HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: rna.538907v1 13/8/1301    most recent Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Sipa, K. Articles by Nawrot, B. Search for Related Content PubMed PubMed Citation Articles by Sipa, K. Articles by Nawrot, B. Social Bookmarking                   What's this?

Effect of base modifications on structure, thermodynamic stability, and gene silencing activity of short interfering RNA

¹ Department of Bioorganic Chemistry, Centre of Molecular and Macromolecular Studies, Polish Academy of Sciences, 90-363 Lodz, Poland
² Institute of Organic Chemistry, Faculty of Chemistry, Technical University of Lodz, Zeromskiego 116, 90-924 Lodz, Poland

	ABSTRACT

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTAL DATA ACKNOWLEDGMENTS REFERENCES

 
A series of nucleobase-modified siRNA duplexes containing "rare" nucleosides, 2-thiouridine (s²U), pseudouridine (), and dihydrouridine (D), were evaluated for their thermodynamic stability and gene silencing activity. The duplexes with modified units at terminal positions exhibited similar stability as the nonmodified reference. Introduction of the s²U or units into the central part of the antisense strand resulted in duplexes with higher melting temperatures (Tm). In contrary, D unit similarly like wobble base pair led to the less stable duplexes (Tm 3.9 and 6.6°C, respectively). Gene-silencing activity of siRNA duplexes directed toward enhanced green fluorescent protein or beta-site APP cleaving enzyme was tested in a dual fluorescence assay. The duplexes with s²U and units at their 3'-ends and with a D unit at their 5'-ends (with respect to the guide strands) were the most potent gene expression inhibitors. Duplexes with s²U and units at their 5'-ends were by 50% less active than the nonmodified counterpart. Those containing a D unit or wobble base pair in the central domain had the lowest Tm, disturbed the A-type helical structure, and had more than three times lower activity than their nonmodified congener. Activity of siRNA containing the wobble base pair could be rescued by placing the thio-nucleoside at the position 3'-adjacent to the mutation site. Thermally stable siRNA molecules containing several s²U units in the antisense strand were biologically as potent as their native counterparts. The present results provide a new chemical tool for modulation of siRNA gene-silencing activity.

Keywords: siRNA; thermodynamic stability; duplex asymmetry; 2-thiouridine; chemical modification

	INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTAL DATA ACKNOWLEDGMENTS REFERENCES

 
Exogenous control of gene expression for functional studies and potentially for therapeutic applications became realistic by the use of RNA interference (RNAi) technology. Shortly after its discovery by Fire et al. (1998) in the Caenorhabditis elegans system, the RNAi machinery was proven to be present and effective in mammalian cells (Elbashir et al. 2001a). Sequence-specific gene silencing may be induced by synthetic, short, double-stranded RNAs (short interfering RNAs, siRNAs) identified in Drosophila melanogaster embryo extracts (Zamore et al. 2000; Elbashir et al. 2001b; Ketting et al. 2001). Most commonly used siRNA duplexes consist of two 21-nucleotide (nt) complementary strands terminated on their 3'-ends with 2-nt overhangs (Elbashir et al. 2001b; Nykanen et al. 2001; Tang et al. 2003). The siRNA duplex is incorporated into nucleoprotein complex RISC (RNA-induced silencing complex) (Martinez et al. 2002). One of the two strands is selected as a guide strand and the other one, designated as the sense or passenger strand, is endonucleolytically cleaved by the RISC (Matranga et al. 2005; Rand et al. 2005). The remaining guide (antisense) strand directs the RISC complex to the complementary mRNA target. Assembly of the active RISC complex is an asymmetrical process in which the selection of a guide siRNA strand is determined by the relative thermodynamic stability of the 3'- and 5'-ends of the siRNA duplex (Khvorova et al. 2003; Schwarz et al. 2003).

Even though siRNAs have been successfully used for gene silencing in vivo, there is still a need to improve their pharmacokinetic features, which predetermine therapeutic utility. One approach to reach this goal is the introduction of chemical modifications into siRNA duplexes (Manoharan 2004; Fougerolles et al. 2005; Uprichard 2005; Nawrot and Sipa 2006). Research efforts in this field are focused on improving three of the most important siRNAs features: (1) cell membrane permeability, (2) in vivo stability, and (3) specificity of their silencing action. Insufficient cross-membrane cellular uptake limits the application of nonmodified oligonucleotides, including siRNA, in medicine. Interestingly, Soutschek and colleagues reported improvement in in vivo cell membrane trafficking of siRNA with a cholesteryl group attached at the 3'-end (Soutschek et al. 2004).

As far as stability is concerned, chemical modifications introduced into internucleotide phosphate bonds, or at the C2' position of the ribose ring, have kindled some hopes for enhanced resistance of siRNA to hydrolysis by cellular nucleases. Encouraging results have been reported concerning replacement of internucleotide phosphates with phosphorothioates (Li et al. 2005) as well as with boranophosphates (Hall et al. 2004). Phosphorothioate-derived oligonucleotides have been reported to stimulate the physical cellular uptake of siRNA in human cells (Overhoff and Sczakiel 2005). The studies performed so far with siRNAs modified at the C2' position included introduction of fluorine (Layzer et al. 2004; Dowler et al. 2006), LNA units (LNA = locked nucleic acid) (Braasch et al. 2003; Elmen et al. 2005), as well as 2'-O-alkyl modifications (2'-O-Me, 2'-O-MOE) (Prakash et al. 2005). Recently 4'-thioribose appeared to be another interesting example of a stabilizing modification (Hoshika et al. 2005; Dande et al. 2006).

