Synthesis and fluorescence properties of 3,6-diaminocarbazole-modified pyrrolidinyl peptide nucleic acid

Abstract

This work aims to explore a new clickable carbazole-based fluorescent label, its incorporation into pyrrolidinyl peptide nucleic acid (acpcPNA), and the interactions between the labeled PNA probe and DNA by fluorescence spectrophotometry and molecular dynamics simulation. A carbazole derivative, namely 3,6-diaminocarbazole (DAC), was synthesized and incorporated into the internal and terminal positions of azide-modified acpcPNA via a sequential reductive alkylation and click reaction previously developed by our group. The DAC-modified acpcPNA can form a stable hybrid with complementary DNA with somewhat lower stability compared to unmodified acpcPNA. Most importantly, the DAC-modified acpcPNA exhibits a remarkable fluorescence increase in the presence of DNA (up to 35.5 fold with complementary DNA). Non-complementary as well as single mismatched DNA targets gave a smaller fluorescence increase (1.1 to 18.6 fold), and the discrimination could be further improved by increasing the temperature to dissociate the mismatched hybrids. Molecular dynamic simulations revealed that the DAC interacts with adjacent nucleobases in single stranded PNA, resulting in quenching of the fluorescence signal. When the PNA formed a hybrid with DNA, the DAC was pushed away from the duplex, resulting in a fluorescence increase. Thus, the DAC-labeled acpcPNA is a potential candidate as a self-reporting probe for determination of DNA sequences.

---

Introduction

Self-reporting fluorescence oligonucleotides have been extensively used as a probe for the determination of DNA sequences. Several approaches have been explored in order to introduce a mechanism for fluorescence changes upon hybridization of the probe with the correct DNA target. Classically, two labels that can interact by static quenching,¹ Fluorescence Resonance Energy Transfer (FRET)^2–4 or excimer/exciplex formation⁵ are used in combination. More recently, there are attempts to develop singly labeled self-reporting probes by attachment of an intercalative or environment sensitive dye to the oligonucleotide.⁶ However, the performance of these singly-labeled probes is generally variable, depending on the sequence and therefore still leaves room for improvements.

Carbazole is an electron rich aromatic molecule that showed many interesting properties such as high thermal stability, excellent photophysical and electroconductive properties and the ease of synthesis and chemical modification at 3-, 6- and 9-positions. Carbazole derivatives, both in monomeric and oligomeric forms, have found widespread applications as photoconductivity polymers,^7–10 polymeric light emitting devices,^11,12 high temperature thermostable polymers,¹³ electrochromic materials,^14–16 dye-sensitized solar cell and dendrimeric electro phosphorescent devices.^17–19 In addition, carbazole derivatives have been used as fluorophores in sensor applications, including DNA detection.²⁰ Carbazole derivatives with cationic groups at 3,6-positions have been reported to bind in the DNA groove of AT-rich sequences.^21–24 However, the use of carbazole derivatives in combination with an oligonucleotide or analogues for sequence-specific detection of DNA sequences is unknown.

Peptide nucleic acid or PNA is a DNA mimic consisting of nucleobases attached onto a peptide-like backbone.^25,26 PNA can hybridize with DNA according to Watson–Crick base pairing rules with high affinity and sequence specificity. The absence of phosphate group in the PNA backbone results in decreasing of the electrostatic repulsion with DNA, and therefore PNA can form a more stable duplex with DNA than DNA itself. In addition, the binding of PNA and DNA is highly sequence specific and is not sensitive to ionic strength. These properties, together with the excellent chemical and biological stabilities, make PNA a promising tool for several applications in medical and biotechnology fields. Recently, we reported a series of new PNA system with restricted conformation flexibility as a result of incorporation of cyclic structures into the PNA backbone.^27–30 Pyrrolidinyl PNA probes labeled with environment sensitive dyes such as pyrene,^31,32 Nile red,^32,33 or other DNA binding dyes such as thiazole orange³⁴ have been successfully applied as self-reporting fluorescence probes for DNA sequence detection.

From the remarkable photophysical properties and DNA binding ability mentioned above, carbazole has a potential to be developed into a probe for DNA detection by fluorescence spectroscopy. In the present work, an electron rich carbazole derivative, namely 3,6-diaminocarbazole (DAC), was incorporated at an internal position of acpcPNA in order to develop a new self-reporting fluorescence hybridization probe. Here we report the synthesis, thermal stabilities and fluorescence properties of DAC-modified acpcPNA probes, as well as their ability to discriminate between complementary and single-mismatched DNA targets. Moreover, molecular dynamics simulation was performed to explain the behavior of DAC-labeled acpcPNA probes.

