinone derivative, in frozen, glassy 30% ethylene glycol/sodium phosphate buffer is shown in Figure 2, spectrum A (the presence of 30% ethylene glycol has very little effect on the PNA/DNA hybrid stability. For example, a PNA(AQ2)/DNA-1X duplex melts at 66°C under these conditions, a suppression of only 3°C relative to the completely aqueous buffer). It exhibits a strong, vibrationally structured emission typical of anthraquinones having n[pi]* triplet states. Spectrum B is the phosphorescence spectrum of single-stranded PNA(AQ1) recorded under identical conditions. Incorporation of the AQ into single-stranded PNA leads to 70% quenching of the emission, indicating that the PNA bases can readily donate an electron to the excited AQ. Hybridization of the PNA with DNA-1X results in nearly complete quenching of the residual AQ phosphorescence (spectrum C). These findings indicate that the AQ is at least partially intercalated within the PNA/DNA helix since, upon excitation, the quinone excited state is readily quenched by an adjacent nucleobase.
Qualitatively similar results are obtained for PNA(AQ2): the quinone phosphorescence is quenched by >95% in the duplex relative to AQC(2). An interesting difference from PNA(AQ1) is observed in the absence of DNA: the phosphorescence is still quenched by 90% for PNA(AQ2), which is significantly greater than for single-stranded PNA(AQ1) (70% quenching).
A computer model for the PNA(AQ1)/DNA-1X duplex is shown in Figure 3. The model shows that there is ample space in which to accommodate the AQ moiety within the helix, i.e., the AQ can intercalate. A similar structure results from PNA(AQ2)/DNA-1X, although the remaining space visible on the DNA side of the intercalation site in Figure 3 is completely filled.
The data described in the previous sections demonstrate that it is possible to modify PNA at internal positions with reporter groups which, upon hybridization with complementary oligodeoxynucleotides, intercalate into the duplex. Anthraquinone was chosen for its relative ease of coupling to the PNA backbone as well as for its documented photochemical reactivity with DNA (15-18).
Thermal denaturation curves measured for these PNA(AQ) conjugates hybridized with their DNA complements reveal the same features as observed for unmodified PNA: there is a strong preference for alignment of the PNA N-terminus with the DNA 3'-terminus and single base mismatches are highly destabilizing ([Delta]Tm = 8-10°C). Thus, replacement of a nucleobase with the non-hydrogen bonding AQ derivative does not alter the general characteristics of PNA/DNA recognition. The observation of discrete melting transitions when the absorbance is monitored at 330 nm, where only the AQ absorbs, is strong evidence in favor of intercalation by the AQ units. Phosphorescence quenching, which requires intimate association of the AQ with the nucleobases, and molecular modeling are also consistent with an intercalative binding mode for the AQ.
Interesting trends are revealed when comparisons are made among the three modified PNA oligomers. The stabilization observed for the PNA(AQ)/DNA hybrids relative to a PNA(Ac)/DNA duplex is substantial: 4.8 and 11.7°C for AQ1 and AQ2, respectively. Moreover, while hybrids of PNA(Ac) with each of the five DNA oligomers listed in Table 1 exhibit little variation in stability, there is a clear preference of the PNA(AQ) oligomers for an abasic site at the opposed position on the DNA strand. The amount of destabilization is independent of the identity of the opposed base for AQ1 while greater destabilization is observed for the purines than the pyrimidines for AQ2, which possesses the longer linker. Thus, the intercalated AQ effectively discriminates between a hydrogen and a nucleobase on the deoxyribose, whereas the acetyl group exhibits no preference.
The melting transitions in Figure 1B are evidence in favor of intercalation of the AQ derivative in the duplex state, leading to hypochromicity in the n[pi]* transition of the AQ. Denaturation of the two strands relieves the stacking interactions and leads to an enhanced absorption, analogous to the enhanced absorption at 260 nm by the nucleobases. There is substantially greater recovery of absorption by PNA(AQ2) than for PNA(AQ1). We attribute this to better stacking of the AQ derivative in the PNA(AQ2)/DNA-1X hybrid (rather than better destacking upon melting) based on the higher Tm for PNA(AQ2) and the modeling experiments, which indicate that structural distortions of the PNA/DNA duplex are required to optimally stack AQ1 into the hybrid. The lower hypochromicity for AQ1 then implies that the quinone is tilted relative to the adjacent base pairs, leading to less stacking: the structural distortion required for optimal stacking would apparently incur too great an energetic penalty.