The last of the above-mentioned problems with the therapeutic application of siRNA is related to the induction of so-called "off target effects," first documented by Jackson and colleagues with cDNA-microarray technology (Jackson et al. 2003). Recent reports (Fedorov et al. 2006; Jackson et al. 2006) provide some evidence that chemical modification of the guide strand of the siRNA duplex can suppress such unintended gene silencing.

Base-modified siRNAs have been investigated so far to a very limited extent (Parrish et al. 2000; Chiu and Rana 2003). The paper by Parrish et al. (2000) describes induction of RNAi in C. elegans by dsRNAs containing 4-thiouridine, 5-bromo-, 5-iodo-, 5-(3-aminoallyl)-uridine, or inosine. While these experiments did not demonstrate any remarkable differences in silencing activity regardless of the introduced modifications, the paper by Chiu and Rana (2003) describes reduced silencing activity of duplexes containing 5-bromouridine, 5-iodouridine, 2,6-diaminopurine, and N-3-methyl-uridine. The patent literature also claims several base-modified nucleosides as components of small interfering nucleic acids, albeit no details are provided with regard to properties of such modified siRNAs (Leake et al. 2004).

In these studies we evaluated the thermodynamic stability and gene-silencing activity of base-modified siRNAs containing three naturally occurring modified nucleosides: 2-thiouridine (s²U), pseudouridine (), and dihydrouridine (D). As has been already demonstrated, the first two nucleosides, due to the preferred C3'-endo conformation of the ribose ring (Agris et al. 1992; Davis et al. 1998), improve the thermodynamic stability of the s²U- (Kumar and Davis 1997; Luyten and Herdewijn 1998; Testa et al. 1999; Shohda et al. 2000) and the -containing double-stranded RNA helices (Davis et al. 1998). Moreover, the s²U unit, when present in the anticodon sequence of transfer RNA, enhances its specificity due to preferred base pairing with A and restricted wobble base pairing with G (Agris et al. 1992). Introduction of the s²U unit into the siRNA structure also seems to be profitable due to increased hydrophobicity of sulfur-modified molecules which might exhibit elevated cellular uptake (Cummins et al. 1995). The dihydrouridine unit destabilizes the RNA structure as it prefers the C2'-endo ribose ring conformation and induces the B-DNA-type sugar ring conformation (Stuart et al. 1996). In addition, the D nucleoside contains a nonaromatic base ring that excludes stabilizing base stacking with neighboring nucleotides within the RNA strand (Westhof et al. 1988; Davis 1998). In these studies modified nucleosides were introduced into functionally important regions of the siRNA duplex, e.g., at position 10 of the antisense strand, which is opposite the cleaved internucleotide bond of a target mRNA sequence (Elbashir et al. 2001c), or at the 3'-ends of the sense and antisense strands (Khvorova et al. 2003; Schwarz et al. 2003). The results presented here throw some light on the area of siRNA structure–activity relationships and provide a new chemical tool for modulation of siRNA-silencing potency.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTAL DATA ACKNOWLEDGMENTS REFERENCES

 
Dual fluorescence reporter system
For silencing activity studies we validated the quantitative reporter system established by Chiu and Rana, based on measurement of the relative fluorescence intensity of the enhanced green fluorescent protein (EGFP) and red fluorescent protein (RFP), expressed in HeLa cells from the pEGFP-C1 and pDsRed1-N1 plasmids, respectively (Chiu and Rana 2002, 2003). Our modifications of the model system included optimization of transfection conditions with commercially available pDsRed2-N1 plasmid or pBACE1–GFP fusion plasmid, provided by Dr Weihong Song, The University of British Columbia, Vancouver, Canada, and a 96-well plate format, compatible with the Synergy HT plate reader (details in Materials and Methods). Correlation between fluorescence intensity and the level of mRNA of fluorescent proteins was confirmed by RT-PCR (data included in Supplemental Material).

It was suggested by Layzer et al. (2004) that transient transfection performed in the Chiu and Rana system may lead to some false positive results of siRNA activity. To eliminate the possible influence of the decay of EGFP gene expression in the time course of the experiment, we performed a time-dependent experiment (HeLa cells cotransfected with pEGFP-C1 and pDsRed2-N1 plasmids) and observed only negligible decrease of the fluorescence intensity over 228 h after transfection (data included in Supplemental Material). Therefore, we conclude that the decrease of the fluorescence level, observed in our 36-h assay, results mostly from the siRNA activity.

Selection of model siRNA for chemical modification
The sequence of siRNA duplex G1, selected by Chiu and Rana (2002), was not suitable for our studies due to the lack of uridine units in positions selected for modifications (at position 19 of the sense strand and at positions 10 and 19 of the antisense strand). Therefore, we selected the G2 siRNA sequence shifted 7 nt upstream from that of G1 (Fig. 1A) and determined the silencing activity in a validated dual fluorescence assay (Materials and Methods). The G2 duplex induced remarkable GFP gene silencing, even at concentrations as low as 1 nM, while the activity of G1 under the same conditions was only moderate (Fig. 1B). The dramatic difference in the silencing potency between G1 and G2 is not surprising in the light of previous reports that a shift of the target sequence by a few nucleotides can strongly affect the siRNA duplex activity (Holen et al. 2002; Harborth et al. 2003). Both siRNA duplexes exhibited noticeable cytotoxicity (MTT assay) at the highest concentrations tested (50 nM), but only a negligible effect was observed at low concentrations (1–5 nM). The cytotoxic "off-target" effects were concentration-dependent and could be diminished by using siRNA in as low as 1 nM concentrations (Persengiev et al. 2004).