Results and discussion

Synthesis of diaminocarbazole (DAC)-labeled PNA

A novel clickable 3,6-diaminocarbazole for labeling of the acpcPNA was first prepared. N⁹-Propargylated, Boc-protected-3,6-diaminocarbazole (5) was prepared in 4 steps with 37% overall yield starting from carbazole (Scheme 1). Controlled nitration of the carbazole (1) with fuming nitric acid in 1,2-dichloroethane at 45 °C afforded the dinitration product 2 in good yield. Subsequent reduction of the dinitro-compound 2 with tin(II) chloride in a mixture of acetic acid and concentrated hydrochloric acid provided the desired diaminocarbazole 3 in excellent yield after simple precipitation. The amino groups in 3 were protected to avoid undesired side reactions at this position during subsequent steps. Boc group was chosen as it could be easily introduced and subsequently cleaved under acidic conditions required for the cleavage of the modified PNA from the solid support. The Boc-protected DAC (4) was next alkylated with propargyl bromide in the presence of K₂CO₃/KI to get the desired clickable diaminocarbazole derivative (5).

Scheme 1 Synthesis of clickable 3,6-diaminocarbazole label (5).

AcpcPNA sequences were synthesized manually by Fmoc solid phase peptide synthesis as described previously.^27,35 In order to provide a handle for attachment of the DAC via the previously reported reductive alkylation-Click chemistry strategy,³⁴ one of the 2-amino cyclopentanecarboxylic acid (ACPC) spacer in the middle of the acpcPNA strand was replaced with 3-aminopyrrolidine-4-carboxylic acid (APC) for internal labeling.³⁵ For terminal labeling, no modification with APC was required, but the last ACPC residue was omitted from the PNA sequence. After end-capping by acetylation (this step was skipped for the terminal modification) and removal of all protecting groups from the nucleobase and the APC spacer, the azide group was introduced by reductive alkylation with 4-azidobutanal. Finally, the clickable DAC label (5) was attached to the azide-modified acpcPNA on the solid support by Cu-catalyzed click chemistry. Four DAC-modified acpcPNA sequences were designed to investigate the DNA binding and fluorescent properties under different sequence context. These include an internally-labeled homothymine (T9^DAC), two internally-labeled mix sequences with different flanking bases (A/T and C/G pairs) (M10AT^DAC and M10GC^DAC) and a terminally-labeled mix sequence (DAC-M10). All modified PNAs were purified by reverse phase HPLC (to >90% purity). The identities of the obtained PNA sequences were confirmed by MALDI-TOF mass spectrometry, which showed the mass signals that are in a good agreement with the expected values. The sequences and characterization data of all PNA involved in this studies are shown in Table 1.

Table 1 Sequence and characterization data of DAC-labeled acpcPNA

PNA	Sequence (N → C)	tR^a (min)	Yield^b (%)	m/z (calcd)	m/z (found)
a HPLC condition: C18 column 4.6 × 50 mm, particle size 3 μm, gradient 0.1% TFA in H2O : MeOH 90 : 10 for 5 min then linear gradient to 10 : 90 over 30 min, flow rate 0.5 mL min^−1.b Isolated yield after HPLC.
T9^DAC	Ac-TTTT^(DAC)TTTTT-LysNH2	32.0	5.8	3513.6	3512.4
M10AT^DAC	Ac-GTAGA^(DAC)TCACT-Lys NH2	29.5	5.2	3892.8	3894.1
M10GC^DAC	Ac-GTAGC^(DAC)GCACT-Lys NH2	27.7	4.4	3893.8	3893.2
DAC-M10	DAC-GTAGATCACT-Lys NH2	29.0	5.4	3738.8	3738.3

---

Thermal stability and fluorescent properties of DAC-labeled acpcPNA

To investigate the binding properties with DNA, the stabilities of complementary and mismatched DAC-modified PNA·DNA hybrids were determined by thermal denaturation experiments. Thermal stability (T_m) and fluorescent properties of DAC-labeled acpcPNA are summarized in Table 2. The complementary DNA hybrid of T9^DAC sequence showed a T_m of 65.2 °C, which was considerably lower than the corresponding DNA hybrid of unmodified T9 acpcPNA (>80 °C). The T_m of the DNA hybrid of the M10AT^DAC sequence of 53.4 °C was also reduced compared to the unmodified PNA (57.0 °C). This suggests that the internal modification with the DAC dye destabilizes the acpcPNA·DNA duplex, probably due to steric effect. This behavior was also observed in our previous studies with the same apc/acpcPNA system with different labels.^34,35 The mix-base sequences PNA M10GC^DAC and DAC-M10 showed T_m values of 59.0 and 58.2 °C, respectively. The T_m value of the terminally DAC-modified DAC-M10 duplex was in fact larger than that of the unmodified duplex (ΔT_m = +1.2 °C). This can be explained by the additional stabilization from the end stacking of the DAC at the terminal base pair of the PNA·DNA duplex as observed in the case of pyrene label.³¹ All single mismatched PNA·DNA hybrids exhibited considerably lower T_m than complementary hybrids (ΔT_m = −27.0 to −41.6 °C for T9^DAC and −18.4 to −28.9 °C for M10GC^DAC/M10AT^DAC and less than −38.2 °C for DAC-M10). The ΔT_m values are comparable to those of unmodified PNA, which indicated that the base pairing specificity of the DAC-labeled acpcPNA remained high.

