Nucleic Acids Research

Synthesis of dithymidine methanephosphonamidates

A one-pot strategy has been chosen for the synthesis of dimeric nucleotides with a modified internucleotide bond. As monomeric units we used 5[prime]-O-DMT-protected 3[prime]-amino-3[prime]-deoxythymidine 3 (10) and 3[prime]-O-acetyl-protected thymidine 1 (Scheme 1). The applied standard protecting groups allowed the use of basic deprotection for the removal of acetyl groups and further functionalization of the 3[prime]-end of the dimer, while the acid-labile dimethoxytrityl group at the 5[prime]-end was suitable for solid phase synthesis of oligomers. Reaction of 3[prime]-O-acetylthymidin-5[prime]-yl-methanephosphonochloridate 2 with 3[prime]-amino-3[prime]-deoxythymidine 3 resulted in a mixture of two diastereomeric dimers, 4a [TnpmT(fast), TLC-fast migrating diastereomer] and 4b [TnpmT(slow), TLC-slow migrating diastereomer] as well as symmetrical bis-(3[prime]-O-acetylthymidyn-5[prime]-yl)-methanephosphonate 4c in the ratio 1:1:2. Diastereomers 4a and 4b and by-product 4c were separated by column chromatography on silica gel in a methanol/chloroform gradient. The total yield of isolated products was <30% and was not optimized. Dinucleotide units for solid phase synthesis were prepared by routine 3[prime]-end deprotection of 4a and 4b under basic conditions (ethanolic ammonium hydroxide), to give 5a and 5b respectively. Subsequent phosphitylation with 2-cyanoethyl tetraisopropylaminephosphorodiamidite afforded phosphoramidities 7a and 7b in 92 and 71% yield respectively (Scheme 1).

Structure of dinucleoside methanephosphonamidates

The structures of dinucleoside methanephosphonamidates 4, 5 and 6 were determined by NMR, MS and CD analysis. Spectral data (³¹P and MS) are given in Table 1. The ³¹P NMR chemical shifts of phosphorus in dinucleoside methanephosphonamidates fall in the range 34-37 p.p.m. and differ slightly for the two diastereomers of each pair of dimers (Table 1), exhibiting an upfield shift for diastereomers 4a and 6a. The same upfield chemical shifts for P-Me protons within each pair of dimers were observed in ¹H NMR spectra of diastereomers originating from 4a (data available in Supplementary Material (PDF 86 KB)).

The structural differences between dithymidine phosphate and both diastereomeric dithymidine methanephosphonamidates are reflected in the CD spectra. Those of 6a and 6b (available in Supplementary Material (PDF 86 KB)) are almost identical in the region between 245 and 300 nm. However, they differ in shape and intensity in the range 210-245 nm. Further studies on the properties of diastereomeric dithymidine methanephosphonamidates are in progress.

Conformation of sugar rings in dinucleotides 6a and 6b

Deoxyribonucleosides adopt predominantly the 2[prime]-endo sugar ring conformation in solution and in the solid state (16). A typical B-DNA helix also exhibits 2[prime]-endo sugar ring puckering (17). The conformational preferences observed for the sugar ring in various 3[prime]-substituted 3[prime]-deoxythymidine derivatives strongly depend on the electronegativity of the 3[prime]-substituent (18). Modification of 2[prime]-deoxyribonucleosides to 3[prime]-amino-2[prime],3[prime]-dideoxyribonucleosides increases the population of the 3[prime]-endo conformer, both in nucleosides (18) and in oligonucleotides originating from these modified units (19). The conformation of the sugar rings of dithymidine methanephosphonamidates 6a and 6b was elucidated routinely using an empirical method of analysis of the vicinal coupling constants ³J_{1[prime]2[prime]} and ³J_{1[prime]2[prime][prime]} (Table 3; 20). The population of the 2[prime]-endo sugar ring conformer (% S) was calculated from the equation % S = (³J_{1[prime]2[prime]} + ³J_{1[prime]2[prime][prime]} - 9.8)/5.9 (20). As shown in Table 3, both diastereomers prefer the same mode of ring pucker. The 5[prime]-terminal nucleoside (Tn) riboses take predominantly the 3[prime]-endo conformation, whereas unmodified sugars at the 3[prime]-terminus (pT) exhibit predominantly the 2[prime]-endo ring pucker. Comparison of these datawith conformational data for the parent dimers: dithymidine phosphate (TpT), dithymidine phosphoramidate (TnpT) (19) and dithymidine methanephosphonate (TpmT) (21) indicates a preference for the 2[prime]-endo sugar ring conformation for dinucleotide TpT and for 3[prime]-terminal nucleosides in TnpT and in TnpmT dimers. Dithymidine methanephosphonates exist in an equimolar equilibrium of both conformational isomers. However, distinct conformational changes are observed for5[prime]-terminal sugar of TnpT and both TnpmT dimers. In these cases the 2[prime]-endo conformer is a minor one. The presence of a less electronegative 3[prime]-amino group in 3[prime]-amino-3[prime]-deoxy modified dinucleotides induces their A-type structure. Most probably, this property promotes oligo(deoxyribonucleoside phosphoramidates) as high affinity complements for hybridization with target RNAs (9,19). In this respect novel oligo(deoxyribonucleoside methanephosphonamidates) are promising compounds for further investigations.