View larger version (29K): [in this window] [in a new window]   FIGURE 1. Silencing activity of siRNA duplexes directed toward EGFP mRNA. (A) Sequences of two siRNA molecules, indicated as G1 and G2 (common region is underlined). (B) Silencing activity of G1 and G2 used in 1–50 nM concentrations. (C) HeLa cells viability 36 h after transfection with siRNAs in concentrations given above. The means ± standard deviation is given.

View larger version (7K): [in this window] [in a new window]   FIGURE 2. Structures of naturally occurring rare nucleosides used for modification of siRNA duplexes. R = ribose residue.

We also prepared a series of G2 siRNA molecules possessing 17 or 15 double-stranded tracts—each strand with two thymidine-nucleotide overhangs (duplexes 17 and 18, respectively). In modified versions of the duplexes 16, 17, and 18, all uridine units in the antisense strands were substituted with the 2-thio-analog yielding siRNA molecules containing 7, 6, and 5 s²U units (Table 1, duplexes 19–21, respectively). We assumed that such s²U-modified siRNA duplexes would exhibit higher specificity and enhanced affinity toward the target sequence.

With the aim of verifying the results obtained for variants of the G2 duplex we screened two other sets of modified duplexes originating from two distinct siRNA molecules—siRNA B (Table 2, duplexes 22–24) and siRNA SYM (Table 2, duplexes 25–33).

Some thermodynamic stabilization was observed for the duplexes modified with the s²U and units in the central position (duplexes 9 and 8, respectively), but only in the increase of the Tm (>2°C for s²U and 1°C for ), while G values were virtually the same as for the G2 counterpart. Such a negligible effect may be explained by the presence of two G:C base pairs at each side of the modification site, limiting the influence of a single-base modification. The stabilizing effect of an s²U unit on the siRNA duplex is well illustrated in the case of duplex 11, in which the thio-modification is situated next to the wobble base pair. In this case Tm and G values exhibit considerable differences in comparison to corresponding values for parent duplex 10 (Tm = 3°C and G° = 1.1 kcal/mol).

Modifications introduced at the ends of the duplexes (at positions 19 of the sense or antisense strands) have very little effect on duplex stability (Tm ± 1°C, G° values for duplexes 1, 2, 4–6, 12, and 13), with only one exception of G° = –3.43 kcal/mol for the duplex ss²U19/aG2 (3). We cannot rationally explain this result in the light of reported data that modifications introduced at the ends of a duplex should have relatively little influence on its overall stability (Nawrot et al. 2004).

Duplexes sG2/as2Uall 19 bp, sG2/as2Uall 17 bp, and sG2/as2Uall 15 bp (19–21) with antisense strands fully modified with s²U were thermally very stable (no transitions completed below 96°C). We assume that such exceptional thermodynamic stability results from increased affinity of s²U-modified oligomers for its complementary strands, originating from the preferential C3'-endo sugar ring puckering (Scheit and Faerber 1975; Agris et al. 1992; Kumar and Davis 1997; Testa et al. 1999). Similarly, the same structural feature makes LNA-modified siRNA duplexes thermodynamically very stable (Elmen et al. 2005). Melting temperatures for duplexes 19–24 could be determined in buffer with no magnesium cations (0.1 mM EDTA, 10 mM Tris-HCl, 100 mM NaCl) and with greater temperature gradient per minute (see Supplemental Table S1 for thermodynamic data of duplexes 1–21 in these conditions).

Circular dichroism spectra of modified siRNA duplexes
Circular dichroism (CD) spectra (Fig. 3) were measured for six selected siRNA duplexes that exhibited the biggest differences in thermodynamic stability or appreciable silencing activity in comparison to the reference G2 duplex. The CD spectrum of the G2 duplex (16) indicates a typical A-shaped structure of the double-stranded RNA with the maximum of the positive Cotton effect at 268 nm and crossover points at 245 and 290 nm. The most significant differences from this spectrum are seen in the CD spectrum of the sG2/aD10 duplex (7) (a shift of the crossover point from 245 to 255 nm and less negative band at 210 nm). The differences observed suggest that the structure of siRNA duplex 7 is distorted from the A-type to the B-type helix typical for double-stranded DNA. Such a change was expected since, as mentioned above, a D nucleoside destabilizes the RNA structure (Dalluge et al. 1996; Stuart et al. 1996). Therefore, its insertion in the central part of the siRNA duplex has significant structural consequences.

View larger version (17K): [in this window] [in a new window]   FIGURE 3. Circular dichroism (CD) spectra of duplex G2 19 bp (16 ——) and siRNA duplexes modified at the central domain changing the duplex structure sG2/aD10 (7 ·········), sG2/aW9 (10 – – –), and sG2/aW9s2U10 (11 –··–··–). Reagents and conditions: 10 mM Tris-HCl, 100 mM NaCl, and 10 mM MgCl2 buffer (pH 7.4), Temp. 25°C, duplexes concentration = 1 µM. Sequences of the corresponding duplexes are given in Table 1.

