- 1.1 Introduction
- 1.2 LNA – a Primer
- 1.3 Fluorophore-functionalised LNA – an Overview
- 1.3.1 Introduction
- 1.3.2 N2′-Functionalised 2′-amino-LNA
- 1.3.3 N2′-Functionalised 2′-amino-α-L-LNA
- 1.3.4 C5-Functionalised LNA
- 1.3.5 LNA with Fluorescent Nucleobase Surrogates
- 1.4 Applications of Fluorophore-functionalised LNA
- 1.4.1 Formation of Pyrene Arrays
- 1.4.2 Hybridisation Probes
- 1.4.3 Base Discriminating Fluorescent Probes
- 1.4.4 Excimer-based Probes
- 1.5 Summary and Perspective
CHAPTER 1: Fluorophore-functionalised Locked Nucleic Acids (LNAs)
-
Published:18 Oct 2012
-
P. J. Hrdlicka and M. E. Østergaard, in DNA Conjugates and Sensors, ed. K. R. Fox and T. Brown, The Royal Society of Chemistry, 2012, pp. 1-33.
Download citation file:
Fluorophore-modified oligonucleotides (ONs) are extensively used in mechanistic biological studies, molecular diagnostics, drug research, biotechnology and materials science. In this chapter, we describe the synthesis, properties and applications of ONs modified with various classes of fluorophore-functionalized locked nucleic acid (LNA) monomers, which display photophysical properties that are difficult to mimic with more flexible and synthetically more readily accessible building blocks.
1.1 Introduction
Fluorophore-modified oligonucleotides (ONs) are extensively used in mechanistic biological studies, molecular diagnostics, drug research, biotechnology and materials science.1 Specific applications include their use to monitor the progress of real-time polymerase chain reaction (PCR),2 detect cellular RNA,3 detect single nucleotide polymorphisms (SNPs),4 study RNA-folding,5 monitor enzyme activities,6 and generate self-assembled chromophore arrays.7 These applications have forced chemists to develop fluorophore-functionalised building blocks with emission characteristics that are influenced by factors in their microenvironment.8 For example, hybridisation probes enable detection of nucleic acid targets under conditions where excess probe cannot be washed away, by displaying low fluorescence emission in the absence of target, but prominent emission in the presence of target.9
In this chapter, we will focus on the synthesis, properties and applications of ONs modified with fluorophore-functionalised LNA (locked nucleic acid) monomers, since these materials display photophysical properties that are difficult to mimic with more flexible monomers. As will be discussed in the following sections, these properties are linked to the unique structural characteristics of LNA-type monomers, which offer increased positional control of the fluorophore.
1.2 LNA – a Primer
As part of efforts directed toward developing high-affinity antisense oligonucleotides,10,11 the Wengel12 and Imanishi13 groups independently developed LNA in the late 1990s. LNAs can formally be regarded as conformationally restricted analogues of 2′-O-methyl ribonucleotides in which the methyl group is connected to the 4′-position of the sugar ring (Figure 1.1). The resulting dioxabicyclo-[2.2.1]-heptane skeleton forces the five-membered furanose ring into a C3′-endo conformation, which resembles the conformation that is adopted by ribonucleotides in RNA duplexes (Figure 1.1).14 Incorporation of LNA monomers into ONs gradually tunes the conformation of neighbouring 2′-deoxyribonucleotides from DNA-like C2′-endo conformations toward RNA-like C3′-endo conformations.14 The effect is linked to the limited internal flexibility of LNA nucleotides, which restricts conformational interconversion of neighbouring nucleotides.15 This influences the geometry of LNA-modified duplexes, which display greater RNA character than unmodified reference duplexes.14,16 The four chiral centres in LNA nucleotides give rise to eight possible stereoisomers (the chirality of the 2′- and 4′-positions is interrelated owing to the oxymethylene ring). While six of the eight LNA stereoisomers result in improved RNA affinity relative to unmodified reference strands,17 only α-L-LNA displays hybridisation properties that are comparable to those of LNA (Figure 1.1). Unlike LNA, which is an RNA mimic, α-L-LNA is considered a DNA mimic because duplexes between α-L-LNA-modified DNA strands and complementary DNA/RNA adopt geometries that globally resemble unmodified reference duplexes.18,19
Structures of DNA, RNA LNA and α-L-LNA (upper) and their preferred sugar conformations (lower). Nucleoside numbering of the carbons in the bicyclic ring is shown for LNA.
Structures of DNA, RNA LNA and α-L-LNA (upper) and their preferred sugar conformations (lower). Nucleoside numbering of the carbons in the bicyclic ring is shown for LNA.
Incorporation of LNA monomers into ONs results in significantly improved thermal affinity toward complementary DNA and RNA targets (ΔTm/mod up to +10 °C).20 The stabilising effects of LNA monomers are sequence dependent and either entropy or enthalpy driven, suggesting that preorganisation of the LNA-modified strand or stronger base stacking interactions, respectively, contribute to stabilisation of the duplexes.21 LNA-modified ONs, moreover, display excellent mismatch specificity.22 ONs modified with α-L-LNA monomers have been less systematically studied owing to more limited commercial access, but generally display similar DNA/RNA affinity and mismatch discrimination as conventional LNAs.23 The intriguing biophysical properties of LNA and α-L-LNA have led to the development of numerous LNA analogues.24
LNAs have found widespread use in fundamental research, biotechnology, diagnostics and drug development.25 For example, their ability to stabilise interactions with RNA and provide protection from cellular nucleases has been widely explored in antisense technology.26 Modulation of gene expression, through LNA-mediated targeting of messenger (m)RNA, pre-mRNA or micro (mi)RNA,27–29 has accelerated gene function studies and led to the development of LNA-based antisense drug candidates against diseases of genetic origin.20 Other applications of LNAs include their use as in situ hybridisation probes to monitor spatiotemporal expression patterns of miRNAs30 and as primers to improve allele-selective PCR.31 The readers are directed to other sources for additional background on LNA.25,32
1.3 Fluorophore-functionalised LNA – an Overview
1.3.1 Introduction
Many of the fluorophore-modified oligonucleotides (ONs) used in biotechnology are labelled at the termini through attachment of the fluorophore during solid-phase synthesis, via post-synthetic labelling, or through enzymatic incorporation.33 While these labelling strategies have become largely routine, they often generate fluorescent probes that are insufficiently responsive to changes in their microenvironment and/or hybridisation state for certain applications.8 This is largely due to their inherent flexibility, which leads to poor positional control of the fluorophore (Figure 1.2 – left). To address these limitations, a plethora of fluorophore-modified nucleotide monomers with lower inherent flexibility have been developed, which allow internal labelling of ONs and improved positional control of the fluorophore (Figure 1.2 – centre).1,33 As will be discussed in Section 1.4, ONs modified with such monomers display characteristics that have enabled the development of chromophore arrays and various diagnostic probes (representative monomers are shown in Figure 1.3). Attachment of fluorophores to the conformationally restricted LNA skeleton is poised to result in probes with even greater positional control of the fluorophore (Figure 1.2 – right).
Interplay between monomer flexibility and positional control of fluorophore with different labelling approaches (shown for units with predominantly intercalative or groove binding modes). Solid and dashed lines illustrate primary and alternative binding modes, respectively.
Interplay between monomer flexibility and positional control of fluorophore with different labelling approaches (shown for units with predominantly intercalative or groove binding modes). Solid and dashed lines illustrate primary and alternative binding modes, respectively.
Pyrene-functionalised monomers with intermediate flexibility. Py=pyren-1-yl.