Absolute configuration of dithymidine methanephosphonamidates

Dinucleoside (3[prime]N->P5[prime]) methanephosphonamidates constitute a new class of derivatives for which neither spectral nor crystallographic data are available. Therefore, our assignment of absolute configuration at the phosphorus is based on data reported for the structurally related methanephosphonates (22-25). In only a very few cases was the absolute configuration of the dinucleoside methanephosphonates determined by X-ray analysis (26,27). However, there are some spectral features, common within particular diastereomeric series, which can be used to predict the stereochemistry of chiral phosphorus centers by means of UV, CD and NMR spectroscopy (24). Fast migrating dinucleoside methanephosphonates were always identified as [Rp] diastereomers by means of TLC (protected TpmT) and HPLC (deprotected TpmT). [Rp] diastereomers exhibit an upfield chemical shift of the phosphorus atom in ³¹P NMR (22). The same upfield chemical shift is observed for P-Me group protons in ¹H NMR (22). The profiles of CD spectra for pairs of dimers are similar, but differ in magnitude of molecular ellipticity, with higher values for the fast migrating isomer [Rp] due to the different mode of base stacking and different overall structure (24). Taking into account the above arguments, tentative assignment of the absolute configuration of the phosphorus atom in dinucleoside methanephosphonamidates 4a-7a [TnpmT(fast)] as [Rp] is proposed. This assignment was made on the basis of the following data: (i) the fast migrating protected dimer is the 4a dimer (Table 1); (ii) upfield shifted signals of the phosphorus in ³¹P NMR spectra are obtained for the 4a, 5a, 6a and 7a derivatives (Table 1); (iii) upfield shifted signals of P-Me group protons in ¹H NMR spectra are obtained for the 4a and 6a derivatives (data available in Supplementary Material (PDF 86 KB)); (iv) the CD spectrum of 6a exhibits a slightly higher magnitude of the peaks at 215 and 245 nm.