Cytotoxicity of the base-modified siRNA duplexes
Because chemically modified siRNA duplexes may induce severe cytotoxic effects (Amarzguioui et al. 2003; Czauderna et al. 2003), we checked the influence of duplexes 1–21 on cell viability using the MTT test (Supplemental Fig. S6). We assumed that the cytotoxicity of the siRNA duplexes containing the naturally occurring s²U, , and D units would be similar to that of the parent duplexes. The toxicity of the duplexes 1–21 used at the concentration of 1 nM in HeLa cells was negligible. Fully modified 19-bp duplex sG2/as2Uall (19) showed the biggest, but still minor, effect on cell viability (6% reduction), as compared to the control (cells transfected with plasmids only). The modified series of siRNAs B (duplexes 22–24) and siRNA SYM (duplexes 25–33) did not show any notable cytotoxicity in concentrations used in the study (data not shown).

Effect of base modification on gene silencing activity of siRNA
Modification of duplex termini
We expected that or s²U units introduced at position 19 of the sense strand would "close" the 5'-end of the corresponding duplexes (Davis et al. 1998; Testa et al. 1999), and thus would reduce their silencing activity. Indeed, data presented in Figure 4 show that the siRNAs modified within the s²U and units in position 19 of the sense strand, as in duplexes s19/aG2 (2) and ss2U19/aG2 (3), have lower silencing potency (31% and 24% of EGFP expression in control cells, respectively) compared to the activity of the reference duplex G2 (16), which lowered EGFP expression to 11% of the control value.

View larger version (42K): [in this window] [in a new window]   FIGURE 4. Gene-silencing activity of siRNAs modified with s2U, , or D units at central and terminal domains of G2 duplex. SiRNA activity was tested in a dual fluorescence assay, as described in Materials and Methods. HeLa cells were transfected with siRNA duplexes in 1 nM concentration. A positive control: cells transfected with pDsRed2-N1 and pEGFP-C1 plasmids only. Percentage of EGFP protein expression was normalized to RFP protein level. Error bars correspond to standard deviations of results of at least three independent transfection experiments (±SD). siRNA symbols and abbreviations according to Table 1.

The dihydrouridine unit, incorporated at the 3'-end of the siRNA sense strand, should result in the "opening" of this terminus, and thus have a positive influence on the silencing activity. In contrast, its introduction on the 3'-end of the guide strand might cause the opposite effect. Unexpectedly, the sD19/aG2 duplex (1), containing a D unit at its 5'-end (in respect to the guide strand), was slightly less active (16% of control EGFP) as compared to the reference G2 duplex. In agreement with our expectations a threefold suppression of the silencing potency was observed for the sG2/aD19 duplex (4). Similar effects were reported for the siRNA molecules containing single chemical modifications or mismatches, including wobble base pairs, at the 3'- and/or 5'-terminal positions of the duplexes (Chiu and Rana 2003; Schwarz et al. 2003; Holen et al. 2005; Dande et al. 2006; Dowler et al. 2006).

Modification of position 10 of the antisense strand
We intended to check the impact of base modifications in the central part of the duplex on its activity, due to the direct involvement of this domain in RISC-associated cleavage of the target mRNA (Elbashir et al. 2001c). We assumed that s²U and nucleosides would strengthen antisense siRNA/mRNA interactions and, possibly result in higher potency of such modified constructs. However, only the sG2/as2U10 duplex (9) exhibited tolerance for s²U modification, showing similar activity (12% of control EGFP) to the parent G2 duplex (Fig. 4). This observation is interesting in light of earlier reported data about modifications of the central part of the antisense strand of siRNA duplex not being well tolerated (Hamada et al. 2002; Saxena et al. 2003; Du et al. 2005; Holen et al. 2005). The remaining two duplexes sG2/aD10 (7) and sG2/a10 (8) exhibited three- to fourfold suppressed silencing activity. Interestingly, noticeable enhancement of activity (> 50%) was achieved by introduction of the s²U unit at the 3'-adjacent position of the wobble base pair in the sG2/aW9 (10) duplex (see sG2/aW9s2U10 duplex, 11). It is worth noticing that observed loss and subsequent recovery of silencing activity of siRNAs modified in central positions (duplexes 10 and 11, respectively) correlate with changes in their thermodynamic stability (Table 3) and helical structure (Fig. 3, CD spectra).

We decided to verify this last intriguing result by screening the activity of siRNA B and its modified variants (Table 2, entries 22–24). Duplex B was designed toward mRNA of human and mouse BACE1 protein and thus expression of fusion plasmid pBACE–GFP was monitored in a dual fluorescence assay. Nonmodified duplex B in 5 nM concentration decreased the level of BACE–GFP expression down to 30% of the control. Its variant mutated in position 8 (CU) of the antisense strand (duplex BaW8, 23) was half as active. Introduction of the s²U unit at the 3'-adjacent position to the G–U base pair gave duplex 24 (B aW8s2U9), exhibiting silencing activity comparable with that of the nonmodified duplex 22 (Fig. 5). Therefore, we confirmed that introduction of the s²U unit next to the wobble base pair leads to the recovery of most of the siRNA duplex silencing activity. The observed effect probably originates from the improved A-type helical structure of the s²U-containing RNA duplex.