An overview of the major classes of fluorophore-functionalised LNAs is given in the following sections along with a brief discussion of their hybridisation properties and binding modes (Figure 1.4). The discussion of their binding modes relies on indirect structural data, such as absorption and fluorescence spectra and molecular modelling, owing to the absence of nuclear magnetic resonance (NMR) solution or X-ray crystal structures. Fluorophores have been conjugated to LNA-type monomers through: a) attachment to the sugar moiety, b) attachment to the nucleobase moiety, or c) substitution of the nucleobase (Figure 1.4). Pyrene-functionalised LNAs have been studied in particular detail because:
pyrene moieties engage in π-stacking with nucleobases or other pyrene moieties, enabling array formation, intercalation (stacking area: pyrene ∼184 Å vs. A : T base pair ∼221 Å)34 and/or formation of pyrene–pyrene excimers35
pyrene fluorescence is sensitive to the polarity of the microenvironment36 and the nature of neighbouring nucleobases, which quench pyrene fluorescence via photoinduced electron transfer (guanine moieties are typically the strongest quenchers; G>C∼T>A).37
Fluorophore-functionalised LNA monomers. Py=pyren-1-yl, Per=perylen-3-yl, Cor=coronen-1-yl.
Fluorophore-functionalised LNA monomers. Py=pyren-1-yl, Per=perylen-3-yl, Cor=coronen-1-yl.
1.3.2 N2′-Functionalised 2′-amino-LNA
Hybridisation properties. ONs that are modified with 2′-amino-LNA monomers carrying small groups at the N2′-position (e.g. methyl; benzoyl; 2-aminoethyl; amino acids) display similar affinity toward DNA/RNA targets as conventional LNA.38,39 Large N2′-fluorophores such as pyrene, perylene or coronene (Figure 1.4), on the other hand, have different impacts on target affinity, depending on the orientation, steric bulk and linker chemistry of the fluorophore (Table 1.1).38,40–44 For example, monomers in which the fluorophore is connected via a N2′-acyl linker generally result in greater duplex stabilisation than N2′-alkyl linked monomers. Fluorescence emission profiles (see Section 1.4.2) and molecular modelling studies suggest that the N2′-substituents are directed toward the minor groove (Figure 1.5).38,45,46
Representative thermal denaturation temperatures of duplexes between N2′-fluorophore-functionalised 2′-amino-LNAs and complementary DNA or RNA.
. | . | ΔTm (°C) . | ||||||
---|---|---|---|---|---|---|---|---|
. | . | 5′-GTG ABA TGC . | . | 3′-CAC TAB ACG . | ||||
Monomer . | . | vs. DNA . | . | vs. RNA . | . | vs. DNA . | . | vs. RNA . |
Measured at 1 μM concentration of each strand in medium salt buffer ([Na+]=110 mM, [Cl−]=100 mM, pH 7.0 (NaH2PO4/Na2HPO4)). nd=not determined, nt=no transition. For structures of monomers, see Figure 1.4. | ||||||||
LNA-T47,48 | +5.0 | +9.5 | +6.5 | +9.5 | ||||
F45,49 | +3.0 | +5.0 | +1.0 | nd | ||||
G46 | +2.5 | +7.0 | +6.0 | +9.5 | ||||
H44 | +3.0 | +1.5 | +2.0 | +3.0 | ||||
I42 | +3.0 | +6.5 | +7.0 | +7.5 | ||||
J41 | – 8.0 | – 6.5 | – 6.5 | 0.0 | ||||
K41 | +5.5 | +8.0 | +6.0 | +9.0 | ||||
L41 | +2.5 | nt | nt | nt | ||||
M43 | nd | nd | +6.5 | +3.5 | ||||
N43 | nd | nd | +10.0 | +9.5 |
. | . | ΔTm (°C) . | ||||||
---|---|---|---|---|---|---|---|---|
. | . | 5′-GTG ABA TGC . | . | 3′-CAC TAB ACG . | ||||
Monomer . | . | vs. DNA . | . | vs. RNA . | . | vs. DNA . | . | vs. RNA . |
Measured at 1 μM concentration of each strand in medium salt buffer ([Na+]=110 mM, [Cl−]=100 mM, pH 7.0 (NaH2PO4/Na2HPO4)). nd=not determined, nt=no transition. For structures of monomers, see Figure 1.4. | ||||||||
LNA-T47,48 | +5.0 | +9.5 | +6.5 | +9.5 | ||||
F45,49 | +3.0 | +5.0 | +1.0 | nd | ||||
G46 | +2.5 | +7.0 | +6.0 | +9.5 | ||||
H44 | +3.0 | +1.5 | +2.0 | +3.0 | ||||
I42 | +3.0 | +6.5 | +7.0 | +7.5 | ||||
J41 | – 8.0 | – 6.5 | – 6.5 | 0.0 | ||||
K41 | +5.5 | +8.0 | +6.0 | +9.0 | ||||
L41 | +2.5 | nt | nt | nt | ||||
M43 | nd | nd | +6.5 | +3.5 | ||||
N43 | nd | nd | +10.0 | +9.5 |
Position of fluorophores in DNA duplexes modified with N2′-functionalised 2′-amino-LNA. Two representations of the lowest energy structure of the duplex between 5′-d(TTF AFA FAF CAc G) and complementary DNA, where F is 2′-N-(pyren-1-yl)methyl-2′-amino-LNA-T and c is 5-methylcytosin-1-yl LNA (from ref. 45; copyright 2004 Royal Society of Chemistry).
Position of fluorophores in DNA duplexes modified with N2′-functionalised 2′-amino-LNA. Two representations of the lowest energy structure of the duplex between 5′-d(TTF AFA FAF CAc G) and complementary DNA, where F is 2′-N-(pyren-1-yl)methyl-2′-amino-LNA-T and c is 5-methylcytosin-1-yl LNA (from ref. 45; copyright 2004 Royal Society of Chemistry).
Synthesis. The most convenient synthetic route to N2′-functionalised 2′-amino LNA-T phosphoramidites initiates from commercially available diacetone-α-D-allose, which is converted into glycosyl donor 1 via an optimised multistep reaction sequence involving: O3-benzylation, regioselective 5,6-O-isopropylidene cleavage, oxidative cleavage of the resulting vicinal diol, crossed aldol condensation and Cannizzaro reduction, mesylation of the resulting diol, and acetolysis of the remaining isopropylidene group (Scheme 1.1).50 Glycosylation of 1 under Vorbrüggen conditions,51 followed by O2′-deacylation, O2′-mesylation and intramolecular nucleophilic displacement, results in the formation of anhydronucleoside 2.52 Opening under acidic conditions, followed by O2′-triflation of the resulting threo-configured nucleoside and installation of a 2′-azido group with inversion of configuration, affords nucleoside 3. Azide reduction via a Staudinger reaction, and concurrent intramolecular substitution of the 6′-mesylate, affords 2′-amino-LNA derivative 4.52 A series of non-trivial protecting group manipulations converts 4 into partially protected amino alcohol 5,49,52 which is used as a substrate for N2′-functionalisation via reductive amination or chemoselective N-acylation. O3′-phosphitylation then affords phosphoramidites 6, which are used in machine-assisted solid-phase DNA synthesis.38 A variety of fluorophores (e.g. pyrene, perylene, coronene derivatives) have been attached in this manner despite the resource-intensive synthetic route (10–20% yield from diacetone-α-D-allose over ∼20 steps).38,41–44
Outline of synthetic route to N2′-functionalised 2′-amino-LNA-T phosphoramidites: (a) BnBr, NaH, THF; (b) 80% aq. AcOH; (c) NaIO4, THF/H2O; (d) HCHO, aq. NaOH, 1,4-dioxane; (e) MsCl, pyridine/CH2Cl2; (f) c. H2SO4, Ac2O, AcOH; (g) thymine, BSA, TMSOTf, CH3CN; (h) half sat. NH3/MeOH; (i) MsCl, pyridine; (j) DBU, CH3CN; (k) acetone, dil. aq. H2SO4; (l) Tf2O, DMAP, pyridine/CH2Cl2; (m) NaN3, DMF; (n) PMe3, aq. NaOH, THF; (o) NaOBz, DMF; (p) sat. NH3/MeOH; (q) DMTrCl, pyridine/CH2Cl2; (r) HCOONH4, 20% Pd(OH)2/C, EtOAc; (s) N2′-functionalisation (e.g. ArCOOH, HBTU, EtN(i-Pr)2, DMF or ArCHO, NaBH(OAc)3, ClCH2CH2Cl); (t) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2.