The stereochemistry of the chiral phosphorus center in dinucleoside methanephosphonates was also investigated and predicted on the basis of NMR data with application of a ROESY technique (22,23,25). As was shown for dinucleoside methanephosphonates, nuclear Overhauser effects (NOEs) between H-3[prime], H-4[prime] and H-5[prime] ribose protons and protons of the phosphorus methyl group strongly depend on the absolute configuration of the phosphorus and were used as a criterion to distinguish between [Sp] and [Rp] diastereomers. The absolute phosphorus configuration the NOE of P-Me to H-4[prime] of the 5[prime]-terminal nucleoside was used as diagnostic for [Rp] (22). The presence of a distinct cross-peak between P-Me and H-3[prime] of the 5[prime]-terminal nucleoside was reported to be a criterion for assignment of the [Sp] phosphorus configuration (23), however further investigations of a variety of dimeric methanephosphonates only partially support this assumption (24). In some cases both diastereomers exhibit NOE cross-peaks of P-Me to H-3[prime] of the 5[prime]-terminal nucleoside (22). NOE cross-peaks of P-Me to H-5[prime] and H-5[prime][prime] of the 3[prime]-terminal nucleoside were also detected for the [Rp] dimer (22). Thus, further assignment of absolute configuration of the phosphorus center in dithymidine methanephosphonamidates is based on NOEs between P-Me group protons and protons of the ribose moieties of the fully protected dimers 4a and 4b (Fig. 1a and b respectively). In the extended part of the ROESY spectrum of fast migrating diastereomer 4a (Fig. 1a) the strongest cross-peak originates from interaction of P-Me protons with H-3[prime] of Tn or H-5[prime] of pT sugars protons. Analysis of the overlapping signals shows that the upfield part of the broad signal represents the H-5[prime] proton of pT. Thus, the strongest cross-peak which is present in the spectrum of fast migrating diastereomer 4a originates from the interaction between P-CH₃ and H-5[prime] of pT protons. Also, cross-peaks from P-Me to H-4[prime] of Tn and H-4[prime] of pT are present. A ROESY spectrum of slow migrating diastereomer 4b (Fig. 1b) shows one distinct cross-peak from the P-Me protons to the H-3[prime] and H-4[prime] protons of the Tn ribose moiety. Unfortunately, an overlap of these H-3[prime] and H-4[prime] proton resonances does not allow us to assign NOE interactions for particular nuclei. However, in the ROESY spectrum of fully deprotected dimer 6b (data not shown) two cross-peaks of similar intensity from P-Me to the H-3[prime] and H-4[prime] protons of Tn are present, thus indicating both P-Me-H3[prime] and P-Me-H4[prime] interactions. Although the diagnostic NOEs of dinucleoside methanephosphonamidates are similar to those of dinucleoside methanephosphonates, in the spectra of both diastereomers 4a and 4b strong interactions between P-Me and H-4[prime] of Tn are present. This could be due to the shorter P-N bond (1.73 Å) in methanephosphonamidates than the P-O bond (1.79 Å) in the parent methanephosphonates, causing a decrease in the P-Me-H-4[prime] interatomic distance and, thus, enhancing the NOE interaction. Molecular modeling of dithymidine methanephosphonamidates 6a and 6b with the help of the HyperChem program (MM+ method) results in a structure of the [Rp] diastereomer with the closest P-CH₃-H-5[prime] of pT contact (2.37 Å). In contrast, the model structure of the [Sp] diastereomer has the P-CH₃-H-3[prime] of Tn distance as the closest contact. Thus, assignment of the absolute configuration at the phosphorus atom as [Rp] for fast migrating dimer 4a and [Sp] for slow migrating dimer 4b is proposed.

Figure 1. Extended plot of ROESY spectra of 5[prime]-DMT, 3[prime]-Ac-protected dithymidine methanephosphonamidates 4a [TnpmT(fast), (a)] and 4b [TnpmT(slow), (b)]. Cross-peaks of P-Me and ribose protons are shown. Tn and pT correspond to ribose protons of the 5[prime]-terminal and 3[prime]-terminal nucleoside respectively.

Figure 2. Gel electrophoresis of 5[prime]-³²P-labeled oligo(deoxyribonucleoside methanephosphonamidates) F1-F4 (lanes 2, 4, 6 and 8) and S1-S4 (lanes 3, 5, 7 and 9) (20% polyacrylamide, 7 M urea). A mobility retardation effect is observed due to introduction of a non-ionic modified internucleotide linkage. For comparison dodecathymidylate (lanes 1 and 10) is shown.

Stability of dimers 6a and 6b

Fully deprotected dimers 6a and 6b were tested for their stability at pH 3 and pH 11 at 37°C for 15 h. Stability of dimers 6a and 6b to Penicillum citrinum nuclease P1 and svPDE at 37°C for 15 h was also checked. HPLC profiles of the mixtures before [peak of TnpmT(fast) at R_t = 12.82 min, peak of TnpmT(slow) at R_t = 13.30 min] and after incubation (peaks exclusively at 12.80 or 13.23 min) (according to method A, Materials and Methods) are identical (data not shown). Thus, dithymidine methanephosphonamidates are stable under a broad range of pH conditions and are resistant to enzymatic cleavage.

Oligodeoxyribonucleotides: synthesis, deprotection and purification

Two series of chimeric thymidine dodecamers, containing one to four dithymidine methanephosphonamidate (TnpmT) blocks incorporated into the oligodeoxyribonucleotide chain, were obtained by the phosphoramidite approach (Table 2; 28). Oligomers originating from dimer 4a [TnpmT(fast)] are designated F and those originating from dimer 4b [TnpmT(slow)] are designated S. Average coupling yields of dimeric units were 86-96% per step, as judged by DMT cation assay. Oligomers were purified before and after DMT removal by means of reverse phase HPLC with a triethylammonium bicarbonate (pH 7.0)/acetonitrile gradient. The purity of the oligomers was checked by HPLC analysis on a C18 column according to method B (Materials and Methods) (Table 2) and by 20% polyacrylamide/7 M urea gel electrophoresis of 5[prime]-³²P-labeled samples (Fig. 2). The mobilities of oligomers with modified internucleotide bonds were compared with the mobility of non-modified homopolymer T₁₂. Subsequent mobility retardation of oligomers F1-F4 (lanes 2, 4, 6 and 8) and S1-S4 (lanes 3, 5, 7 and 9) in comparison with the mobility of T₁₂ (lanes 1 and 10) occurs due to increasing participation of hydrophobic, non-ionic methanephosphonamidate internucleotide linkages in the overall structure of oligomers within a particular series. The determined purity of all oligomers was [ge]98%. Oligomers of both series show the correct molecular mass values as determined by MALDI-TOF mass spectrometry (Table 2).