View larger version (24K): [in this window] [in a new window]   FIGURE 5. Gene-silencing activity of siRNA B modified with wobble and/or s2U unit at central domain. SiRNA activity was tested in a dual fluorescence assay, as described in Materials and Methods. HeLa cells were transfected with siRNA duplexes in 5 nM concentration. A positive control: cells transfected with pDsRed2-N1 and pBACE–GFP plasmids only. Percentage of BACE–GFP protein expression was normalized to RFP protein level. Error bars correspond to standard deviations of results of at least three independent transfection experiments (±SD). siRNA symbols and abbreviations according to Table 2.

These promising data were verified by structure–activity relationship studies of the siRNA duplex SYM (25), containing at each terminus the same signature of Watson–Crick hydrogen bonds (three A–U bp, followed by one G–C bp and one A–U bp) (Table 2). We assumed that such a duplex, with similar thermodynamic stability of both ends, should be an accurate model to study asymmetry inducing modifications. All modified duplexes, screened in this test, containing a D nucleoside at their 5'-end and s²U unit(s) at their 3'-end with respect to the guide strand (Table 2, sequences 26–33), showed 25%–50% higher BACE1 silencing activity in comparison with the unmodified symmetrical duplex 25 (Fig. 6). Introduction of a single dihydrouridine unit at the 3'-end of the sense strand (as in 27) induced stronger enhancement of the silencing activity (40% more than nonmodified 25) than a single or double 2-thiouridine at the 3'-end of the antisense strand. Duplexes 26 and 28 were 25% more active than reference 25. Simultaneous introduction of s²U and D units in the same sense strand caused the strongest of the observed improvements (duplex 29 was 50% more active than the parent one, 25). Introduction of further s²U units at the 3'-end of the duplex, as in duplexes 31–33, did not cause any further increase of the silencing activity (Fig. 6).

View larger version (33K): [in this window] [in a new window]   FIGURE 6. Gene-silencing activity of siRNA SYM modified with D and/or s2U units at terminal domains. SiRNA activity was tested in a dual fluorescence assay, as described in Materials and Methods. HeLa cells were transfected with siRNA duplexes in 1 nM concentration. A positive control: cells transfected with pDsRed2-N1 and pBACE–GFP plasmids only. Percentage of BACE–GFP protein expression was normalized to RFP protein level. Error bars correspond to standard deviations of results of at least three independent transfection experiments (±SD). siRNA symbols and abbreviations according to Table 2.

Duplexes of various lengths with antisense strands fully modified with s²U
Taking into account the improved affinity of s²U-modified strands for complementary templates and, consequently, their enhanced target specificity, we also prepared a series of siRNA duplexes with 19, 17, or 15 bp (16–18) and their modified counterparts (19–21), containing antisense strands fully modified with seven, six, or five s²U units, respectively. As shown by hybridization studies, duplexes aG2/as2Uall 19 bp (19), aG2/as2Uall 17 bp (20), and aG2/as2Uall 15 bp (21) exhibit extremely high thermodynamic stability (the relevant melting profiles did not reach a plateau up to 96°C). We asked the question whether such stable duplexes can be effectively unwound by the dsRNA helicase engaged in the RISC complex (Chiu et al. 2005). It has already been reported that thermodynamically stable duplexes containing LNA-modified units are active in RNAi-induced gene silencing (Braasch et al. 2003; Elmen et al. 2005), although their ethylene bridged nucleic acid analogs completely lost their silencing potency (Hamada et al. 2002). Martinez and Tuschl (2004) proved that a 15-nt sequence of the antisense strand is required for effective cleavage catalyzed by the RISC complex. On the basis of these data we assumed that shortened s²U-modified siRNA duplexes might still exhibit remarkable silencing activity.

Indeed, as we demonstrate here, the duplexes 19, 20, and 21 are as active in EGFP silencing as the parent G2 duplexes of the same length (Fig. 7). Trimming of the siRNA duplexes 16 and 19 on their 3'-ends (with respect to the guide strands) by two nucleotides caused a twofold decrease in the silencing activity. Further shortening of the length of the duplexes decreased their efficiency dramatically, for both the unmodified (G2 15 bp, 18) and s²U-modified (sG2/as2Uall 15 bp, 21) cases. Interestingly, the shortest modified duplex 21 was slightly more active (79% of control EGFP) than its counterpart (92% of control EGFP), which might originate from its enhanced binding affinity to the target mRNA.

View larger version (23K): [in this window] [in a new window]   FIGURE 7. Silencing activity of siRNA duplexes G2 varying in length: 19 bp (16), 17 bp (17), 15 bp (18), and their modified analogs containing, in the antisense strand, seven (sG2/as2Uall 19 bp, 19), six (sG2/as2Uall 17 bp, 20), or five (sG2/as2Uall 15 bp, 21) s2U units, respectively. Silencing activity of siRNA duplexes 19–21 in HeLa cells was determined in a dual fluorescence assay, with siRNAs used at 20 nM concentration. Error bars correspond to the standard deviations of results of at least three independent transfection experiments (±SD).