Table 2 Thermal stability (T_m) and fluorescence properties of DAC-labeled acpcPNA^a

Entry	PNA	DNA (3′ → 5′)	Tm (°C)	F/F0	ΦF	ΦF (ds)/ΦF (ss)^b	Notes^c
a All measurements were performed in 10 mM sodium phosphate buffer (pH 7.0), [PNA] = 1.0 μM, [DNA] = 1.2 μM, λex 315 nm, λem 450 nm.b Quantum yields were measured in phosphate buffer using quinine sulfate (0.540) as a reference.c ss = single stranded PNA, comp = complementary, sm = single mismatched.
1	T9^DAC	None	—	—	0.009	—	ss
2		AAAAAAAAA	65.2	35.5	0.242	26.7	comp
3		AAAACAAAA	33.7	18.6	0.139	15.3	smC
4		AAAAGAAAA	23.6	6.0	0.041	4.5	smG
5		AAAATAAAA	38.2	15.7	0.160	17.6	smT
6	M10AT^DAC	None	—	—	0.007	—	ss
7		AGTGATCTAC	53.4	13.2	0.083	11.9	comp
8		AGTGCTCTAC	29.0	2.7	0.012	1.7	smC
9		AGTGGTCTAC	24.5	1.7	0.014	2.0	smG
10		AGTGTTCTAC	35.0	6.5	0.034	4.8	smT
11	M10GC^DAC	None	—	—	0.011	—	ss
12		AGTGCGCTAC	59.0	10.2	0.117	11.1	comp
13		AGTGATCTAC	30.8	4.9	0.099	9.4	smA
14		AGTGGTCTAC	33.4	7.0	0.085	8.0	smG
15		AGTGTTCTAC	39.2	9.1	0.104	9.9	smT

---

Next, fluorescence experiments were performed with all PNA sequences and their DNA hybrids. The responsiveness of the PNA probe was reported in terms of fluorescence intensity ratio (F/F₀; where F = fluorescence intensity of PNA·DNA hybrid, F₀ = fluorescence intensity of single stranded PNA; λ_ex 315 nm; λ_em 450 nm). While free 3,6-diaminocarbazole showed a high Φ_F of 0.489, all internally-labeled single stranded PNA probes exhibited very low Φ_F values of 0.01 or less, regardless of the identity of the neighboring nucleobase (Table 2, entries 1, 6, 11). This suggests that the DAC label is quenched upon incorporation into the PNA, probably by interactions with the nucleobases similar to pyrene.³¹ The sequences M10AT^DAC and T9^DAC showed lower quantum yields than M10GC^DAC, which means that A/T are better quenchers than C/G bases for DAC label. Nevertheless, judging from the uniformly low quantum yields of all PNA probes, the difference in quenching efficiency by different nucleobases is much less than pyrene, which emphasizes the important feature of the DAC label.³¹

A remarkable increase in fluorescence signal was observed upon hybridization with complementary DNA, as shown by the fluorescence intensity ratio (F/F₀) between 10.2 and 35.5 (Table 2, entries 2, 7, 12). The quantum yield of the DAC-labeled PNA·DNA duplexes was also consistently increased by >10-fold to around 0.08–0.24, which is still considerably lower than the free DAC (3). This indicates that the interaction between the DAC and the neighboring nucleobases is reduced in PNA·DNA duplexes due to the formation of base pairs. This behavior is similar to pyrene label and a similar mechanism of fluorescence increase is proposed.³¹ It should be noted that free DAC also interact with single stranded DNA, resulting in fluorescence quenching, which is consistent with the proposed mechanism (see ESI Fig. S22†). CD spectra of M10GC^DAC and its complementary DNA hybrid (see ESI Fig. S23†) also clearly confirms the formation of PNA·DNA hybrids as shown by the change in CD signals at 210 and 260 nm. No CD signal was observed in the UV absorption region of DAC, suggesting that the DAC label was not well-oriented and thus did not appreciably interact with the PNA·DNA duplex.

Single mismatched PNA·DNA hybrids generally showed smaller fluorescence increase, but some variations in the sequence context were observed (Fig. 1). The M10AT^DAC showed a better mismatch discrimination than the T9^DAC and M10GC^DAC probes. Possible explanation is that certain mismatched PNA·DNA hybrids (Table 2, entries 3, 5, 10, 13–15) are still sufficiently stable at 20 °C – the temperature at which the fluorescent experiments were carried out. Completely unrelated DNA target provided a signal that was indistinguishable from the single stranded probes in all cases. For terminal labeled PNA probe DAC-M10, no fluorescence change could be observed for both complementary and mismatched DNA targets (Fig. 1b). Since the single stranded PNA probe exhibited low fluorescence similar to internally labeled PNA probes, it can be concluded that the quenching by neighboring nucleobases still operates. T_m measurement confirms the strong binding between the DAC-M10 probe and DNA in a sequence specific fashion (58 °C for complementary duplexes and <20 °C for mismatched duplexes). The absence of fluorescence increase upon duplex formation was therefore explained by the interaction between the DAC and the terminal GC base pair of the PNA·DNA duplex similar to the pyrene label.³¹