Outline of synthetic route to N2′-functionalised 2′-amino-LNA-T phosphoramidites: (a) BnBr, NaH, THF; (b) 80% aq. AcOH; (c) NaIO4, THF/H2O; (d) HCHO, aq. NaOH, 1,4-dioxane; (e) MsCl, pyridine/CH2Cl2; (f) c. H2SO4, Ac2O, AcOH; (g) thymine, BSA, TMSOTf, CH3CN; (h) half sat. NH3/MeOH; (i) MsCl, pyridine; (j) DBU, CH3CN; (k) acetone, dil. aq. H2SO4; (l) Tf2O, DMAP, pyridine/CH2Cl2; (m) NaN3, DMF; (n) PMe3, aq. NaOH, THF; (o) NaOBz, DMF; (p) sat. NH3/MeOH; (q) DMTrCl, pyridine/CH2Cl2; (r) HCOONH4, 20% Pd(OH)2/C, EtOAc; (s) N2′-functionalisation (e.g. ArCOOH, HBTU, EtN(i-Pr)2, DMF or ArCHO, NaBH(OAc)3, ClCH2CH2Cl); (t) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2.
1.3.3 N2′-Functionalised 2′-amino-α-L-LNA
Hybridisation properties. ONs that are modified with 2′-amino-α-L-LNA monomers carrying small non-aromatic units at the N2′-position, such as ethyl or acetyl groups, have detrimental impact on duplex stability (ΔTm down to –17 °C per modification).53 In stark contrast, ONs modified with pyrene-functionalised 2′-amino-α-L-LNA monomers (Figure 1.4) display exceptional thermal affinity toward DNA targets (ΔTm up to +19 °C per modification, Table 1.2), which far exceeds that of conventional α-L-LNA.53 The linker between the fluorophore and sugar skeleton has considerable influence on duplex thermostability; short N2′-acyl linkers are favoured over N2′-alkyl and longer N2′-acyl linkers (Table 1.2).
Representative thermal denaturation temperatures of duplexes between N2′-fluorophore-functionalised 2′-amino-α-L-LNAs and DNA or RNA complements.
. | . | ΔTm (°C) . | ||||||
---|---|---|---|---|---|---|---|---|
. | . | 5′-GTG ABA TGC . | . | 3′-CAC TAB ACG . | ||||
Monomer . | . | vs. DNA . | . | vs. RNA . | . | vs. DNA . | . | vs. RNA . |
See Table 1.1 for experimental conditions. For structures of monomers, see Figure 1.4. Tm values are from ref. 53. | ||||||||
α-L-LNA-T | +6.0 | +8.5 | +8.0 | +10.0 | ||||
W | +14.0 | +5.0 | +15.5 | +7.5 | ||||
X | +19.0 | +10.0 | +19.5 | +11.5 | ||||
Y | +15.5 | +9.5 | +16.5 | +12.0 | ||||
Z | +6.0 | +7.0 | +6.5 | +6.5 |
. | . | ΔTm (°C) . | ||||||
---|---|---|---|---|---|---|---|---|
. | . | 5′-GTG ABA TGC . | . | 3′-CAC TAB ACG . | ||||
Monomer . | . | vs. DNA . | . | vs. RNA . | . | vs. DNA . | . | vs. RNA . |
See Table 1.1 for experimental conditions. For structures of monomers, see Figure 1.4. Tm values are from ref. 53. | ||||||||
α-L-LNA-T | +6.0 | +8.5 | +8.0 | +10.0 | ||||
W | +14.0 | +5.0 | +15.5 | +7.5 | ||||
X | +19.0 | +10.0 | +19.5 | +11.5 | ||||
Y | +15.5 | +9.5 | +16.5 | +12.0 | ||||
Z | +6.0 | +7.0 | +6.5 | +6.5 |
The pronounced DNA selectivity, along with hybridisation-induced bathochromic shifts of pyrene absorption maxima and increased intensity of circular dichroism (CD) signals in the pyrene region, strongly support an intercalative binding mode for the pyrene moieties of monomers W–Y.53 Closer analysis of the molecular arrangement in these monomers reveals that the attachment points of the nucleobase and pyrene moieties are restricted relative to each other as a consequence of the 2-oxo-5-azabicyclo[2.2.1]heptane skeleton (Figure 1.6). This, together with the short rigid linker between the bicyclic skeleton and pyrene moiety, results in forced intercalation of the fluorophore into the duplex core. Results from molecular modelling studies provide additional support for this hypothesis (Figure 1.6).
Binding modes of N2′-functionalised 2′-amino-α-L-LNA (reproduced with permission from ref. 53; copyright 2009 American Chemical Society).
Binding modes of N2′-functionalised 2′-amino-α-L-LNA (reproduced with permission from ref. 53; copyright 2009 American Chemical Society).
Synthesis. The synthesis of N2′-functionalised 2′-amino-α-L-LNA-T monomers initiates from inexpensive diacetone-α-D-glucose, which is converted into methyl furanoside 7 in a similar manner to that discussed for 2′-amino-LNA, except that the 1,2-O-isopropylidene group is cleaved using hydrogen chloride in methanol (Scheme 1.2).54 Furanoside 7 is converted into an inseparable anomeric mixture of nucleoside 8 via a reaction sequence entailing O2-triflation, installation of a C2-azido group with inversion of configuration (i.e. azido group pointing ‘up’), acetolysis and Vorbrüggen glycosylation. Attempts to develop a route in which a C2′-azido group is installed at the nucleoside level were unsuccessful.54 Treatment of 8 under Staudinger conditions results in an anomeric mixture of bicyclic nucleosides from which 2′-amino-α-L-LNA nucleoside 9 is isolated in moderate yield. Subsequent protecting group manipulations furnish key intermediate 10, which is used as a substrate for N2′-functionalisation.53 While a handful of N2′-pyrene- and coronene-functionalised 2′-amino-α-L-LNA monomers have been prepared in this manner, the time- and resource-intensive route has prevented full exploration of this compound class (∼20 steps; <4% overall yield from diacetone-α-D-glucose).53–55
Outline of synthetic route to N2′-functionalised 2′-amino-α-L-LNA-T phosphoramidites: (a) BnBr, n-Bu4NI, NaH, THF; (b) 80% aq. AcOH; (c) NaIO4, THF/H2O; (d) HCHO, aq. NaOH, 1,4-dioxane; (e) MsCl, pyridine; (f) CH3COCl, MeOH; (g) Tf2O, pyridine/CH2Cl2; (h) NaN3, 15-crown-5, DMF; (i) c. H2SO4, Ac2O, AcOH; (j) thymine, BSA, TMSOTf, ClCH2CH2Cl; (k) PMe3, aq. NaOH, THF; (l) (CF3CO)2O, pyridine/CH2Cl2; (m) KOAc, 18-crown-6, 1,4-dioxane; (n) sat. NH3/MeOH; (o) BCl3, hexane/CH2Cl2; (p) DMTrCl, DMAP, pyridine; (q) aq. NaOH, EtOH/pyridine; (r) N2′-functionalisation (e.g. PyCHO, NaBH(OAc)3, ClCH2CH2Cl or PyCOOH, HATU, EtN(i-Pr)2, DMF); (s) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2.