Structures of oligodeoxyribonucleotides

The structures of oligodeoxyribonucleotides F1-F4, S1-S4, the reference T₁₂ and their heteroduplexes with dA₁₂ or A₁₂ were investigated by CD measurements. The respective CD spectra are available in Supplementary Material (PDF 86 KB). The spectra of single-stranded oligomers are almost identical. The CD profiles for DNA/DNA duplexes also do not differ for the particular modifications. They show positive peaks at 280 nm and intense negative ones at 250 nm and belong to the B-DNA family. Modified oligomers as well as the T₁₂ oligomer hybridized with RNA exhibit positive bands around 265 nm and negative ones at 250 nm. Their structure can be assigned to the A-DNA family of duplexes from the shape of the spectra (17). Unlike double-stranded DNA/DNA, in this case the presence of four modified linkages (as in the F4 and S4 oligomers) causes a significant increase in the intensity of the 265 nm bands. This difference probably reflects the more A-type character of F4/RNA and S4/RNA duplexes in comparison with the duplex T₁₂/RNA.

Enzymatic degradation of oligo(thymidine methanephosphonamidates)

As indicated above, the methanephosphonamidate internucleotide bond in dinucleotides 6a and 6b is stable against nucleases P1 and svPDE. Exhaustive enzymatic digestion of oligomers F1-F4 and S1-S4 with Penicillum citrinum nuclease P1 (18 h, 37°C) and calf intestine alkaline phosphatase (3 h, 37°C) in 100 mM Tris-HCl, pH 7.2, and 1 mM Zn²⁺ buffer resulted in thymidine and dimer 6a [TnpmT(fast)] or 6b [TnpmT(slow)] respectively. The digests were analyzed by HPLC on a C18 column with an acetonitrile/0.1 M ammonium acetate (pH 7.0) gradient. Two main peaks detected at 10.05 and 12.87 min or alternatively 13.53 min were assigned as the peaks of thymidine and dimer 6a or dimer 6b respectively, by means of co-injection with genuine samples (data not shown). Numeric peak integration was in accordance with the calculated values. Shorter incubation times caused only partial digestion of the oligomers, resulting in HPLC profiles showing several peaks with R_t longer than that of thymidine. The ratio of the peaks of thymidine to the sum of the peaks with longer retention times (assigned as TnpmT, pTnpmT, TnpmTpTnpmT, pTnpmTpTnpmT etc.) was in agreement with the expected values for particular oligomers. Relatively slow degradation of modified oligomers is a consequence of the presence of alternative phospodiester and methanephosphonamidate bonds, thus making phosphodiester bonds less accessible to the endonuclease.

Figure 3. Gel electrophoresis of products of 3[prime]-exonuclease degradation of oligo(deoxyribonucleoside methanephosphonamidates) F1-F4 (lanes 2-5) and S1-S4 (lanes 6-9) (20% polyacrylamide, 7 M urea). Incubation with 3[prime]-exonuclease of a 50% saline solution of human plasma was for 4 h at 37°C. Resistance of the methanephosphonamidate linkage toward 3[prime]-exonuclease degradation results in shorter oligonucleotides possessing TnpmT dimers at their 3[prime]-ends (lanes 2-5 and 6-9). For comparison, degradation of dodecathymidylate T₁₂ is shown (lane 1). Bands of lower intensity in lanes 2-9 correspond to the products of degradation due to the `hopping' properties of 3[prime]-exonuclease.