Fig. 1 Fluorescence spectra of DAC-labeled acpcPNA with complementary, single mismatch, non-complementary and single strand form of (a) M10AT^DAC (b) DAC-M10. Conditions: 10 mM sodium phosphate buffer, pH 7.0 at 25 °C, [PNA] = 1.0 μM and [DNA] = 1.2 μM, excitation wavelength = 315 nm. Complementary DNA = AGTGATCTAC, mismatch C DNA = AGTGCTCTAC, mismatch G DNA = AGTGGTCTAC and mismatch T DNA = AGTGTTCTAC.

These results clearly demonstrate that the fluorescence of internally DAC-labeled acpcPNA is low in single stranded PNA probes due to quenching by neighboring nucleobases, especially T and A bases. This interaction diminished – resulting in a large fluorescence increase – when the PNA·DNA duplexes are formed, provided that the DAC label does not locate at the end of the duplex which allows interactions with the terminal base pairs.

Improving single mismatch discrimination of the DAC-labeled acpcPNA probes

Although the T_m measurement indicated the high specificity as shown by a large decrease in the T_m values between complementary and single mismatch PNA·DNA duplexes (ΔT_m ranging from −18.4 to −31.5 °C), but the fluorescence spectra of certain mismatch duplexes still showed relatively high fluorescence signal. Based on a hypothesis that the high fluorescence was due to the high stability of some mismatched duplexes at room temperature, the temperature was increased to improve the single mismatch discrimination by fluorescence spectrophotometry. Single mismatch hybrids should dissociate more easily than complementary hybrids, and therefore the temperature increase should be able to enhance the mismatch discrimination ability of DAC-labeled acpcPNA. The experiment was carried out only with the complementary hybrids and mismatched hybrids that does not show satisfactory discrimination in Table 2. The fluorescence spectra were measured at different temperature from 20 °C to 90 °C (see ESI†). Upon heating, the duplex dissociated with concomitant decrease in the fluorescence intensity. The plots between the fluorescence signal at 450 nm and temperature appeared in the form of reverse S-curves (Fig. 2a).

Fig. 2 (a) Fluorescence melting curves of M10GC^DAC (b) fluorescence spectra of M10GC^DAC at RT (c) fluorescence spectra of M10GC^DAC at 50 °C. All experiments was performed in 10 mM phosphate-buffered pH 7.0, [PNA] = 1.0 μM, [DNA] = 1.2 μM, λ_ex 315 nm, PMT voltage = medium.

Since the maximal difference will be obtained at the temperature whereby the mismatched duplexes were dissociated as much as possible and the complementary duplexes are still stable, it is important to vary the temperature to find the optimal value case by case. The specific temperature that provided maximum differentiation in the case of T9^DAC, M10AT^DAC and M10GC^DAC was 55, 45 and 50 °C, respectively. The mismatched hybrids of T9^DAC and M10GC^DAC were more stable than those of M10AT^DAC and therefore required a higher temperature to complete the separation. At the aforementioned optimal temperature, the fluorescence ratio between complementary and single mismatch hybrids improved from 1.0–2.5 folds to 6.0–21.1 folds (Fig. 2b and c and Table 3). The results confirm that the specificity of DAC-labeled acpcPNA in distinguishing between complementary and single-mismatched DNA by DAC-labeled PNA probes can be improved by simply increasing the temperature without requiring more complex operations such as enzymatic digestion.

Table 3 Comparison of F/F₀ of DAC-labeled acpcPNA at room temperature and high temperature^a

Entry	PNA	DNA (3′ → 5′)	F/F0 (25 °C)	Fcomp/Fmm (25 °C)	F/F0 (high temp)^b	Fcomp/Fmm (high temp)^b
a All measurements were performed in 10 mM sodium phosphate buffer (pH 7.0), [PNA] = 1.0 μM, [DNA] = 1.2 μM, λex 315 nm, λem 450 nm.b High temperature: T9^DAC = 55 °C; M10AT^DAC = 45 °C; M10GC^DAC = 50 °C.
1	T9^DAC	AAAAAAAAA	35.4	—	35.7	—
2		AAAACAAAA	18.6	1.9	2.5	14.2
3		AAAAGAAAA	6.0	5.9	—	—
4		AAAATAAAA	15.7	2.3	1.7	21.1
5		TTTTTTTTT	1.1	32.4	—	—
6	M10AT^DAC	AGTGATCTAC	13.2	—	12.9	—
7		AGTGCTCTAC	2.6	5.2	1.3	10.0
8		AGTGGTCTAC	1.6	8.2	—	—
9		AGTGTTCTAC	5.7	2.3	1.2	10.6
10		GGGGGGGGGG	1.2	10.9	—	—
11	M10GC^DAC	AGTGCGCTAC	10.1	—	10.2	—
12		AGTGATCTAC	5.3	1.9	—	—
13		AGTGGTCTAC	7.1	1.4	1.0	10.3
14		AGTGTTCTAC	9.0	1.1	1.7	6.0
15		GGGGGGGGGG	0.9	11.1	—	—

---

Next, to support the proposed mechanism of fluorescence change of the DAC-labeled PNA probe, molecular dynamics simulation and energy calculation were performed on both single stranded and duplex PNA probe.