Outline of synthetic route to N2′-functionalised 2′-amino-α-L-LNA-T phosphoramidites: (a) BnBr, n-Bu4NI, NaH, THF; (b) 80% aq. AcOH; (c) NaIO4, THF/H2O; (d) HCHO, aq. NaOH, 1,4-dioxane; (e) MsCl, pyridine; (f) CH3COCl, MeOH; (g) Tf2O, pyridine/CH2Cl2; (h) NaN3, 15-crown-5, DMF; (i) c. H2SO4, Ac2O, AcOH; (j) thymine, BSA, TMSOTf, ClCH2CH2Cl; (k) PMe3, aq. NaOH, THF; (l) (CF3CO)2O, pyridine/CH2Cl2; (m) KOAc, 18-crown-6, 1,4-dioxane; (n) sat. NH3/MeOH; (o) BCl3, hexane/CH2Cl2; (p) DMTrCl, DMAP, pyridine; (q) aq. NaOH, EtOH/pyridine; (r) N2′-functionalisation (e.g. PyCHO, NaBH(OAc)3, ClCH2CH2Cl or PyCOOH, HATU, EtN(i-Pr)2, DMF); (s) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2.
1.3.4 C5-Functionalised LNA
Hybridisation properties. ONs modified with C5-functionalised LNA-U monomers carrying small substituents (e.g. ethynyl, 3-aminopropyn-1-yl, amino acids), generally display higher affinity toward DNA/RNA targets than conventional LNAs.47,48 In contrast, larger and more hydrophobic substituents, such as fatty acids, cholesterol or pyrene derivatives (see Figure 1.4), are detrimental to duplex thermostability; similar observations have been made with the corresponding C5-functionalised α-L-LNA-U monomers (Table 1.3).47,48,56 Fluorescence emission spectra suggest that the C5-substituent is directed toward the major groove (see Section 1.4.3).56 We stipulate that interactions between H6 and H3′ (or H2′ in α-L-LNA monomers) hinder rotation around the glycosidic bond, resulting in greater positional control of the fluorophore relative to corresponding C5-functionalised DNA monomers (Figure 1.7).
Representative thermal denaturation temperatures of duplexes between C5-fluorophore-functionalised LNA/α-L-LNA and DNA or RNA complements.
. | . | ΔTm (°C) . | ||||||
---|---|---|---|---|---|---|---|---|
. | . | 5′-GTG ABA TGC . | . | 3′-CAC TAB ACG . | ||||
Monomer . | . | vs. DNA . | . | vs. RNA . | . | vs. DNA . | . | vs. RNA . |
See Table 1.1 for experimental conditions. For structures of monomers, see Figure 1.4. | ||||||||
LNA-T47,48 | +5.0 | +9.5 | +6.5 | +9.5 | ||||
O56 | – 1.0 | +1.5 | 0 | +2.5 | ||||
P56 | – 6.5 | – 4.0 | – 4.0 | 0.0 | ||||
Q47 | – 10.5 | – 2.0 | nd | nd | ||||
R47 | – 5.5 | – 1.5 | nd | nd |
. | . | ΔTm (°C) . | ||||||
---|---|---|---|---|---|---|---|---|
. | . | 5′-GTG ABA TGC . | . | 3′-CAC TAB ACG . | ||||
Monomer . | . | vs. DNA . | . | vs. RNA . | . | vs. DNA . | . | vs. RNA . |
See Table 1.1 for experimental conditions. For structures of monomers, see Figure 1.4. | ||||||||
LNA-T47,48 | +5.0 | +9.5 | +6.5 | +9.5 | ||||
O56 | – 1.0 | +1.5 | 0 | +2.5 | ||||
P56 | – 6.5 | – 4.0 | – 4.0 | 0.0 | ||||
Q47 | – 10.5 | – 2.0 | nd | nd | ||||
R47 | – 5.5 | – 1.5 | nd | nd |
Illustration of interactions between H6 and H3′ (LNA) or H2′ (α-L-LNA) in C5-functionalised LNA/α-L-LNA, which hinder rotation about the glycosidic bond.
Illustration of interactions between H6 and H3′ (LNA) or H2′ (α-L-LNA) in C5-functionalised LNA/α-L-LNA, which hinder rotation about the glycosidic bond.
Synthesis. The synthesis of C5-functionalised LNA-U monomers initiates from glycosyl donor 1,50 which is converted into fully deprotected LNA uridine diol 12 via: a) Lewis acid-catalysed glycosylation with persilylated uracil, b) tandem O2′-deacylation and intramolecular nucleophilic displacement furnishing the LNA skeleton, and c) protecting group manipulations (Scheme 1.3).57,58 C5-Iodination of 12 followed by O5′-dimethoxytritylation provides access to key intermediate 13, which is coupled to fluorophore-modified terminal alkynes via the Sonogashira approach.56,58 Subsequent O3′-phosphitylation provides desired phosphoramidite 14. Interestingly, the synthesis of C5-functionalised LNA-U monomers only requires two additional steps relative to conventional LNA monomers, i.e. C5-iodination and C5-functionalisation. As a result, these monomers are the most readily accessible functionalised LNA monomers (∼15% overall yield from diacetone allose; ∼15 steps). The synthesis of C5-fluorophore-functionalised α-L-LNA-U monomers proceeds from diacetone-α-D-glucose via a similar route.56
Outline of synthetic route to C5-functionalised LNA-U phosphoramidites: (a) uracil, BSA, TMSOTf, CH3CN; (b) aq. NaOH, 1,4-dioxane; (c) NaOBz, DMF; (d) aq. NaOH, THF; (e) 88% HCOOH, 20% Pd(OH)2/C, THF/MeOH; (f) I2, CAN, AcOH; (g) DMTrCl, pyridine; (h) C5′-functionalisation (e.g. PyCONH2CH2C≡CH, Pd(PPh3)4, CuI, Et3N, DMF); (i) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2.
Outline of synthetic route to C5-functionalised LNA-U phosphoramidites: (a) uracil, BSA, TMSOTf, CH3CN; (b) aq. NaOH, 1,4-dioxane; (c) NaOBz, DMF; (d) aq. NaOH, THF; (e) 88% HCOOH, 20% Pd(OH)2/C, THF/MeOH; (f) I2, CAN, AcOH; (g) DMTrCl, pyridine; (h) C5′-functionalisation (e.g. PyCONH2CH2C≡CH, Pd(PPh3)4, CuI, Et3N, DMF); (i) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2.
1.3.5 LNA with Fluorescent Nucleobase Surrogates
Hybridisation properties. ONs modified with LNA or α-L-LNA based C-glycosides S, T or U (see Figure 1.4) display strongly reduced thermal affinity toward complementary DNA relative to unmodified reference strands (Table 1.4).59–61 However, these probes display interesting universal hybridisation characteristics, i.e. they exhibit virtually identical DNA/RNA target affinity regardless of the nucleotide opposite of the modification site (Table 1.4). These characteristics strongly suggest that the pyrene moieties of these C-glycosides act as nucleobase surrogates, which force the opposing nucleotide out from the duplex core in a similar manner to that reported for DNA-based monomers.62,63 Development of universal hybridisation probes has been a longstanding goal because of their potential application as degenerate PCR primers and microarray probes when the identity of one or more nucleotides in a target sequence is unknown.64
Thermal denaturation temperatures (Tms are shown) of duplexes between centrally modified ONs and DNA targets.