Human plasma 3[prime]-exonuclease digestion

Dodecathymidylates F1-F4 and S1-S4 and the reference T₁₂ were used to study their stability to 3[prime]-exonuclease from human plasma (13). Substrates labeled at the 5[prime]-end with [³²P]phosphate were incubated in 50% saline solution of human plasma for 4 h and then analyzed on a 20% polyacrylamide gel containing 7 M urea (Fig. 3). Non-modified oligomer T₁₂ is digested to a ladder of products from T₁₁ to T₂ (lane 1). The oligomers F1-F4 (lanes 2-5) and S1-S4 (lanes 6-9) containing modified internucleotide bonds are degraded only partially. Degradation proceeds from the 3[prime]-terminus of the oligomers and is stopped by the presence of a modified internucleotide linkage. Thus, the remaining non-degraded oligomers contain the modified unit TnpmT at their 3[prime]-ends. For example, both the F1 and S1 oligomers (lanes 2 and 6 respectively) are digested to oligomers containing from n - 1 to n - 5 nucleotides. It is noteworthy that in the course of digestion, which starts from the 3[prime]-end of the oligomers, the ratio of charged internucleotide linkages to methanephosphonamidate linkages drops. Therefore, the products originating from modified oligomers F1-F4 (lanes 2-5) and S1-S4 (lanes 6-9) exhibit gradually decreased mobility on a gel. As a result, the bands representing degraded oligomers overlap with bands of non-degraded ones. Overexposed gels additionally show ladders of shorter oligomers. These short 5[prime]-labeled oligomers consist of the products of enzymatic degradation of modified oligomers F1-F4 and S1-S4 due to the `hopping' properties (29,36) of human plasma 3[prime]-exonuclease with modified internucleotide linkages.

Table 4. Hybridization properties of oligo(deoxyribonucleoside methane-phosphonamidates) F2-F4 and S2-S4 with dA₁₂ and A₁₂. Temperature depressions caused by introduction of each modified internucleotide linkage and for two different complementary sequences are shown

Oligonucleotide	Target	Tma	[Delta]Tmb	[Delta]Tmc
T12	dA12	39.0
F2	dA12	33.7	2.6
F3	dA12	31.3	2.4
F4	dA12	29.3	2.0
S2	dA12	27.0	6.0
S3	dA12	21.4	5.6
S4	dA12	15.3	6.1
T12	A12	27.3		11.7
F2	A12	23.6	1.8	10.1
F3	A12	22.3	1.3	9.0
F4	A12	21.3	1.0	8.0
S2	A12	17.3	5.0	9.7
S3	A12	13.1	4.2	8.3
S4	A12

^aT_m was determined with an accuracy of ±0.5°C
^bT_m decrease caused by an introduction of the subsequent additional methanephosphonamidate linkage into the phosphate backbone
^cDifference in T_m values for two complementary heteroduplexes formed by dA₁₂ and A₁₂ respectively.

Hybridization properties

DNA/DNA duplexes. As was shown by the two TpmT oligonucleotides with either the [Rp] or [Sp] phosphorus configuration, the binding affinity to complementary oligonucleotides significantly depends on the stereochemistry of the methanephosphonate units (6,25). Melting temperature measurements of duplexes obtained by hybridization of modified oligomers F1-F4 and S1-S4 with the dA₁₂ target were performed at various oligonucleotide concentrations (1 × 10^-6 - 6 × 10^-6 M). The resulting T_m values as plots of T_m versus duplex concentration are included in Supplementary Material (PDF 86 KB). The thermal stability of all duplexes possessing a modified internucleotide linkage incorporated into the oligonucleotide chain is lower than that of the unmodified T₁₂/dA₁₂ duplex and strongly depends on the type and number of introduced internucleotide bond modifications (Table 4). The F-type modification [originating from TnpmT(fast)] lowers the T_m by ~2.4°C, whereas the S-type modification [originating from TnpmT(slow)] lowers the T_m by ~6.0°C. Since, for self-complementary octamers containing one methanephosphonate linkage introduced into the oligonucleotide chain, depression of T_m for the [Rp] diastereomer is smaller than for the [Sp] diastereomer (25), the results obtained are additional support for our assignment of the absolute phosphorus configurations in dithymidine methanephosphonamidates as [Rp] for 6a [TnpmT(fast)] and [Sp] for 6b [TnpmT(slow)]. Thermodynamic parameters ([Delta]H°, [Delta]S° and [Delta]G° at 25°C) obtained from the linear dependence 1/T_m versus ln[total strand concentration] (30) show no statistically significant differences for the studied oligomers (data included in Supplementary Material (PDF 86 KB)).