Molecular modeling

MD structures. For both simulated systems (ssPNA and PNA·DNA duplex), the structure of DAC started by pointing to bulk solvent was rolled and close to PNA strand after 0.5 ns and then fluctuated in such position along 10 ns of simulation time, as displayed in Fig. 3. The root mean squared deviations (RMSDs) of the MD structures with respect to their starting structures were 4.42 ± 0.67 Å and 4.02 ± 0.40 Å for ssPNA and PNA·DNA systems, respectively, averaged over the last 2000 structures, see Fig. S25† for the RMSD profiles. Clearly, single stranded PNA provides larger fluctuation because of a less restriction of its structure. The average structure revealed that the DAC is closer to the nucleobases in the single stranded PNA than that in the PNA·DNA duplex (Fig. 4), which should lead to a larger interaction between DAC and PNA for the single stranded PNA system.

Fig. 3 MD snapshots of a single stranded PNA and the corresponding PNA·DNA duplex.

Fig. 4 Average MD structures of single stranded PNA and PNA·DNA duplex.

Energy calculation. In order to investigate the DAC–PNA interactions between DAC and nucleobases in single stranded PNA as well as PNA·DNA duplexes in more details, we have calculated the binding energy between the DAC and three closest PNA nucleobases, whereas other fragments were omitted. The calculations were performed using Gaussian 09 program with the density functional theory B3LYP/6-31G* level based on the average MD structures for ssPNA and PNA·DNA systems. As shown in Table 4, the binding energy of single stranded PNA system is significantly higher than that of PNA·DNA duplex. This confirms that there are stronger interactions between DAC and nucleobases in single stranded PNA than in PNA·DNA duplex, and further supports the mechanism of fluorescence change proposed above.

Table 4 The binding energy (ΔE_binding) between DAC and three closest PNA nucleobases^a

System	ΔEbinding (kcal mol^−1)
a The calculation is expressed as ΔEbinding = EDAC–PNA − EDAC − EPNA, where EDAC–PNA, EDAC and EPNA are the total energy of DAC and three PNA nucleotides, DAC moiety and three PNA nucleotides, respectively.
ssPNA	−11.40
PNA·DNA	−3.86

---

Experimental section

General procedures

Carbazole (reagent grade for synthesis) was purchased from Merck. All reagent grade chemicals and solvents were purchased from standard suppliers and were used without further purification. ¹H and ¹³C NMR spectra were recorded on Bruker Avance 400 NMR spectrometer operating at 400 MHz for ¹H and 100 MHz for ¹³C. RP-HPLC experiments were performed on a Waters Delta 600 HPLC system. Oligonucleotides were obtained from Pacific Science (Thailand) or BioDesign (Thailand).

N⁹-Propargyl-3,6-bis(tert-butoxycarbonylamino)carbazole (5)

A reaction mixture containing 3,6-bis(tert-butoxycarbonylamino)carbazole (4) (0.80 g, 2.00 mmol) (see ESI†), propargyl bromide (0.28 mL, 3.00 mmol), KI (0.33 g, 2.00 mmol) and K₂CO₃ (2.77 g, 20.0 mmol) in dry DMF (5 mL) was stirred under N₂ atmosphere at room temperature overnight. The reaction mixture was diluted with water and extracted with CH₂Cl₂. The combined organic extracts were washed with water and brine, dried over anhydrous Na₂SO₄ and evaporated to dryness. The crude product was purified by column chromatography on silica gel to obtain the pure compound as a yellow solid. ¹H NMR (400 MHz, DMSO-d₆): δ 1.49 (s, 18H), 3.20 (t, J = 2.2 Hz, 1H), 5.18 (s, 2H), 7.41 (d, J = 8.7 Hz, 2H), 7.49 (d, J = 8.7 Hz, 2H), 8.21 (s, 2H), 9.30 (s, 2H) ¹³C NMR (100 MHz, DMSO-d₆): δ 153.2, 136.2, 131.9, 122.3, 118.2, 109.5, 79.2, 78.7, 74.3, 31.9, 28.2, 27.9; IR: 3397, 3254, 2974, 2112 (C≡C), 1720, 1702 (C=O), 1524, 1490, 1048, 1023, 864, 792, 578 cm⁻¹; HRMS (ESI+): m/z calcd for C₂₅H₂₉N₃O₄Na⁺: 458.2056 [M + Na]⁺ found: 458.2092, m/z calcd for C₂₅H₂₉N₃O₄K⁺: [M + K]⁺ 474.1795 found: 474.1826.