. | . | Tm (°C) . | ||||||
---|---|---|---|---|---|---|---|---|
. | . | 5′-GTG ABA TGC: 3′-CAC TYT ACG . | ||||||
Monomer . | Y: . | A . | . | C . | . | G . | . | T . |
See Table 1.1 for experimental conditions. For structures of monomers, see Figure 1.4. | ||||||||
DNA-T59 | 28 | 11 | 12 | 19 | ||||
S59 | 18 | 17 | 18 | 19 | ||||
T60 | 21 | 22 | 27 | 23 | ||||
U61 | 21 | 20 | 19 | 21 |
Synthesis. The synthetic route to the representative C-glycoside LNA monomer S initiates from diacetone-α-D-allose (Scheme 1.4).59 The starting material is converted to methyl furanoside 16 in an equivalent manner to that discussed for 7 (Scheme 1.2) with the exception that a para-methoyxybenzyl (PMB) group is used for protection of the O3-position rather than a regular benzyl group. PMB, which can be oxidatively cleaved using 2,3-dichloro-5,6-dicyano-1,4-benzoquinone (DDQ), is used to minimise cleavage of the benzylic O4–C1 bond during O3-deprotection. Base-induced cyclisation of methyl furanoside 16 followed by several protecting group manipulation steps provides dioxabicyclo-[2.2.1]-heptane 17. Acidic hydrolysis of the unstable acetal generates γ-hydroxy-aldehyde 18, which is stereospecifically converted into 19 upon treatment with Grignard reagents. Subsequent reformation of the bicyclic ring under Mitsunobu conditions followed by protecting group manipulations, including the aforementioned cleavage of the PMB group, provides phosphoramidite 20 (<5% overall yield, ∼16 steps). Different aryl groups have been introduced in this manner.59 The corresponding phosphoramidite of α-L-LNA monomer T is obtained in a related manner although additional inversion and protection/deprotection steps are needed.60
Outline of synthetic route to LNA monomers with nucleobase surrogates: (a) p-MeOC6H4CH2Cl, NaH, DMF; (b) 70% AcOH; (c) NaIO4, H2O; (d) HCHO, aq. NaOH, THF/H2O; (e) MsCl, pyridine; (f) HCl/CH3OH/H2O; (g) NaH, DMF (isolation of major anomer); (h) KOAc, 18-crown-6, 1,4-dioxane; (i) sat. NH3 in MeOH; (j) p-MeOC6H4CH2Cl, NaH, THF; (k) 80% AcOH; (l) ArMgBr, THF; (m) TMAD, Bu3P, benzene; (n) DDQ, CH2Cl2/H2O; (o) DMTrCl, pyridine; (p) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2.
Outline of synthetic route to LNA monomers with nucleobase surrogates: (a) p-MeOC6H4CH2Cl, NaH, DMF; (b) 70% AcOH; (c) NaIO4, H2O; (d) HCHO, aq. NaOH, THF/H2O; (e) MsCl, pyridine; (f) HCl/CH3OH/H2O; (g) NaH, DMF (isolation of major anomer); (h) KOAc, 18-crown-6, 1,4-dioxane; (i) sat. NH3 in MeOH; (j) p-MeOC6H4CH2Cl, NaH, THF; (k) 80% AcOH; (l) ArMgBr, THF; (m) TMAD, Bu3P, benzene; (n) DDQ, CH2Cl2/H2O; (o) DMTrCl, pyridine; (p) NC(CH2)2OP(Cl)N(i-Pr)2, EtN(i-Pr)2, CH2Cl2.
1.4 Applications of Fluorophore-functionalised LNA
1.4.1 Formation of Pyrene Arrays
The use of nucleic acids as scaffolds for programmable arrangement of chromophores has received considerable attention owing to the prospect of developing DNA-based light harvesting antenna systems.7,65,66 Pyrene arrays have often been studied as simple model systems toward this end.
Formation of pyrene arrays in the major groove of DNA duplexes has been realised using ONs with five sequential incorporations of 5-(pyren-1-yl)-2′-deoxyuridine monomer A (see Figure 1.3), as evidenced by hybridisation-induced excitonic coupling of pyrene signals in CD spectra.67 The presence of mismatched base pairs in the proximity of the array results in electronic decoupling of the pyrene moieties, which implies interesting diagnostic applications for detection of single nucleotide polymorphisms.
Formation of pyrene arrays in the minor groove has been realised using duplexes between RNA targets and 2′-O-methyl RNA strands that are sequentially modified with 2′-O-(pyren-1-yl)methyluridine monomer B (see Figure 1.3).68 The use of RNA duplexes, rather than enzymatically more stable DNA duplexes, is necessary to minimise undesired intercalation of the pyrene moieties.69 In an alternative approach, an interstrand array of pyrene moieties was generated in the minor groove of RNA duplexes, which are modified with 2′-O-(pyren-1-yl)methyl uridine/adenosine pairs.70
Several pyrene arrays, which are based on functionalised LNA monomers, have been developed. The first example involves DNA duplexes with interstrand arrangements of 2′-N-(pyren-1-yl)methyl-2′-amino-LNA monomer F (Figures 1.4 and 1.8).45 The thermostability values of singly modified and reference DNA duplexes are virtually identical. Placement of two F monomers in a ‘–1 interstrand zipper arrangement’ leads to considerable duplex stabilisation and excimer formation, whereas ‘+1 interstrand monomer arrangements’ do not (Tm=35 °C vs. 30 °C, respectively, Figure 1.8). These trends are augmented by incorporation of additional –1 zipper cassettes. The lowest energy structures from force-field calculations reveal structural features that are consistent with these characteristics, as pyrene moieties in –1 zipper arrangements stack pair wise across the minor groove (Figure 1.8).45 More brightly fluorescent versions of this ‘communication system’, based on N2′-(phenylethynyl)pyrene-functionalised 2′-amino-LNA monomers M or N (Figure 1.4), have been developed.43 The predictability of the original communication system has been used to signal stepwise self-assembly of branched oligonucleotides into higher order structures.71,72 The extraordinary duplex thermostability, directional preference and robustness of the communication system underscore the advantages of functionalised LNA.
Formation of pyrene interstrand arrays using 2′-N-(pyren-1-yl)methyl-2′-amino-LNA monomer F (for structure, see Figure 1.4). Left: Tm-values and fluorescence characteristics of DNA duplexes modified with monomer F. Right: lowest energy structure from force field calculations on the duplex with three –1 interstrand arrangements of monomer F (Tm=77 °C) (from ref. 45; copyright 2004 RSC Publishing).
Formation of pyrene interstrand arrays using 2′-N-(pyren-1-yl)methyl-2′-amino-LNA monomer F (for structure, see Figure 1.4). Left: Tm-values and fluorescence characteristics of DNA duplexes modified with monomer F. Right: lowest energy structure from force field calculations on the duplex with three –1 interstrand arrangements of monomer F (Tm=77 °C) (from ref. 45; copyright 2004 RSC Publishing).
In another example, 2′-N-(pyren-1-yl)acetyl-2′-amino-α-L-LNA monomer Y has been used to generate regular arrangements of pyrene moieties inside the core of DNA duplexes (Figure 1.9).73 ONs modified with 5′-(YΦ)-steps, i.e. an arrangement where monomer Y is 3′-flanked by abasic site monomer Φ, display remarkable affinity toward DNA targets with abasic sites in the +1 interstrand position relative to Y (Figure 1.9). Thus, a 13-mer DNA duplex containing four of these 5′-(YΦ):3′-(AΦ)- cassettes, each separated by one base pair, displays a Tm of ∼60 °C, whereas the corresponding reference duplex, where the 5′-(YΦ):3′-(AΦ)-units are replaced by 5′-(TA):3′-(AT) base pairs, has a Tm of ∼35 °C (Figure 1.9). Hybridisation-induced bathochromic shifts of pyrene absorption maxima, along with data from thermodynamic studies and force field calculations, suggest that monomer Y forces the pyrene moiety into the void formed by the Φ:Φ pair, where it engages in efficient π–π stacking with neighbouring base pairs (Figure 1.9).73 Similar, albeit far less thermostable, pyrene arrangements have been generated using the C-glycoside LNA monomer U.61
Regular arrangements of pyrene moieties in duplex cores using 5′-(YΦ):3′-(AΦ)-units. Molecular modelling structure depicts 13-mer DNA duplex containing two separated 5′-(YΦ):3′-(AΦ)-units (reproduced with permission from ref. 73; copyright 2008 American Chemical Society).