The phosphorescence data also indicate that the AQ chromophore in PNA(AQ1) cannot optimally stack with adjacent bases, even in the single stranded PNA. The phosphorescence of the AQ is quenched by 90% in single stranded PNA(AQ2) but only by 70% in PNA(AQ1). Evidently, the flexible linker in PNA(AQ2) permits the AQ to effectively stack with the adjacent bases, whether in a single stranded or duplex environment. In both cases, hybridization with DNA-1X leads to ~95% quenching, so there must be substantial interaction between the AQ and the bases at the intercalation site.
Phosphorescence quenching of AQ upon intercalation into DNA is the first step in an intriguing sequence of events which lead to cleavage of the DNA at low oxidation potential trap sites. The mechanism for this irreversible damage of the DNA involves electron transfer from an adjacent base to the triplet state AQ at the intercalation site, then migration of the `hole' along the DNA until it is trapped (19). We have recently shown that a similar process occurs in the PNA(AQ)/DNA duplexes: irradiation of the intercalated AQ leads to cleavage of the DNA strand at sites which are >20 Å away from the intercalation site. This damage only occurs when the PNA(AQ) is hybridized with the DNA complement, leading to effective detection of the DNA sequence (20).
The results described above illustrate the exciting potential of modified PNA oligonucleotides. Reagents that typically intercalate in duplex DNA do not intercalate in DNA/PNA hybrids. However, covalent attachment of the candidate intercalator overcomes this prohibition. The ease of synthesizing internally modified PNA, the stability of the resulting hybrid duplexes, and their ability to mimic the reactivity of duplex DNA will contribute to the growing application of PNA in chemistry, biochemistry and medicine.
We wish to thank Prof. W. David Wilson and Dr Daoyuan Ding of Georgia State University for the computer models of PNA(AQ)/DNA and Prof. Loren Williams of Georgia Institute of Technology for many helpful discussions. This work was supported by funding from the National Institutes of Health (PHS NRSA GM16498-01 to B.A. and grant RO1 GM28190 to G.B.S.) for which we are grateful.
Nucleic Acids Research
Pages
Introduction
Materials And Methods
General
Materials
PNA synthesis
Formation of stable PNA/DNA hybrids: thermal denaturation studies
Phosphorescence quenching
Computer modeling of PNA(AQ)/DNA hybrids
Results
Melting behavior of PNA/DNA hybrid duplexes
Phosphorescence of PNA(AQ)/DNA hybrid duplexes
Modeling of PNA(AQ)/DNA hybrid duplexes
Discussion
Acknowledgements
References
3,6-Diaza-(N3-2-anthraquinoyl)-N6-Boc-hexanoic acid (monomer AQ1).Anthraquinone-2-carboxylic acid (2 g, 7.9 mmol), DCC (1.7 g, 8.3 mmol), O-(7-benzotriazole-1-yl)-1,1,3,3-tetramethyluronium-hexafluorophosphate (HOBT, 1.08 g, 8.0 mmol) and methyl [N-(2-Boc-aminoethyl)]glycinate (2 g, 8.6 mmol) were dissolved in DMF (25 ml) and stirred at room temperature overnight. The reaction mixture was filtered and the filtrate was washed with CH2Cl2 (2 × 25 ml). The solution was extracted with diluted NaHCO3 (3 × 25 ml), 2 M NaHSO4 (2 × 25 ml) and brine. The organic phase was dried with MgSO4, filtered and evaporated to dryness under reduced pressure. The yellow foam was dissolved in THF (10 ml) and 1 M LiOH (30 ml) was added. The mixture was stirred for 2 h. THF was removed from the solution under reduced pressure and the pH was adjusted to 2.8 with 2 M NaHSO4. The precipitate was extracted with CH2Cl2 (2 × 25 ml), evaporated to dryness and then redissolved in ethyl acetate (3 ml). The ethyl acetate solution was poured into hexane (150 ml) and the precipitate that formed was collected, 3.1 g (87%). 1H-NMR DMSO-d6 d: 7.81-8.30 (m, 7 H, aromatics); 6.98 and 7.72 (m, 1 H, BocNH); 4.17 and 3.97 (s, 2 H, CH2O); 3.52 and 3.24 (m, 2 H, CH2); 3.21 and 3.02 (m, 2 H, CH2); 1.17 and 1.19 (s, 9 H, Boc).
REFERENCES
This page is run by Oxford University Press, Great Clarendon Street, Oxford OX2 6DP, as part of the OUP Journals
Comments and feedback: www-admin{at}oup.co.uk
Last modification: 23 Jan 1998
Copyright© Oxford University Press, 1998.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  A. RAY and B. NORDÉN Peptide nucleic acid (PNA): its medical and biotechnical applications and promise for the future FASEB J, June 1, 2000; 14(9): 1041 - 1060. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||