DNA/RNA duplexes. The stabilities of DNA/RNA duplexes are much lower than those of DNA/DNA (31). As expected, the melting temperatures of duplexes originating from F2-F4 or S2-S3 oligomers and the A₁₂ target oligoribonucleotide are lower by 8.0-11.7°C (Table 4) as compared with those with the dA₁₂ target oligonucleotide. This lowering of the melting temperature is smaller when oligo(deoxyribonucleoside methanephosphonamidates) are hybridized with RNA than with DNA, as compared with non-modified oligonucleotide T₁₂. Here the F-type modification lowers T_m values by only 1.0-1.8°C, whereas the S-type modification lowers T_m values by 4.2-5.0°C. As was shown by Gryaznov et al. (9), a decathymidylate possessing phosphoramidate linkages in alternate positions exhibits a lower binding affinity for DNA targets but hybridizes better with RNA targets. In our case an introduced methanephosphonamidate linkage decreases the binding affinity of modified oligomers for complementary RNA (to a lesser extent) and DNA (to a greater extent) targets, with less effect for F-type modifications (6). The difference in hybridization properties of oligomers originating from both [TnpmT(fast) and TnpmT(slow)] diastereomeric units could be caused by different steric perturbations (6) as well as different hydration patterns (12,35) of duplexes caused by the presence of a methyl group in the modified phosphate backbone.

Figure 4. Thermal denaturation profiles of complexes of oligo(deoxyribonucleoside methanephosphonamidates) F4 and S4 and dodecathymidylate T₁₂ hybridized with the d(A₂₁C₄T₂₁) hairpin (HP) oligomer. The transitions of hairpin to single-stranded oligomers are observed in the temperature range 60-90°C. The clear transition of triplex to duplex is observed in the range 3-35°C for the F4/HP complex (insert).

Triplex formation

Since it is well known that homopolymers of thymidine have the ability to bind to poly(dA)·poly(T) duplex oligonucleotides by Hoogsteen hydrogen bonds giving rise to stable triplex structures of the T*dA·T type (32,33), melting temperature experiments were performed with dodecathymidylates T₁₂, F4 and S4 and the duplex oligonucleotide hairpin oligomer d(A₂₁C₄T₂₁). An increase in UV absorption at temperatures >60°C represents transition of the hairpin structure to single-stranded oligomer (Fig. 4). All the melting profiles for T₁₂/d(A₂₁C₄T₂₁), F4/d(A₂₁C₄T₂₁) and S4/d(A₂₁C₄T₂₁) triplexes are similar in this region. However, the profiles of those melting curves are significantly different at lower temperatures (3-50°C, insert in Fig. 4). The curve observed for oligonucleotide F4 complexed with hairpin oligomer d(A₂₁C₄T₂₁) clearly shows an upper part of triplex-duplex transition, in contrast to the lack of such clear transitions for the remaining complexes. Despite a low binding affinity of the dodecathymidylates, due to the length of the chain, for hairpin d(A₂₁C₄T₂₁), F-type oligodeoxyribonucleotide modification enhances triplex formation (34). In contrast, S-type modification, present in oligomer S4, does not improve its affinity for the hairpin d(A₂₁C₄T₂₁) and decreases triplex stability, as compared with the unmodified triplex structure. Heating and cooling curves (0.2°C/min) recorded for T₁₂/d(A₂₁C₄T₂₁) or F4/d(A₂₁C₄T₂₁) triplexes do not exhibit any significant hysteresis for triplex-duplex transitions (data not shown). Further hybridization studies on longer chimeric oligo(deoxyribonucleoside methanephosphonamidates) concerning their tendency to form triplex structures are in progress.