View larger version (22K): [in this window] [in a new window]   FIGURE 8. The differences between a free Gibbs' energy (Gas 37°C) calculated for the 3'- and 5'-ends of different length in the siRNA duplexes G2 (A) and SYM (B). The duplex polarity is assigned according to the antisense strand polarity.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTAL DATA ACKNOWLEDGMENTS REFERENCES

In our studies, we observed that, within a set of s²U, , and D-modified duplexes G2 (duplexes 1–15), the most active were duplexes with a s²U unit incorporated at the 3'-end of the guide strand, especially in combination with the sense strand modified with a dihydrouridine unit in position 19. This observation was confirmed with the SYM duplex (25), possessing five subsequent Watson–Crick base pairs at both duplex ends of the same pattern (the same order of U–A and C–G base pairs). Thermodynamic symmetry of this model duplex was confirmed by the rather low value of the G^as _37°C parameters (Fig. 8).

In the modified SYM duplexes 25–33, the presence of the s²U unit(s) at the 3'-end and/or that of the dihydrouridine unit at the 5'-end resulted in a remarkable increase of siRNA duplex activity (25%–50%). Therefore, we assume that such an "opening" of the 5'-end of the siRNA by the D modification and its "closing" at the 3'-end by the s²U unit(s) are advantageous for duplexes containing symmetrical sequences at their termini (with respect to base-pairing interactions) and exhibiting moderate or low silencing activity. One can imagine the usefulness of such modifications for duplexes of moderate or low activity, designed for fixed sequences, e.g., SNP-detecting siRNAs. Furthermore, our calculations indicate that siRNA duplex asymmetry is defined mostly by the relative thermodynamic stability of its three terminal base pairs. This conclusion is based on the observation that, for the highly active G2 duplex, the calculated G^as _37°C values are meaningful when not more than three base pairs are taken into account. These data may help in the rational design (including chemical modification) of siRNA duplexes.

Modifications at the duplex termini have an impact on siRNA activity probably not only because of induction of duplex asymmetry but as a result of other effects, which should be taken into account, including interactions between guide strand and complementary mRNA as well as between RNA and proteins of RNAi machinery. For example, the expected lower potency of the sG2/aD19 duplex 4, containing a dihydrouridine unit in position 19 of the antisense strand, may result from weakened binding of this modified strand to the target mRNA. The dihydrouridine unit changes the ribose ring conformation, and therefore may also influence binding of the modified 3'-end of the antisense strand to the PAZ domain of the Argonaute protein (Lingel et al. 2003; Yan et al. 2003; Ma et al. 2004).

We did not modify the 5'-end of the guide strand, because previous studies reported that modification of this end of the duplex impairs siRNA-silencing activity (Chiu and Rana 2003). The recent structural studies by Ma et al. (2005) provide an explanation for the functional importance of the 5'-end of the antisense strand, as it participates in nucleation with messenger RNA. In this process the PIWI domain of the Argonaute protein binds to the siRNA antisense strand via interactions between the phosphate groups of the RNA chain and the amino acid residues of the protein. In light of these new data we suppose that modifications within the nucleobases at the 5'-end of the antisense strand should not alter interactions with RISC. It was recently shown by Jackson et al. (2006) and Fedorov et al. (2006) that subtle chemical modifications can limit siRNA off-target effects. These studies demonstrate that a single 2'-O-Me modification introduced at position 2 of the antisense strand does not interfere with duplex potency and, importantly, limits unintended "off-target" silencing. According to the authors a possible explanation for the observed effect is the alteration of the binding of the guide strand to the PIWI domain of the RISC by the 2'-O-methyl group. In our opinion it is worth noticing that the 2'-O-Me modification also decreases the free energy of hybridization (due to the C3'-endo conformation) and thus should enhance the binding affinity to mRNA within the seed region. Our s²U modification exhibits similar hybridization features, however, should not cause any steric hindrance when bound to the RISC. Therefore, using the s²U unit might be valuable in further studies on the mechanism of limiting nonintended silencing by modification at position 2 of the guide strand.

In addition, sulfur in position 2 of the nucleobase ring enhances favorable base stacking, causing further enhancement of affinity for the target sequence (Cummins et al. 1995). Another feature of the s²U modification is its restricted wobble base pairing, resulting in improved specificity for the complementary template (Agris et al. 1992). This feature might be beneficial for intended target recognition by limiting the affinity for "off-target" transcripts. Therefore, in our opinion, further studies on duplexes containing an s²U modification at the 5'-ends of their antisense strands are worthwhile.

We intended to check also the impact on activity of base modifications positioned in the central part of the siRNA duplex, due to the direct involvement of this domain in RISC-associated cleavage of the target mRNA (Elbashir et al. 2001c). The structure of this region of siRNA is also crucial, as demonstrated recently, due to the cleavage of the sense strand of siRNA before its dissociation from the duplex (Matranga et al. 2005; Rand et al. 2005). Our results show that 2-thiouridine modifications within the central part of the antisense strand, which improve duplex stability, do not interfere with either the helical structure or the silencing potency. However, it is not clear why pseudouridine at position 10 of the guide strand led to reduced activity of the sG2/a10 construct, when thermodynamic and CD data suggested that the structure of this duplex is not distorted (see Table 3 and Supplemental Fig. S5). In contrast, dihydrouridine placed at position 10 or the CU mutation at position 9 as in G2 relative duplex 10 or at position 8 as in duplex 23 of the antisense strand (leading to wobble base pairing with both the sense strand of siRNA and a target mRNA sequence) caused much lower silencing activity. The presence of the D or wobble base pair in the central domain of the guide strand probably significantly disturbs the A-type helical structure of the guide/mRNA complex. Furthermore, wobble base pairing at the central position of the siRNA duplex may also disturb the efficiency of the cleavage of the sense strand and, in this way, disturbs RISC assembly (Matranga et al. 2005; Rand et al. 2005). Similar observations of reduction of the silencing activity of siRNA modified with wobble base pairs in the center of the duplex were reported at the time our project was underway (Holen et al. 2005). Interestingly, the noticeable recovery of activity of both wobble-mutated duplexes 10 and 23 could be achieved by introduction of the s²U unit at the 3'-adjacent position of the wobble base pair. Improvement of silencing activity of the duplex 11 correlates with its enhanced thermodynamic stability (Table 3) and with the minimal disruption of its structural integrity, as it could be concluded from the corresponding CD spectrum (Fig. 3). In our opinion this observation strongly supports the idea that the conformational changes induced by the s²U nucleoside leading to A-type helical structure are advantageous for the siRNA duplex silencing activity.