Regular arrangements of pyrene moieties in duplex cores using 5′-(YΦ):3′-(AΦ)-units. Molecular modelling structure depicts 13-mer DNA duplex containing two separated 5′-(YΦ):3′-(AΦ)-units (reproduced with permission from ref. 73; copyright 2008 American Chemical Society).
1.4.2 Hybridisation Probes
Hybridisation probes are designed to display: a) low fluorescence in the absence of target, typically through quenching interactions between the fluorophore and nucleobase moieties,37 and b) prominent fluorescence in the presence of target, by placing the fluorophore in a less quenching environment. Early examples of hybridisation probes include ONs modified with cyanine,74,75 fluorescein76,77 or pyrene moieties.78 For example, the fluorescence intensity of RNA strands modified with 2′-O-(pyren-1-yl)methyluridine monomer B (see Figure 1.3) increases up to 30-fold upon hybridisation with RNA, while much smaller increases are observed with DNA targets. The fluorescence properties are moderately influenced by the nature of the 3′-flanking nucleobase [emission quantum yield ΦF=0.10/0.10/0.16/0.24 for duplexes between complementary RNA and 5′-r(ACA BXC AGU GUU GAU) where X=G/A/U/C, respectively].78 In contrast, incorporation of monomer B into DNA strands produces probes with far more variable emission profiles.79,80 These differences are due to different pyrene binding modes; NMR solution structures of RNA:RNA and DNA:DNA duplexes modified with monomer B show that the pyrene moieties predominantly are located in the minor groove and duplex core, respectively.69 This underlines just how interrelated are the photophysical properties and positional control of the fluorophore.
In contrast, ONs modified with C5-pyrene-functionalised triazole-linked 2′-deoxyuridine monomer C (see Figure 1.3) result in 9- to 23-fold increases in fluorescence upon hybridisation with DNA, as well as RNA, targets. Accordingly, the data suggest that placement of the pyrene moiety in the major groove81 dominates over intercalative binding modes, which would result in quenching of fluorescence. These hybridisation probes only display mild sequence limitations; prominent hybridisation-induced increases in fluorescence intensity are observed when monomer C is flanked by A/C/G, while flanking Ts result in inadequate quenching of single stranded probes and low increases. The moderate quantum yields (ΦF≤0.16) are the main drawback of these probes. Interestingly, ONs modified with the corresponding C5-functionalised LNA monomer R (see Figure 1.4) display markedly larger hybridisation-induced intensity increases (up to 51- to 42-fold increases vs. DNA and RNA, respectively).47 The formation of more brightly fluorescent duplexes suggests that the LNA skeleton plays an active role in directing the pyrene moiety into the non-quenching major groove. Similar observations have been made in comparative studies of other C5-functionalised DNA/LNA monomers (see Section 1.4.3).56
DNA strands modified with 2′-N-(pyren-1-yl)-carbonyl-2′-amino-LNA monomer G (see Figure 1.4) result in prominent increases in fluorescence intensity upon hybridisation with DNA and RNA targets (typically 2- and 10-fold), leading to the formation of brightly fluorescent duplexes (ΦF=0.28–0.99, Figure 1.10).46,82,83 Successful design of these ‘Glowing LNA’84 requires incorporation of at least two G monomers separated by at least one nucleotide, because this decreases the fluorescence intensity of single stranded probes through pyrene–nucleobase interactions (Figure 1.10). Molecular modelling studies suggest that the bicyclic skeleton and short rigid amide linker of monomer G force the pyrene moiety into the non-quenching minor groove of duplexes.46 The importance of positional control is emphasised by the fact that ONs modified with the slightly more flexible 2′-N-(pyren-1-yl)-methyl-2′-amino-LNA monomer F do not display reliable hybridisation-induced increases in fluorescence intensity and result in the formation of far less brightly fluorescent duplexes. Incorporation of monomer G into RNA or 2′-O-methyl-RNA strands yields probes that display even larger increases in fluorescence intensity upon target binding, because the single stranded probes (SSPs) are more efficiently quenched, while duplexes display even higher quantum yields (Figure 1.10).83
Principle of hybridisation probes and quencher-free molecular beacons modified with 2′-N-(pyren-1-yl)carbonyl-2′-amino-LNA-T monomer G (upper panel), and quantum yields of Glowing LNA probes with different backbone chemistries in the absence (SSP) or presence of complementary DNA/RNA (adapted with permission from ref. 83; copyright 2010 American Chemical Society).
Principle of hybridisation probes and quencher-free molecular beacons modified with 2′-N-(pyren-1-yl)carbonyl-2′-amino-LNA-T monomer G (upper panel), and quantum yields of Glowing LNA probes with different backbone chemistries in the absence (SSP) or presence of complementary DNA/RNA (adapted with permission from ref. 83; copyright 2010 American Chemical Society).
Monomer G has also been evaluated as a building block in ‘quencher-free’ molecular beacons (MBs).83 Unlike conventional MBs, which are end-functionalised with fluorophore–quencher pairs, leading to binding-induced dequenching of fluorescence emission,85 quencher-free MBs rely on the presence of microenvironment-sensitive monomers in the target loop to signal binding-induced changes in the secondary structure.86 Hybridisation of a 29-mer quencher-free MB with four separated G monomers in the 15-mer target loop (Figure 1.10) to complementary DNA/RNA targets results in 3- to 9-fold increases in signal intensity and formation of brightly fluorescent duplexes (ΦF=0.45–0.62).83 The quencher-free MBs display high biostability (>48 h) and greater thermal discrimination of singly mismatched DNA/RNA targets than their linear counterparts, which facilitates their use for imaging of cellular RNA.83
The interesting characteristics of multilabelled Glowing LNA stimulated the development of other N2′-fluorophore-functionalised 2′-amino-LNA monomers.40–44 For example, ONs modified with 2′-N-(perylen-3-yl)carbonyl-2′-amino-LNA monomer I (see Figure 1.4) display a similar pattern of hybridisation-induced increases in fluorescence intensity to that of the original Glowing LNA probes.42 While red-shifted emission relative to the original Glowing LNA probes is observed (∼490 nm vs. ∼400 nm), the emission increases and duplex quantum yields are lower (ΦF=0.09–0.50). Optimised probes based on monomer I were, nonetheless, also used for cellular RNA imaging.
ONs modified with the larger 2′-N-4-(coronen-1-yl)methyl-2′-amino-LNA monomer H (see Figure 1.4) also display red-shifted emission (∼430 nm) but only display minor changes in signal intensity upon hybridisation to DNA/RNA targets.44 Loss of positional control, due to the more flexible linker connecting the coronene and sugar moieties, is one of the likely reasons for these photophysical characteristics.
ONs modified with various N2′-(phenylethynyl)pyrenecarbonyl-functionalised 2′-amino-LNA monomers J–L (see Figure 1.4) have been shown to display strongly red-shifted fluorescence emission (∼420–520 nm, depending on the number of phenylethynyl substituents).41 High duplex quantum yields and modest increases in signal intensity are observed for some of these probes, but the trends are less predictable than with the original Glowing LNA probes.
