3'-deoxyguanosine (3’-dG)–(2'-5' linked) is deoxy at the 3’-position of the ribose, instead of at the usual 2’-position (note: the 3’-deoxynucleotides of A, C, and T are also available from Gene Link). 3’-deoxynucleotides are used to substitute 2’-5’ phosphodiester linkages for the usual 3’-5’ phosphodiester linkages at some or all positions of an oligo. Oligonucleotides containing all, or primarily, 2’,5’-phosphodiester linkages selectively bind to complementary single-stranded 3’,5’-RNA over comparable 3,5’-DNA (1). This property means that DNA oligos containing such linkages could be useful in either anti-sense applications or as ssRNA-specific probes.

In an important fundamental study, Bhan et al. (2) observed the following regarding 2’-5’ DNA oligos:

(a) 2’-5’ phosphodiester oligos bind to complementary ssRNA, but not to complementary ssDNA, while the corresponding 3’-5 phosphodiester oligos bind to both ssRNA and ssDNA.2’-5’ oligos are thus selective for complementary ssRNA.

(b) The Tm of 2’-5’ DNA:RNA duplexes is only modestly lower than that of standard 3’-5’ DNA:RNA duplexes (DeltaTm ~ – 0.2degC per 2’-5’ linkage substitution).

(c) Phosphorothiolation (which confers nuclease resistance) of 2’-5’ linkages does not appear to significantly lower the Tm of 2’-5’ DNA:RNA duplexes (by contrast, phosphorothiolation of 3’-5’ linkages does significantly lower the Tm of 3’-5’ DNA:RNA duplexes).

(d) 2’-5’ oligos, whether phosphorothiolated or not, do not support RNAse H activity when bound to complementary RNA. However,

(e) Substitution of six or seven contiguous 3’-5’ phosphorothiolate linkages into a 2’-5’ phosphorothiolated oligo at an appropriate place (that is, making a 2’-5’/3’-5’ phosphorothiolated chimeric oligo) restores the oligo’s ability to support RNAse H activity.

(f) 2’-5’ oligos, whether phosphorothiolated or not, show little or no non-sequence specific binding to cellular proteins (by contrast, 3’-5’ oligos show considerable levels of such binding).

(g) Using Chinese hamster ovary cells transfected with the human 5a-reductase-II (5aR-II) gene, a 2’-5’/3’-5’ chimeric phosphorothiolated anti-sense oligo, with the 3’-5 linkages (seven) in the center to activate RNAse H activity, and complementary to the 5’-UTR of 5aR-II, showed comparable effectiveness (~ 70% at 100 nM oligo conc.) at inhibiting 5aR-II protein expression in a dose-dependent manner as the corresponding 3’-5’ anti-sense oligo.

In summary, this research suggests that 2’-5’/3’-5’ phosphorothiolated chimeric oligos, in which 6-7 of the linkages are 3’-5’ to ensure that it can support RNAse H activity, have considerable potential as anti-sense reagents, due to their high selectivity for complementary RNA targets, and minimal non-sequence specific binding to cellular proteins.