It is interesting to notice that the unimpaired silencing activity of duplexes 19–21, which contain antisense strands fully modified with s²U units and which are extremely stable to thermodynamic denaturation, suggests that the proteins of the RISC complex possess an unusual ability to facilitate unwinding and dissociation of such thermodynamically very stable siRNA duplexes.

The usefulness of similar 2-thiosubstituted nucleobase modifications as candidates for application in RNAi was recently mentioned by the Egli group (Diop-Frimpong et al. 2005). Their structural studies show that a DNA duplex modified with a 2'-O-[2(methoxy)ethyl]-2-thiothymidine unit exhibits favorable features for base pairing and induces only small structural perturbations within the A-form double-stranded helix. Therefore, combinations of chemical modification involving sulfur in position 2 of the nucleobase and 2'-modification at the ribose moiety are potential objects for future studies on chemically modified siRNA duplexes.

In conclusion, we have demonstrated that the naturally occurring modified nucleosides, 2-thiouridine, pseudouridine, and dihydrouridine, when introduced into an siRNA duplex, modulate its silencing potency. The extent of this effect depends on the modification position and is most advantageous when the s²U, , and D nucleosides participate in an enhancement of the siRNA duplex asymmetry. Therefore, the modified nucleosides described here may be considered as useful units for improving siRNA activity and specificity.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTAL DATA ACKNOWLEDGMENTS REFERENCES

 
Preparation of siRNAs
Oligoribonucleotides were prepared according to the routine phosphoramidite approach (Caruthers 1985) using standard LCA CPG glass support and commercially available nucleoside phosphoramidites (Glen Research). Phosphoramidites of suitably protected modified units (s²U, , D) were prepared as described previously (Guenther et al. 1994; Agris et al. 1995). Oligonucleotide synthesis was performed on a 1 µM scale on an Applied Biosystems 394 instrument under conditions recommended by the synthesizer supplier, except for the synthesis of s²U-containing oligomers. In these cases the oxidation step was performed with 1 M tBuOOH (Fluka) in acetonitrile/methylene chloride (95/5) for 6 min (Kumar and Davis 1995; Sochacka 2001). Oligonucleotides were cleaved from the solid support as 5'-DMT-derivatives, and then deprotected and purified according to the procedure described elsewhere (McSwinggen and Beigelman 2003). Briefly, support-bound oligonucleotides were treated with 33% ethanolic methylamine and DMSO, 1:1 mixture (v:v) for 15 min at 65°C and then with TEA x 3HF for 15 min at 65°C. The reaction mixture was frozen for 30 min at – 20°C, quenched with cold 1.5 M ammonium bicarbonate and poured into a conditioned SepPak (Waters) cartridge. Shorter oligomers were eluted with 14% CH₃CN in 50 mM NaOAc and 50 mM NaCl. The remaining oligomer was treated with 2% aqueous TFA for 15 min at room temperature and washed with water, 1 M NaCl, and water. The product was eluted from the cartridge with 30% aqueous solution (aq.) CH₃CN, and after partial solvent removal the RNA aqueous solution was frozen and kept at – 20°C. The concentration of oligomers was determined spectrophotometrically by UV absorbance at _max in water using the extinction coefficients calculated according to the method published earlier (Brown and Brown 1991). The structure of the oligonucleotides was confirmed by MALDI-TOF mass spectrometry (data given in Table 1) and their purity was assessed by PAGE (20% acrylamide/7 M urea, ³²P-labeled).

Assembly of siRNA duplexes was done in water by mixing equimolar amounts of complementary sense and antisense oligonucleotides (final duplex concentration was 20 µM), heating the mixture at 95°C for 2 min, and slow (3 h) cooling down to room temperature. Purity of the resulting duplexes was confirmed by PAGE (20% polyacrylamide, nondenaturing conditions, ³²P-labeled), and formation of duplexes was monitored by 4% agarose electrophoresis.

Melting profiles and thermodynamic calculations
All absorption measurements were carried out in a 1-cm path length cell with a UV/Vis 916 spectrophotometer equipped with a Peltier Thermocell (GBC). Complementary oligonucleotides were mixed in 10 mM Tris-HCl, 100 mM NaCl, 10 mM MgCl₂ buffer (pH 7.4) at final concentration of duplexes of 1 µM, heated at 96°C, and strands association was achieved by cooling down to 5°C, with a temperature gradient of 0.4°C/min. Melting profiles were recorded while heating from 5°C to 96°C with a temperature gradient of 0.2°C/min. The melting temperatures were calculated by the first-order derivative method. Thermodynamic parameters were obtained by fitting of the recorded curves (MeltWin software, version 3.5). Each set of experimental data was fitted twice.