Synthesis, purification and characterization of DAC-modified acpcPNA

The apc-modified acpcPNA was synthesized via solid phase peptide synthesis using the four Fmoc-protected pyrrolidine monomers (Fmoc-A^Bz-OPfp, Fmoc-T-OPfp, Fmoc-C^Bz-OPfp, and Fmoc-G^Ibu-OPfp) and spacers [Fmoc-(1S,2S)-ACPC-OPfp, Fmoc-(3R,4S)-APC(Tfa)-OPfp] at 1.5 μmol scale on Tentagel S-RAM resin according to the previously published protocol.^27,28,35 Lysine was included at the C-terminus in the form of Fmoc-Lys(Boc)-OPfp to improve the water solubility of the PNA. After completion of the PNA synthesis, the N-terminal Fmoc group was removed and the free amino group was capped by acetylation. The apc-modified acpcPNA was split to 0.5 μmol portions for further labeling experiments. The nucleobase protecting groups (Bz, Ibu) and APC spacer protection (Tfa) were removed by heating with 1 : 1 aqueous ammonia/dioxane at 60 °C overnight. The apc-modified acpcPNA (0.5 μmol) was alkylated at the pyrrolidine ring nitrogen of the APC spacer with 4-azidobutanal (15 μmol, 30 equiv.) in the presence of sodium cyanoborohydride (30 μmol, 60 equiv.) and acetic acid (30 μmol, 60 equiv.) in methanol (100 μL) at room temperature overnight.³⁴ The azide-modified acpcPNA was further reacted with N⁹-propargyl-3,6-bis(Boc-amino)carbazole (5) (7.5 μmol, 15 equiv.) while still attached on the solid support in the presence of tris[(1-benzyl-1H-1,2,3-triazol-4yl)methyl]amine (TBTA, 30 μmol, 60 equiv.), tetrakis(acetonitrile)copper(I) hexafluorophosphate (15 μmol, 30 equiv.) and (+)-sodium-L-ascorbate (60 μmol, 120 equiv.) in 3 : 1 (v/v) DMSO : ^tBuOH at room temperature overnight.³⁴ After the reaction was completed as monitored by MALDI-TOF analyses (12 h), the diaminocarbazole-modified acpcPNA was cleaved from the resin with 50% trifluoroacetic acid (TFA) in dichloromethane and 10% triisopropylsilane (TIS) (500 μL × 30 min × 3). The combined cleavage solution was dried under a stream of nitrogen gas and the crude PNA was precipitated by addition of diethyl ether (Scheme 2). The crude PNA was purified by reverse-phase HPLC with using mobile phase A : B (A: 0.1% TFA in H₂O; B: 0.1% TFA in MeOH) 9 : 1 for 5 min followed by a linear gradient to A : B 1 : 9 over 55 min. The purified PNA was characterized by MALDI-TOF mass spectrometry (Microflex, Bruker Daltonics) by using α-cyano-4-hydroxy-cinnamic acid (CCA) as a matrix.

Scheme 2 Synthesis of DAC-modified acpcPNA.

Spectroscopic experiments

All samples were prepared at PNA concentration = 1.0 μM and DNA concentration = 1.2 μM in 10 mM phosphate buffer pH 7.0. UV-visible spectra and thermal stabilities (T_m) were measured on a CARY 100 Bio UV-vis spectrophotometer at 260 nm from 20–90 °C at a rate of 1 °C min⁻¹. The melting temperatures were determined from first derivative plots. Fluorescence spectra were measured on Cary Eclipes Fluorescence Spectrophotometer at excitation wavelength of 315 nm. Both excitation and emission slits were set to 5 nm and PMT voltage = medium.

Fluorescence quantum yield (Φ_F) measurement

The fluorescence quantum yield (Φ_F) of 3,6-diaminocarbazole (3), single stranded-PNA and PNA·DNA duplexes [T9^DAC, M10AT^DAC and M10GC^DAC] were calculated using quinine sulfate (Φ_F = 0.540) as a ref. 36 Absorption and fluorescence spectra were measured on CARY 100 Bio UV-vis spectrophotometer (Varian) and CARY Eclipse Fluorescence spectrophotometer (Varian/Agilent Technologies), respectively. Seven concentrations (1.0, 2.0, 2.5, 5.0, 7.5, 10.0, 12.0 μM) of quinine sulfate samples in phosphate buffer were prepared from a stock solution (1.0 mM quinine sulfate in 0.1 M H₂SO₄). The absorbance at 315 nm was less than 0.1 for all concentrations. The magnitudes of integrated fluorescence intensity were plotted against the absorbance of the solution absorbance. The quantum yield of sample was then calculated according to eqn (1).³¹

Φ_F(sample) = Φ_F(standard)[m(sample)/m(standard)][η(sample)²/η(standard)²] (1)

where m is the slope from the plot of integrated fluorescence intensity vs. absorbance and η is the refractive index of the solvent.