CONCLUSIONS

Diastereomeric dithymidine methanephosphonamidates 4a and 4b were obtained by reaction of 3[prime]-amino-3[prime]-deoxythymidine with 3[prime]-O-acetylthymidin-5-yl-methanephosphonchloridate. Successful separation of the diastereomers was achieved by silica gel column chromatography with a methanol/chloroform gradient, giving rise to fast migrating diastereomer 4a [TnpmT(fast)] and slow migrating diastereomer 4b [TnpmT(slow)]. By comparison with the stereochemistry of dinucleoside methanephosphonates tentative assignment of the absolute configuration at the phosphorus atom of [Rp] for diastereomer 4a and [Sp] for diastereomer 4b is proposed. Selective removal of the 3[prime]-terminal protecting group from 4a and 4b and subsequent phosphitylation with 2-cyanoethyl tetraisopropylphosphordiamidite gave 7a and 7b respectively, which were used as building blocks for automated solid phase synthesis of oligodeoxyribonucleotides. Chimeric dodecathymidylates F1-F4 and S1-S4, possessing one to four modified linkages originating from diastereomers TnpmT(fast) or TnpmT(slow) respectively, were obtained. Stability of dimeric methanephosphonamidates 6a and 6b to nuclease P1 and svPDE as well stability of dodecathymidylates F1-F4 and S1-S4 to nuclease P1 and 3[prime]-exonuclease from human plasma show a complete resistance of the methanephosphonamidate linkage to nucleolytic degradation. The binding affinities of chimeric dodecathymidylates for single-stranded DNA and RNA as well as double-stranded DNA strongly depend on the stereochemistry of the methanephosphonamidate units. In general, both chimeric dodecathymidylate series possess lower binding affinities for complementary single-stranded oligonucleotides in comparison with the binding properties of non-modified dodecathymidylate. However, chimeric dodecathymidylates F1-F4, originating from diastereomer TnpmT(fast), exhibit better hybridization properties in comparison with dodecathymidylates S1-S4, originating from diastereomer TnpmT(slow). Moreover, chimeric dodecathymidylate F4, possessing four methanephosphonamidate linkages in alternate positions, exhibits a higher binding affinity for hairpin d(A₂₁C₄T₂₁) in comparison with the T₁₂ and S4 oligomers. Modification of the oligodeoxyribonucleotide phosphate backbone with F-type methanephosphonamidate linkages enhances triplex stability.

In summary, the stability against enzymatic degradation and the configurational differentiation with respect to binding affinity of [Rp] and [Sp] oligo(deoxyribonucleoside methanephosphonamidates) for single-stranded DNA and RNA as well double-stranded DNA could be of interest for further investigation.

ACKNOWLEDGEMENTS

This project was supported by The State Committee for Scientific Research (KBN) grant 4P05F 02310 (to W.J.S.). The authors are indebted to Mrs Wieslawa Goss for excellent technical assistance.

See Supplementary Material (PDF 86 KB) available in NAR Online.

REFERENCES

1. Uhlmann,E. and Peyman,A. (1990) Chem. Rev., 90, 544-584.

2. Crooke,S.T. and Lebleu,B. (1993) Antisense Research and Applications. CRC Press, Boca Raton, FL.

3. Lesnikowski,Z.J., Jaworska,M. and Stec,W.J. (1988) Nucleic Acids Res., 16, 11675-11689. MEDLINE Abstract

4. Lesnikowski,Z.J., Jaworska,M. and Stec,W.J. (1990) Nucleic Acids Res., 18, 2109-2115. MEDLINE Abstract

5. Agrawal,S., Temsamani,J., Galbraith,W. and Tang,J. (1995)Clin. Pharmacokinet., 28, 7-16. MEDLINE Abstract

6. Reynolds,M.A., Hogrefe,R.I., Jaeger,J.A., Schwartz,D.A., Riley,T.A., Marvin,W.B., Daily,W.J., Vaghefi,M.M., Beck,T.A., Knowles,S.K., Klem,R.E. et al). (1996) Nucleic Acids Res. 24, 4584-4591. MEDLINE Abstract

7. Agrawal,S. and Zhang,R. (1997) Ciba Foundn Symp., 209, 60-75.

8. Zhou-Sun,B., Sun,J., Gryaznov,S.M., Liquier,J., Garestier,T., Hélène,C. and Taillandier,E. (1997) Nucleic Acids Res., 25, 1782-1787. MEDLINE Abstract

9. Gryaznov,S.M. and Chen,J. (1994) J. Am. Chem. Soc., 116, 3143-3144.

10. Chen,J., Schultz,R.G., Lloyd,D.H. and Gryaznov,S.M. (1995)Nucleic Acids Res., 23, 2661-2668. MEDLINE Abstract

11. Gryaznov,S.M., Skorski,T., Cucco,C., Nieborowska-Skorska,M., Chiu,C.Y., Lloyd,D.H., Chen,J., Koziolkiewicz,M. and Calabretta,B. (1996) Nucleic Acids Res., 24, 1508-1514.

12. Barsky,D., Colvin,M.E., Zon,G. and Gryaznov,S.M. (1997)Nucleic Acids Res., 25, 830-935. MEDLINE Abstract

13. Koziolkiewicz,M., Wójcik, M., Kobylanska,A., Karwowski,B., Rebowska,B., Guga,P. and Stec,W.J. (1997) Antisense Nucleic Acids Drug Dev., 7, 43-48.