The pyren-1-ylcarbonyl moiety of Glowing LNA monomer G has been conjugated to ring-expanded LNA monomers, but multilabelled probes were not studied, which prevents direct comparison.87
In summary, the above examples demonstrate that attachment of the pyren-1-yl-carbonyl moiety to the 2′-amino-LNA skeleton (i.e. monomer G) produces hybridisation probes with desirable characteristics, presumably by striking an optimal balance between steric and electronic requirements.
1.4.3 Base Discriminating Fluorescent Probes
Fluorophore-functionalised LNA monomers have also found use as building blocks for base discriminating fluorescent (BDF) probes. Unlike hybridisation probes, BDF probes display duplex emission characteristics that are strongly dependent on the nature of the nucleotide opposite the BDF monomer (Figure 1.11).88,89 Accordingly, these probes are useful for detection of SNPs, which are the most frequently occurring genetic variations in the human genome and important biomedical markers.4
ONs modified with C5-[3-(1-pyrenecarboxamido)propynyl]-functionalised 2′-deoxyuridine monomer D (see Figure 1.3) are interesting BDF probes, which have been used for detection of SNP sites in human breast cancers.90,91 Prominent signal increases are observed upon hybridisation to complementary DNA (2- to 10-fold), while hybridisation to DNA strands with SNP-sites opposite monomer D results in much smaller increases (Figures 1.11 and 1.12). These trends presumably correspond to different fluorophore binding modes: a) in unhybridised probes, interactions between the fluorophore and nucleobase moieties ensure low emission levels; b) upon duplex formation with complementary DNA, the fluorophore is placed in the non-quenching major groove resulting in high levels of emission; c) duplex formation with SNP-containing DNA targets results in a change in nucleobase orientation from the anti to syn conformation, resulting in intercalation of the fluorophore and nucleobase-mediated quenching.90 Considering that guanine moieties are the most efficient quenchers of pyrene fluorescence, it is not surprising that the emission characteristics of these probes are influenced by neighbouring nucleotides. Probes in which monomer D is flanked by cytosine or guanine units discriminate SNPs more efficiently than probes with ADA/TDT contexts (emission decreases by 86–92% vs. 54–86% relative to matched duplexes).56
Fluorescence emission spectra of probes modified with C5-[3-(1-pyrenecarboxamido)propynyl]-functionalised DNA (left), LNA (centre) or α-L-LNA (right) monomers. Probe sequence: 5′-d(CG CAA GBG ACC GC), where B=monomer D, P or O. Spectra are for probes in absence (SSP) or presence of complementary DNA (cDNA) or mismatched DNA (mmDNA, mismatched nucleotide across from modification listed in parenthesis). λex=344 nm, T=5 °C (adapted with permission from ref. 56; copyright 2011 Wiley).
Fluorescence emission spectra of probes modified with C5-[3-(1-pyrenecarboxamido)propynyl]-functionalised DNA (left), LNA (centre) or α-L-LNA (right) monomers. Probe sequence: 5′-d(CG CAA GBG ACC GC), where B=monomer D, P or O. Spectra are for probes in absence (SSP) or presence of complementary DNA (cDNA) or mismatched DNA (mmDNA, mismatched nucleotide across from modification listed in parenthesis). λex=344 nm, T=5 °C (adapted with permission from ref. 56; copyright 2011 Wiley).
The corresponding LNA and α-L-LNA monomers P and O (see Figure 1.4) were explored on the basis of the hypothesis that the extreme sugar pucker of the bicyclic skeletons would influence the nucleobase orientation (see Figure 1.7) and, accordingly, the position of the fluorophore. Indeed, ONs modified with LNA monomer P display similar hybridisation-induced increases in emission (2- to 15-fold), 5–60% larger duplex quantum yields (ΦF=0.44–0.67) and improved SNP-discrimination when the monomers are flanked by C/G (86–97% decreased emission relative to matched duplex, Figure 1.12).56 ONs modified with α-L-LNA monomer O display slightly larger hybridisation-induced signal increases (4- to 10-fold), 20–50% larger duplex quantum yields (ΦF=0.50–0.80) and, with the exception of G-mismatches, improved SNP-discrimination with probes having AOA or TOT sequence contexts (emission decreased by 65–93% relative to matched duplex).56 The larger duplex quantum yields observed with LNA and α-L-LNA monomers O and P most likely reflect greater fluorophore occupancy in the major groove. Higher duplex quantum yields have also been observed for ONs modified with C5-pyrene-functionalised triazole-linked LNA monomers Q and R relative to their DNA counterparts (see Figure 1.4).47
1.4.4 Excimer-based Probes
Excimers, i.e. electronically excited π-stacking dimers of identical fluorophores, emit highly Stokes-shifted fluorescence (e.g. ∼120 nm for pyrene excimers35 ). The presence of excimer emission provides valuable structural insight as the fluorophores adopt co-planar arrangements with a separation of ∼3.4 Å. As discussed in previous sections, fluorophore-functionalised LNA monomers offer great positional control of fluorophores and have accordingly been used in the development of probes that rely on excimer signals for detection of nucleic acid targets. ONs with two next-nearest neighbour incorporations of 2′-N-(pyren-1-yl)acetyl-2′-amino-α-L-LNA monomer Y are an example of this (Figure 1.13).92 Duplexes with matched DNA/RNA targets predominantly display pyrene monomer fluorescence, whereas duplexes with mismatched base pairs near the modified region (positions 4–7, Figure 1.13) display intense excimers (2- to 11-fold increase). Similar to BDF probes, and unlike hybridisation probes, this assay does not require stringent temperature control. In other words, SNPs are discriminated even if mismatched duplexes are formed. As discussed previously, the pyrene moiety of monomer Y is forced into the core of matched duplexes, which prevents excimer formation due to spatial separation of the pyrene moieties (Figure 1.13). Mismatched duplexes have a more dynamic duplex geometry, which allows the pyrene moieties to adopt extrahelical orientations suitable for formation of pyrene–pyrene excimers (Figure 1.13).92
Principle of SNP detection using ONs with two next-nearest neighbour incorporation of 2′-N-(pyren-1-yl)acetyl-2′-amino-α-L-LNA monomer Y (see Figure 1.4 for structure of monomer) (reproduced with permission from ref. 92; copyright 2007 Wiley).
Principle of SNP detection using ONs with two next-nearest neighbour incorporation of 2′-N-(pyren-1-yl)acetyl-2′-amino-α-L-LNA monomer Y (see Figure 1.4 for structure of monomer) (reproduced with permission from ref. 92; copyright 2007 Wiley).
Another class of probes that rely on excimers to signal the presence of SNPs are the so-called dual probes, where two end-labelled probes are assembled into a ternary complex upon target binding.93 This brings the fluorophores into close proximity, leading to excimer formation (Figure 1.14). The use of two shorter probes for target binding, rather than a long probe, results in improved thermal discrimination of mismatches and reduces false positive signals. Another advantage of this approach is that the large Stokes shifts allow for clear fluorescent distinction between unhybridised probes and target complexes. LNA-modified ONs, which are end-labelled with 2′-N-(pyren-1-ylmethyl)-2′-amino-LNA monomer F (Figure 1.4), have been used in dual probe assays.94 As expected, pyrene monomer fluorescence is predominantly observed in the absence of target, while binding to complementary DNA/RNA targets results in excimer formation (Figure 1.14). In contrast, hybridisation to singly mismatched DNA/RNA targets leads to significantly less intense excimer emission (Figure 1.14). Thermal denaturation experiments have demonstrated that the decreased excimer intensity is a result of: a) mismatch-induced structural perturbation of the ternary complex leading to inefficient pyrene–pyrene stacking (positions 8–11, Figure 1.14), and/or b) lack of ternary complex formation due to efficient thermal mismatch discrimination (positions 2–7, Figure 1.14), which is augmented by the short probes and high content of conventional LNA.94 In contrast, the corresponding ‘covalently linked dual probe’ only allows for fluorescent discrimination of SNPs in a narrow region (positions 7–10) through local disruption of excimers.94
Left: principle of SNP-discrimination using excimer-forming dual probes. Centre: fluorescence emission spectra of duplexes between LNA-based dual probes end-functionalised with monomer F and complementary (red) or singly mismatched DNA target (black). Right: probe/target sequences; discrimination of SNPs in positions 1–12 (λem=480 nm) (adapted with permission from ref. 94; copyright 2007 Wiley).