Molecular dynamics (MD) simulation. MD simulations were performed on single stranded acpcPNA with a sequence of GTAGA̲T̲CACT whereby the DAC was linked to the apc spacer at the underlined position, and the corresponding acpcPNA·DNA duplex. Because the three-dimensional structures of ssPNA and PNA·DNA duplex were not available, the starting structure of PNA·DNA duplex was built up based on the corresponding sequence of DNA·DNA duplex (dGTAGATCACT·dAGTGATCTAC) which was generated using the NAB module of the AMBER 12 package.³⁷ The first strand of the duplex was replaced with PNA strand in an antiparallel direction yielding the PNA·DNA duplex.³⁸ The structure of diaminocarbazole (DAC) linked to backbone of PNA was first optimized with B3LYP/6-31G* calculation using Gaussian 09 (ref. 39) before adding to the PNA strand. The initial structure of single stranded acpcPNA was obtained in a similar manner. The force field parameters of the PNA and DAC were generated using the standard procedure.³⁸ To setup the simulations, the starting structure of each system was immersed in a rectangular water box with TIP3P water molecules⁴⁰ extended by 10 Å in each direction from the solute and then neutralized by adding Na⁺ counterions (for PNA·DNA system). The MD simulations were carried out using AMBER 12 suite of program with parm10 force field complemented with the prepared parameters of PNA and DAC, SHAKE algorithm⁴¹ employed for all hydrogen atoms, a 9 Å cutoff applied to the non-bonding Lennard-Jones interactions, the particle mesh Ewald method used to account for long-range interactions,⁴² time step of 1 fs and NPT ensemble (P = 1 atm and T = 300 K). The simulation protocol consists of heating, equilibration and production steps, and was described in details elsewhere.³⁸ The MD trajectories were generated for 10 ns and the coordinates were stored every 1 ps, totally 10 000 structures.

Conclusion

In conclusion, we successfully synthesized novel internally and terminally 3,6-diamino-carbazole-labeled acpcPNA and evaluated their binding and fluorescence properties. Thermal denaturation experiments revealed the high stability (T_m 53.4–65.2 °C) and high specificity (ΔT_m > 20 °C) of PNA·DNA hybrid similar to unmodified acpcPNA. Single stranded DAC-labeled acpcPNA showed low fluorescence signals due to quenching of the DAC label with neighboring nucleobases. After hybridization with DNA, the internally DAC-labeled acpcPNA showed moderate to large fluorescence increase (10.1–35.5 folds) compared to the single stranded PNA. The specificity was further improved to allow discrimination between complementary and single mismatched DNA targets by increasing the temperature. Terminally DAC-labeled acpcPNA showed no significant fluorescence change after hybridization with DNA. A model was proposed to explain the mechanism of fluorescence change, which was further supported by MD simulations and energy calculation. These results demonstrate the potential of DAC as a fluorescence label to be used in combination with oligonucleotide probes for detection of specific DNA sequences.

Acknowledgements

We acknowledge Naresuan University (NU grant No. P2558C371 to CS) and The Thailand Research Fund (DPG5780002, to TV and KS) for financial supports and National e-Science Infrastructure Consortium (URL: http://www.e-science.in.th) for computing resources. Partial technical supports by Ms Boonsong Ditmangklo (Chulalongkorn University) are also gratefully acknowledged.