CD spectra recording
CD spectra were recorded on a CD6 dichrograph (Jobin-Yvon) at 25°C in the same buffer as in melting experiments at a duplex concentration of 1 µM using a 5-mm path length cell, 2-nm bandwidth, and 1–2-sec integration time. Each spectrum was smoothed with a 25-point algorithm (included in the manufacturer's software, version 2.2) after averaging of at least three scans.

Dual fluorescence model
Cell cultures and transfections
HeLa (human cervix carcinoma) cells were cultured in RPMI 1640 medium (Gibco; BRL) supplemented with 10% heat-inactivated FBS (Gibco, BRL, Paisley) and with antibiotics (penicillin 100 U/mL, streptomycin 100 µg/mL, Polfa) at 37°C and 5% CO₂. Twenty-four hours prior to the experiment, cells were plated in 96-well black wall plates with a transparent bottom (Perkin-Elmer) at a density of 15,000 cells per well. One hour before treatment, the cell medium was replaced with one free of antibiotics (100 µL/well). The cells were cotransfected with plasmid DNA (pDsRed2-N1, 15 ng/well, and pEGFP-C1, 30 ng/well, BD Biosciences or p-BACE–GFP; Qing et al. 2004), 70 ng/well, provided by Dr Weihong Song, The University of British Columbia, Vancouver, Canada, and siRNA (1–50 nM final concentrations) dissolved in OPTI-MEM medium (50 µL/well, Gibco) and complexed with transfection reagent (Lipofectamine 2000, Invitrogen) in the proportion 1 µL of lipofectamine per 1 µg of nucleic acid. The cells were incubated in the transfection mixture for 5 h and then the mixture was replaced with fresh, culturing medium with antibiotics. After 36-h incubation at 37°C in atmosphere of 5% CO₂ cells were washed three times with phosphate saline buffer (PBS) and lysed overnight with NP-40 buffer (150 mM NaCl, 1% IGEPAL, 50 mM Tris-HCl [pH 7.0], 1 mM PMSF) at 37°C. Cell lysates were used for fluorescence determination.

Quantification of siRNA activity
Fluorescence values of enhanced green fluorescent protein (EGFP) and red fluorescent protein (RFP) fluorophores were measured in cell-culturing plates using Synergy HT (BIO-TEK) reader, and data quantification was done with KC4 software. Excitation and emission wavelengths for both fluorescent proteins were as follows (the bandwidth is given after the slash): _Ex = 485/20 nm and _Em = 528/20 nm for GFP and _Ex = 530/25 nm and _Em = 590/30 nm for RFP. The siRNA activity was calculated as a ratio of GFP to RFP fluorescence values. Each transfection experiment was performed in eight wells. A mean value of fluorescence was calculated after discarding the most extreme values. The mean value of background fluorescence (CL, cells treated with lipofectamine only) was subtracted from the mean sample fluorescence value. The level of GFP fluorescence of control transfected cells (treated with pDsRed2-N1 and the corresponding pEGFP-C1 or p-BACE–GFP plasmids only, control) was taken as a reference (100%). Each siRNA activity value given on the plots is an average of mean values from at least three independent experiments and the standard deviations are given as error bars.

Cytotoxicity
The cytotoxicity of siRNA duplexes for HeLa cells was measured using the MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; Sigma) assay (activity of the mitochondrial respiratory chain) (Hansen et al. 1989). Cells were plated and transfected as described above (siRNAs at 1 nM concentration). As a background control, the cells were treated with lipofectamine only. After 36 h of incubation at 37°C and 5% CO₂, MTT solution (5 mg/mL) in PBS was added to each well and incubated for additional 2 h at 37°C. Finally, 95 µL of lysis buffer (20% SDS, 50% aqueous dimethylformamide, pH 4.5) was added to each well and incubated overnight at 37°C. Absorbance of a given sample was measured at 570 nm, with the reference wavelength 630 nm (plate reader Synergy HT, BIO-TEK). The percentage of living cells (P_LC) was calculated from the equation: P_LC = (A_Spl – A_CL)/(A_Mock – A_CL) x 100% where A_Spl is the absorbance of a given sample of cells treated with siRNA, A_CL is the background absorbance (CL control), A_Mock is the absorbance of the cells untreated with siRNAs (control transfected cells). Data points represent means of at least four measurements.

	SUPPLEMENTAL DATA

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION MATERIALS AND METHODS SUPPLEMENTAL DATA ACKNOWLEDGMENTS REFERENCES

 
The following Supplemental Material is available upon request from the corresponding author (e-mail: bnawrot@bio.cbmm.lodz.pl): correlation of the fluorescence intensity and the level of mRNA of EGFP and RFP; time course of the expression of pEGFP-C1 plasmid in HeLa cells; MALDI-TOF MS of oligonucleotides sD19, as2U19, a10, and as2Uall 17 bp; CD spectra of duplexes 8, 9, and 13; cytotoxicity of siRNA duplexes 1–21; thermal stability of siRNA duplexes 1–21 in buffer with no magnesium ions added.

	ACKNOWLEDGMENTS