Left: principle of SNP-discrimination using excimer-forming dual probes. Centre: fluorescence emission spectra of duplexes between LNA-based dual probes end-functionalised with monomer F and complementary (red) or singly mismatched DNA target (black). Right: probe/target sequences; discrimination of SNPs in positions 1–12 (λem=480 nm) (adapted with permission from ref. 94; copyright 2007 Wiley).
Invader LNA probes, i.e. DNA duplexes modified with one or more +1 interstrand zipper arrangements of 2′-N-(pyren-1-yl)methyl-2′-amino-α-L-LNA-T monomer W (see Figure 1.4), enable detection of double stranded DNA (dsDNA) targets through changes in excimer emission (Figure 1.15).95,96 This specific monomer arrangement (termed an energetic hotspot) forces the pyrene moieties into the same region within a duplex core, leading to excimer formation, partial duplex unwinding and decreased thermostability.95,96 Other interstrand monomer arrangements induce much higher duplex thermostabilisation, highlighting the importance of positional control to obtain functional control.95,96 In contrast, the two strands that comprise an Invader LNA probe display very high affinity toward complementary DNA (ΔTm/mod up to +15 °C).53 The large differences in thermostability between probe–target and Invader probe duplexes generate an energetically favourable gradient for recognition of mixed-sequence dsDNA targets at physiologically relevant ionic strengths (Figure 1.15).95,96 The recognition process is monitored by a decrease in excimer emission as the pyrene moieties are forced apart (Figure 1.15). Addition of a 13-mer Invader LNA to an equimolar quantity of complementary dsDNA target at 110 mM NaCl, pH 7 and low experimental temperatures results in ∼50% recognition within ∼40 min. Recognition of dsDNA even occurs in buffers of high ionic strength (710 mM NaCl, pH 7, t50% ∼125 min). The use of Invader LNAs with more than one energetic hotspot accelerates dsDNA recognition. Addition of Invader LNAs to non-isosequential dsDNA targets results in much slower and less complete decay of the excimer signal, which demonstrates the specificity of the recognition process.96 Subsequent studies have shown that the free energy for dsDNA recognition can be increased through changes in the linker chemistry of the Invader LNA monomers.97
Left: illustration of Invader LNA concept. Right: time course of fluorescence emission upon addition of Invader LNA to a complementary dsDNA target (λex=335 nm, 20 °C). Invader LNA: 5′-d(GGT AWA TAT AGG C):3′-d(CCA TAW ATA TCC G) (from ref. 96; copyright 2010 RSC Publishing).
Left: illustration of Invader LNA concept. Right: time course of fluorescence emission upon addition of Invader LNA to a complementary dsDNA target (λex=335 nm, 20 °C). Invader LNA: 5′-d(GGT AWA TAT AGG C):3′-d(CCA TAW ATA TCC G) (from ref. 96; copyright 2010 RSC Publishing).
1.5 Summary and Perspective
The following classes of fluorophore-conjugated LNAs have been highlighted in the previous sections:
N2′-functionalised 2′-amino-LNAs, which direct fluorophores into the minor groove, without major impacts on duplex thermostability. These characteristics have led to the development of pyrene arrays; communication systems signalling self-assembly of higher order nucleic acid structures; brightly fluorescent hybridisation probes for RNA imaging; and dual probes for SNP-discrimination.
N2′-functionalised 2′-amino-α-L-LNAs, which direct aromatic fluorophores into duplex cores, resulting in extraordinary stabilisation through π-stacking with neighbouring nucleobases. These monomers have been used in the development of regular pyrene arrangements inside duplex cores; excimer-based probes for discrimination of SNPs; and novel strategies for detection of mixed-sequence dsDNA under physiologically relevant conditions.
C5-functionalised LNAs and α-L-LNAs, which direct fluorophores into the major groove, despite marked duplex destabilisation. These monomers have been used to generate base discriminating fluorescent probes for detection of SNPs.
C-glycoside LNAs and α-L-LNAs, which force fluorescent nucleobase surrogates into the duplex core. Although strong duplex destabilisation ensues, these probes display interesting universal hybridisation characteristics and have been used to form regular pyrene arrangements.
A unifying element of these different classes of fluorophore-functionalised LNA is that functional control is achieved through positional control via the combined use of conformationally restricted bicyclic skeletons and short linkers for attachment of fluorophores (see Figure 1.2). Several classes of fluorophore-functionalised LNA remain unexplored. For example, attachment of fluorophores to the C8-position of LNA purines98 is poised to afford building blocks that display interesting BDF-characteristics;99 the extreme sugar pucker of the LNA skeleton will restrict the rotational freedom of the nucleobase, leading to stricter positional control of the fluorophore, when ompared with corresponding DNA monomers. LNA monomers that are functionalised via the extra ring are another class of interesting building blocks.100–104 The attached fluorophores would be expected to point toward the minor groove, and potentially afford probes with similar characteristics to N2′-functionalised 2′-amino-LNA monomers. Moreover, recent reports on C5′-branched LNA and α-L-LNA monomers have demonstrated that small substituents are well tolerated in the major groove of nucleic acid duplexes, which suggests the 5′-position as another interesting attachment point for fluorophores.105,106
The long synthetic routes to fluorophore-functionalised LNA monomers, however, remain a major limitation, and development of synthetically more feasible analogues is therefore needed. As discussed herein, flexible fluorophore-functionalised monomers rarely display comparable target affinity and/or emission profiles relative to their LNA-based counterparts. In a rare exception to this, however, we recently demonstrated that O2′-intercalator-functionalised uridine and N2′-intercalator-functionalised 2′-N-methyl-2′-aminouridine monomers (see Figure 1.2) largely mimic the remarkable DNA hybridisation properties of N2′-pyrene-functionalised 2′-amino-α-L-LNAs (average ΔTm/mod ∼8 °C).107 Conformational restriction of the sugar moiety presumably plays a less important role in this case because intercalation of the fluorophore is the dominating binding mode. Gratifyingly, these monomers have been demonstrated to be equipotent mimics of the original Invader LNA monomers, which will allow more extensive exploration of this novel strategy of dsDNA targeting.97
An interesting approach to more readily accessible 2′-amino-LNA monomers has been proposed by Wengel's group. Inspired by the ability of conventional LNA monomers increasingly to tune the conformation of neighbouring 2′-deoxyribonucleotides toward RNA-like C3′-endo conformations during duplex formation,14 ONs were modified with 2′-N-pyren-1-ylmethyl-2′-N-methylaminouridine monomer E (see Figure 1.2) and surrounded by conventional LNA nucleotides.108 Unfortunately, the probes display markedly lower target affinity than the corresponding reference DNA–LNA mixmers, presumably because LNA-induced conformational tuning of monomer E into a C3′-endo conformation is not energetically favourable. However, other combinations of fluorophore-functionalised DNA-monomers and LNA-backbones are likely to afford readily accessible mimics of fluorophore-functionalised LNAs in the future.109
The unique photophysical properties of fluorophore-functionalised LNAs will undoubtedly continue to stimulate development of new building blocks, leading to applications in life sciences, sensor technology and materials science.