14. Agarwal,K.L. and Riftina,F. (1979) Nucleic Acids Res., 6, 3009-3024. MEDLINE Abstract

15. Zon,G. and Stec,W.J. (1991) In Eckstein,F. (ed.), Oligonucleotides and Analogues. A Practical Approach. IRL Press at Oxford University Press, Oxford, pp. 87-108.

16. Altona,C. and Sundaralingam,M. (1973) J. Am. Chem. Soc., 95, 2333-2344. MEDLINE Abstract

17. Saenger,W. (1984) Principles in Nucleic Acid Structure. Springer-Verlag, New York, NY.

18. Thibaudeau,C., Plavec,J., Garg,N., Papchikhin,A. and Chattopadhyaya,J. (1994) J. Am. Chem. Soc., 116, 4038-4043.

19. Ding,D., Gryaznov,S.M., Lloyd,D.H., Chandrasekaran,S., Yao,S., Ratmeyer,L., Pan,Y. and Wilson,W.D. (1996) Nucleic Acids Res., 24, 354-360. MEDLINE Abstract

20. Rinkel,L.J. and Altona,C. (1987) J. Biomol. Struct. Dyn., 4, 621-649. MEDLINE Abstract

21. Wozniak,L.A., Wieczorek,M., Pyzowski,J., Majzner,W. and Stec,W.J. (1998) J. Org. Chem., in press.

22. Löschner,T. and Engels,J.W. (1990) Nucleic Acids Res., 18, 5083-5088. MEDLINE Abstract

23. Wang,C., Wang,L., Yang,X., Jiang,T. and Zhang,L. (1993) Nucleic Acids Res., 21, 3245-3248. MEDLINE Abstract

24. Lebedev,A.V., Frauendorf,A., Vyazovkina,E.V. and Engels,J.W. (1993) Tetrahedron, 49, 1043-1052.

25. Bower,M., Summers,M.F., Powell,C., Shinozuka,K., Regan,J.B., Zon,G. and Wilson,W.D. (1987) Nucleic Acids Res., 15, 4915-4930. MEDLINE Abstract

26. Chacko,K.K., Lindner,K. and Saenger,W. (1983) Nucleic Acids Res., 18, 2109-2115.

27. Han,F., Watt,W., Callahan,L., Kezdy,F.J. and Agarwal,K.L. (1986)Proc. Natl. Acad. Sci. USA, 83, 1617-1621.

28. Caruthers,M.H. (1985) Science, 230, 281-285. MEDLINE Abstract

29. Stec,W.J., Zon,G., Egan,W. and Stec,B. (1984) J. Am. Chem. Soc., 106, 6077-6079.

30. Breslauer,K.J. (1994) In Agrawal,S. (ed.), Methods in Molecular Biology, Vol. 26, Protocols for Oligonucleotide Conjugates. Humana Press, Totowa, NJ, pp. 347-372.

31. Clark,C.L., Cecil,P.K., Singh,D. and Gray,D.M. (1997) Nucleic Acids Res., 25, 4098-4105. MEDLINE Abstract

32. Thuong,N.T. and Hélène,C. (1993) Angew. Chem. Int. Edn English, 32, 666-690.

33. Giovannangeli,C. and Hélène,C. (1997) Antisense Nucleic Acids Drug Dev., 7, 413-421.

34. Hausheer,F.H., Singh,U.C., Saxe,J.D., Colvin,O.M. and T'so,P.O.P. (1990) Anti-Cancer Drug Design, 5, 159-167. MEDLINE Abstract

35. Tereshko,V., Gryaznov,S. and Egli,M. (1998) J. Am. Chem. Soc., 120, 269-283.

36. Ozaki,H., Kitamura,M., Yamana,K., Murakami,A. and Shimidzu,T. (1990) Bull. Chem. Soc. Jpn, 63, 1929-1936.

---

*To whom correspondence should be addressed. Tel: +48 42 81 97 44; Fax: +48 42 81 54 83; Email: bnawrot@lodz.pdi.net

---

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: 19 May 1998
Copyright©Oxford University Press, 1998.
CiteULike    Connotea    Del.icio.us    What's this?

This Article Abstract Print PDF (129K) Supplementary Data 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 (12) Request Permissions Commercial Re-use Guidelines for Open Access NAR Content Google Scholar Articles by Nawrot, B. Articles by Stec, W. J. Search for Related Content PubMed PubMed Citation Articles by Nawrot, B. Articles by Stec, W. J. Social Bookmarking          What's this?