References

1. M. K. Johansson, H. Fidder, D. Dick and R. M. Cook, J. Am. Chem. Soc., 2002, 124, 6950–6956.
2. L. Giestas, G. N. M. Ferreira, P. V. Baptista and J. C. Lima, Sens. Actuators, B, 2008, 134, 146–157.
3. E. G. Matveeva, Z. Gryczynski, D. R. Stewart and I. Gryczynski, J. Lumin., 2010, 130, 698–702.
4. P. Rashatasakhon, K. Vongnam, W. Siripornnoppakhun, T. Vilaivan and M. Sukwattanasinitt, Talanta, 2012, 88, 593–598.
5. N. Maneelun and T. Vilaivan, Tetrahedron, 2013, 69, 10805–10810.
6. B. S. Gaylord, A. J. Heeger and G. C. Bazan, J. Am. Chem. Soc., 2003, 125, 896–900.
7. Z.-K. Xu, B.-K. Zhu and Y.-Y. Xu, Chem. Mater., 1998, 10, 1350–1354.
8. D. Li-yun, J. De-sheng, H. Jun, D. Lei, L. Cheng, W. Jun-tao and L. Bin, Trans. Nonferrous Met. Soc. China, 2006, 16, s191–s194.
9. D. W. Kim, H. Moon, S. Y. Park and S. I. Hong, React. Funct. Polym., 1999, 42, 73–86.
10. I. F. Perepichka, D. D. Mysyk and N. I. Sokolov, Synth. Met., 1999, 101, 9–10.
11. X. Pan, S. Liu, H. S. O. Chan and S. C. Ng, Macromolecules, 2005, 38, 7629–7635.
12. J. Sun, H.-J. Jiang, J.-L. Zhang, Y. Tao and R.-F. Chen, New J. Chem., 2013, 37, 977–985.
13. J. P. Chen and A. Natansohn, Macromolecules, 1999, 32, 3171–3177.
14. A. Oral, S. Koyuncu and I. Kaya, Synth. Met., 2009, 159, 1620–1627.
15. S. Koyuncu, I. Kaya, F. B. Koyuncu and E. Ozdemir, Synth. Met., 2009, 159, 1034–1042.
16. F. B. Koyuncu, S. Koyuncu and E. Ozdemir, Electrochim. Acta, 2010, 55, 4935–4941.
17. C.-H. Tao, N. Zhu and V. W.-W. Yam, J. Photochem. Photobiol., A, 2009, 207, 94–101.
18. P. Thongkasee, A. Thangthong, N. Janthasing, T. Sudyoadsuk, S. Namuangruk, T. Keawin, S. Jungsuttiwong and V. Promarak, ACS Appl. Mater. Interfaces, 2014, 6, 8212–8222.
19. G. Sathiyan, E. K. T. Sivakumar, R. Ganesamoorthy, R. Thangamuthu and P. Sakthivel, Tetrahedron Lett., 2016, 57, 243–252.
20. B. Dumat, G. Bordeau, E. Faurel-Paul, F. Mahuteau-Betzer, N. Saettel, M. Bombled, G. Metgé, F. Charra, C. Fiorini-Debuisschert and M.-P. Teulade-Fichou, Biochimie, 2011, 93, 1209–1218.
21. F. A. Tanious, D. Y. Ding, D. A. Patrick, R. R. Tidwell and W. D. Wilson, Biochemistry, 1997, 36, 15315–15325.
22. F. A. Tanious, D. Y. Ding, D. A. Patrick, C. Bailly, R. R. Tidwell and W. D. Wilson, Biochemistry, 2000, 39, 12091–12101.
23. F. A. Tanious, W. D. Wilson, D. A. Patrick, R. R. Tidwell, P. Colson, C. Houssier, C. Tardy and C. Bailly, Eur. J. Biochem., 2001, 268, 3455–3464.
24. T. Jia, J. Xiang, J. Wang, P. Guoa and J. Yu, Org. Biomol. Chem., 2013, 11, 5512–5520.
25. P. E. Nielsen, M. Egholm, R. H. Berg and O. Buchardt, Science, 1991, 254, 1497–1500.
26. M. Egholm, O. Buchardt, L. Christensen, C. Behrens, S. M. Freier, D. A. Driver, D. A. R. H. Berg, S. K. Kim, B. Norden and P. E. Nielsen, Nature, 1993, 365, 566–568.
27. T. Vilaivan and C. Srisuwannaket, Org. Lett., 2006, 8, 1897–1900.
28. C. Vilaivan, C. Srisuwannaket, C. Ananthanawat, C. Suparpprom, J. Kawakami, Y. Yamaguchi, Y. Tanaka and T. Vilaivan, Artif. DNA, 2011, 2, 50–59.
29. W. Mansawat, C. Vilaivan, A. Balazs, D. J. Aitken and T. Vilaivan, Org. Lett., 2012, 14, 1440–1443.
30. T. Vilaivan, Acc. Chem. Res., 2015, 48, 1645–1656.
31. C. Boonlua, B. Ditmangklo, N. Reenabthue, C. Suparpprom, N. Poomsuk, K. Siriwong and T. Vilaivan, RSC Adv., 2014, 4, 8817–8827.
32. S. Sezi, R. Varghese, T. Vilaivan and H.-A. Wagenknecht, ChemistryOpen, 2012, 1, 173–176.
33. N. Yotapan, C. Charoenpakdee, P. Wathanathavorn, B. Ditmangklo, H.-A. Wagenknecht and T. Vilaivan, Beilstein J. Org. Chem., 2014, 10, 2166–2174.
34. B. Ditmangklo, C. Boonlua, C. Suparpprom and T. Vilaivan, Bioconjugate Chem., 2013, 24, 614–625.
35. N. Reenabthue, C. Boonlua, C. Vilaivan, T. Vilaivan and C. Suparpprom, Bioorg. Med. Chem. Lett., 2011, 21, 6465–6469.
36. W. H. Melhuish, J. Phys. Chem., 1961, 65, 229–235.
37. AMBER 12, University of California, San Francisco, 2012.
38. N. Poomsuk and K. Siriwong, Chem. Phys. Lett., 2012, 588, 237–241.
39. GAUSSIAN 09 (Revision B.01), Gaussian, Inc., Wallingford, Connecticut, 2009.
40. W. L. Jorgensen, J. Am. Chem. Soc., 1981, 103, 335–340.
41. J. P. Ryckaert, G. Cicotti and J. C. Berendsen, J. Comput. Phys., 1997, 23, 327–341.
42. T. Darden, D. York and L. Pedersen, J. Chem. Phys., 1993, 98, 10089–10092.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra15861g

---